From 1c6924f18bd8fdad63c4f191c00605092c17fa6e Mon Sep 17 00:00:00 2001 From: Quincey Koziol Date: Sat, 12 Nov 2016 01:05:47 -0800 Subject: Refactor H5O code to clean up message allocation, align cache deserialize code with revise_chunks changes, and remove unused "message locking" code. --- src/H5Oalloc.c | 859 +++++++++++++++++++++++++++++++++---------------------- src/H5Ocache.c | 289 ++++++++++--------- src/H5Ochunk.c | 58 +++- src/H5Omessage.c | 173 ----------- src/H5Opkg.h | 20 +- src/H5Oprivate.h | 3 - test/ohdr.c | 142 --------- 7 files changed, 727 insertions(+), 817 deletions(-) diff --git a/src/H5Oalloc.c b/src/H5Oalloc.c index 33d2e9d..d22fd92 100644 --- a/src/H5Oalloc.c +++ b/src/H5Oalloc.c @@ -67,8 +67,11 @@ static herr_t H5O_alloc_null(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t null_idx const H5O_msg_class_t *new_type, void *new_native, size_t new_size); static htri_t H5O_alloc_extend_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned chunkno, size_t size, size_t *msg_idx); +static herr_t H5O__alloc_find_best_nonnull(const H5F_t *f, const H5O_t *oh, size_t *size, + H5O_msg_alloc_info_t *found_msg); static herr_t H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new_idx); +static herr_t H5O__alloc_find_best_null(const H5O_t *oh, size_t size, size_t *mesg_idx); static htri_t H5O_move_cont(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned cont_u); static htri_t H5O_move_msgs_forward(H5F_t *f, hid_t dxpl_id, H5O_t *oh); static htri_t H5O_merge_null(H5F_t *f, hid_t dxpl_id, H5O_t *oh); @@ -707,17 +710,10 @@ done: /*------------------------------------------------------------------------- - * Function: H5O_alloc_new_chunk + * Function: H5O__alloc_find_best_nonnull * - * Purpose: Allocates a new chunk for the object header, including - * file space. - * - * One of the other chunks will get an object continuation - * message. If there isn't room in any other chunk for the - * object continuation message, then some message from - * another chunk is moved into this chunk to make room. - * - * SIZE need not be aligned. + * Purpose: Find the best fit non-null message for a given size of message + * to allocate. * * Note: The algorithm for finding a message to replace with a * continuation message is still fairly limited. It's possible @@ -735,50 +731,34 @@ done: * * Failure: Negative * - * Programmer: Robb Matzke - * matzke@llnl.gov - * Aug 7 1997 + * Programmer: Quincey Koziol + * koziol@lbl.gov + * Oct 21 2016 * *------------------------------------------------------------------------- */ static herr_t -H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new_idx) +H5O__alloc_find_best_nonnull(const H5F_t *f, const H5O_t *oh, size_t *size, + H5O_msg_alloc_info_t *found_msg) { - /* Struct for storing information about "best" messages to allocate from */ - typedef struct { - int msgno; /* Index in message array */ - size_t gap_size; /* Size of any "gap" in the chunk immediately after message */ - size_t null_size; /* Size of any null message in the chunk immediately after message */ - size_t total_size; /* Total size of "available" space around message */ - unsigned null_msgno; /* Message index of null message immediately after message */ - } alloc_info; - H5O_mesg_t *curr_msg; /* Pointer to current message to operate on */ - H5O_chunk_proxy_t *chk_proxy; /* Chunk that message is in */ - size_t cont_size; /*continuation message size */ - size_t multi_size = 0; /* Size of all the messages in the last chunk */ - int found_null = (-1); /* Best fit null message */ - alloc_info found_attr = {-1, 0, 0, 0, 0}; /* Best fit attribute message */ - alloc_info found_other = {-1, 0, 0, 0, 0}; /* Best fit other message */ - size_t idx; /* Message number */ - uint8_t *p = NULL; /*ptr into new chunk */ - H5O_cont_t *cont = NULL; /*native continuation message */ - unsigned chunkno; /* Chunk allocated */ - haddr_t new_chunk_addr; + size_t cont_size; /* Continuation message size */ + size_t multi_size; /* Size of all the messages in the last chunk */ unsigned u; /* Local index variable */ herr_t ret_value = SUCCEED; /* Return value */ - FUNC_ENTER_NOAPI_NOINIT + FUNC_ENTER_STATIC - /* check args */ + /* Check args */ + HDassert(f); HDassert(oh); - HDassert(size > 0); - size = H5O_ALIGN_OH(oh, size); + HDassert(size); + HDassert(*size > 0); + HDassert(found_msg); /* - * Find the smallest null message that will hold an object - * continuation message. Failing that, find the smallest message - * that could be moved to make room for the continuation message. + * Find the smallest message that could be moved to make room for the + * continuation message. * * Don't ever move continuation message from one chunk to another. * @@ -787,23 +767,10 @@ H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new * */ cont_size = H5O_ALIGN_OH(oh, (size_t)(H5F_SIZEOF_ADDR(f) + H5F_SIZEOF_SIZE(f))); + multi_size = 0; for(u = 0, curr_msg = &oh->mesg[0]; u < oh->nmesgs; u++, curr_msg++) { - if(curr_msg->type->id == H5O_NULL_ID) { - if(cont_size == curr_msg->raw_size) { - found_null = (int)u; - break; - } /* end if */ - else if(curr_msg->raw_size > cont_size && - (found_null < 0 || curr_msg->raw_size < oh->mesg[found_null].raw_size)) - found_null = (int)u; - } /* end if */ - else if(curr_msg->type->id == H5O_CONT_ID) { - /* Don't consider continuation messages (for now) */ - } /* end if */ - else if(curr_msg->locked) { - /* Don't consider locked messages */ - } /* end if */ - else { + /* Don't consider continuation messages (for now) */ + if(H5O_CONT_ID != curr_msg->type->id) { unsigned msg_chunkno = curr_msg->chunkno; /* Chunk that the message is in */ uint8_t *end_chunk_data = (oh->chunk[msg_chunkno].image + oh->chunk[msg_chunkno].size) - (H5O_SIZEOF_CHKSUM_OH(oh) + oh->chunk[msg_chunkno].gap); /* End of message data in chunk */ uint8_t *end_msg = curr_msg->raw + curr_msg->raw_size; /* End of current message */ @@ -837,34 +804,51 @@ H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new /* Check if message is large enough to hold continuation info */ if(total_size >= cont_size) { - if(curr_msg->type->id == H5O_ATTR_ID) { - if(found_attr.msgno < 0 || total_size < found_attr.total_size) { - found_attr.msgno = (int)u; - found_attr.gap_size = gap_size; - found_attr.null_size = null_size; - found_attr.total_size = total_size; - found_attr.null_msgno = null_msgno; - } /* end if */ - } /* end if */ + hbool_t better = FALSE; /* Whether the current message is better than a previous one */ + + /* Check for first message that can be moved */ + if(found_msg->msgno < 0) + better = TRUE; else { - if(found_other.msgno < 0 || total_size < found_other.total_size) { - found_other.msgno = (int)u; - found_other.gap_size = gap_size; - found_other.null_size = null_size; - found_other.total_size = total_size; - found_other.null_msgno = null_msgno; - } /* end if */ + /* Prioritize moving non-attributes above attributes */ + /* (Even attributes with an otherwise better fit */ + if(found_msg->id == H5O_ATTR_ID && curr_msg->type->id != H5O_ATTR_ID) + better = TRUE; + /* Either two attributes, or two non-attributes */ + else { + /* Take a smaller one */ + if(total_size < found_msg->total_size) + better = TRUE; + /* If they are the same size, choose the earliest one + * in the chunk array */ + /* (Could also bias toward message earlier / later + * chunk in, but shouldn't be a big deal - QAK, 2016/10/21) + */ + else if(total_size == found_msg->total_size) { + if(msg_chunkno < found_msg->chunkno) + better = TRUE; + } /* end else-if */ + } /* end else */ } /* end else */ + + /* If we found a better message, keep its info */ + if(better) { + found_msg->msgno = (int)u; + found_msg->id = curr_msg->type->id; + found_msg->chunkno = msg_chunkno; + found_msg->gap_size = gap_size; + found_msg->null_size = null_size; + found_msg->total_size = total_size; + found_msg->null_msgno = null_msgno; + } /* end if */ } /* end if */ - else if(found_null < 0 && found_attr.msgno < 0 && found_other.msgno < 0 && msg_chunkno == oh->nchunks - 1) + else if(found_msg->msgno < 0 && msg_chunkno == oh->nchunks - 1) /* Keep track of the total size of smaller messages in the last * chunk, in case we need to move more than 1 message. */ multi_size += curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh); - } /* end else */ + } /* end if */ } /* end for */ - if(found_null >= 0 || found_attr.msgno >= 0 || found_other.msgno >= 0) - multi_size = 0; /* * If we must move some other message to make room for the null @@ -876,17 +860,56 @@ H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new * If all else fails, move every message in the last chunk. * */ - if(multi_size == 0) { - if(found_null < 0) { - if(found_other.msgno < 0) - found_other = found_attr; - - HDassert(found_other.msgno >= 0); - size += (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + oh->mesg[found_other.msgno].raw_size; - } /* end if */ - } /* end if */ + if(found_msg->msgno < 0) + *size += multi_size; else - size += multi_size; + *size += (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + oh->mesg[found_msg->msgno].raw_size; + +done: + FUNC_LEAVE_NOAPI(ret_value) +} /* H5O__alloc_find_best_nonnull() */ + + +/*------------------------------------------------------------------------- + * Function: H5O__alloc_chunk + * + * Purpose: Allocates and initializes a new chunk for the object header, + * including file space. + * + * Return: Success: SUCCEED, with chunk number for the + * new chunk and a pointer to the location in its + * image where the first message should be placed. + * + * Failure: Negative + * + * Programmer: Quincey Koziol + * koziol@lbl.gov + * Oct 21 2016 + * + *------------------------------------------------------------------------- + */ +herr_t +H5O__alloc_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, + size_t found_null, const H5O_msg_alloc_info_t *found_msg, size_t *new_idx) +{ + H5O_mesg_t *curr_msg; /* Pointer to current message to operate on */ + H5O_chunk_proxy_t *chk_proxy; /* Chunk that message is in */ + size_t cont_size; /*continuation message size */ + size_t idx; /* Message number */ + uint8_t *p = NULL; /* Pointer into new chunk image */ + H5O_cont_t *cont = NULL; /*native continuation message */ + unsigned chunkno; /* Chunk allocated */ + haddr_t new_chunk_addr; /* Address of new chunk in file */ + unsigned u; /* Local index variable */ + herr_t ret_value = SUCCEED; /* Return value */ + + FUNC_ENTER_PACKAGE + + /* check args */ + HDassert(f); + HDassert(oh); + HDassert(found_msg); + HDassert(new_idx); /* * The total chunk size must include the requested space plus enough @@ -944,133 +967,136 @@ H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new HGOTO_ERROR(H5E_RESOURCE, H5E_NOSPACE, FAIL, "can't allocate more space for messages") /* Check if we need to move multiple messages, in order to make room for the new message */ - if(multi_size > 0) { - /* Move all non-null messages in the last chunk to the new chunk. This - * should be extremely rare so we don't care too much about minimizing - * the space used. - */ - H5O_mesg_t *null_msg; /* Pointer to new null message */ - - /* Protect last chunk */ - if(NULL == (chk_proxy = H5O_chunk_protect(f, dxpl_id, oh, chunkno - 1))) - HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to load object header chunk") - - /* Copy each message to the new location */ - for(u = 0, curr_msg = &oh->mesg[0]; u < oh->nmesgs; u++, curr_msg++) - if(curr_msg->chunkno == chunkno - 1) { - if(curr_msg->type->id == H5O_NULL_ID) { - /* Delete the null message */ - if(u < oh->nmesgs - 1) - HDmemmove(curr_msg, curr_msg + 1, ((oh->nmesgs - 1) - u) * sizeof(H5O_mesg_t)); - oh->nmesgs--; + cont_size = H5O_ALIGN_OH(oh, (size_t)(H5F_SIZEOF_ADDR(f) + H5F_SIZEOF_SIZE(f))); + if(found_null >= oh->nmesgs) { + if(found_msg->msgno < 0) { + /* Move all non-null messages in the last chunk to the new chunk. This + * should be extremely rare so we don't care too much about minimizing + * the space used. + */ + H5O_mesg_t *null_msg; /* Pointer to new null message */ + + /* Protect last chunk */ + if(NULL == (chk_proxy = H5O_chunk_protect(f, dxpl_id, oh, chunkno - 1))) + HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to load object header chunk") + + /* Copy each message to the new location */ + for(u = 0, curr_msg = &oh->mesg[0]; u < oh->nmesgs; u++, curr_msg++) + if(curr_msg->chunkno == chunkno - 1) { + if(curr_msg->type->id == H5O_NULL_ID) { + /* Delete the null message */ + if(u < oh->nmesgs - 1) + HDmemmove(curr_msg, curr_msg + 1, ((oh->nmesgs - 1) - u) * sizeof(H5O_mesg_t)); + oh->nmesgs--; + } /* end if */ + else { + HDassert(curr_msg->type->id != H5O_CONT_ID); + + /* Copy the raw data */ + HDmemcpy(p, curr_msg->raw - (size_t)H5O_SIZEOF_MSGHDR_OH(oh), + curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh)); + + /* Update the message info */ + curr_msg->chunkno = chunkno; + curr_msg->raw = p + H5O_SIZEOF_MSGHDR_OH(oh); + + /* Account for copied message in new chunk */ + p += (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + curr_msg->raw_size; + size -= (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + curr_msg->raw_size; + } /* end else */ } /* end if */ - else { - HDassert(curr_msg->type->id != H5O_CONT_ID); - /* Copy the raw data */ - HDmemcpy(p, curr_msg->raw - (size_t)H5O_SIZEOF_MSGHDR_OH(oh), - curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh)); + /* Create a null message spanning the entire last chunk */ + found_null = oh->nmesgs++; + null_msg = &(oh->mesg[found_null]); + null_msg->type = H5O_MSG_NULL; + null_msg->dirty = TRUE; + null_msg->native = NULL; + null_msg->raw = oh->chunk[chunkno - 1].image + + ((chunkno == 1) ? H5O_SIZEOF_HDR(oh) : H5O_SIZEOF_CHKHDR_OH(oh)) + - H5O_SIZEOF_CHKSUM_OH(oh) + H5O_SIZEOF_MSGHDR_OH(oh); + null_msg->raw_size = oh->chunk[chunkno - 1].size + - ((chunkno == 1) ? (size_t)H5O_SIZEOF_HDR(oh) : (size_t)H5O_SIZEOF_CHKHDR_OH(oh)) + - (size_t)H5O_SIZEOF_MSGHDR_OH(oh); + null_msg->chunkno = chunkno - 1; - /* Update the message info */ - curr_msg->chunkno = chunkno; - curr_msg->raw = p + H5O_SIZEOF_MSGHDR_OH(oh); + HDassert(null_msg->raw_size >= cont_size); - /* Account for copied message in new chunk */ - p += (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + curr_msg->raw_size; - size -= (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + curr_msg->raw_size; - } /* end else */ - } /* end if */ + /* Remove any gap in the chunk */ + oh->chunk[chunkno - 1].gap = 0; - /* Create a null message spanning the entire last chunk */ - found_null = (int)oh->nmesgs++; - null_msg = &(oh->mesg[found_null]); - null_msg->type = H5O_MSG_NULL; - null_msg->dirty = TRUE; - null_msg->native = NULL; - null_msg->raw = oh->chunk[chunkno - 1].image - + ((chunkno == 1) ? H5O_SIZEOF_HDR(oh) : H5O_SIZEOF_CHKHDR_OH(oh)) - - H5O_SIZEOF_CHKSUM_OH(oh) + H5O_SIZEOF_MSGHDR_OH(oh); - null_msg->raw_size = oh->chunk[chunkno - 1].size - - ((chunkno == 1) ? (size_t)H5O_SIZEOF_HDR(oh) : (size_t)H5O_SIZEOF_CHKHDR_OH(oh)) - - (size_t)H5O_SIZEOF_MSGHDR_OH(oh); - null_msg->chunkno = chunkno - 1; - - HDassert(null_msg->raw_size >= cont_size); - - /* Remove any gap in the chunk */ - oh->chunk[chunkno - 1].gap = 0; - - /* Release chunk, marking it dirty */ - if(H5O_chunk_unprotect(f, dxpl_id, chk_proxy, TRUE) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") - } else if(found_null < 0) { - /* Move message (that will be replaced with continuation message) - * to new chunk, if necessary. - */ - H5O_mesg_t *null_msg; /* Pointer to new null message */ - - /* Protect chunk */ - if(NULL == (chk_proxy = H5O_chunk_protect(f, dxpl_id, oh, oh->mesg[found_other.msgno].chunkno))) - HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to load object header chunk") - - /* Create null message for space that message to copy currently occupies */ - found_null = (int)oh->nmesgs++; - null_msg = &(oh->mesg[found_null]); - null_msg->type = H5O_MSG_NULL; - null_msg->native = NULL; - null_msg->raw = oh->mesg[found_other.msgno].raw; - null_msg->raw_size = oh->mesg[found_other.msgno].raw_size; - null_msg->chunkno = oh->mesg[found_other.msgno].chunkno; - - /* Copy the message to move (& its prefix) to its new location */ - HDmemcpy(p, oh->mesg[found_other.msgno].raw - H5O_SIZEOF_MSGHDR_OH(oh), - oh->mesg[found_other.msgno].raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh)); - - /* Switch moved message to point to new location */ - oh->mesg[found_other.msgno].raw = p + H5O_SIZEOF_MSGHDR_OH(oh); - oh->mesg[found_other.msgno].chunkno = chunkno; - - /* Account for copied message in new chunk */ - p += (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + oh->mesg[found_other.msgno].raw_size; - size -= (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + oh->mesg[found_other.msgno].raw_size; - - /* Add any available space after the message to move to the new null message */ - if(found_other.gap_size > 0) { - /* Absorb a gap after the moved message */ - HDassert(oh->chunk[null_msg->chunkno].gap == found_other.gap_size); - null_msg->raw_size += found_other.gap_size; - oh->chunk[null_msg->chunkno].gap = 0; + /* Release chunk, marking it dirty */ + if(H5O_chunk_unprotect(f, dxpl_id, chk_proxy, TRUE) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") } /* end if */ - else if(found_other.null_size > 0) { - H5O_mesg_t *old_null_msg = &oh->mesg[found_other.null_msgno]; /* Pointer to NULL message to eliminate */ + else { + /* Move message (that will be replaced with continuation message) + * to new chunk, if necessary. + */ + H5O_mesg_t *null_msg; /* Pointer to new null message */ + + /* Protect chunk */ + if(NULL == (chk_proxy = H5O_chunk_protect(f, dxpl_id, oh, oh->mesg[found_msg->msgno].chunkno))) + HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to load object header chunk") - /* Absorb a null message after the moved message */ - HDassert((null_msg->raw + null_msg->raw_size) == (old_null_msg->raw - H5O_SIZEOF_MSGHDR_OH(oh))); - null_msg->raw_size += found_other.null_size; + /* Create null message for space that message to copy currently occupies */ + found_null = oh->nmesgs++; + null_msg = &(oh->mesg[found_null]); + null_msg->type = H5O_MSG_NULL; + null_msg->native = NULL; + null_msg->raw = oh->mesg[found_msg->msgno].raw; + null_msg->raw_size = oh->mesg[found_msg->msgno].raw_size; + null_msg->chunkno = oh->mesg[found_msg->msgno].chunkno; + + /* Copy the message to move (& its prefix) to its new location */ + HDmemcpy(p, oh->mesg[found_msg->msgno].raw - H5O_SIZEOF_MSGHDR_OH(oh), + oh->mesg[found_msg->msgno].raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh)); + + /* Switch moved message to point to new location */ + oh->mesg[found_msg->msgno].raw = p + H5O_SIZEOF_MSGHDR_OH(oh); + oh->mesg[found_msg->msgno].chunkno = chunkno; + + /* Account for copied message in new chunk */ + p += (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + oh->mesg[found_msg->msgno].raw_size; + size -= (size_t)H5O_SIZEOF_MSGHDR_OH(oh) + oh->mesg[found_msg->msgno].raw_size; + + /* Add any available space after the message to move to the new null message */ + if(found_msg->gap_size > 0) { + /* Absorb a gap after the moved message */ + HDassert(oh->chunk[null_msg->chunkno].gap == found_msg->gap_size); + null_msg->raw_size += found_msg->gap_size; + oh->chunk[null_msg->chunkno].gap = 0; + } /* end if */ + else if(found_msg->null_size > 0) { + H5O_mesg_t *old_null_msg = &oh->mesg[found_msg->null_msgno]; /* Pointer to NULL message to eliminate */ - /* Release any information/memory for message */ - H5O_msg_free_mesg(old_null_msg); + /* Absorb a null message after the moved message */ + HDassert((null_msg->raw + null_msg->raw_size) == (old_null_msg->raw - H5O_SIZEOF_MSGHDR_OH(oh))); + null_msg->raw_size += found_msg->null_size; - /* Remove null message from list of messages */ - if(found_other.null_msgno < (oh->nmesgs - 1)) - HDmemmove(old_null_msg, old_null_msg + 1, ((oh->nmesgs - 1) - found_other.null_msgno) * sizeof(H5O_mesg_t)); + /* Release any information/memory for message */ + H5O_msg_free_mesg(old_null_msg); - /* Decrement # of messages */ - /* (Don't bother reducing size of message array for now -QAK) */ - oh->nmesgs--; + /* Remove null message from list of messages */ + if(found_msg->null_msgno < (oh->nmesgs - 1)) + HDmemmove(old_null_msg, old_null_msg + 1, ((oh->nmesgs - 1) - found_msg->null_msgno) * sizeof(H5O_mesg_t)); - /* Adjust message index for new NULL message */ - found_null--; - } /* end if */ + /* Decrement # of messages */ + /* (Don't bother reducing size of message array for now -QAK) */ + oh->nmesgs--; + + /* Adjust message index for new NULL message */ + found_null--; + } /* end if */ - /* Mark the new null message as dirty */ - null_msg->dirty = TRUE; + /* Mark the new null message as dirty */ + null_msg->dirty = TRUE; - /* Release chunk, marking it dirty */ - if(H5O_chunk_unprotect(f, dxpl_id, chk_proxy, TRUE) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") + /* Release chunk, marking it dirty */ + if(H5O_chunk_unprotect(f, dxpl_id, chk_proxy, TRUE) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") + } /* end if */ } /* end if */ - HDassert(found_null >= 0); /* Create null message for [rest of] space in new chunk */ /* (account for chunk's magic # & checksum) */ @@ -1083,7 +1109,7 @@ H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new oh->mesg[idx].chunkno = chunkno; /* Insert the new chunk into the cache */ - if(H5O_chunk_add(f, dxpl_id, oh, chunkno) < 0) + if(H5O_chunk_add(f, dxpl_id, oh, chunkno, oh->mesg[found_null].chunkno) < 0) HGOTO_ERROR(H5E_OHDR, H5E_CANTINSERT, FAIL, "can't add new chunk to cache") /* Initialize the continuation information */ @@ -1094,7 +1120,7 @@ H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new cont->chunkno = chunkno; /* Split the null message and point at continuation message */ - if(H5O_alloc_null(f, dxpl_id, oh, (size_t)found_null, H5O_MSG_CONT, cont, cont_size) < 0) + if(H5O_alloc_null(f, dxpl_id, oh, found_null, H5O_MSG_CONT, cont, cont_size) < 0) HGOTO_ERROR(H5E_OHDR, H5E_CANTINSERT, FAIL, "can't split null message") /* Set new message index value */ @@ -1102,10 +1128,162 @@ H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new done: FUNC_LEAVE_NOAPI(ret_value) +} /* H5O__alloc_chunk() */ + + +/*------------------------------------------------------------------------- + * Function: H5O_alloc_new_chunk + * + * Purpose: Allocates a new chunk for the object header, including + * file space. + * + * One of the other chunks will get an object continuation + * message. If there isn't room in any other chunk for the + * object continuation message, then some message from + * another chunk is moved into this chunk to make room. + * + * SIZE need not be aligned. + * + * Note: The algorithm for finding a message to replace with a + * continuation message is still fairly limited. It's possible + * that two (or more) messages smaller than a continuation message + * might occupy a chunk and need to be moved in order to make + * room for the continuation message. + * + * Also, we aren't checking for NULL messages in front of another + * message right now... + * + * Return: Success: Index number of the null message for the + * new chunk. The null message will be at + * least SIZE bytes not counting the message + * ID or size fields. + * + * Failure: Negative + * + * Programmer: Robb Matzke + * matzke@llnl.gov + * Aug 7 1997 + * + *------------------------------------------------------------------------- + */ +static herr_t +H5O_alloc_new_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, size_t *new_idx) +{ + size_t cont_size; /*continuation message size */ + size_t idx; /* Message number */ + H5O_msg_alloc_info_t found_msg; /* Best fit non-null message */ + herr_t ret_value = SUCCEED; /* Return value */ + + FUNC_ENTER_NOAPI_NOINIT + + /* check args */ + HDassert(oh); + HDassert(size > 0); + size = H5O_ALIGN_OH(oh, size); + + /* Find the smallest null message that could hold a continuation message */ + idx = oh->nmesgs; + cont_size = H5O_ALIGN_OH(oh, (size_t)(H5F_SIZEOF_ADDR(f) + H5F_SIZEOF_SIZE(f))); + if(H5O__alloc_find_best_null(oh, cont_size, &idx) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_NOTFOUND, FAIL, "error while locating best null header message") + + /* If we couldn't find a null message, locate the best message to move to new chunk */ + if(idx >= oh->nmesgs) { + found_msg.msgno = -1; + if(H5O__alloc_find_best_nonnull(f, oh, &size, &found_msg) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_NOTFOUND, FAIL, "error while locating best non-null header message") + } /* end if */ + + /* Allocate and initialize new chunk in the file */ + if(H5O__alloc_chunk(f, dxpl_id, oh, size, idx, &found_msg, new_idx) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTALLOC, FAIL, "can't allocate new object header chunk") + +done: + FUNC_LEAVE_NOAPI(ret_value) } /* H5O_alloc_new_chunk() */ /*------------------------------------------------------------------------- + * Function: H5O__alloc_find_best_null + * + * Purpose: Find the best fit null message for a given size of message + * to allocate. + * + * Return: Non-negative on success/Negative on failure + * + * Programmer: Quincey Koziol + * koziol@lbl.gov + * Oct 21 2016 + * + *------------------------------------------------------------------------- + */ +static herr_t +H5O__alloc_find_best_null(const H5O_t *oh, size_t size, size_t *mesg_idx) +{ + size_t idx; /* Index of message which fits allocation */ + int found_null; /* Best fit null message */ + herr_t ret_value = SUCCEED; /* Return value */ + + FUNC_ENTER_STATIC + + /* check args */ + HDassert(oh); + HDassert(size > 0); + HDassert(mesg_idx); + + /* Find the smallest null message that could hold the new object header message */ + found_null = -1; + for(idx = 0; idx < oh->nmesgs; idx++) { + if(H5O_NULL_ID == oh->mesg[idx].type->id) { + /* If we found an exact fit, use it */ + if(oh->mesg[idx].raw_size == size) { + /* Keep first exact fit */ + if(found_null < 0) + found_null = idx; + else + /* If we've got more than one exact fit, choose the one in the earliest chunk */ + if(oh->mesg[idx].chunkno < oh->mesg[found_null].chunkno) { + found_null = idx; + + /* If we found an exact fit in object header chunk #0, we can get out */ + /* (Could extend this to look for earliest message in + * chunk #0 - QAK, 2016/10/21) + */ + if(0 == oh->mesg[idx].chunkno) + break; + } /* end if */ + } /* end if */ + /* Look for null message that's larger than needed */ + else if(oh->mesg[idx].raw_size > size) { + /* Keep first one found */ + if(found_null < 0) + found_null = idx; + else + /* Check for better fit */ + if(oh->mesg[idx].raw_size < oh->mesg[found_null].raw_size) + found_null = idx; + else { + /* If they are the same size, choose the one in the earliest chunk */ + if(oh->mesg[idx].raw_size == oh->mesg[found_null].raw_size) { + if(oh->mesg[idx].chunkno < oh->mesg[found_null].chunkno) + found_null = idx; + } /* end if */ + } /* end else */ + } /* end else-if */ + /* Ignore too-small null messages */ + else + ; + } /* end if */ + } /* end for */ + if(found_null >= 0) + *mesg_idx = found_null; + +done: + FUNC_LEAVE_NOAPI(ret_value) +} /* H5O__alloc_find_best_null() */ + + +/*------------------------------------------------------------------------- * Function: H5O_alloc * * Purpose: Allocate enough space in the object header for this message. @@ -1143,10 +1321,10 @@ H5O_alloc(H5F_t *f, hid_t dxpl_id, H5O_t *oh, const H5O_msg_class_t *type, HGOTO_ERROR(H5E_OHDR, H5E_CANTINIT, FAIL, "object header message is too large") aligned_size = H5O_ALIGN_OH(oh, raw_size); - /* look for a null message which is large enough */ - for(idx = 0; idx < oh->nmesgs; idx++) - if(H5O_NULL_ID == oh->mesg[idx].type->id && oh->mesg[idx].raw_size >= aligned_size) - break; + /* Find the smallest null message that could hold the new object header message */ + idx = oh->nmesgs; + if(H5O__alloc_find_best_null(oh, aligned_size, &idx) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_NOTFOUND, FAIL, "error while locating best null header message") /* if we didn't find one, then allocate more header space */ if(idx >= oh->nmesgs) { @@ -1307,12 +1485,6 @@ H5O_move_cont(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned cont_u) nonnull_size = 0; for(v = 0, curr_msg = &oh->mesg[0]; v < oh->nmesgs; v++, curr_msg++) { if(curr_msg->chunkno == deleted_chunkno) { - /* If there's a locked message, we can't move all messages out of - * chunk to delete, so get out now. - */ - if(curr_msg->locked) - HGOTO_DONE(FALSE) - /* Find size of all non-null messages in the chunk pointed to by the continuation message */ if(curr_msg->type->id != H5O_NULL_ID) { HDassert(curr_msg->type->id != H5O_CONT_ID); @@ -1552,152 +1724,149 @@ H5O_move_msgs_forward(H5F_t *f, hid_t dxpl_id, H5O_t *oh) } /* end else-if */ } /* end if */ - /* Don't let locked messages be moved into earlier chunk */ - if(!curr_msg->locked) { - /* Loop over messages again, looking for large enough null message in earlier chunk */ - for(v = 0, null_msg = &oh->mesg[0]; v < oh->nmesgs; v++, null_msg++) { - if(H5O_NULL_ID == null_msg->type->id && curr_msg->chunkno > null_msg->chunkno - && curr_msg->raw_size <= null_msg->raw_size) { - unsigned old_chunkno; /* Old message information */ - uint8_t *old_raw; - - /* Keep old information about non-null message */ - old_chunkno = curr_msg->chunkno; - old_raw = curr_msg->raw; - - /* Protect chunks */ - if(NULL == (null_chk_proxy = H5O_chunk_protect(f, dxpl_id, oh, null_msg->chunkno))) - HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to load object header chunk") - if(NULL == (curr_chk_proxy = H5O_chunk_protect(f, dxpl_id, oh, curr_msg->chunkno))) - HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to load object header chunk") + /* Loop over messages again, looking for large enough null message in earlier chunk */ + for(v = 0, null_msg = &oh->mesg[0]; v < oh->nmesgs; v++, null_msg++) { + if(H5O_NULL_ID == null_msg->type->id && curr_msg->chunkno > null_msg->chunkno + && curr_msg->raw_size <= null_msg->raw_size) { + unsigned old_chunkno; /* Old message information */ + uint8_t *old_raw; + + /* Keep old information about non-null message */ + old_chunkno = curr_msg->chunkno; + old_raw = curr_msg->raw; + + /* Protect chunks */ + if(NULL == (null_chk_proxy = H5O_chunk_protect(f, dxpl_id, oh, null_msg->chunkno))) + HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to load object header chunk") + if(NULL == (curr_chk_proxy = H5O_chunk_protect(f, dxpl_id, oh, curr_msg->chunkno))) + HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to load object header chunk") + + /* Copy raw data for non-null message to new chunk */ + HDmemcpy(null_msg->raw - H5O_SIZEOF_MSGHDR_OH(oh), curr_msg->raw - H5O_SIZEOF_MSGHDR_OH(oh), curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh)); + + /* Point non-null message at null message's space */ + curr_msg->chunkno = null_msg->chunkno; + curr_msg->raw = null_msg->raw; + curr_chk_dirtied = TRUE; + + /* Change information for null message */ + if(curr_msg->raw_size == null_msg->raw_size) { + /* Point null message at old non-null space */ + /* (Instead of freeing it and allocating new message) */ + null_msg->chunkno = old_chunkno; + null_msg->raw = old_raw; + + /* Mark null message dirty */ + null_msg->dirty = TRUE; + null_chk_dirtied = TRUE; + + /* Release current chunk, marking it dirty */ + if(H5O_chunk_unprotect(f, dxpl_id, curr_chk_proxy, curr_chk_dirtied) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") + curr_chk_proxy = NULL; + curr_chk_dirtied = FALSE; + + /* Check for gap in null message's chunk */ + if(oh->chunk[old_chunkno].gap > 0) { + /* Eliminate the gap in the chunk */ + if(H5O_eliminate_gap(oh, &null_chk_dirtied, null_msg, + ((oh->chunk[old_chunkno].image + oh->chunk[old_chunkno].size) - (H5O_SIZEOF_CHKSUM_OH(oh) + oh->chunk[old_chunkno].gap)), + oh->chunk[old_chunkno].gap) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTREMOVE, FAIL, "can't eliminate gap in chunk") + } /* end if */ - /* Copy raw data for non-null message to new chunk */ - HDmemcpy(null_msg->raw - H5O_SIZEOF_MSGHDR_OH(oh), curr_msg->raw - H5O_SIZEOF_MSGHDR_OH(oh), curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh)); + /* Release null chunk, marking it dirty */ + if(H5O_chunk_unprotect(f, dxpl_id, null_chk_proxy, null_chk_dirtied) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") + null_chk_proxy = NULL; + null_chk_dirtied = FALSE; + } /* end if */ + else { + size_t new_null_msg; /* Message index for new null message */ - /* Point non-null message at null message's space */ - curr_msg->chunkno = null_msg->chunkno; - curr_msg->raw = null_msg->raw; - curr_chk_dirtied = TRUE; + /* Check if null message is large enough to still exist */ + if((null_msg->raw_size - curr_msg->raw_size) < (size_t)H5O_SIZEOF_MSGHDR_OH(oh)) { + size_t gap_size = null_msg->raw_size - curr_msg->raw_size; /* Size of gap produced */ - /* Change information for null message */ - if(curr_msg->raw_size == null_msg->raw_size) { - /* Point null message at old non-null space */ - /* (Instead of freeing it and allocating new message) */ - null_msg->chunkno = old_chunkno; - null_msg->raw = old_raw; + /* Adjust the size of the null message being eliminated */ + null_msg->raw_size = curr_msg->raw_size; /* Mark null message dirty */ null_msg->dirty = TRUE; null_chk_dirtied = TRUE; - /* Release current chunk, marking it dirty */ - if(H5O_chunk_unprotect(f, dxpl_id, curr_chk_proxy, curr_chk_dirtied) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") - curr_chk_proxy = NULL; - curr_chk_dirtied = FALSE; - - /* Check for gap in null message's chunk */ - if(oh->chunk[old_chunkno].gap > 0) { - /* Eliminate the gap in the chunk */ - if(H5O_eliminate_gap(oh, &null_chk_dirtied, null_msg, - ((oh->chunk[old_chunkno].image + oh->chunk[old_chunkno].size) - (H5O_SIZEOF_CHKSUM_OH(oh) + oh->chunk[old_chunkno].gap)), - oh->chunk[old_chunkno].gap) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTREMOVE, FAIL, "can't eliminate gap in chunk") - } /* end if */ + /* Add the gap to the chunk */ + if(H5O_add_gap(f, oh, null_msg->chunkno, &null_chk_dirtied, v, null_msg->raw + null_msg->raw_size, gap_size) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTINSERT, FAIL, "can't insert gap in chunk") - /* Release null chunk, marking it dirty */ - if(H5O_chunk_unprotect(f, dxpl_id, null_chk_proxy, null_chk_dirtied) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") - null_chk_proxy = NULL; - null_chk_dirtied = FALSE; + /* Re-use message # for new null message taking place of non-null message */ + new_null_msg = v; } /* end if */ else { - size_t new_null_msg; /* Message index for new null message */ - - /* Check if null message is large enough to still exist */ - if((null_msg->raw_size - curr_msg->raw_size) < (size_t)H5O_SIZEOF_MSGHDR_OH(oh)) { - size_t gap_size = null_msg->raw_size - curr_msg->raw_size; /* Size of gap produced */ + /* Adjust null message's size & offset */ + null_msg->raw += curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh); + null_msg->raw_size -= curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh); - /* Adjust the size of the null message being eliminated */ - null_msg->raw_size = curr_msg->raw_size; - - /* Mark null message dirty */ - null_msg->dirty = TRUE; - null_chk_dirtied = TRUE; + /* Mark null message dirty */ + null_msg->dirty = TRUE; + null_chk_dirtied = TRUE; - /* Add the gap to the chunk */ - if(H5O_add_gap(f, oh, null_msg->chunkno, &null_chk_dirtied, v, null_msg->raw + null_msg->raw_size, gap_size) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTINSERT, FAIL, "can't insert gap in chunk") + /* Create new null message for previous location of non-null message */ + if(oh->nmesgs >= oh->alloc_nmesgs) { + if(H5O_alloc_msgs(oh, (size_t)1) < 0) + HGOTO_ERROR(H5E_RESOURCE, H5E_NOSPACE, FAIL, "can't allocate more space for messages") - /* Re-use message # for new null message taking place of non-null message */ - new_null_msg = v; - } /* end if */ - else { - /* Adjust null message's size & offset */ - null_msg->raw += curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh); - null_msg->raw_size -= curr_msg->raw_size + (size_t)H5O_SIZEOF_MSGHDR_OH(oh); - - /* Mark null message dirty */ - null_msg->dirty = TRUE; - null_chk_dirtied = TRUE; - - /* Create new null message for previous location of non-null message */ - if(oh->nmesgs >= oh->alloc_nmesgs) { - if(H5O_alloc_msgs(oh, (size_t)1) < 0) - HGOTO_ERROR(H5E_RESOURCE, H5E_NOSPACE, FAIL, "can't allocate more space for messages") - - /* "Retarget" 'curr_msg' pointer into newly re-allocated array of messages */ - curr_msg = &oh->mesg[u]; - } /* end if */ - - /* Get message # for new null message */ - new_null_msg = oh->nmesgs++; - } /* end else */ - - /* Release null message's chunk, marking it dirty */ - if(H5O_chunk_unprotect(f, dxpl_id, null_chk_proxy, null_chk_dirtied) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") - null_chk_proxy = NULL; - null_chk_dirtied = FALSE; - - /* Initialize new null message to take over non-null message's location */ - oh->mesg[new_null_msg].type = H5O_MSG_NULL; - oh->mesg[new_null_msg].native = NULL; - oh->mesg[new_null_msg].raw = old_raw; - oh->mesg[new_null_msg].raw_size = curr_msg->raw_size; - oh->mesg[new_null_msg].chunkno = old_chunkno; - - /* Mark new null message dirty */ - oh->mesg[new_null_msg].dirty = TRUE; - curr_chk_dirtied = TRUE; - - /* Check for gap in new null message's chunk */ - if(oh->chunk[old_chunkno].gap > 0) { - /* Eliminate the gap in the chunk */ - if(H5O_eliminate_gap(oh, &curr_chk_dirtied, &oh->mesg[new_null_msg], - ((oh->chunk[old_chunkno].image + oh->chunk[old_chunkno].size) - (H5O_SIZEOF_CHKSUM_OH(oh) + oh->chunk[old_chunkno].gap)), - oh->chunk[old_chunkno].gap) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTREMOVE, FAIL, "can't eliminate gap in chunk") + /* "Retarget" 'curr_msg' pointer into newly re-allocated array of messages */ + curr_msg = &oh->mesg[u]; } /* end if */ - /* Release new null message's chunk, marking it dirty */ - if(H5O_chunk_unprotect(f, dxpl_id, curr_chk_proxy, curr_chk_dirtied) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") - curr_chk_proxy = NULL; - curr_chk_dirtied = FALSE; + /* Get message # for new null message */ + new_null_msg = oh->nmesgs++; } /* end else */ - /* Indicate that we packed messages */ - packed_msg = TRUE; + /* Release null message's chunk, marking it dirty */ + if(H5O_chunk_unprotect(f, dxpl_id, null_chk_proxy, null_chk_dirtied) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") + null_chk_proxy = NULL; + null_chk_dirtied = FALSE; + + /* Initialize new null message to take over non-null message's location */ + oh->mesg[new_null_msg].type = H5O_MSG_NULL; + oh->mesg[new_null_msg].native = NULL; + oh->mesg[new_null_msg].raw = old_raw; + oh->mesg[new_null_msg].raw_size = curr_msg->raw_size; + oh->mesg[new_null_msg].chunkno = old_chunkno; + + /* Mark new null message dirty */ + oh->mesg[new_null_msg].dirty = TRUE; + curr_chk_dirtied = TRUE; - /* Break out of loop */ - /* (If it's possible to move message to even earlier chunk - * we'll get it on the next pass - QAK) - */ - break; - } /* end if */ - } /* end for */ - } /* end if */ + /* Check for gap in new null message's chunk */ + if(oh->chunk[old_chunkno].gap > 0) { + /* Eliminate the gap in the chunk */ + if(H5O_eliminate_gap(oh, &curr_chk_dirtied, &oh->mesg[new_null_msg], + ((oh->chunk[old_chunkno].image + oh->chunk[old_chunkno].size) - (H5O_SIZEOF_CHKSUM_OH(oh) + oh->chunk[old_chunkno].gap)), + oh->chunk[old_chunkno].gap) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTREMOVE, FAIL, "can't eliminate gap in chunk") + } /* end if */ + + /* Release new null message's chunk, marking it dirty */ + if(H5O_chunk_unprotect(f, dxpl_id, curr_chk_proxy, curr_chk_dirtied) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect object header chunk") + curr_chk_proxy = NULL; + curr_chk_dirtied = FALSE; + } /* end else */ + + /* Indicate that we packed messages */ + packed_msg = TRUE; + + /* Break out of loop */ + /* (If it's possible to move message to even earlier chunk + * we'll get it on the next pass - QAK) + */ + break; + } /* end if */ + } /* end for */ /* If we packed messages, get out of loop and start over */ /* (Don't know if this has any benefit one way or the other -QAK) */ @@ -1715,10 +1884,14 @@ H5O_move_msgs_forward(H5F_t *f, hid_t dxpl_id, H5O_t *oh) ret_value = (htri_t)did_packing; done: - if(null_chk_proxy && H5O_chunk_unprotect(f, dxpl_id, null_chk_proxy, null_chk_dirtied) < 0) - HDONE_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect null object header chunk") - if(curr_chk_proxy && H5O_chunk_unprotect(f, dxpl_id, curr_chk_proxy, curr_chk_dirtied) < 0) - HDONE_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect current object header chunk") + if(ret_value < 0) { + if(null_chk_proxy && H5O_chunk_unprotect(f, dxpl_id, null_chk_proxy, null_chk_dirtied) < 0) + HDONE_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect null object header chunk") + if(curr_chk_proxy && H5O_chunk_unprotect(f, dxpl_id, curr_chk_proxy, curr_chk_dirtied) < 0) + HDONE_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to unprotect current object header chunk") + } /* end if */ + else + HDassert(!null_chk_proxy && !curr_chk_proxy); FUNC_LEAVE_NOAPI(ret_value) } /* H5O_move_msgs_forward() */ diff --git a/src/H5Ocache.c b/src/H5Ocache.c index 0bb0bdf..f1a128d 100644 --- a/src/H5Ocache.c +++ b/src/H5Ocache.c @@ -88,9 +88,6 @@ static herr_t H5O__cache_chk_serialize(const H5F_t *f, void *image, size_t len, static herr_t H5O__cache_chk_free_icr(void *thing); static herr_t H5O__cache_chk_clear(const H5F_t *f, void *thing, hbool_t about_to_destroy); -/* Chunk proxy routines */ -static herr_t H5O__chunk_proxy_dest(H5O_chunk_proxy_t *chunk_proxy); - /* Chunk routines */ static herr_t H5O__chunk_deserialize(H5O_t *oh, haddr_t addr, size_t len, const uint8_t *image, H5O_common_cache_ud_t *udata, hbool_t *dirty); @@ -99,6 +96,7 @@ static herr_t H5O__chunk_serialize(const H5F_t *f, H5O_t *oh, unsigned chunkno); /* Misc. routines */ static herr_t H5O__add_cont_msg(H5O_cont_msgs_t *cont_msg_info, const H5O_cont_t *cont); +static herr_t H5O_decode_prefix(H5F_t *f, H5O_t *oh, const uint8_t *buf, void *_udata); /*********************/ @@ -160,99 +158,48 @@ H5FL_SEQ_DEFINE(H5O_cont_t); /*------------------------------------------------------------------------- - * Function: H5O__cache_get_load_size() + * Function: H5O_decode_prefix * - * Purpose: Tell the metadata cache how much data to read from file in - * the first speculative read for the object header. Note that we do - * not have to be concerned about reading past the end of file, as the - * cache will clamp the read to avoid this if needed. + * Purpose: To decode the object header prefix. + * The coding is extracted fromt H5O__cache_deserialize() to this routine. * - * Return: Success: SUCCEED - * Failure: FAIL + * Return: Non-negative on success/Negative on failure * - * Programmer: John Mainzer - * 7/28/14 + * Programmer: Vailin Choi; Aug 2015 * *------------------------------------------------------------------------- */ static herr_t -H5O__cache_get_load_size(const void H5_ATTR_UNUSED *_udata, size_t *image_len) -{ - FUNC_ENTER_STATIC_NOERR - - /* Check arguments */ - HDassert(image_len); - - *image_len = H5O_SPEC_READ_SIZE; - - FUNC_LEAVE_NOAPI(SUCCEED) -} /* end H5O__cache_get_load_size() */ - - -/*------------------------------------------------------------------------- - * Function: H5O__cache_deserialize - * - * Purpose: Attempt to deserialize the object header contained in the - * supplied buffer, load the data into an instance of H5O_t, and - * return a pointer to the new instance. - * - * Note that the object header is read with with a speculative read. - * If the initial read is too small, make note of this fact and return - * without error. H5C_load_entry() will note the size discrepency - * and retry the deserialize operation with the correct size read. - * - * Return: Success: Pointer to in core representation - * Failure: NULL - * - * Programmer: John Mainzer - * 7/28/14 - * - *------------------------------------------------------------------------- - */ -static void * -H5O__cache_deserialize(const void *_image, size_t len, void *_udata, - hbool_t *dirty) +H5O_decode_prefix(H5F_t *f, H5O_t *oh, const uint8_t *buf, void *_udata) { - H5O_t *oh = NULL; /* Object header read in */ - H5O_cache_ud_t *udata = (H5O_cache_ud_t *)_udata; /* User data for callback */ - const uint8_t *image = (const uint8_t *)_image; /* Pointer into buffer to decode */ - size_t prefix_size; /* Size of object header prefix */ - size_t buf_size; /* Size of prefix+chunk #0 buffer */ - void * ret_value = NULL; /* Return value */ + H5O_cache_ud_t *udata = (H5O_cache_ud_t *)_udata; /* User data for callback */ + const uint8_t *p = buf; /* Pointer into buffer to decode */ + size_t prefix_size; /* Size of object header prefix */ + herr_t ret_value = SUCCEED; /* Return value */ - FUNC_ENTER_STATIC + FUNC_ENTER_NOAPI_NOINIT /* Check arguments */ - HDassert(image); - HDassert(len > 0); + HDassert(f); + HDassert(oh); + HDassert(buf); HDassert(udata); - HDassert(udata->common.f); - HDassert(udata->common.cont_msg_info); - HDassert(dirty); - - /* Allocate space for the object header data structure */ - if(NULL == (oh = H5FL_CALLOC(H5O_t))) - HGOTO_ERROR(H5E_RESOURCE, H5E_NOSPACE, NULL, "memory allocation failed") - - /* File-specific, non-stored information */ - oh->sizeof_size = H5F_SIZEOF_SIZE(udata->common.f); - oh->sizeof_addr = H5F_SIZEOF_ADDR(udata->common.f); /* Check for presence of magic number */ /* (indicates version 2 or later) */ - if(!HDmemcmp(image, H5O_HDR_MAGIC, (size_t)H5_SIZEOF_MAGIC)) { + if(!HDmemcmp(p, H5O_HDR_MAGIC, (size_t)H5_SIZEOF_MAGIC)) { /* Magic number */ - image += H5_SIZEOF_MAGIC; + p += H5_SIZEOF_MAGIC; /* Version */ - oh->version = *image++; - if(H5O_VERSION_2 != oh->version) - HGOTO_ERROR(H5E_OHDR, H5E_VERSION, NULL, "bad object header version number") + oh->version = *p++; + if(H5O_VERSION_2 != oh->version) + HGOTO_ERROR(H5E_OHDR, H5E_VERSION, FAIL, "bad object header version number") /* Flags */ - oh->flags = *image++; - if(oh->flags & ~H5O_HDR_ALL_FLAGS) - HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, NULL, "unknown object header status flag(s)") + oh->flags = *p++; + if(oh->flags & ~H5O_HDR_ALL_FLAGS) + HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, FAIL, "unknown object header status flag(s)") /* Number of links to object (unless overridden by refcount message) */ oh->nlink = 1; @@ -261,13 +208,13 @@ H5O__cache_deserialize(const void *_image, size_t len, void *_udata, if(oh->flags & H5O_HDR_STORE_TIMES) { uint32_t tmp; /* Temporary value */ - UINT32DECODE(image, tmp); + UINT32DECODE(p, tmp); oh->atime = (time_t)tmp; - UINT32DECODE(image, tmp); + UINT32DECODE(p, tmp); oh->mtime = (time_t)tmp; - UINT32DECODE(image, tmp); + UINT32DECODE(p, tmp); oh->ctime = (time_t)tmp; - UINT32DECODE(image, tmp); + UINT32DECODE(p, tmp); oh->btime = (time_t)tmp; } /* end if */ else @@ -275,11 +222,10 @@ H5O__cache_deserialize(const void *_image, size_t len, void *_udata, /* Attribute fields */ if(oh->flags & H5O_HDR_ATTR_STORE_PHASE_CHANGE) { - UINT16DECODE(image, oh->max_compact); - UINT16DECODE(image, oh->min_dense); - + UINT16DECODE(p, oh->max_compact); + UINT16DECODE(p, oh->min_dense); if(oh->max_compact < oh->min_dense) - HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, NULL, "bad object header attribute phase change values") + HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, FAIL, "bad object header attribute phase change values") } /* end if */ else { oh->max_compact = H5O_CRT_ATTR_MAX_COMPACT_DEF; @@ -289,44 +235,44 @@ H5O__cache_deserialize(const void *_image, size_t len, void *_udata, /* First chunk size */ switch(oh->flags & H5O_HDR_CHUNK0_SIZE) { case 0: /* 1 byte size */ - oh->chunk0_size = *image++; + oh->chunk0_size = *p++; break; case 1: /* 2 byte size */ - UINT16DECODE(image, oh->chunk0_size); + UINT16DECODE(p, oh->chunk0_size); break; case 2: /* 4 byte size */ - UINT32DECODE(image, oh->chunk0_size); + UINT32DECODE(p, oh->chunk0_size); break; case 3: /* 8 byte size */ - UINT64DECODE(image, oh->chunk0_size); + UINT64DECODE(p, oh->chunk0_size); break; default: - HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, NULL, "bad size for chunk 0") + HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, FAIL, "bad size for chunk 0") } /* end switch */ if(oh->chunk0_size > 0 && oh->chunk0_size < H5O_SIZEOF_MSGHDR_OH(oh)) - HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, NULL, "bad object header chunk size") + HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, FAIL, "bad object header chunk size") } /* end if */ else { /* Version */ - oh->version = *image++; + oh->version = *p++; if(H5O_VERSION_1 != oh->version) - HGOTO_ERROR(H5E_OHDR, H5E_VERSION, NULL, "bad object header version number") + HGOTO_ERROR(H5E_OHDR, H5E_VERSION, FAIL, "bad object header version number") /* Flags */ oh->flags = H5O_CRT_OHDR_FLAGS_DEF; /* Reserved */ - image++; + p++; /* Number of messages */ - UINT16DECODE(image, udata->v1_pfx_nmesgs); + UINT16DECODE(p, udata->v1_pfx_nmesgs); /* Link count */ - UINT32DECODE(image, oh->nlink); + UINT32DECODE(p, oh->nlink); /* Reset unused time fields */ oh->atime = oh->mtime = oh->ctime = oh->btime = 0; @@ -336,24 +282,110 @@ H5O__cache_deserialize(const void *_image, size_t len, void *_udata, oh->min_dense = 0; /* First chunk size */ - UINT32DECODE(image, oh->chunk0_size); - - if((udata->v1_pfx_nmesgs > 0 && - oh->chunk0_size < H5O_SIZEOF_MSGHDR_OH(oh)) || - (udata->v1_pfx_nmesgs == 0 && oh->chunk0_size > 0)) - HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, NULL, "bad object header chunk size") + UINT32DECODE(p, oh->chunk0_size); + if((udata->v1_pfx_nmesgs > 0 && oh->chunk0_size < H5O_SIZEOF_MSGHDR_OH(oh)) || + (udata->v1_pfx_nmesgs == 0 && oh->chunk0_size > 0)) + HGOTO_ERROR(H5E_OHDR, H5E_BADVALUE, FAIL, "bad object header chunk size") /* Reserved, in version 1 (for 8-byte alignment padding) */ - image += 4; + p += 4; } /* end else */ /* Determine object header prefix length */ - prefix_size = (size_t)(image - (const uint8_t *)_image); + prefix_size = (size_t)(p - buf); HDassert((size_t)prefix_size == (size_t)(H5O_SIZEOF_HDR(oh) - H5O_SIZEOF_CHKSUM_OH(oh))); +done: + FUNC_LEAVE_NOAPI(ret_value) +} /* H5O_decode_prefix() */ + + +/*------------------------------------------------------------------------- + * Function: H5O__cache_get_load_size() + * + * Purpose: Tell the metadata cache how much data to read from file in + * the first speculative read for the object header. Note that we do + * not have to be concerned about reading past the end of file, as the + * cache will clamp the read to avoid this if needed. + * + * Return: Success: SUCCEED + * Failure: FAIL + * + * Programmer: John Mainzer + * 7/28/14 + * + *------------------------------------------------------------------------- + */ +static herr_t +H5O__cache_get_load_size(const void H5_ATTR_UNUSED *_udata, size_t *image_len) +{ + FUNC_ENTER_STATIC_NOERR + + /* Check arguments */ + HDassert(image_len); + + *image_len = H5O_SPEC_READ_SIZE; + + FUNC_LEAVE_NOAPI(SUCCEED) +} /* end H5O__cache_get_load_size() */ + + +/*------------------------------------------------------------------------- + * Function: H5O__cache_deserialize + * + * Purpose: Attempt to deserialize the object header contained in the + * supplied buffer, load the data into an instance of H5O_t, and + * return a pointer to the new instance. + * + * Note that the object header is read with with a speculative read. + * If the initial read is too small, make note of this fact and return + * without error. H5C_load_entry() will note the size discrepency + * and retry the deserialize operation with the correct size read. + * + * Return: Success: Pointer to in core representation + * Failure: NULL + * + * Programmer: John Mainzer + * 7/28/14 + * + *------------------------------------------------------------------------- + */ +static void * +H5O__cache_deserialize(const void *_image, size_t len, void *_udata, + hbool_t *dirty) +{ + H5O_t *oh = NULL; /* Object header read in */ + H5O_cache_ud_t *udata = (H5O_cache_ud_t *)_udata; /* User data for callback */ + const uint8_t *image = (const uint8_t *)_image; /* Pointer into buffer to decode */ + size_t buf_size; /* Size of prefix+chunk #0 buffer */ + void * ret_value = NULL; /* Return value */ + + FUNC_ENTER_STATIC + + /* Check arguments */ + HDassert(image); + HDassert(len > 0); + HDassert(udata); + HDassert(udata->common.f); + HDassert(udata->common.cont_msg_info); + HDassert(dirty); + + /* Allocate space for the object header data structure */ + if(NULL == (oh = H5FL_CALLOC(H5O_t))) + HGOTO_ERROR(H5E_RESOURCE, H5E_NOSPACE, NULL, "memory allocation failed") + + /* File-specific, non-stored information */ + oh->sizeof_size = H5F_SIZEOF_SIZE(udata->common.f); + oh->sizeof_addr = H5F_SIZEOF_ADDR(udata->common.f); + + /* Decode header prefix */ + if(H5O_decode_prefix(udata->common.f, oh, image, udata) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTINIT, NULL, "can't deserialize object header prefix") + /* Compute the size of the buffer used */ buf_size = oh->chunk0_size + (size_t)H5O_SIZEOF_HDR(oh); + /* Check to see if the buffer provided is large enough to contain both * the prefix and the first chunk. If it isn't, make note of the desired * size, but otherwise do nothing. H5C_load_entry() will notice the @@ -768,32 +800,35 @@ H5O__cache_chk_deserialize(const void *image, size_t len, void *_udata, if(H5O__chunk_deserialize(udata->oh, udata->common.addr, udata->size, (const uint8_t *)image, &(udata->common), dirty) < 0) HGOTO_ERROR(H5E_OHDR, H5E_CANTINIT, NULL, "can't deserialize object header chunk") - /* Set the fields for the chunk proxy */ - chk_proxy->oh = udata->oh; + /* Set the chunk number for the chunk proxy */ H5_CHECKED_ASSIGN(chk_proxy->chunkno, unsigned, udata->oh->nchunks - 1, size_t); } /* end if */ else { /* Sanity check */ HDassert(udata->chunkno < udata->oh->nchunks); - /* Set the fields for the chunk proxy */ - chk_proxy->oh = udata->oh; + /* Set the chunk number for the chunk proxy */ chk_proxy->chunkno = udata->chunkno; /* Sanity check that the chunk representation we have in memory is * the same as the one being brought in from disk. */ - HDassert(0 == HDmemcmp(image, chk_proxy->oh->chunk[chk_proxy->chunkno].image, chk_proxy->oh->chunk[chk_proxy->chunkno].size)); + HDassert(0 == HDmemcmp(image, udata->oh->chunk[chk_proxy->chunkno].image, udata->oh->chunk[chk_proxy->chunkno].size)); } /* end else */ /* Increment reference count of object header */ if(H5O_inc_rc(udata->oh) < 0) HGOTO_ERROR(H5E_OHDR, H5E_CANTINC, NULL, "can't increment reference count on object header") + chk_proxy->oh = udata->oh; /* Set return value */ ret_value = chk_proxy; done: + if(NULL == ret_value) + if(chk_proxy && H5O__chunk_dest(chk_proxy) < 0) + HDONE_ERROR(H5E_OHDR, H5E_CANTRELEASE, NULL, "unable to destroy object header chunk") + FUNC_LEAVE_NOAPI(ret_value) } /* end H5O__cache_chk_deserialize() */ @@ -920,7 +955,7 @@ H5O__cache_chk_free_icr(void *_thing) HDassert(chk_proxy->cache_info.type == H5AC_OHDR_CHK); /* Destroy object header chunk proxy */ - if(H5O__chunk_proxy_dest(chk_proxy) < 0) + if(H5O__chunk_dest(chk_proxy) < 0) HGOTO_ERROR(H5E_OHDR, H5E_CANTRELEASE, FAIL, "unable to destroy object header chunk proxy") done: @@ -1453,39 +1488,3 @@ done: FUNC_LEAVE_NOAPI(ret_value) } /* H5O__chunk_serialize() */ - -/*------------------------------------------------------------------------- - * Function: H5O__chunk_proxy_dest - * - * Purpose: Destroy a chunk proxy object - * - * Return: Success: SUCCEED - * Failure: FAIL - * - * Programmer: Quincey Koziol - * koziol@hdfgroup.org - * July 13, 2008 - * - *------------------------------------------------------------------------- - */ -static herr_t -H5O__chunk_proxy_dest(H5O_chunk_proxy_t *chk_proxy) -{ - herr_t ret_value = SUCCEED; /* Return value */ - - FUNC_ENTER_STATIC - - /* Check arguments */ - HDassert(chk_proxy); - - /* Decrement reference count of object header */ - if(chk_proxy->oh && H5O_dec_rc(chk_proxy->oh) < 0) - HGOTO_ERROR(H5E_OHDR, H5E_CANTDEC, FAIL, "can't decrement reference count on object header") - - /* Release the chunk proxy object */ - chk_proxy = H5FL_FREE(H5O_chunk_proxy_t, chk_proxy); - -done: - FUNC_LEAVE_NOAPI(ret_value) -} /* H5O__chunk_proxy_dest() */ - diff --git a/src/H5Ochunk.c b/src/H5Ochunk.c index 18561b3..f6eb135 100644 --- a/src/H5Ochunk.c +++ b/src/H5Ochunk.c @@ -93,7 +93,8 @@ H5FL_DEFINE(H5O_chunk_proxy_t); *------------------------------------------------------------------------- */ herr_t -H5O_chunk_add(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx) +H5O_chunk_add(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx, + unsigned cont_chunkno) { H5O_chunk_proxy_t *chk_proxy = NULL; /* Proxy for chunk, to mark it dirty in the cache */ herr_t ret_value = SUCCEED; /* Return value */ @@ -110,14 +111,16 @@ H5O_chunk_add(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx) if(NULL == (chk_proxy = H5FL_CALLOC(H5O_chunk_proxy_t))) HGOTO_ERROR(H5E_RESOURCE, H5E_NOSPACE, FAIL, "memory allocation failed") - /* Set the values in the chunk proxy */ - chk_proxy->oh = oh; - chk_proxy->chunkno = idx; - /* Increment reference count on object header */ if(H5O_inc_rc(oh) < 0) HGOTO_ERROR(H5E_OHDR, H5E_CANTINC, FAIL, "can't increment reference count on object header") + /* Set the values in the chunk proxy */ + chk_proxy->f = f; + chk_proxy->oh = oh; + chk_proxy->chunkno = idx; + chk_proxy->cont_chunkno = cont_chunkno; + /* Insert the chunk proxy into the cache */ if(H5AC_insert_entry(f, dxpl_id, H5AC_OHDR_CHK, oh->chunk[idx].addr, chk_proxy, H5AC__NO_FLAGS_SET) < 0) HGOTO_ERROR(H5E_OHDR, H5E_CANTINSERT, FAIL, "unable to cache object header chunk") @@ -126,8 +129,8 @@ H5O_chunk_add(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx) done: if(ret_value < 0) - if(chk_proxy) - chk_proxy = H5FL_FREE(H5O_chunk_proxy_t, chk_proxy); + if(chk_proxy && H5O__chunk_dest(chk_proxy) < 0) + HDONE_ERROR(H5E_OHDR, H5E_CANTRELEASE, FAIL, "unable to destroy object header chunk") FUNC_LEAVE_NOAPI_TAG(ret_value, FAIL) } /* end H5O_chunk_add() */ @@ -172,6 +175,7 @@ H5O_chunk_protect(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx) HGOTO_ERROR(H5E_OHDR, H5E_CANTINC, NULL, "can't increment reference count on object header") /* Set chunk proxy fields */ + chk_proxy->f = f; chk_proxy->oh = oh; chk_proxy->chunkno = idx; } /* end if */ @@ -200,8 +204,8 @@ H5O_chunk_protect(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx) done: /* Cleanup on error */ if(!ret_value) - if(0 == idx && chk_proxy) - chk_proxy = H5FL_FREE(H5O_chunk_proxy_t, chk_proxy); + if(0 == idx && chk_proxy && H5O__chunk_dest(chk_proxy) < 0) + HDONE_ERROR(H5E_OHDR, H5E_CANTRELEASE, NULL, "unable to destroy object header chunk") FUNC_LEAVE_NOAPI_TAG(ret_value, NULL) } /* end H5O_chunk_protect() */ @@ -409,3 +413,39 @@ done: FUNC_LEAVE_NOAPI_TAG(ret_value, FAIL) } /* end H5O_chunk_delete() */ + +/*------------------------------------------------------------------------- + * Function: H5O__chunk_dest + * + * Purpose: Destroy a chunk proxy object + * + * Return: Success: SUCCEED + * Failure: FAIL + * + * Programmer: Quincey Koziol + * koziol@hdfgroup.org + * July 13, 2008 + * + *------------------------------------------------------------------------- + */ +herr_t +H5O__chunk_dest(H5O_chunk_proxy_t *chk_proxy) +{ + herr_t ret_value = SUCCEED; /* Return value */ + + FUNC_ENTER_PACKAGE + + /* Check arguments */ + HDassert(chk_proxy); + + /* Decrement reference count of object header */ + if(chk_proxy->oh && H5O_dec_rc(chk_proxy->oh) < 0) + HGOTO_ERROR(H5E_OHDR, H5E_CANTDEC, FAIL, "can't decrement reference count on object header") + + /* Release the chunk proxy object */ + chk_proxy = H5FL_FREE(H5O_chunk_proxy_t, chk_proxy); + +done: + FUNC_LEAVE_NOAPI(ret_value) +} /* H5O__chunk_dest() */ + diff --git a/src/H5Omessage.c b/src/H5Omessage.c index d42896c..7063ef6 100644 --- a/src/H5Omessage.c +++ b/src/H5Omessage.c @@ -2254,176 +2254,3 @@ done: FUNC_LEAVE_NOAPI(ret_value) } /* end H5O_flush_msgs() */ - -/*------------------------------------------------------------------------- - * Function: H5O_msg_chunkno - * - * Purpose: Queries the object header chunk index for a message. - * - * Return: Success: >=0 value indicating the chunk number for - * the message - * Failure: <0 - * - * Programmer: Quincey Koziol - * koziol@hdfgroup.org - * Apr 22 2010 - * - *------------------------------------------------------------------------- - */ -int -H5O_msg_get_chunkno(const H5O_loc_t *loc, unsigned type_id, hid_t dxpl_id) -{ - H5O_t *oh = NULL; /* Object header to use */ - const H5O_msg_class_t *type; /* Actual H5O class type for the ID */ - H5O_mesg_t *idx_msg; /* Pointer to message to modify */ - unsigned idx; /* Index of message to modify */ - int ret_value = -1; /* Return value */ - - FUNC_ENTER_NOAPI(FAIL) - - /* check args */ - HDassert(loc); - HDassert(loc->file); - HDassert(H5F_addr_defined(loc->addr)); - HDassert(type_id < NELMTS(H5O_msg_class_g)); - type = H5O_msg_class_g[type_id]; /* map the type ID to the actual type object */ - HDassert(type); - - /* Get the object header */ - if(NULL == (oh = H5O_protect(loc, dxpl_id, H5AC__READ_ONLY_FLAG))) - HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to protect object header") - - /* Locate message of correct type */ - for(idx = 0, idx_msg = &oh->mesg[0]; idx < oh->nmesgs; idx++, idx_msg++) - if(type == idx_msg->type) - break; - if(idx == oh->nmesgs) - HGOTO_ERROR(H5E_OHDR, H5E_NOTFOUND, FAIL, "message type not found") - - /* Set return value */ - H5_CHECKED_ASSIGN(ret_value, int, idx_msg->chunkno, unsigned); - -done: - if(oh && H5O_unprotect(loc, dxpl_id, oh, H5AC__NO_FLAGS_SET) < 0) - HDONE_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to release object header") - - FUNC_LEAVE_NOAPI(ret_value) -} /* end H5O_msg_get_chunkno() */ - - -/*------------------------------------------------------------------------- - * Function: H5O_msg_lock - * - * Purpose: Locks a message into a particular chunk, preventing it from - * being moved into another chunk. - * - * Return: Non-negative on success/Negative on failure - * - * Programmer: Quincey Koziol - * koziol@hdfgroup.org - * Apr 22 2010 - * - *------------------------------------------------------------------------- - */ -herr_t -H5O_msg_lock(const H5O_loc_t *loc, unsigned type_id, hid_t dxpl_id) -{ - H5O_t *oh = NULL; /* Object header to use */ - const H5O_msg_class_t *type; /* Actual H5O class type for the ID */ - H5O_mesg_t *idx_msg; /* Pointer to message to modify */ - unsigned idx; /* Index of message to modify */ - herr_t ret_value = SUCCEED; /* Return value */ - - FUNC_ENTER_NOAPI(FAIL) - - /* check args */ - HDassert(loc); - HDassert(loc->file); - HDassert(H5F_addr_defined(loc->addr)); - HDassert(type_id < NELMTS(H5O_msg_class_g)); - type = H5O_msg_class_g[type_id]; /* map the type ID to the actual type object */ - HDassert(type); - - /* Get the object header */ - if(NULL == (oh = H5O_protect(loc, dxpl_id, H5AC__READ_ONLY_FLAG))) - HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to protect object header") - - /* Locate message of correct type */ - for(idx = 0, idx_msg = &oh->mesg[0]; idx < oh->nmesgs; idx++, idx_msg++) - if(type == idx_msg->type) - break; - if(idx == oh->nmesgs) - HGOTO_ERROR(H5E_OHDR, H5E_NOTFOUND, FAIL, "message type not found") - - /* Fail if the message is already locked */ - if(idx_msg->locked) - HGOTO_ERROR(H5E_OHDR, H5E_CANTLOCK, FAIL, "message already locked") - - /* Make the message locked */ - idx_msg->locked = TRUE; - -done: - if(oh && H5O_unprotect(loc, dxpl_id, oh, H5AC__NO_FLAGS_SET) < 0) - HDONE_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to release object header") - - FUNC_LEAVE_NOAPI(ret_value) -} /* end H5O_msg_lock() */ - - -/*------------------------------------------------------------------------- - * Function: H5O_msg_unlock - * - * Purpose: Unlocks a message, allowing it to be moved into another chunk. - * - * Return: Non-negative on success/Negative on failure - * - * Programmer: Quincey Koziol - * koziol@hdfgroup.org - * Apr 22 2010 - * - *------------------------------------------------------------------------- - */ -herr_t -H5O_msg_unlock(const H5O_loc_t *loc, unsigned type_id, hid_t dxpl_id) -{ - H5O_t *oh = NULL; /* Object header to use */ - const H5O_msg_class_t *type; /* Actual H5O class type for the ID */ - H5O_mesg_t *idx_msg; /* Pointer to message to modify */ - unsigned idx; /* Index of message to modify */ - herr_t ret_value = SUCCEED; /* Return value */ - - FUNC_ENTER_NOAPI(FAIL) - - /* check args */ - HDassert(loc); - HDassert(loc->file); - HDassert(H5F_addr_defined(loc->addr)); - HDassert(type_id < NELMTS(H5O_msg_class_g)); - type = H5O_msg_class_g[type_id]; /* map the type ID to the actual type object */ - HDassert(type); - - /* Get the object header */ - if(NULL == (oh = H5O_protect(loc, dxpl_id, H5AC__READ_ONLY_FLAG))) - HGOTO_ERROR(H5E_OHDR, H5E_CANTPROTECT, FAIL, "unable to protect object header") - - /* Locate message of correct type */ - for(idx = 0, idx_msg = &oh->mesg[0]; idx < oh->nmesgs; idx++, idx_msg++) - if(type == idx_msg->type) - break; - if(idx == oh->nmesgs) - HGOTO_ERROR(H5E_OHDR, H5E_NOTFOUND, FAIL, "message type not found") - - /* Fail if the message is not locked */ - if(!idx_msg->locked) - HGOTO_ERROR(H5E_OHDR, H5E_CANTUNLOCK, FAIL, "message not locked") - - /* Make the message unlocked */ - idx_msg->locked = FALSE; - -done: - if(oh && H5O_unprotect(loc, dxpl_id, oh, H5AC__NO_FLAGS_SET) < 0) - HDONE_ERROR(H5E_OHDR, H5E_CANTUNPROTECT, FAIL, "unable to release object header") - - FUNC_LEAVE_NOAPI(ret_value) -} /* end H5O_msg_unlock() */ - diff --git a/src/H5Opkg.h b/src/H5Opkg.h index a719872..659bf6e 100644 --- a/src/H5Opkg.h +++ b/src/H5Opkg.h @@ -245,7 +245,6 @@ struct H5O_msg_class_t { struct H5O_mesg_t { const H5O_msg_class_t *type; /*type of message */ hbool_t dirty; /*raw out of date wrt native */ - hbool_t locked; /*message is locked into chunk */ uint8_t flags; /*message flags */ H5O_msg_crt_idx_t crt_idx; /*message creation index */ unsigned chunkno; /*chunk number for this mesg */ @@ -254,6 +253,17 @@ struct H5O_mesg_t { size_t raw_size; /*size with alignment */ }; +/* Struct for storing information about "best" message to move to new chunk */ +typedef struct H5O_msg_alloc_info_t { + int msgno; /* Index in message array */ + unsigned id; /* Message type ID on disk */ + unsigned chunkno; /* Index in chunk array */ + size_t gap_size; /* Size of any "gap" in the chunk immediately after message */ + size_t null_size; /* Size of any null message in the chunk immediately after message */ + size_t total_size; /* Total size of "available" space around message */ + unsigned null_msgno; /* Message index of null message immediately after message */ +} H5O_msg_alloc_info_t; + typedef struct H5O_chunk_t { haddr_t addr; /*chunk file address */ size_t size; /*chunk size */ @@ -369,8 +379,10 @@ typedef struct H5O_chunk_proxy_t { H5AC_info_t cache_info; /* Information for metadata cache functions, _must_ be */ /* first field in structure */ + H5F_t *f; /* Pointer to file for object header/chunk */ H5O_t *oh; /* Object header for this chunk */ unsigned chunkno; /* Chunk number for this chunk */ + unsigned cont_chunkno; /* Chunk number for the chunk containing the continuation message that points to this chunk */ } H5O_chunk_proxy_t; /* Callback information for loading object header chunk from disk */ @@ -559,7 +571,8 @@ H5_DLL herr_t H5O_msg_iterate_real(H5F_t *f, H5O_t *oh, const H5O_msg_class_t *t const H5O_mesg_operator_t *op, void *op_data, hid_t dxpl_id); /* Object header chunk routines */ -H5_DLL herr_t H5O_chunk_add(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx); +H5_DLL herr_t H5O_chunk_add(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx, + unsigned cont_chunkno); H5_DLL H5O_chunk_proxy_t *H5O_chunk_protect(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx); H5_DLL herr_t H5O_chunk_unprotect(H5F_t *f, hid_t dxpl_id, @@ -567,6 +580,7 @@ H5_DLL herr_t H5O_chunk_unprotect(H5F_t *f, hid_t dxpl_id, H5_DLL herr_t H5O_chunk_update_idx(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx); H5_DLL herr_t H5O_chunk_resize(H5O_t *oh, H5O_chunk_proxy_t *chk_proxy); H5_DLL herr_t H5O_chunk_delete(H5F_t *f, hid_t dxpl_id, H5O_t *oh, unsigned idx); +H5_DLL herr_t H5O__chunk_dest(H5O_chunk_proxy_t *chunk_proxy); /* Collect storage info for btree and heap */ H5_DLL herr_t H5O_attr_bh_info(H5F_t *f, hid_t dxpl_id, H5O_t *oh, @@ -574,6 +588,8 @@ H5_DLL herr_t H5O_attr_bh_info(H5F_t *f, hid_t dxpl_id, H5O_t *oh, /* Object header allocation routines */ H5_DLL herr_t H5O_alloc_msgs(H5O_t *oh, size_t min_alloc); +H5_DLL herr_t H5O__alloc_chunk(H5F_t *f, hid_t dxpl_id, H5O_t *oh, size_t size, + size_t found_null, const H5O_msg_alloc_info_t *found_msg, size_t *new_idx); H5_DLL herr_t H5O_alloc(H5F_t *f, hid_t dxpl_id, H5O_t *oh, const H5O_msg_class_t *type, const void *mesg, size_t *mesg_idx); H5_DLL herr_t H5O_condense_header(H5F_t *f, H5O_t *oh, hid_t dxpl_id); diff --git a/src/H5Oprivate.h b/src/H5Oprivate.h index 0acac8c..e8ba267 100644 --- a/src/H5Oprivate.h +++ b/src/H5Oprivate.h @@ -898,9 +898,6 @@ H5_DLL void* H5O_msg_decode(H5F_t *f, hid_t dxpl_id, H5O_t *open_oh, unsigned type_id, const unsigned char *buf); H5_DLL herr_t H5O_msg_delete(H5F_t *f, hid_t dxpl_id, H5O_t *open_oh, unsigned type_id, void *mesg); -H5_DLL int H5O_msg_get_chunkno(const H5O_loc_t *loc, unsigned type_id, hid_t dxpl_id); -H5_DLL herr_t H5O_msg_lock(const H5O_loc_t *loc, unsigned type_id, hid_t dxpl_id); -H5_DLL herr_t H5O_msg_unlock(const H5O_loc_t *loc, unsigned type_id, hid_t dxpl_id); /* Object metadata flush/refresh routines */ H5_DLL herr_t H5O_flush_common(H5O_loc_t *oloc, hid_t obj_id, hid_t dxpl_id); diff --git a/test/ohdr.c b/test/ohdr.c index 5642c41..e613a98 100644 --- a/test/ohdr.c +++ b/test/ohdr.c @@ -811,148 +811,6 @@ main(void) FAIL_STACK_ERROR PASSED(); - /* - * Test moving message to first chunk - */ - TESTING("locking messages"); - HDmemset(&oh_loc, 0, sizeof(oh_loc)); - if(H5O_create(f, H5AC_ind_read_dxpl_id, (size_t)64, (size_t)0, H5P_GROUP_CREATE_DEFAULT, &oh_loc/*out*/) < 0) - FAIL_STACK_ERROR - if(1 != H5O_link(&oh_loc, 1, H5AC_ind_read_dxpl_id)) - FAIL_STACK_ERROR - - /* Create second object header, to guarantee that first object header uses multiple chunks */ - HDmemset(&oh_loc2, 0, sizeof(oh_loc2)); - if(H5O_create(f, H5AC_ind_read_dxpl_id, (size_t)64, (size_t)0, H5P_GROUP_CREATE_DEFAULT, &oh_loc2/*out*/) < 0) - FAIL_STACK_ERROR - if(1 != H5O_link(&oh_loc2, 1, H5AC_ind_read_dxpl_id)) - FAIL_STACK_ERROR - - /* Fill object header with messages, creating multiple chunks */ - for(i = 0; i < 10; i++) { - time_new = (i + 1) * 1000 + 10; - if(H5O_msg_create(&oh_loc, H5O_MTIME_NEW_ID, 0, 0, &time_new, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - } /* end for */ - - /* Get # of object header chunks */ - if(H5O_get_hdr_info(&oh_loc, H5AC_ind_read_dxpl_id, &hdr_info) < 0) - FAIL_STACK_ERROR - if(hdr_info.nchunks != 2) - TEST_ERROR - - /* Add message to lock to object header */ - time_new = 11111111; - if(H5O_msg_create(&oh_loc, H5O_MTIME_ID, 0, 0, &time_new, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - - /* Verify chunk index for message */ - if((chunkno = H5O_msg_get_chunkno(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id)) < 0) - FAIL_STACK_ERROR - if(chunkno != 1) - TEST_ERROR - - /* Lock the message into the chunk */ - if(H5O_msg_lock(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - - /* Attempt to lock the message twice */ - H5E_BEGIN_TRY { - ret = H5O_msg_lock(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id); - } H5E_END_TRY; - if(ret >= 0) - TEST_ERROR - - /* Delete all the other messages, which would move the message into - * chunk #0, if it wasn't locked - */ - if(H5O_msg_remove(&oh_loc, H5O_MTIME_NEW_ID, H5O_ALL, TRUE, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - - /* Verify chunk index for message */ - if((chunkno = H5O_msg_get_chunkno(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id)) < 0) - FAIL_STACK_ERROR - if(chunkno != 1) - TEST_ERROR - - /* Unlock the message */ - if(H5O_msg_unlock(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - - /* Attempt to unlock the message twice */ - H5E_BEGIN_TRY { - ret = H5O_msg_unlock(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id); - } H5E_END_TRY; - if(ret >= 0) - TEST_ERROR - - /* Close object headers */ - if(H5O_close(&oh_loc2) < 0) - FAIL_STACK_ERROR - if(H5O_close(&oh_loc) < 0) - FAIL_STACK_ERROR - - /* Open first object header */ - HDmemset(&oh_loc, 0, sizeof(oh_loc)); - if(H5O_create(f, H5AC_ind_read_dxpl_id, (size_t)64, (size_t)0, H5P_GROUP_CREATE_DEFAULT, &oh_loc/*out*/) < 0) - FAIL_STACK_ERROR - if(1 != H5O_link(&oh_loc, 1, H5AC_ind_read_dxpl_id)) - FAIL_STACK_ERROR - - /* Create second object header, to guarantee that first object header uses multiple chunks */ - HDmemset(&oh_loc2, 0, sizeof(oh_loc2)); - if(H5O_create(f, H5AC_ind_read_dxpl_id, (size_t)64, (size_t)0, H5P_GROUP_CREATE_DEFAULT, &oh_loc2/*out*/) < 0) - FAIL_STACK_ERROR - if(1 != H5O_link(&oh_loc2, 1, H5AC_ind_read_dxpl_id)) - FAIL_STACK_ERROR - - /* Add message to move to object header */ - time_new = 11111111; - if(H5O_msg_create(&oh_loc, H5O_MTIME_ID, 0, 0, &time_new, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - - /* Verify chunk index for message */ - if((chunkno = H5O_msg_get_chunkno(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id)) < 0) - FAIL_STACK_ERROR - if(chunkno != 0) - TEST_ERROR - - /* Lock the message into the chunk */ - if(H5O_msg_lock(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - - /* Fill object header with messages, creating multiple chunks */ - /* (would normally move locked message to new chunk) */ - for(i = 0; i < 10; i++) { - time_new = (i + 1) * 1000 + 10; - if(H5O_msg_create(&oh_loc, H5O_MTIME_NEW_ID, 0, 0, &time_new, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - } /* end for */ - - /* Get # of object header chunks */ - if(H5O_get_hdr_info(&oh_loc, H5AC_ind_read_dxpl_id, &hdr_info) < 0) - FAIL_STACK_ERROR - if(hdr_info.nchunks != 2) - TEST_ERROR - - /* Verify chunk index for message */ - if((chunkno = H5O_msg_get_chunkno(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id)) < 0) - FAIL_STACK_ERROR - if(chunkno != 0) - TEST_ERROR - - /* Unlock the message */ - if(H5O_msg_unlock(&oh_loc, H5O_MTIME_ID, H5AC_ind_read_dxpl_id) < 0) - FAIL_STACK_ERROR - - /* Close object headers */ - if(H5O_close(&oh_loc2) < 0) - FAIL_STACK_ERROR - if(H5O_close(&oh_loc) < 0) - FAIL_STACK_ERROR - - PASSED(); - /* Close the file we created */ if(H5Fclose(file) < 0) TEST_ERROR -- cgit v0.12 2 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387 4388 4389 4390 4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 4409 4410 4411 4412 4413 4414 4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460 4461 4462 4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510 4511 4512 4513 4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526 4527 4528 4529 4530 4531 4532 4533 4534 4535 4536 4537 4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548 4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715 4716 4717 4718 4719 4720 4721 4722 4723 4724 4725 4726 4727 4728 4729 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786 4787 4788 4789 4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814 4815 4816 4817 4818 4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860 4861 4862 4863 4864 4865 4866 4867 4868 4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896 4897 4898 4899 4900 4901 4902 4903 4904 4905 4906 4907 4908 4909 4910 4911 4912 4913 4914 4915 4916 4917 4918 4919 4920 4921 4922 4923 4924 4925 4926 4927 4928 4929 4930 4931 4932 4933 4934 4935 4936 4937 4938 4939 4940 4941 4942 4943 4944 4945 4946 4947 4948 4949 4950 4951 4952 4953 4954 4955 4956 4957 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970 4971 4972 4973 4974 4975 4976 4977 4978 4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990 4991 4992 4993 4994 4995 4996 4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015 5016 5017 5018 5019 5020 5021 5022 5023 5024 5025 5026 5027 5028 5029 5030 5031 5032 5033 5034 5035 5036 5037 5038 5039 5040 5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056 5057 5058 5059 5060 5061 5062 5063 5064 5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081 5082 5083 5084 5085 5086 5087 5088 5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099 5100 5101 5102 5103 5104 5105 5106 5107 5108 5109 5110 5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144 5145 5146 5147 5148 5149 5150 5151 5152 5153 5154 5155 5156 5157 5158 5159 5160 5161 5162 5163 5164 5165 5166 5167 5168 5169 5170 5171 5172 5173 5174 5175 5176 5177 5178 5179 5180 5181 5182 5183 5184 5185 5186 5187 5188 5189 5190 5191 5192 5193 5194 5195 5196 5197 5198 5199 5200 5201 5202 5203 5204 5205 5206 5207 5208 5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225 5226 5227 5228 5229 5230 5231 5232 5233 5234 5235 5236 5237 5238 5239 5240 5241 5242 5243 5244 5245 5246 5247 5248 5249 5250 5251 5252 5253 5254 5255 5256 5257 5258 5259 5260 5261 5262 5263 5264 5265 5266 5267 5268 5269 5270 5271 5272 5273 5274 5275 5276 5277 5278 5279 5280 5281 5282 5283 5284 5285 5286 5287 5288 5289 5290 5291 5292 5293 5294 5295 5296 5297 5298 5299 5300 5301 5302 5303 5304 5305 5306 5307 5308 5309 5310 5311 5312 5313 5314 5315 5316 5317 5318 5319 5320 5321 5322 5323 5324 5325 5326 5327 5328 5329 5330 5331 5332 5333 5334 5335 5336 5337 5338 5339 5340 5341 5342 5343 5344 5345 5346 5347 5348 5349 5350 5351 5352 5353 5354 5355 5356 5357 5358 5359 5360 5361 5362 5363 5364 5365 5366 5367 5368 5369 5370 5371 5372 5373 5374 5375 5376 5377 5378 5379 5380 5381 5382 5383 5384 5385 5386 5387 5388 5389 5390 5391 5392 5393 5394 5395 5396 5397 5398 5399 5400 5401 5402 5403 5404 5405 5406 5407 5408 5409 5410 5411 5412 5413 5414 5415 5416 5417 5418 5419 5420 5421 5422 5423 5424 5425 5426 5427 5428 5429 5430 5431 5432 5433 5434 5435 5436 5437 5438 5439 5440 5441 5442 5443 5444 5445 5446 5447 5448 5449 5450 5451 5452 5453 5454 5455 5456 5457 5458 5459 5460 5461 5462 5463 5464 5465 5466 5467 5468 5469 5470 5471 5472 5473 5474 5475 5476 5477 5478 5479 5480 5481 5482 5483 5484 5485 5486 5487 5488 5489 5490 5491 5492 5493 5494 5495 5496 5497 5498 5499 5500 5501 5502 5503 5504 5505 5506 5507 5508 5509 5510 5511 5512 5513 5514 5515 5516 5517 5518 5519 5520 5521 5522 5523 5524 5525 5526 5527 5528 5529 5530 5531 5532 5533 5534 5535 5536 5537 5538 5539 5540 5541 5542 5543 5544 5545 5546 5547 5548 5549 5550 5551 5552 5553 5554 5555 5556 5557 5558 5559 5560 5561 5562 5563 5564 5565 5566 5567 5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578 5579 5580 5581 5582 5583 5584 5585 5586 5587 5588 5589 5590 5591 5592 5593 5594 5595 5596 5597 5598 5599 5600 5601 5602 5603 5604 5605 5606 5607 5608 5609 5610 5611 5612 5613 5614 5615 5616 5617 5618 5619 5620 5621 5622 5623 5624 5625 5626 5627 5628 5629 5630 5631 5632 5633 5634 5635 5636 5637 5638 5639 5640 5641 5642 5643 5644 5645 5646 5647 5648 5649 5650 5651 5652 5653 5654 5655 5656 5657 5658 5659 5660 5661 5662 5663 5664 5665 5666 5667 5668 5669 5670 5671 5672 5673 5674 5675 5676 5677 5678 5679 5680 5681 5682 5683 5684 5685 5686 5687 5688 5689 5690 5691 5692 5693 5694 5695 5696 5697 5698 5699 5700 5701 5702 5703 5704 5705 5706 5707 5708 5709 5710 5711 5712 5713 5714 5715 5716 5717 5718 5719 5720 5721 5722 5723 5724 5725 5726 5727 5728 5729 5730 5731 5732 5733 5734 5735 5736 5737 5738 5739 5740 5741 5742 5743 5744 5745 5746 5747 5748 5749 5750 5751 5752 5753 5754 5755 5756 5757 5758 5759 5760 5761 5762 5763 5764 5765 5766 5767 5768 5769 5770 5771 5772 5773 5774 5775 5776 5777 5778 5779 5780 5781 5782 5783 5784 5785 5786 5787 5788 5789 5790 5791 5792 5793 5794 5795 5796 5797 5798 5799 5800 5801 5802 5803 5804 5805 5806 5807 5808 5809 5810 5811 5812 5813 5814 5815 5816 5817 5818 5819 5820 5821 5822 5823 5824 5825 5826 5827 5828 5829 5830 5831 5832 5833 5834 5835 5836 5837 5838 5839 5840 5841 5842 5843 5844 5845 5846 5847 5848 5849 5850 5851 5852 5853 5854 5855 5856 5857 5858 5859 5860 5861 5862 5863 5864 5865 5866 5867 5868 5869 5870 5871 5872 5873 5874 5875 5876 5877 5878 5879 5880 5881 5882 5883 5884 5885 5886 5887 5888 5889 5890 5891 5892 5893 5894 5895 5896 5897 5898 5899 5900 5901 5902 5903 5904 5905 5906 5907 5908 5909 5910 5911 5912 5913 5914 5915 5916 5917 5918 5919 5920 5921 5922 5923 5924 5925 5926 5927 5928 5929 5930 5931 5932 5933 5934 5935 5936 5937 5938 5939 5940 5941 5942 5943 5944 5945 5946 5947 5948 5949 5950 5951 5952 5953 5954 5955 5956 5957 5958 5959 5960 5961 5962 5963 5964 5965 5966 5967 5968 5969 5970 5971 5972 5973 5974 5975 5976 5977 5978 5979 5980 5981 5982 5983 5984 5985 5986 5987 5988 5989 5990 5991 5992 5993 5994 5995 5996 5997 5998 5999 6000 6001 6002 6003 6004 6005 6006 6007 6008 6009 6010 6011 6012 6013 6014 6015 6016 6017 6018 6019 6020 6021 6022 6023 6024 6025 6026 6027 6028 6029 6030 6031 6032 6033 6034 6035 6036 6037 6038 6039 6040 6041 6042 6043 6044 6045 6046 6047 6048 6049 6050 6051 6052 6053 6054 6055 6056 6057 6058 6059 6060 6061 6062 6063 6064 6065 6066 6067 6068 6069 6070 6071 6072 6073 6074 6075 6076 6077 6078 6079 6080 6081 6082 6083 6084 6085 6086 6087 6088 6089 6090 6091 6092 6093 6094 6095 6096 6097 6098 6099 6100 6101 6102 6103 6104 6105 6106 6107 6108 6109 6110 6111 6112 6113 6114 6115 6116 6117 6118 6119 6120 6121 6122 6123 6124 6125 6126 6127 6128 6129 6130 6131 6132 6133 6134 6135 6136 6137 6138 6139 6140 6141 6142 6143 6144 6145 6146 6147 6148 6149 6150 6151 6152 6153 6154 6155 6156 6157 6158 6159 6160 6161 6162 6163 6164 6165 6166 6167 6168 6169 6170 6171 6172 6173 6174 6175 6176 6177 6178 6179 6180 6181 6182 6183 6184 6185 6186 6187 6188 6189 6190 6191 6192 6193 6194 6195 6196 6197 6198 6199 6200 6201 6202 6203 6204 6205 6206 6207 6208 6209 6210 6211 6212 6213 6214 6215 6216 6217 6218 6219 6220 6221 6222 6223 6224 6225 6226 6227 6228 6229 6230 6231 6232 6233 6234 6235 6236 6237 6238 6239 6240 6241 6242 6243 6244 6245 6246 6247 6248 6249 6250 6251 6252 6253 6254 6255 6256 6257 6258 6259 6260 6261 6262 6263 6264 6265 6266 6267 6268 6269 6270 6271 6272 6273 6274 6275 6276 6277 6278 6279 6280 6281 6282 6283 6284 6285 6286 6287 6288 6289 6290 6291 6292 6293 6294 6295 6296 6297 6298 6299 6300 6301 6302 6303 6304 6305 6306 6307 6308 6309 6310 6311 6312 6313 6314 6315 6316 6317 6318 6319 6320 6321 6322 6323 6324 6325 6326 6327 6328 6329 6330 6331 6332 6333 6334 6335 6336 6337 6338 6339 6340 6341 6342 6343 6344 6345 6346 6347 6348 6349 6350 6351 6352 6353 6354 6355 6356 6357 6358 6359 6360 6361 6362 6363 6364 6365 6366 6367 6368 6369 6370 6371 6372 6373 6374 6375 6376 6377 6378 6379 6380 6381 6382 6383 6384 6385 6386 6387 6388 6389 6390 6391 6392 6393 6394 6395 6396 6397 6398 6399 6400 6401 6402 6403 6404 6405 6406 6407 6408 6409 6410 6411 6412 6413 6414 6415 6416 6417 6418 6419 6420 6421 6422 6423 6424 6425 6426 6427 6428 6429 6430 6431 6432 6433 6434 6435 6436 6437 6438 6439 6440 6441 6442 6443 6444 6445 6446 6447 6448 6449 6450 6451 6452 6453 6454 6455 6456 6457 6458 6459 6460 6461 6462 6463 6464 6465 6466 6467 6468 6469 6470 6471 6472 6473 6474 6475 6476 6477 6478 6479 6480 6481 6482 6483 6484 6485 6486 6487 6488 6489 6490 6491 6492 6493 6494 6495 6496 6497 6498 6499 6500 6501 6502 6503 6504 6505 6506 6507 6508 6509 6510 6511 6512 6513 6514 6515 6516 6517 6518 6519 6520 6521 6522 6523 6524 6525 6526 6527 6528 6529 6530 6531 6532 6533 6534 6535 6536 6537 6538 6539 6540 6541 6542 6543 6544 6545 6546 6547 6548 6549 6550 6551 6552 6553 6554 6555 6556 6557 6558 6559 6560 6561 6562 6563 6564 6565 6566 6567 6568 6569 6570 6571 6572 6573 6574 6575 6576 6577 6578 6579 6580 6581 6582 6583 6584 6585 6586 6587 6588 6589 6590 6591 6592 6593 6594 6595 6596 6597 6598 6599 6600 6601 6602 6603 6604 6605 6606 6607 6608 6609 6610 6611 6612 6613 6614 6615 6616 6617 6618 6619 6620 6621 6622 6623 6624 6625 6626 6627 6628 6629 6630 6631 6632 6633 6634 6635 6636 6637 6638 6639 6640 6641 6642 6643 6644 6645 6646 6647 6648 6649 6650 6651 6652 6653 6654 6655 6656 6657 6658 6659 6660 6661 6662 6663 6664 6665 6666 6667 6668 6669 6670 6671 6672 6673 6674 6675 6676 6677 6678 6679 6680 6681 6682 6683 6684 6685 6686 6687 6688 6689 6690 6691 6692 6693 6694 6695 6696 6697 6698 6699 6700 6701 6702 6703 6704 6705 6706 6707 6708 6709 6710 6711 6712 6713 6714 6715 6716 6717 6718 6719 6720 6721 6722 6723 6724 6725 6726 6727 6728 6729 6730 6731 6732 6733 6734 6735 6736 6737 6738 6739 6740 6741 6742 6743 6744 6745 6746 6747 6748 6749 6750 6751 6752 6753 6754 6755 6756 6757 6758 6759 6760 6761 6762 6763 6764 6765 6766 6767 6768 6769 6770 6771 6772 6773 6774 6775 6776 6777 6778 6779 6780 6781 6782 6783 6784 6785 6786 6787 6788 6789 6790 6791 6792 6793 6794 6795 6796 6797 6798 6799 6800 6801 6802 6803 6804 6805 6806 6807 6808 6809 6810 6811 6812 6813 6814 6815 6816 6817 6818 6819 6820 6821 6822 6823 6824 6825 6826 6827 6828 6829 6830 6831 6832 6833 6834 6835 6836 6837 6838 6839 6840 6841 6842 6843 6844 6845 6846 6847 6848 6849 6850 6851 6852 6853 6854 6855 6856 6857 6858 6859 6860 6861 6862 6863 6864 6865 6866 6867 6868 6869 6870 6871 6872 6873 6874 6875 6876 6877 6878 6879 6880 6881 6882 6883 6884 6885 6886 6887 6888 6889 6890 6891 6892 6893 6894 6895 6896 6897 6898 6899 6900 6901 6902 6903 6904 6905 6906 6907 6908 6909 6910 6911 6912 6913 6914 6915 6916 6917 6918 6919 6920 6921 6922 6923 6924 6925 6926 6927 6928 6929 6930 6931 6932 6933 6934 6935 6936 6937 6938 6939 6940 6941 6942 6943 6944 6945 6946 6947 6948 6949 6950 6951 6952 6953 6954 6955 6956 6957 6958 6959 6960 6961 6962 6963 6964 6965 6966 6967 6968 6969 6970 6971 6972 6973 6974 6975 6976 6977 6978 6979 6980 6981 6982 6983 6984 6985 6986 6987 6988 6989 6990 6991 6992 6993 6994 6995 6996 6997 6998 6999 7000 7001 7002 7003 7004 7005 7006 7007 7008 7009 7010 7011 7012 7013 7014 7015 7016 7017 7018 7019 7020 7021 7022 7023 7024 7025 7026 7027 7028 7029 7030 7031 7032 7033 7034 7035 7036 7037 7038 7039 7040 7041 7042 7043 7044 7045 7046 7047 7048 7049 7050 7051 7052 7053 7054 7055 7056 7057 7058 7059 7060 7061 7062 7063 7064 7065 7066 7067 7068 7069 7070 7071 7072 7073 7074 7075 7076 7077 7078 7079 7080 7081 7082 7083 7084 7085 7086 7087 7088 7089 7090 7091 7092 7093 7094 7095 7096 7097 7098 7099 7100 7101 7102 7103 7104 7105 7106 7107 7108 7109 7110 7111 7112 7113 7114 7115 7116 7117 7118 7119 7120 7121 7122 7123 7124 7125 7126 7127 7128 7129 7130 7131 7132 7133 7134 7135 7136 7137 7138 7139 7140 7141 7142 7143 7144 7145 7146 7147 7148 7149 7150 7151 7152 7153 7154 7155 7156 7157 7158 7159 7160 7161 7162 7163 7164 7165 7166 7167 7168 7169 7170 7171 7172 7173 7174 7175 7176 7177 7178 7179 7180 7181 7182 7183 7184 7185 7186 7187 7188 7189 7190 7191 7192 7193 7194 7195 7196 7197 7198 7199 7200 7201 7202 7203 7204 7205 7206 7207 7208 7209 7210 7211 7212 7213 7214 7215 7216 7217 7218 7219 7220 7221 7222 7223 7224 7225 7226 7227 7228 7229 7230 7231 7232 7233 7234 7235 7236 7237 7238 7239 7240 7241 7242 7243 7244 7245 7246 7247 7248 7249 7250 7251 7252 7253 7254 7255 7256 7257 7258 7259 7260 7261 7262 7263 7264 7265 7266 7267 7268 7269 7270 7271 7272 7273 7274 7275 7276 7277 7278 7279 7280 7281 7282 7283 7284 7285 7286 7287 7288 7289 7290 7291 7292 7293 7294 7295 7296 7297 7298 7299 7300 7301 7302 7303 7304 7305 7306 7307 7308 7309 7310 7311 7312 7313 7314 7315 7316 7317 7318 7319 7320 7321 7322 7323 7324 7325 7326 7327 7328 7329 7330 7331 7332 7333 7334 7335 7336 7337 7338 7339 7340 7341 7342 7343 7344 7345 7346 7347 7348 7349 7350 7351 7352 7353 7354 7355 7356 7357 7358 7359 7360 7361 7362 7363 7364 7365 7366 7367 7368 7369 7370 7371 7372 7373 7374 7375 7376 7377 7378 7379 7380 7381 7382 7383 7384 7385 7386 7387 7388 7389 7390 7391 7392 7393 7394 7395 7396 7397 7398 7399 7400 7401 7402 7403 7404 7405 7406 7407 7408 7409 7410 7411 7412 7413 7414 7415 7416 7417 7418 7419 7420 7421 7422 7423 7424 7425 7426 7427 7428 7429 7430 7431 7432 7433 7434 7435 7436 7437 7438 7439 7440 7441 7442 7443 7444 7445 7446 7447 7448 7449 7450 7451 7452 7453 7454 7455 7456 7457 7458 7459 7460 7461 7462 7463 7464 7465 7466 7467 7468 7469 7470 7471 7472 7473 7474 7475 7476 7477 7478 7479 7480 7481 7482 7483 7484 7485 7486 7487 7488 7489 7490 7491 7492 7493 7494 7495 7496 7497 7498 7499 7500 7501 7502 7503 7504 7505 7506 7507 7508 7509 7510 7511 7512 7513 7514 7515 7516 7517 7518 7519 7520 7521 7522 7523 7524 7525 7526 7527 7528 7529 7530 7531 7532 7533 7534 7535 7536 7537 7538 7539 7540 7541 7542 7543 7544 7545 7546 7547 7548 7549 7550 7551 7552 7553 7554 7555 7556 7557 7558 7559 7560 7561 7562 7563 7564 7565 7566 7567 7568 7569 7570 7571 7572 7573 7574 7575 7576 7577 7578 7579 7580 7581 7582 7583 7584 7585 7586 7587 7588 7589 7590 7591 7592 7593 7594 7595 7596 7597 7598 7599 7600 7601 7602 7603 7604 7605 7606 7607 7608 7609 7610 7611 7612 7613 7614 7615 7616 7617 7618 7619 7620 7621 7622 7623 7624 7625 7626 7627 7628 7629 7630 7631 7632 7633 7634 7635 7636 7637 7638 7639 7640 7641 7642 7643 7644 7645 7646 7647 7648 7649 7650 7651 7652 7653 7654 7655 7656 7657 7658 7659 7660 7661 7662 7663 7664 7665 7666 7667 7668 7669 7670 7671 7672 7673 7674 7675 7676 7677 7678 7679 7680 7681 7682 7683 7684 7685 7686 7687 7688 7689 7690 7691 7692 7693 7694 7695 7696 7697 7698 7699 7700 7701 7702 7703 7704 7705 7706 7707 7708 7709 7710 7711 7712 7713 7714 7715 7716 7717 7718 7719 7720 7721 7722 7723 7724 7725 7726 7727 7728 7729 7730 7731 7732 7733 7734 7735 7736 7737 7738 7739 7740 7741 7742 7743 7744 7745 7746 7747 7748 7749 7750 7751 7752 7753 7754 7755 7756 7757 7758 7759 7760 7761 7762 7763 7764 7765 7766 7767 7768 7769 7770 7771 7772 7773 7774 7775 7776 7777 7778 7779 7780 7781 7782 7783 7784 7785 7786 7787 7788 7789 7790 7791 7792 7793 7794 7795 7796 7797 7798 7799 7800 7801 7802 7803 7804 7805 7806 7807 7808 7809 7810 7811 7812 7813 7814 7815 7816 7817 7818 7819 7820 7821 7822 7823 7824 7825 7826 7827 7828 7829 7830 7831 7832 7833 7834 7835 7836 7837 7838 7839 7840 7841 7842 7843 7844 7845 7846 7847 7848 7849 7850 7851 7852 7853 7854 7855 7856 7857 7858 7859 7860 7861 7862 7863 7864 7865 7866 7867 7868 7869 7870 7871 7872 7873 7874 7875 7876 7877 7878 7879 7880 7881 7882 7883 7884 7885 7886 7887 7888 7889 7890 7891 7892 7893 7894 7895 7896 7897 7898 7899 7900 7901 7902 7903 7904 7905 7906 7907 7908 7909 7910 7911 7912 7913 7914 7915 7916 7917 7918 7919 7920 7921 7922 7923 7924 7925 7926 7927 7928 7929 7930 7931 7932 7933 7934 7935 7936 7937 7938 7939 7940 7941 7942 7943 7944 7945 7946 7947 7948 7949 7950 7951 7952 7953 7954 7955 7956 7957 7958 7959 7960 7961 7962 7963 7964 7965 7966 7967 7968 7969 7970 7971 7972 7973 7974 7975 7976 7977 7978 7979 7980 7981 7982 7983 7984 7985 7986 7987 7988 7989 7990 7991 7992 7993 7994 7995 7996 7997 7998 7999 8000 8001 8002 8003 8004 8005 8006 8007 8008 8009 8010 8011 8012 8013 8014 8015 8016 8017 8018 8019 8020 8021 8022 8023 8024 8025 8026 8027 8028 8029 8030 8031 8032 8033 8034 8035 8036 8037 8038 8039 8040 8041 8042 8043 8044 8045 8046 8047 8048 8049 8050 8051 8052 8053 8054 8055 8056 8057 8058 8059 8060 8061 8062 8063 8064 8065 8066 8067 8068 8069 8070 8071 8072 8073 8074 8075 8076 8077 8078 8079 8080 8081 8082 8083 8084 8085 8086 8087 8088 8089 8090 8091 8092 8093 8094 8095 8096 8097 8098 8099 8100 8101 8102 8103 8104 8105 8106 8107 8108 8109 8110 8111 8112 8113 8114 8115 8116 8117 8118 8119 8120 8121 8122 8123 8124 8125 8126 8127 8128 8129 8130 8131 8132 8133 8134 8135 8136 8137 8138 8139 8140 8141 8142 8143 8144 8145 8146 8147 8148 8149 8150 8151 8152 8153 8154 8155 8156 8157 8158 8159 8160 8161 8162 8163 8164 8165 8166 8167 8168 8169 8170 8171 8172 8173 8174 8175 8176 8177 8178 8179 8180 8181 8182 8183 8184 8185 8186 8187 8188 8189 8190 8191 8192 8193 8194 8195 8196 8197 8198 8199 8200 8201 8202 8203 8204 8205 8206 8207 8208 8209 8210 8211 8212 8213 8214 8215 8216 8217 8218 8219 8220 8221 8222 8223 8224 8225 8226 8227 8228 8229 8230 8231 8232 8233 8234 8235 8236 8237 8238 8239 8240 8241 8242 8243 8244 8245 8246 8247 8248 8249 8250 8251 8252 8253 8254 8255 8256 8257 8258 8259 8260 8261 8262 8263 8264 8265 8266 8267 8268 8269 8270 8271 8272 8273 8274 8275 8276 8277 8278 8279 8280 8281 8282 8283 8284 8285 8286 8287 8288 8289 8290 8291 8292 8293 8294 8295 8296 8297 8298 8299 8300 8301 8302 8303 8304 8305 8306 8307 8308 8309 8310 8311 8312 8313 8314 8315 8316 8317 8318 8319 8320 8321 8322 8323 8324 8325 8326 8327 8328 8329 8330 8331 8332 8333 8334 8335 8336 8337 8338 8339 8340 8341 8342 8343 8344 8345 8346 8347 8348 8349 8350 8351 8352 8353 8354 8355 8356 8357 8358 8359 8360 8361 8362 8363 8364 8365 8366 8367 8368 8369 8370 8371 8372 8373 8374 8375 8376 8377 8378 8379 8380 8381 8382 8383 8384 8385 8386 8387 8388 8389 8390 8391 8392 8393 8394 8395 8396 8397 8398 8399 8400 8401 8402 8403 8404 8405 8406 8407 8408 8409 8410 8411 8412 8413 8414 8415 8416 8417 8418 8419 8420 8421 8422 8423 8424 8425 8426 8427 8428 8429 8430 8431 8432 8433 8434 8435 8436 8437 8438 8439 8440 8441 8442 8443 8444 8445 8446 8447 8448 8449 8450 8451 8452 8453 8454 8455 8456 8457 8458 8459 8460 8461 8462 8463 8464 8465 8466 8467 8468 8469 8470 8471 8472 8473 8474 8475 8476 8477 8478 8479 8480 8481 8482 8483 8484 8485 8486 8487 8488 8489 8490 8491 8492 8493 8494 8495 8496 8497 8498 8499 8500 8501 8502 8503 8504 8505 8506 8507 8508 8509 8510 8511 8512 8513 8514 8515 8516 8517 8518 8519 8520 8521 8522 8523 8524 8525 8526 8527 8528 8529 8530 8531 8532 8533 8534 8535 8536 8537 8538 8539 8540 8541 8542 8543 8544 8545 8546 8547 8548 8549 8550 8551 8552 8553 8554 8555 8556 8557 8558 8559 8560 8561 8562 8563 8564 8565 8566 8567 8568 8569 8570 8571 8572 8573 8574 8575 8576 8577 8578 8579 8580 8581 8582 8583 8584 8585 8586 8587 8588 8589 8590 8591 8592 8593 8594 8595 8596 8597 8598 8599 8600 8601 8602 8603 8604 8605 8606 8607 8608 8609 8610 8611 8612 8613 8614 8615 8616 8617 8618 8619 8620 8621 8622 8623 8624 8625 8626 8627 8628 8629 8630 8631 8632 8633 8634 8635 8636 8637 8638 8639 8640 8641 8642 8643 8644 8645 8646 8647 8648 8649 8650 8651 8652 8653 8654 8655 8656 8657 8658 8659 8660 8661 8662 8663 8664 8665 8666 8667 8668 8669 8670 8671 8672 8673 8674 8675 8676 8677 8678 8679 8680 8681 8682 8683 8684 8685 8686 8687 8688 8689 8690 8691 8692 8693 8694 8695 8696 8697 8698 8699 8700 8701 8702 8703 8704 8705 8706 8707 8708 8709 8710 8711 8712 8713 8714 8715 8716 8717 8718 8719 8720 8721 8722 8723 8724 8725 8726 8727 8728 8729 8730 8731 8732 8733 8734 8735 8736 8737 8738 8739 8740 8741 8742 8743 8744 8745 8746 8747 8748 8749 8750 8751 8752 8753 8754 8755 8756 8757 8758 8759 8760 8761 8762 8763 8764 8765 8766 8767 8768 8769 8770 8771 8772 8773 8774 8775 8776 8777 8778 8779 8780 8781 8782 8783 8784 8785 8786 8787 8788 8789 8790 8791 8792 8793 8794 8795 8796 8797 8798 8799 8800 8801 8802 8803 8804 8805 8806 8807 8808 8809 8810 8811 8812 8813 8814 8815 8816 8817 8818 8819 8820 8821 8822 8823 8824 8825 8826 8827 8828 8829 8830 8831 8832 8833 8834 8835 8836 8837 8838 8839 8840 8841 8842 8843 8844 8845 8846 8847 8848 8849 8850 8851 8852 8853 8854 8855 8856 8857 8858 8859 8860 8861 8862 8863 8864 8865 8866 8867 8868 8869 8870 8871 8872 8873 8874 8875 8876 8877 8878 8879 8880 8881 8882 8883 8884 8885 8886 8887 8888 8889 8890 8891 8892 8893 8894 8895 8896 8897 8898 8899 8900 8901 8902 8903 8904 8905 8906 8907 8908 8909 8910 8911 8912 8913 8914 8915 8916 8917 8918 8919 8920 8921 8922 8923 8924 8925 8926 8927 8928 8929 8930 8931 8932 8933 8934 8935 8936 8937 8938 8939 8940 8941 8942 8943 8944 8945 8946 8947 8948 8949 8950 8951 8952 8953 8954 8955 8956 8957 8958 8959 8960 8961 8962 8963 8964 8965 8966 8967 8968 8969 8970 8971 8972 8973 8974 8975 8976 8977 8978 8979 8980 8981 8982 8983 8984 8985 8986 8987 8988 8989 8990 8991 8992 8993 8994 8995 8996 8997 8998 8999 9000 9001 9002 9003 9004 9005 9006 9007 9008 9009 9010 9011 9012 9013 9014 9015 9016 9017 9018 9019 9020 9021 9022 9023 9024 9025 9026 9027 9028 9029 9030 9031 9032 9033 9034 9035 9036 9037 9038 9039 9040 9041 9042 9043 9044 9045 9046 9047 9048 9049 9050 9051 9052 9053 9054 9055 9056 9057 9058 9059 9060 9061 9062 9063 9064 9065 9066 9067 9068 9069 9070 9071 9072 9073 9074 9075 9076 9077 9078 9079 9080 9081 9082 9083 9084 9085 9086 9087 9088 9089 9090 9091 9092 9093 9094 9095 9096 9097 9098 9099 9100 9101 9102 9103 9104 9105 9106 9107 9108 9109 9110 9111 9112 9113 9114 9115 9116 9117 9118 9119 9120 9121 9122 9123 9124 9125 9126 9127 9128 9129 9130 9131 9132 9133 9134 9135 9136 9137 9138 9139 9140 9141 9142 9143 9144 9145 9146 9147 9148 9149 9150 9151 9152 9153 9154 9155 9156 9157 9158 9159 9160 9161 9162 9163 9164 9165 9166 9167 9168 9169 9170 9171 9172 9173 9174 9175 9176 9177 9178 9179 9180 9181 9182 9183 9184 9185 9186 9187 9188 9189 9190 9191 9192 9193 9194 9195 9196 9197 9198 9199 9200 9201 9202 9203 9204 9205 9206 9207 9208 9209 9210 9211 9212 9213 9214 9215 9216 9217 9218 9219 9220 9221 9222 9223 9224 9225 9226 9227 9228 9229 9230 9231 9232 9233 9234 9235 9236 9237 9238 9239 9240 9241 9242 9243 9244 9245 9246 9247 9248 9249 9250 9251 9252 9253 9254 9255 9256 9257 9258 9259 9260 9261 9262 9263 9264 9265 9266 9267 9268 9269 9270 9271 9272 9273 9274 9275 9276 9277 9278 9279 9280 9281 9282 9283 9284 9285 9286 9287 9288 9289 9290 9291 9292 9293 9294 9295 9296 9297 9298 9299 9300 9301 9302 9303 9304 9305 9306 9307 9308 9309 9310 9311 9312 9313 9314 9315 9316 9317 9318 9319 9320 9321 9322 9323 9324 9325 9326 9327 9328 9329 9330 9331 9332 9333 9334 9335 9336 9337 9338 9339 9340 9341 9342 9343 9344 9345 9346 9347 9348 9349 9350 9351 9352 9353 9354 9355 9356 9357 9358 9359 9360 9361 9362 9363 9364 9365 9366 9367 9368 9369 9370 9371 9372 9373 9374 9375 9376 9377 9378 9379 9380 9381 9382 9383 9384 9385 9386 9387 9388 9389 9390 9391 9392 9393 9394 9395 9396 9397 9398 9399 9400 9401 9402 9403 9404 9405 9406 9407 9408 9409 9410 9411 9412 9413 9414 9415 9416 9417 9418 9419 9420 9421 9422 9423 9424 9425 9426 9427 9428 9429 9430 9431 9432 9433 9434 9435 9436 9437 9438 9439 9440 9441 9442 9443 9444 9445 9446 9447 9448 9449 9450 9451 9452 9453 9454 9455 9456 9457 9458 9459 9460 9461 9462 9463 9464 9465 9466 9467 9468 9469 9470 9471 9472 9473 9474 9475 9476 9477 9478 9479 9480 9481 9482 9483 9484 9485 9486 9487 9488 9489 9490 9491 9492 9493 9494 9495 9496 9497 9498 9499 9500 9501 9502 9503 9504 9505 9506 9507 9508 9509 9510 9511 9512 9513 9514 9515 9516 9517 9518 9519 9520 9521 9522 9523 9524 9525 9526 9527 9528 9529 9530 9531 9532 9533 9534 9535 9536 9537 9538 9539 9540 9541 9542 9543 9544 9545 9546 9547 9548 9549 9550 9551 9552 9553 9554 9555 9556 9557 9558 9559 9560 9561 9562 9563 9564 9565 9566 9567 9568 9569 9570 9571 9572 9573 9574 9575 9576 9577 9578 9579 9580 9581 9582 9583 9584 9585 9586 9587 9588 9589 9590 9591 9592 9593 9594 9595 9596 9597 9598 9599 9600 9601 9602 9603 9604 9605 9606 9607 9608 9609 9610 9611 9612 9613 9614 9615 9616 9617 9618 9619 9620 9621 9622 9623 9624 9625 9626 9627 9628 9629 9630 9631 9632 9633 9634 9635 9636 9637 9638 9639 9640 9641 9642 9643 9644 9645 9646 9647 9648 9649 9650 9651 9652 9653 9654 9655 9656 9657 9658 9659 9660 9661 9662 9663 9664 9665 9666 9667 9668 9669 9670 9671 9672 9673 9674 9675 9676 9677 9678 9679 9680 9681 9682 9683 9684 9685 9686 9687 9688 9689 9690 9691 9692 9693 9694 9695 9696 9697 9698 9699 9700 9701 9702 9703 9704 9705 9706 9707 9708 9709 9710 9711 9712 9713 9714 9715 9716 9717 9718 9719 9720 9721 9722 9723 9724 9725 9726 9727 9728 9729 9730 9731 9732 9733 9734 9735 9736 9737 9738 9739 9740 9741 9742 9743 9744 9745 9746 9747 9748 9749 9750 9751 9752 9753 9754 9755 9756 9757 9758 9759 9760 9761 9762 9763 9764 9765 9766 9767 9768 9769 9770 9771 9772 9773 9774 9775 9776 9777 9778 9779 9780 9781 9782 9783 9784 9785 9786 9787 9788 9789 9790 9791 9792 9793 9794 9795 9796 9797 9798 9799 9800 9801 9802 9803 9804 9805 9806 9807 9808 9809 9810 9811 9812 9813 9814 9815 9816 9817 9818 9819 9820 9821 9822 9823 9824 9825 9826 9827 9828 9829 9830 9831 9832 9833 9834 9835 9836 9837 9838 9839 9840 9841 9842 9843 9844 9845 9846 9847 9848 9849 9850 9851 9852 9853 9854 9855 9856 9857 9858 9859 9860 9861 9862 9863 9864 9865 9866 9867 9868 9869 9870 9871 9872 9873 9874 9875 9876 9877 9878 9879 9880 9881 9882 9883 9884 9885 9886 9887 9888 9889 9890 9891 9892 9893 9894 9895 9896 9897 9898 9899 9900 9901 9902 9903 9904 9905 9906 9907 9908 9909 9910 9911 9912 9913 9914 9915 9916 9917 9918 9919 9920 9921 9922 9923 9924 9925 9926 9927 9928 9929 9930 9931 9932 9933 9934 9935 9936 9937 9938 9939 9940 9941 9942 9943 9944 9945 9946 9947 9948 9949 9950 9951 9952 9953 9954 9955 9956 9957 9958 9959 9960 9961 9962 9963 9964 9965 9966 9967 9968 9969 9970 9971 9972 9973 9974 9975 9976 9977 9978 9979 9980 9981 9982 9983 9984 9985 9986 9987 9988 9989 9990 9991 9992 9993 9994 9995 9996 9997 9998 9999 10000 10001 10002 10003 10004 10005 10006 10007 10008 10009 10010 10011 10012 10013 10014 10015 10016 10017 10018 10019 10020 10021 10022 10023 10024 10025 10026 10027 10028 10029 10030 10031 10032 10033 10034 10035 10036 10037 10038 10039 10040 10041 10042 10043 10044 10045 10046 10047 10048 10049 10050 10051 10052 10053 10054 10055 10056 10057 10058 10059 10060 10061 10062 10063 10064 10065 10066 10067 10068 10069 10070 10071 10072 10073 10074 10075 10076 10077 10078 10079 10080 10081 10082 10083 10084 10085 10086 10087 10088 10089 10090 10091 10092 10093 10094 10095 10096 10097 10098 10099 10100 10101 10102 10103 10104 10105 10106 10107 10108 10109 10110 10111 10112 10113 10114 10115 10116 10117 10118 10119 10120 10121 10122 10123 10124 10125 10126 10127 10128 10129 10130 10131 10132 10133 10134 10135 10136 10137 10138 10139 10140 10141 10142 10143 10144 10145 10146 10147 10148 10149 10150 10151 10152 10153 10154 10155 10156 10157 10158 10159 10160 10161 10162 10163 10164 10165 10166 10167 10168 10169 10170 10171 10172 10173 10174 10175 10176 10177 10178 10179 10180 10181 10182 10183 10184 10185 10186 10187 10188 10189 10190 10191 10192 10193 10194 10195 10196 10197 10198 10199 10200 10201 10202 10203 10204 10205 10206 10207 10208 10209 10210 10211 10212 10213 10214 10215 10216 10217 10218 10219 10220 10221 10222 10223 10224 10225 10226 10227 10228 10229 10230 10231 10232 10233 10234 10235 10236 10237 10238 10239 10240 10241 10242 10243 10244 10245 10246 10247 10248 10249 10250 10251 10252 10253 10254 10255 10256 10257 10258 10259 10260 10261 10262 10263 10264 10265 10266 10267 10268 10269 10270 10271 10272 10273 10274 10275 10276 10277 10278 10279 10280 10281 10282 10283 10284 10285 10286 10287 10288 10289 10290 10291 10292 10293 10294 10295 10296 10297 10298 10299 10300 10301 10302 10303 10304 10305 10306 10307 10308 10309 10310 10311 10312 10313 10314 10315 10316 10317 10318 10319 10320 10321 10322 10323 10324 10325 10326 10327 10328 10329 10330 10331 10332 10333 10334 10335 10336 10337 10338 10339 10340 10341 10342 10343 10344 10345 10346 10347 10348 10349 10350 10351 10352 10353 10354 10355 10356 10357 10358 10359 10360 10361 10362 10363 10364 10365 10366 10367 10368 10369 10370 10371 10372 10373 10374 10375 10376 10377 10378 10379 10380 10381 10382 10383 10384 10385 10386 10387 10388 10389 10390 10391 10392 10393 10394 10395 10396 10397 10398 10399 10400 10401 10402 10403 10404 10405 10406 10407 10408 10409 10410 10411 10412 10413 10414 10415 10416 10417 10418 10419 10420 10421 10422 10423 10424 10425 10426 10427 10428 10429 10430 10431 10432 10433 10434 10435 10436 10437 10438 10439 10440 10441 10442 10443 10444 10445 10446 10447 10448 10449 10450 10451 10452 10453 10454 10455 10456 10457 10458 10459 10460 10461 10462 10463 10464 10465 10466 10467 10468 10469 10470 10471 10472 10473 10474 10475 10476 10477 10478 10479 10480 10481 10482 10483 10484 10485 10486 10487 10488 10489 10490 10491 10492 10493 10494 10495 10496 10497 10498 10499 10500 10501 10502 10503 10504 10505 10506 10507 10508 10509 10510 10511 10512 10513 10514 10515 10516 10517 10518 10519 10520 10521 10522 10523 10524 10525 10526 10527 10528 10529 10530 10531 10532 10533 10534 10535 10536 10537 10538 10539 10540 10541 10542 10543 10544 10545 10546 10547 10548 10549 10550 10551 10552 10553 10554 10555 10556 10557 10558 10559 10560 10561 10562 10563 10564 10565 10566 10567 10568 10569 10570 10571 10572 10573 10574 10575 10576 10577 10578 10579 10580 10581 10582 10583 10584 10585 10586 10587 10588 10589 10590 10591 10592 10593 10594 10595 10596 10597 10598 10599 10600 10601 10602 10603 10604 10605 10606 10607 10608 10609 10610 10611 10612 10613 10614 10615 10616 10617 10618 10619 10620 10621 10622 10623 10624 10625 10626 10627 10628 10629 10630 10631 10632 10633 10634 10635 10636 10637 10638 10639 10640 10641 10642 10643 10644 10645 10646 10647 10648 10649 10650 10651 10652 10653 10654 10655 10656 10657 10658 10659 10660 10661 10662 10663 10664 10665 10666 10667 10668 10669 10670 10671 10672 10673 10674 10675 10676 10677 10678 10679 10680 10681 10682 10683 10684 10685 10686 10687 10688 10689 10690 10691 10692 10693 10694 10695 10696 10697 10698 10699 10700 10701 10702 10703 10704 10705 10706 10707 10708 10709 10710 10711 10712 10713 10714 10715 10716 10717 10718 10719 10720 10721 10722 10723 10724 10725 10726 10727 10728 10729 10730 10731 10732 10733 10734 10735 10736 10737 10738 10739 10740 10741 10742 10743 10744 10745 10746 10747 10748 10749 10750 10751 10752 10753 10754 10755 10756 10757 10758 10759 10760 10761 10762 10763 10764 10765 10766 10767 10768 10769 10770 10771 10772 10773 10774 10775 10776 10777 10778 10779 10780 10781 10782 10783 10784 10785 10786 10787 10788 10789 10790 10791 10792 10793 10794 10795 10796 10797 10798 10799 10800 10801 10802 10803 10804 10805 10806 10807 10808 10809 10810 10811 10812 10813 10814 10815 10816 
\documentclass[b5paper]{book}
\usepackage{hyperref}
\usepackage{makeidx}
\usepackage{amssymb}
\usepackage{color}
\usepackage{alltt}
\usepackage{graphicx}
\usepackage{layout}
\def\union{\cup}
\def\intersect{\cap}
\def\getsrandom{\stackrel{\rm R}{\gets}}
\def\cross{\times}
\def\cat{\hspace{0.5em} \| \hspace{0.5em}}
\def\catn{$\|$}
\def\divides{\hspace{0.3em} | \hspace{0.3em}}
\def\nequiv{\not\equiv}
\def\approx{\raisebox{0.2ex}{\mbox{\small $\sim$}}}
\def\lcm{{\rm lcm}}
\def\gcd{{\rm gcd}}
\def\log{{\rm log}}
\def\ord{{\rm ord}}
\def\abs{{\mathit abs}}
\def\rep{{\mathit rep}}
\def\mod{{\mathit\ mod\ }}
\renewcommand{\pmod}[1]{\ ({\rm mod\ }{#1})}
\newcommand{\floor}[1]{\left\lfloor{#1}\right\rfloor}
\newcommand{\ceil}[1]{\left\lceil{#1}\right\rceil}
\def\Or{{\rm\ or\ }}
\def\And{{\rm\ and\ }}
\def\iff{\hspace{1em}\Longleftrightarrow\hspace{1em}}
\def\implies{\Rightarrow}
\def\undefined{{\rm ``undefined"}}
\def\Proof{\vspace{1ex}\noindent {\bf Proof:}\hspace{1em}}
\let\oldphi\phi
\def\phi{\varphi}
\def\Pr{{\rm Pr}}
\newcommand{\str}[1]{{\mathbf{#1}}}
\def\F{{\mathbb F}}
\def\N{{\mathbb N}}
\def\Z{{\mathbb Z}}
\def\R{{\mathbb R}}
\def\C{{\mathbb C}}
\def\Q{{\mathbb Q}}
\definecolor{DGray}{gray}{0.5}
\newcommand{\emailaddr}[1]{\mbox{$<${#1}$>$}}
\def\twiddle{\raisebox{0.3ex}{\mbox{\tiny $\sim$}}}
\def\gap{\vspace{0.5ex}}
\makeindex
\begin{document}
\frontmatter
\pagestyle{empty}
\title{Multi--Precision Math}
\author{\mbox{
%\begin{small}
\begin{tabular}{c}
Tom St Denis \\
Algonquin College \\
\\
Mads Rasmussen \\
Open Communications Security \\
\\
Greg Rose \\
QUALCOMM Australia \\
\end{tabular}
%\end{small}
}
}
\maketitle
This text has been placed in the public domain.  This text corresponds to the v0.39 release of the
LibTomMath project.

This text is formatted to the international B5 paper size of 176mm wide by 250mm tall using the \LaTeX{}
{\em book} macro package and the Perl {\em booker} package.

\tableofcontents
\listoffigures
\chapter*{Prefaces}
When I tell people about my LibTom projects and that I release them as public domain they are often puzzled.
They ask why I did it and especially why I continue to work on them for free.  The best I can explain it is ``Because I can.''
Which seems odd and perhaps too terse for adult conversation. I often qualify it with ``I am able, I am willing.'' which
perhaps explains it better.  I am the first to admit there is not anything that special with what I have done.  Perhaps
others can see that too and then we would have a society to be proud of.  My LibTom projects are what I am doing to give
back to society in the form of tools and knowledge that can help others in their endeavours.

I started writing this book because it was the most logical task to further my goal of open academia.  The LibTomMath source
code itself was written to be easy to follow and learn from.  There are times, however, where pure C source code does not
explain the algorithms properly.  Hence this book.  The book literally starts with the foundation of the library and works
itself outwards to the more complicated algorithms.  The use of both pseudo--code and verbatim source code provides a duality
of ``theory'' and ``practice'' that the computer science students of the world shall appreciate.  I never deviate too far
from relatively straightforward algebra and I hope that this book can be a valuable learning asset.

This book and indeed much of the LibTom projects would not exist in their current form if it was not for a plethora
of kind people donating their time, resources and kind words to help support my work.  Writing a text of significant
length (along with the source code) is a tiresome and lengthy process.  Currently the LibTom project is four years old,
comprises of literally thousands of users and over 100,000 lines of source code, TeX and other material.  People like Mads and Greg
were there at the beginning to encourage me to work well.  It is amazing how timely validation from others can boost morale to
continue the project. Definitely my parents were there for me by providing room and board during the many months of work in 2003.

To my many friends whom I have met through the years I thank you for the good times and the words of encouragement.  I hope I
honour your kind gestures with this project.

Open Source.  Open Academia.  Open Minds.

\begin{flushright} Tom St Denis \end{flushright}

\newpage
I found the opportunity to work with Tom appealing for several reasons, not only could I broaden my own horizons, but also
contribute to educate others facing the problem of having to handle big number mathematical calculations.

This book is Tom's child and he has been caring and fostering the project ever since the beginning with a clear mind of
how he wanted the project to turn out. I have helped by proofreading the text and we have had several discussions about
the layout and language used.

I hold a masters degree in cryptography from the University of Southern Denmark and have always been interested in the
practical aspects of cryptography.

Having worked in the security consultancy business for several years in S\~{a}o Paulo, Brazil, I have been in touch with a
great deal of work in which multiple precision mathematics was needed. Understanding the possibilities for speeding up
multiple precision calculations is often very important since we deal with outdated machine architecture where modular
reductions, for example, become painfully slow.

This text is for people who stop and wonder when first examining algorithms such as RSA for the first time and asks
themselves, ``You tell me this is only secure for large numbers, fine; but how do you implement these numbers?''

\begin{flushright}
Mads Rasmussen

S\~{a}o Paulo - SP

Brazil
\end{flushright}

\newpage
It's all because I broke my leg. That just happened to be at about the same time that Tom asked for someone to review the section of the book about
Karatsuba multiplication. I was laid up, alone and immobile, and thought ``Why not?'' I vaguely knew what Karatsuba multiplication was, but not
really, so I thought I could help, learn, and stop myself from watching daytime cable TV, all at once.

At the time of writing this, I've still not met Tom or Mads in meatspace. I've been following Tom's progress since his first splash on the
sci.crypt Usenet news group. I watched him go from a clueless newbie, to the cryptographic equivalent of a reformed smoker, to a real
contributor to the field, over a period of about two years. I've been impressed with his obvious intelligence, and astounded by his productivity.
Of course, he's young enough to be my own child, so he doesn't have my problems with staying awake.

When I reviewed that single section of the book, in its very earliest form, I was very pleasantly surprised. So I decided to collaborate more fully,
and at least review all of it, and perhaps write some bits too. There's still a long way to go with it, and I have watched a number of close
friends go through the mill of publication, so I think that the way to go is longer than Tom thinks it is. Nevertheless, it's a good effort,
and I'm pleased to be involved with it.

\begin{flushright}
Greg Rose, Sydney, Australia, June 2003.
\end{flushright}

\mainmatter
\pagestyle{headings}
\chapter{Introduction}
\section{Multiple Precision Arithmetic}

\subsection{What is Multiple Precision Arithmetic?}
When we think of long-hand arithmetic such as addition or multiplication we rarely consider the fact that we instinctively
raise or lower the precision of the numbers we are dealing with.  For example, in decimal we almost immediate can
reason that $7$ times $6$ is $42$.  However, $42$ has two digits of precision as opposed to one digit we started with.
Further multiplications of say $3$ result in a larger precision result $126$.  In these few examples we have multiple
precisions for the numbers we are working with.  Despite the various levels of precision a single subset\footnote{With the occasional optimization.}
 of algorithms can be designed to accomodate them.

By way of comparison a fixed or single precision operation would lose precision on various operations.  For example, in
the decimal system with fixed precision $6 \cdot 7 = 2$.

Essentially at the heart of computer based multiple precision arithmetic are the same long-hand algorithms taught in
schools to manually add, subtract, multiply and divide.

\subsection{The Need for Multiple Precision Arithmetic}
The most prevalent need for multiple precision arithmetic, often referred to as ``bignum'' math, is within the implementation
of public-key cryptography algorithms.   Algorithms such as RSA \cite{RSAREF} and Diffie-Hellman \cite{DHREF} require
integers of significant magnitude to resist known cryptanalytic attacks.  For example, at the time of this writing a
typical RSA modulus would be at least greater than $10^{309}$.  However, modern programming languages such as ISO C \cite{ISOC} and
Java \cite{JAVA} only provide instrinsic support for integers which are relatively small and single precision.

\begin{figure}[!here]
\begin{center}
\begin{tabular}{|r|c|}
\hline \textbf{Data Type} & \textbf{Range} \\
\hline char  & $-128 \ldots 127$ \\
\hline short & $-32768 \ldots 32767$ \\
\hline long  & $-2147483648 \ldots 2147483647$ \\
\hline long long & $-9223372036854775808 \ldots 9223372036854775807$ \\
\hline
\end{tabular}
\end{center}
\caption{Typical Data Types for the C Programming Language}
\label{fig:ISOC}
\end{figure}

The largest data type guaranteed to be provided by the ISO C programming
language\footnote{As per the ISO C standard.  However, each compiler vendor is allowed to augment the precision as they
see fit.}  can only represent values up to $10^{19}$ as shown in figure \ref{fig:ISOC}. On its own the C language is
insufficient to accomodate the magnitude required for the problem at hand.  An RSA modulus of magnitude $10^{19}$ could be
trivially factored\footnote{A Pollard-Rho factoring would take only $2^{16}$ time.} on the average desktop computer,
rendering any protocol based on the algorithm insecure.  Multiple precision algorithms solve this very problem by
extending the range of representable integers while using single precision data types.

Most advancements in fast multiple precision arithmetic stem from the need for faster and more efficient cryptographic
primitives.  Faster modular reduction and exponentiation algorithms such as Barrett's algorithm, which have appeared in
various cryptographic journals, can render algorithms such as RSA and Diffie-Hellman more efficient.  In fact, several
major companies such as RSA Security, Certicom and Entrust have built entire product lines on the implementation and
deployment of efficient algorithms.

However, cryptography is not the only field of study that can benefit from fast multiple precision integer routines.
Another auxiliary use of multiple precision integers is high precision floating point data types.
The basic IEEE \cite{IEEE} standard floating point type is made up of an integer mantissa $q$, an exponent $e$ and a sign bit $s$.
Numbers are given in the form $n = q \cdot b^e \cdot -1^s$ where $b = 2$ is the most common base for IEEE.  Since IEEE
floating point is meant to be implemented in hardware the precision of the mantissa is often fairly small
(\textit{23, 48 and 64 bits}).  The mantissa is merely an integer and a multiple precision integer could be used to create
a mantissa of much larger precision than hardware alone can efficiently support.  This approach could be useful where
scientific applications must minimize the total output error over long calculations.

Yet another use for large integers is within arithmetic on polynomials of large characteristic (i.e. $GF(p)[x]$ for large $p$).
In fact the library discussed within this text has already been used to form a polynomial basis library\footnote{See \url{http://poly.libtomcrypt.org} for more details.}.

\subsection{Benefits of Multiple Precision Arithmetic}
\index{precision}
The benefit of multiple precision representations over single or fixed precision representations is that
no precision is lost while representing the result of an operation which requires excess precision.  For example,
the product of two $n$-bit integers requires at least $2n$ bits of precision to be represented faithfully.  A multiple
precision algorithm would augment the precision of the destination to accomodate the result while a single precision system
would truncate excess bits to maintain a fixed level of precision.

It is possible to implement algorithms which require large integers with fixed precision algorithms.  For example, elliptic
curve cryptography (\textit{ECC}) is often implemented on smartcards by fixing the precision of the integers to the maximum
size the system will ever need.  Such an approach can lead to vastly simpler algorithms which can accomodate the
integers required even if the host platform cannot natively accomodate them\footnote{For example, the average smartcard
processor has an 8 bit accumulator.}.  However, as efficient as such an approach may be, the resulting source code is not
normally very flexible.  It cannot, at runtime, accomodate inputs of higher magnitude than the designer anticipated.

Multiple precision algorithms have the most overhead of any style of arithmetic.  For the the most part the
overhead can be kept to a minimum with careful planning, but overall, it is not well suited for most memory starved
platforms.  However, multiple precision algorithms do offer the most flexibility in terms of the magnitude of the
inputs.  That is, the same algorithms based on multiple precision integers can accomodate any reasonable size input
without the designer's explicit forethought.  This leads to lower cost of ownership for the code as it only has to
be written and tested once.

\section{Purpose of This Text}
The purpose of this text is to instruct the reader regarding how to implement efficient multiple precision algorithms.
That is to not only explain a limited subset of the core theory behind the algorithms but also the various ``house keeping''
elements that are neglected by authors of other texts on the subject.  Several well reknowned texts \cite{TAOCPV2,HAC}
give considerably detailed explanations of the theoretical aspects of algorithms and often very little information
regarding the practical implementation aspects.

In most cases how an algorithm is explained and how it is actually implemented are two very different concepts.  For
example, the Handbook of Applied Cryptography (\textit{HAC}), algorithm 14.7 on page 594, gives a relatively simple
algorithm for performing multiple precision integer addition.  However, the description lacks any discussion concerning
the fact that the two integer inputs may be of differing magnitudes.  As a result the implementation is not as simple
as the text would lead people to believe.  Similarly the division routine (\textit{algorithm 14.20, pp. 598}) does not
discuss how to handle sign or handle the dividend's decreasing magnitude in the main loop (\textit{step \#3}).

Both texts also do not discuss several key optimal algorithms required such as ``Comba'' and Karatsuba multipliers
and fast modular inversion, which we consider practical oversights.  These optimal algorithms are vital to achieve
any form of useful performance in non-trivial applications.

To solve this problem the focus of this text is on the practical aspects of implementing a multiple precision integer
package.  As a case study the ``LibTomMath''\footnote{Available at \url{http://math.libtomcrypt.com}} package is used
to demonstrate algorithms with real implementations\footnote{In the ISO C programming language.} that have been field
tested and work very well.  The LibTomMath library is freely available on the Internet for all uses and this text
discusses a very large portion of the inner workings of the library.

The algorithms that are presented will always include at least one ``pseudo-code'' description followed
by the actual C source code that implements the algorithm.  The pseudo-code can be used to implement the same
algorithm in other programming languages as the reader sees fit.

This text shall also serve as a walkthrough of the creation of multiple precision algorithms from scratch.  Showing
the reader how the algorithms fit together as well as where to start on various taskings.

\section{Discussion and Notation}
\subsection{Notation}
A multiple precision integer of $n$-digits shall be denoted as $x = (x_{n-1}, \ldots, x_1, x_0)_{ \beta }$ and represent
the integer $x \equiv \sum_{i=0}^{n-1} x_i\beta^i$.  The elements of the array $x$ are said to be the radix $\beta$ digits
of the integer.  For example, $x = (1,2,3)_{10}$ would represent the integer
$1\cdot 10^2 + 2\cdot10^1 + 3\cdot10^0 = 123$.

\index{mp\_int}
The term ``mp\_int'' shall refer to a composite structure which contains the digits of the integer it represents, as well
as auxilary data required to manipulate the data.  These additional members are discussed further in section
\ref{sec:MPINT}.  For the purposes of this text a ``multiple precision integer'' and an ``mp\_int'' are assumed to be
synonymous.  When an algorithm is specified to accept an mp\_int variable it is assumed the various auxliary data members
are present as well.  An expression of the type \textit{variablename.item} implies that it should evaluate to the
member named ``item'' of the variable.  For example, a string of characters may have a member ``length'' which would
evaluate to the number of characters in the string.  If the string $a$ equals ``hello'' then it follows that
$a.length = 5$.

For certain discussions more generic algorithms are presented to help the reader understand the final algorithm used
to solve a given problem.  When an algorithm is described as accepting an integer input it is assumed the input is
a plain integer with no additional multiple-precision members.  That is, algorithms that use integers as opposed to
mp\_ints as inputs do not concern themselves with the housekeeping operations required such as memory management.  These
algorithms will be used to establish the relevant theory which will subsequently be used to describe a multiple
precision algorithm to solve the same problem.

\subsection{Precision Notation}
The variable $\beta$ represents the radix of a single digit of a multiple precision integer and
must be of the form $q^p$ for $q, p \in \Z^+$.  A single precision variable must be able to represent integers in
the range $0 \le x < q \beta$ while a double precision variable must be able to represent integers in the range
$0 \le x < q \beta^2$.  The extra radix-$q$ factor allows additions and subtractions to proceed without truncation of the
carry.  Since all modern computers are binary, it is assumed that $q$ is two.

\index{mp\_digit} \index{mp\_word}
Within the source code that will be presented for each algorithm, the data type \textbf{mp\_digit} will represent
a single precision integer type, while, the data type \textbf{mp\_word} will represent a double precision integer type.  In
several algorithms (notably the Comba routines) temporary results will be stored in arrays of double precision mp\_words.
For the purposes of this text $x_j$ will refer to the $j$'th digit of a single precision array and $\hat x_j$ will refer to
the $j$'th digit of a double precision array.  Whenever an expression is to be assigned to a double precision
variable it is assumed that all single precision variables are promoted to double precision during the evaluation.
Expressions that are assigned to a single precision variable are truncated to fit within the precision of a single
precision data type.

For example, if $\beta = 10^2$ a single precision data type may represent a value in the
range $0 \le x < 10^3$, while a double precision data type may represent a value in the range $0 \le x < 10^5$.  Let
$a = 23$ and $b = 49$ represent two single precision variables.  The single precision product shall be written
as $c \leftarrow a \cdot b$ while the double precision product shall be written as $\hat c \leftarrow a \cdot b$.
In this particular case, $\hat c = 1127$ and $c = 127$.  The most significant digit of the product would not fit
in a single precision data type and as a result $c \ne \hat c$.

\subsection{Algorithm Inputs and Outputs}
Within the algorithm descriptions all variables are assumed to be scalars of either single or double precision
as indicated.  The only exception to this rule is when variables have been indicated to be of type mp\_int.  This
distinction is important as scalars are often used as array indicies and various other counters.

\subsection{Mathematical Expressions}
The $\lfloor \mbox{ } \rfloor$ brackets imply an expression truncated to an integer not greater than the expression
itself.  For example, $\lfloor 5.7 \rfloor = 5$.  Similarly the $\lceil \mbox{ } \rceil$ brackets imply an expression
rounded to an integer not less than the expression itself.  For example, $\lceil 5.1 \rceil = 6$.  Typically when
the $/$ division symbol is used the intention is to perform an integer division with truncation.  For example,
$5/2 = 2$ which will often be written as $\lfloor 5/2 \rfloor = 2$ for clarity.  When an expression is written as a
fraction a real value division is implied, for example ${5 \over 2} = 2.5$.

The norm of a multiple precision integer, for example $\vert \vert x \vert \vert$, will be used to represent the number of digits in the representation
of the integer.  For example, $\vert \vert 123 \vert \vert = 3$ and $\vert \vert 79452 \vert \vert = 5$.

\subsection{Work Effort}
\index{big-Oh}
To measure the efficiency of the specified algorithms, a modified big-Oh notation is used.  In this system all
single precision operations are considered to have the same cost\footnote{Except where explicitly noted.}.
That is a single precision addition, multiplication and division are assumed to take the same time to
complete.  While this is generally not true in practice, it will simplify the discussions considerably.

Some algorithms have slight advantages over others which is why some constants will not be removed in
the notation.  For example, a normal baseline multiplication (section \ref{sec:basemult}) requires $O(n^2)$ work while a
baseline squaring (section \ref{sec:basesquare}) requires $O({{n^2 + n}\over 2})$ work.  In standard big-Oh notation these
would both be said to be equivalent to $O(n^2)$.  However,
in the context of the this text this is not the case as the magnitude of the inputs will typically be rather small.  As a
result small constant factors in the work effort will make an observable difference in algorithm efficiency.

All of the algorithms presented in this text have a polynomial time work level.  That is, of the form
$O(n^k)$ for $n, k \in \Z^{+}$.  This will help make useful comparisons in terms of the speed of the algorithms and how
various optimizations will help pay off in the long run.

\section{Exercises}
Within the more advanced chapters a section will be set aside to give the reader some challenging exercises related to
the discussion at hand.  These exercises are not designed to be prize winning problems, but instead to be thought
provoking.  Wherever possible the problems are forward minded, stating problems that will be answered in subsequent
chapters.  The reader is encouraged to finish the exercises as they appear to get a better understanding of the
subject material.

That being said, the problems are designed to affirm knowledge of a particular subject matter.  Students in particular
are encouraged to verify they can answer the problems correctly before moving on.

Similar to the exercises of \cite[pp. ix]{TAOCPV2} these exercises are given a scoring system based on the difficulty of
the problem.  However, unlike \cite{TAOCPV2} the problems do not get nearly as hard.  The scoring of these
exercises ranges from one (the easiest) to five (the hardest).  The following table sumarizes the
scoring system used.

\begin{figure}[here]
\begin{center}
\begin{small}
\begin{tabular}{|c|l|}
\hline $\left [ 1 \right ]$ & An easy problem that should only take the reader a manner of \\
                            & minutes to solve.  Usually does not involve much computer time \\
                            & to solve. \\
\hline $\left [ 2 \right ]$ & An easy problem that involves a marginal amount of computer \\
                     & time usage.  Usually requires a program to be written to \\
                     & solve the problem. \\
\hline $\left [ 3 \right ]$ & A moderately hard problem that requires a non-trivial amount \\
                     & of work.  Usually involves trivial research and development of \\
                     & new theory from the perspective of a student. \\
\hline $\left [ 4 \right ]$ & A moderately hard problem that involves a non-trivial amount \\
                     & of work and research, the solution to which will demonstrate \\
                     & a higher mastery of the subject matter. \\
\hline $\left [ 5 \right ]$ & A hard problem that involves concepts that are difficult for a \\
                     & novice to solve.  Solutions to these problems will demonstrate a \\
                     & complete mastery of the given subject. \\
\hline
\end{tabular}
\end{small}
\end{center}
\caption{Exercise Scoring System}
\end{figure}

Problems at the first level are meant to be simple questions that the reader can answer quickly without programming a solution or
devising new theory.  These problems are quick tests to see if the material is understood.  Problems at the second level
are also designed to be easy but will require a program or algorithm to be implemented to arrive at the answer.  These
two levels are essentially entry level questions.

Problems at the third level are meant to be a bit more difficult than the first two levels.  The answer is often
fairly obvious but arriving at an exacting solution requires some thought and skill.  These problems will almost always
involve devising a new algorithm or implementing a variation of another algorithm previously presented.  Readers who can
answer these questions will feel comfortable with the concepts behind the topic at hand.

Problems at the fourth level are meant to be similar to those of the level three questions except they will require
additional research to be completed.  The reader will most likely not know the answer right away, nor will the text provide
the exact details of the answer until a subsequent chapter.

Problems at the fifth level are meant to be the hardest
problems relative to all the other problems in the chapter.  People who can correctly answer fifth level problems have a
mastery of the subject matter at hand.

Often problems will be tied together.  The purpose of this is to start a chain of thought that will be discussed in future chapters.  The reader
is encouraged to answer the follow-up problems and try to draw the relevance of problems.

\section{Introduction to LibTomMath}

\subsection{What is LibTomMath?}
LibTomMath is a free and open source multiple precision integer library written entirely in portable ISO C.  By portable it
is meant that the library does not contain any code that is computer platform dependent or otherwise problematic to use on
any given platform.

The library has been successfully tested under numerous operating systems including Unix\footnote{All of these
trademarks belong to their respective rightful owners.}, MacOS, Windows, Linux, PalmOS and on standalone hardware such
as the Gameboy Advance.  The library is designed to contain enough functionality to be able to develop applications such
as public key cryptosystems and still maintain a relatively small footprint.

\subsection{Goals of LibTomMath}

Libraries which obtain the most efficiency are rarely written in a high level programming language such as C.  However,
even though this library is written entirely in ISO C, considerable care has been taken to optimize the algorithm implementations within the
library.  Specifically the code has been written to work well with the GNU C Compiler (\textit{GCC}) on both x86 and ARM
processors.  Wherever possible, highly efficient algorithms, such as Karatsuba multiplication, sliding window
exponentiation and Montgomery reduction have been provided to make the library more efficient.

Even with the nearly optimal and specialized algorithms that have been included the Application Programing Interface
(\textit{API}) has been kept as simple as possible.  Often generic place holder routines will make use of specialized
algorithms automatically without the developer's specific attention.  One such example is the generic multiplication
algorithm \textbf{mp\_mul()} which will automatically use Toom--Cook, Karatsuba, Comba or baseline multiplication
based on the magnitude of the inputs and the configuration of the library.

Making LibTomMath as efficient as possible is not the only goal of the LibTomMath project.  Ideally the library should
be source compatible with another popular library which makes it more attractive for developers to use.  In this case the
MPI library was used as a API template for all the basic functions.  MPI was chosen because it is another library that fits
in the same niche as LibTomMath.  Even though LibTomMath uses MPI as the template for the function names and argument
passing conventions, it has been written from scratch by Tom St Denis.

The project is also meant to act as a learning tool for students, the logic being that no easy-to-follow ``bignum''
library exists which can be used to teach computer science students how to perform fast and reliable multiple precision
integer arithmetic.  To this end the source code has been given quite a few comments and algorithm discussion points.

\section{Choice of LibTomMath}
LibTomMath was chosen as the case study of this text not only because the author of both projects is one and the same but
for more worthy reasons.  Other libraries such as GMP \cite{GMP}, MPI \cite{MPI}, LIP \cite{LIP} and OpenSSL
\cite{OPENSSL} have multiple precision integer arithmetic routines but would not be ideal for this text for
reasons that will be explained in the following sub-sections.

\subsection{Code Base}
The LibTomMath code base is all portable ISO C source code.  This means that there are no platform dependent conditional
segments of code littered throughout the source.  This clean and uncluttered approach to the library means that a
developer can more readily discern the true intent of a given section of source code without trying to keep track of
what conditional code will be used.

The code base of LibTomMath is well organized.  Each function is in its own separate source code file
which allows the reader to find a given function very quickly.  On average there are $76$ lines of code per source
file which makes the source very easily to follow.  By comparison MPI and LIP are single file projects making code tracing
very hard.  GMP has many conditional code segments which also hinder tracing.

When compiled with GCC for the x86 processor and optimized for speed the entire library is approximately $100$KiB\footnote{The notation ``KiB'' means $2^{10}$ octets, similarly ``MiB'' means $2^{20}$ octets.}
 which is fairly small compared to GMP (over $250$KiB).  LibTomMath is slightly larger than MPI (which compiles to about
$50$KiB) but LibTomMath is also much faster and more complete than MPI.

\subsection{API Simplicity}
LibTomMath is designed after the MPI library and shares the API design.  Quite often programs that use MPI will build
with LibTomMath without change. The function names correlate directly to the action they perform.  Almost all of the
functions share the same parameter passing convention.  The learning curve is fairly shallow with the API provided
which is an extremely valuable benefit for the student and developer alike.

The LIP library is an example of a library with an API that is awkward to work with.  LIP uses function names that are often ``compressed'' to
illegible short hand.  LibTomMath does not share this characteristic.

The GMP library also does not return error codes.  Instead it uses a POSIX.1 \cite{POSIX1} signal system where errors
are signaled to the host application.  This happens to be the fastest approach but definitely not the most versatile.  In
effect a math error (i.e. invalid input, heap error, etc) can cause a program to stop functioning which is definitely
undersireable in many situations.

\subsection{Optimizations}
While LibTomMath is certainly not the fastest library (GMP often beats LibTomMath by a factor of two) it does
feature a set of optimal algorithms for tasks such as modular reduction, exponentiation, multiplication and squaring.  GMP
and LIP also feature such optimizations while MPI only uses baseline algorithms with no optimizations.  GMP lacks a few
of the additional modular reduction optimizations that LibTomMath features\footnote{At the time of this writing GMP
only had Barrett and Montgomery modular reduction algorithms.}.

LibTomMath is almost always an order of magnitude faster than the MPI library at computationally expensive tasks such as modular
exponentiation.  In the grand scheme of ``bignum'' libraries LibTomMath is faster than the average library and usually
slower than the best libraries such as GMP and OpenSSL by only a small factor.

\subsection{Portability and Stability}
LibTomMath will build ``out of the box'' on any platform equipped with a modern version of the GNU C Compiler
(\textit{GCC}).  This means that without changes the library will build without configuration or setting up any
variables.  LIP and MPI will build ``out of the box'' as well but have numerous known bugs.  Most notably the author of
MPI has recently stopped working on his library and LIP has long since been discontinued.

GMP requires a configuration script to run and will not build out of the box.   GMP and LibTomMath are still in active
development and are very stable across a variety of platforms.

\subsection{Choice}
LibTomMath is a relatively compact, well documented, highly optimized and portable library which seems only natural for
the case study of this text.  Various source files from the LibTomMath project will be included within the text.  However,
the reader is encouraged to download their own copy of the library to actually be able to work with the library.

\chapter{Getting Started}
\section{Library Basics}
The trick to writing any useful library of source code is to build a solid foundation and work outwards from it.  First,
a problem along with allowable solution parameters should be identified and analyzed.  In this particular case the
inability to accomodate multiple precision integers is the problem.  Futhermore, the solution must be written
as portable source code that is reasonably efficient across several different computer platforms.

After a foundation is formed the remainder of the library can be designed and implemented in a hierarchical fashion.
That is, to implement the lowest level dependencies first and work towards the most abstract functions last.  For example,
before implementing a modular exponentiation algorithm one would implement a modular reduction algorithm.
By building outwards from a base foundation instead of using a parallel design methodology the resulting project is
highly modular.  Being highly modular is a desirable property of any project as it often means the resulting product
has a small footprint and updates are easy to perform.

Usually when I start a project I will begin with the header files.  I define the data types I think I will need and
prototype the initial functions that are not dependent on other functions (within the library).  After I
implement these base functions I prototype more dependent functions and implement them.   The process repeats until
I implement all of the functions I require.  For example, in the case of LibTomMath I implemented functions such as
mp\_init() well before I implemented mp\_mul() and even further before I implemented mp\_exptmod().  As an example as to
why this design works note that the Karatsuba and Toom-Cook multipliers were written \textit{after} the
dependent function mp\_exptmod() was written.  Adding the new multiplication algorithms did not require changes to the
mp\_exptmod() function itself and lowered the total cost of ownership (\textit{so to speak}) and of development
for new algorithms.  This methodology allows new algorithms to be tested in a complete framework with relative ease.

\begin{center}
\begin{figure}[here]
\includegraphics{pics/design_process.ps}
\caption{Design Flow of the First Few Original LibTomMath Functions.}
\label{pic:design_process}
\end{figure}
\end{center}

Only after the majority of the functions were in place did I pursue a less hierarchical approach to auditing and optimizing
the source code.  For example, one day I may audit the multipliers and the next day the polynomial basis functions.

It only makes sense to begin the text with the preliminary data types and support algorithms required as well.
This chapter discusses the core algorithms of the library which are the dependents for every other algorithm.

\section{What is a Multiple Precision Integer?}
Recall that most programming languages, in particular ISO C \cite{ISOC}, only have fixed precision data types that on their own cannot
be used to represent values larger than their precision will allow. The purpose of multiple precision algorithms is
to use fixed precision data types to create and manipulate multiple precision integers which may represent values
that are very large.

As a well known analogy, school children are taught how to form numbers larger than nine by prepending more radix ten digits.  In the decimal system
the largest single digit value is $9$.  However, by concatenating digits together larger numbers may be represented.  Newly prepended digits
(\textit{to the left}) are said to be in a different power of ten column.  That is, the number $123$ can be described as having a $1$ in the hundreds
column, $2$ in the tens column and $3$ in the ones column.  Or more formally $123 = 1 \cdot 10^2 + 2 \cdot 10^1 + 3 \cdot 10^0$.  Computer based
multiple precision arithmetic is essentially the same concept.  Larger integers are represented by adjoining fixed
precision computer words with the exception that a different radix is used.

What most people probably do not think about explicitly are the various other attributes that describe a multiple precision
integer.  For example, the integer $154_{10}$ has two immediately obvious properties.  First, the integer is positive,
that is the sign of this particular integer is positive as opposed to negative.  Second, the integer has three digits in
its representation.  There is an additional property that the integer posesses that does not concern pencil-and-paper
arithmetic.  The third property is how many digits placeholders are available to hold the integer.

The human analogy of this third property is ensuring there is enough space on the paper to write the integer.  For example,
if one starts writing a large number too far to the right on a piece of paper they will have to erase it and move left.
Similarly, computer algorithms must maintain strict control over memory usage to ensure that the digits of an integer
will not exceed the allowed boundaries.  These three properties make up what is known as a multiple precision
integer or mp\_int for short.

\subsection{The mp\_int Structure}
\label{sec:MPINT}
The mp\_int structure is the ISO C based manifestation of what represents a multiple precision integer.  The ISO C standard does not provide for
any such data type but it does provide for making composite data types known as structures.  The following is the structure definition
used within LibTomMath.

\index{mp\_int}
\begin{figure}[here]
\begin{center}
\begin{small}
%\begin{verbatim}
\begin{tabular}{|l|}
\hline
typedef struct \{ \\
\hspace{3mm}int used, alloc, sign;\\
\hspace{3mm}mp\_digit *dp;\\
\} \textbf{mp\_int}; \\
\hline
\end{tabular}
%\end{verbatim}
\end{small}
\caption{The mp\_int Structure}
\label{fig:mpint}
\end{center}
\end{figure}

The mp\_int structure (fig. \ref{fig:mpint}) can be broken down as follows.

\begin{enumerate}
\item The \textbf{used} parameter denotes how many digits of the array \textbf{dp} contain the digits used to represent
a given integer.  The \textbf{used} count must be positive (or zero) and may not exceed the \textbf{alloc} count.

\item The \textbf{alloc} parameter denotes how
many digits are available in the array to use by functions before it has to increase in size.  When the \textbf{used} count
of a result would exceed the \textbf{alloc} count all of the algorithms will automatically increase the size of the
array to accommodate the precision of the result.

\item The pointer \textbf{dp} points to a dynamically allocated array of digits that represent the given multiple
precision integer.  It is padded with $(\textbf{alloc} - \textbf{used})$ zero digits.  The array is maintained in a least
significant digit order.  As a pencil and paper analogy the array is organized such that the right most digits are stored
first starting at the location indexed by zero\footnote{In C all arrays begin at zero.} in the array.  For example,
if \textbf{dp} contains $\lbrace a, b, c, \ldots \rbrace$ where \textbf{dp}$_0 = a$, \textbf{dp}$_1 = b$, \textbf{dp}$_2 = c$, $\ldots$ then
it would represent the integer $a + b\beta + c\beta^2 + \ldots$

\index{MP\_ZPOS} \index{MP\_NEG}
\item The \textbf{sign} parameter denotes the sign as either zero/positive (\textbf{MP\_ZPOS}) or negative (\textbf{MP\_NEG}).
\end{enumerate}

\subsubsection{Valid mp\_int Structures}
Several rules are placed on the state of an mp\_int structure and are assumed to be followed for reasons of efficiency.
The only exceptions are when the structure is passed to initialization functions such as mp\_init() and mp\_init\_copy().

\begin{enumerate}
\item The value of \textbf{alloc} may not be less than one.  That is \textbf{dp} always points to a previously allocated
array of digits.
\item The value of \textbf{used} may not exceed \textbf{alloc} and must be greater than or equal to zero.
\item The value of \textbf{used} implies the digit at index $(used - 1)$ of the \textbf{dp} array is non-zero.  That is,
leading zero digits in the most significant positions must be trimmed.
   \begin{enumerate}
   \item Digits in the \textbf{dp} array at and above the \textbf{used} location must be zero.
   \end{enumerate}
\item The value of \textbf{sign} must be \textbf{MP\_ZPOS} if \textbf{used} is zero;
this represents the mp\_int value of zero.
\end{enumerate}

\section{Argument Passing}
A convention of argument passing must be adopted early on in the development of any library.  Making the function
prototypes consistent will help eliminate many headaches in the future as the library grows to significant complexity.
In LibTomMath the multiple precision integer functions accept parameters from left to right as pointers to mp\_int
structures.  That means that the source (input) operands are placed on the left and the destination (output) on the right.
Consider the following examples.

\begin{verbatim}
   mp_mul(&a, &b, &c);   /* c = a * b */
   mp_add(&a, &b, &a);   /* a = a + b */
   mp_sqr(&a, &b);       /* b = a * a */
\end{verbatim}

The left to right order is a fairly natural way to implement the functions since it lets the developer read aloud the
functions and make sense of them.  For example, the first function would read ``multiply a and b and store in c''.

Certain libraries (\textit{LIP by Lenstra for instance}) accept parameters the other way around, to mimic the order
of assignment expressions.  That is, the destination (output) is on the left and arguments (inputs) are on the right.  In
truth, it is entirely a matter of preference.  In the case of LibTomMath the convention from the MPI library has been
adopted.

Another very useful design consideration, provided for in LibTomMath, is whether to allow argument sources to also be a
destination.  For example, the second example (\textit{mp\_add}) adds $a$ to $b$ and stores in $a$.  This is an important
feature to implement since it allows the calling functions to cut down on the number of variables it must maintain.
However, to implement this feature specific care has to be given to ensure the destination is not modified before the
source is fully read.

\section{Return Values}
A well implemented application, no matter what its purpose, should trap as many runtime errors as possible and return them
to the caller.  By catching runtime errors a library can be guaranteed to prevent undefined behaviour.  However, the end
developer can still manage to cause a library to crash.  For example, by passing an invalid pointer an application may
fault by dereferencing memory not owned by the application.

In the case of LibTomMath the only errors that are checked for are related to inappropriate inputs (division by zero for
instance) and memory allocation errors.  It will not check that the mp\_int passed to any function is valid nor
will it check pointers for validity.  Any function that can cause a runtime error will return an error code as an
\textbf{int} data type with one of the following values (fig \ref{fig:errcodes}).

\index{MP\_OKAY} \index{MP\_VAL} \index{MP\_MEM}
\begin{figure}[here]
\begin{center}
\begin{tabular}{|l|l|}
\hline \textbf{Value} & \textbf{Meaning} \\
\hline \textbf{MP\_OKAY} & The function was successful \\
\hline \textbf{MP\_VAL}  & One of the input value(s) was invalid \\
\hline \textbf{MP\_MEM}  & The function ran out of heap memory \\
\hline
\end{tabular}
\end{center}
\caption{LibTomMath Error Codes}
\label{fig:errcodes}
\end{figure}

When an error is detected within a function it should free any memory it allocated, often during the initialization of
temporary mp\_ints, and return as soon as possible.  The goal is to leave the system in the same state it was when the
function was called.  Error checking with this style of API is fairly simple.

\begin{verbatim}
   int err;
   if ((err = mp_add(&a, &b, &c)) != MP_OKAY) {
      printf("Error: %s\n", mp_error_to_string(err));
      exit(EXIT_FAILURE);
   }
\end{verbatim}

The GMP \cite{GMP} library uses C style \textit{signals} to flag errors which is of questionable use.  Not all errors are fatal
and it was not deemed ideal by the author of LibTomMath to force developers to have signal handlers for such cases.

\section{Initialization and Clearing}
The logical starting point when actually writing multiple precision integer functions is the initialization and
clearing of the mp\_int structures.  These two algorithms will be used by the majority of the higher level algorithms.

Given the basic mp\_int structure an initialization routine must first allocate memory to hold the digits of
the integer.  Often it is optimal to allocate a sufficiently large pre-set number of digits even though
the initial integer will represent zero.  If only a single digit were allocated quite a few subsequent re-allocations
would occur when operations are performed on the integers.  There is a tradeoff between how many default digits to allocate
and how many re-allocations are tolerable.  Obviously allocating an excessive amount of digits initially will waste
memory and become unmanageable.

If the memory for the digits has been successfully allocated then the rest of the members of the structure must
be initialized.  Since the initial state of an mp\_int is to represent the zero integer, the allocated digits must be set
to zero.  The \textbf{used} count set to zero and \textbf{sign} set to \textbf{MP\_ZPOS}.

\subsection{Initializing an mp\_int}
An mp\_int is said to be initialized if it is set to a valid, preferably default, state such that all of the members of the
structure are set to valid values.  The mp\_init algorithm will perform such an action.

\index{mp\_init}
\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_init}. \\
\textbf{Input}.   An mp\_int $a$ \\
\textbf{Output}.  Allocate memory and initialize $a$ to a known valid mp\_int state.  \\
\hline \\
1.  Allocate memory for \textbf{MP\_PREC} digits. \\
2.  If the allocation failed return(\textit{MP\_MEM}) \\
3.  for $n$ from $0$ to $MP\_PREC - 1$ do  \\
\hspace{3mm}3.1  $a_n \leftarrow 0$\\
4.  $a.sign \leftarrow MP\_ZPOS$\\
5.  $a.used \leftarrow 0$\\
6.  $a.alloc \leftarrow MP\_PREC$\\
7.  Return(\textit{MP\_OKAY})\\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_init}
\end{figure}

\textbf{Algorithm mp\_init.}
The purpose of this function is to initialize an mp\_int structure so that the rest of the library can properly
manipulte it.  It is assumed that the input may not have had any of its members previously initialized which is certainly
a valid assumption if the input resides on the stack.

Before any of the members such as \textbf{sign}, \textbf{used} or \textbf{alloc} are initialized the memory for
the digits is allocated.  If this fails the function returns before setting any of the other members.  The \textbf{MP\_PREC}
name represents a constant\footnote{Defined in the ``tommath.h'' header file within LibTomMath.}
used to dictate the minimum precision of newly initialized mp\_int integers.  Ideally, it is at least equal to the smallest
precision number you'll be working with.

Allocating a block of digits at first instead of a single digit has the benefit of lowering the number of usually slow
heap operations later functions will have to perform in the future.  If \textbf{MP\_PREC} is set correctly the slack
memory and the number of heap operations will be trivial.

Once the allocation has been made the digits have to be set to zero as well as the \textbf{used}, \textbf{sign} and
\textbf{alloc} members initialized.  This ensures that the mp\_int will always represent the default state of zero regardless
of the original condition of the input.

\textbf{Remark.}
This function introduces the idiosyncrasy that all iterative loops, commonly initiated with the ``for'' keyword, iterate incrementally
when the ``to'' keyword is placed between two expressions.  For example, ``for $a$ from $b$ to $c$ do'' means that
a subsequent expression (or body of expressions) are to be evaluated upto $c - b$ times so long as $b \le c$.  In each
iteration the variable $a$ is substituted for a new integer that lies inclusively between $b$ and $c$.  If $b > c$ occured
the loop would not iterate.  By contrast if the ``downto'' keyword were used in place of ``to'' the loop would iterate
decrementally.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_init.c
\vspace{-3mm}
\begin{alltt}
016
017   /* init a new mp_int */
018   int mp_init (mp_int * a)
019   \{
020     int i;
021
022     /* allocate memory required and clear it */
023     a->dp = OPT_CAST(mp_digit) XMALLOC (sizeof (mp_digit) * MP_PREC);
024     if (a->dp == NULL) \{
025       return MP_MEM;
026     \}
027
028     /* set the digits to zero */
029     for (i = 0; i < MP_PREC; i++) \{
030         a->dp[i] = 0;
031     \}
032
033     /* set the used to zero, allocated digits to the default precision
034      * and sign to positive */
035     a->used  = 0;
036     a->alloc = MP_PREC;
037     a->sign  = MP_ZPOS;
038
039     return MP_OKAY;
040   \}
041   #endif
042
\end{alltt}
\end{small}

One immediate observation of this initializtion function is that it does not return a pointer to a mp\_int structure.  It
is assumed that the caller has already allocated memory for the mp\_int structure, typically on the application stack.  The
call to mp\_init() is used only to initialize the members of the structure to a known default state.

Here we see (line 23) the memory allocation is performed first.  This allows us to exit cleanly and quickly
if there is an error.  If the allocation fails the routine will return \textbf{MP\_MEM} to the caller to indicate there
was a memory error.  The function XMALLOC is what actually allocates the memory.  Technically XMALLOC is not a function
but a macro defined in ``tommath.h``.  By default, XMALLOC will evaluate to malloc() which is the C library's built--in
memory allocation routine.

In order to assure the mp\_int is in a known state the digits must be set to zero.  On most platforms this could have been
accomplished by using calloc() instead of malloc().  However,  to correctly initialize a integer type to a given value in a
portable fashion you have to actually assign the value.  The for loop (line 29) performs this required
operation.

After the memory has been successfully initialized the remainder of the members are initialized
(lines 33 through 34) to their respective default states.  At this point the algorithm has succeeded and
a success code is returned to the calling function.  If this function returns \textbf{MP\_OKAY} it is safe to assume the
mp\_int structure has been properly initialized and is safe to use with other functions within the library.

\subsection{Clearing an mp\_int}
When an mp\_int is no longer required by the application, the memory that has been allocated for its digits must be
returned to the application's memory pool with the mp\_clear algorithm.

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_clear}. \\
\textbf{Input}.   An mp\_int $a$ \\
\textbf{Output}.  The memory for $a$ shall be deallocated.  \\
\hline \\
1.  If $a$ has been previously freed then return(\textit{MP\_OKAY}). \\
2.  for $n$ from 0 to $a.used - 1$ do \\
\hspace{3mm}2.1  $a_n \leftarrow 0$ \\
3.  Free the memory allocated for the digits of $a$. \\
4.  $a.used \leftarrow 0$ \\
5.  $a.alloc \leftarrow 0$ \\
6.  $a.sign \leftarrow MP\_ZPOS$ \\
7.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_clear}
\end{figure}

\textbf{Algorithm mp\_clear.}
This algorithm accomplishes two goals.  First, it clears the digits and the other mp\_int members.  This ensures that
if a developer accidentally re-uses a cleared structure it is less likely to cause problems.  The second goal
is to free the allocated memory.

The logic behind the algorithm is extended by marking cleared mp\_int structures so that subsequent calls to this
algorithm will not try to free the memory multiple times.  Cleared mp\_ints are detectable by having a pre-defined invalid
digit pointer \textbf{dp} setting.

Once an mp\_int has been cleared the mp\_int structure is no longer in a valid state for any other algorithm
with the exception of algorithms mp\_init, mp\_init\_copy, mp\_init\_size and mp\_clear.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_clear.c
\vspace{-3mm}
\begin{alltt}
016
017   /* clear one (frees)  */
018   void
019   mp_clear (mp_int * a)
020   \{
021     int i;
022
023     /* only do anything if a hasn't been freed previously */
024     if (a->dp != NULL) \{
025       /* first zero the digits */
026       for (i = 0; i < a->used; i++) \{
027           a->dp[i] = 0;
028       \}
029
030       /* free ram */
031       XFREE(a->dp);
032
033       /* reset members to make debugging easier */
034       a->dp    = NULL;
035       a->alloc = a->used = 0;
036       a->sign  = MP_ZPOS;
037     \}
038   \}
039   #endif
040
\end{alltt}
\end{small}

The algorithm only operates on the mp\_int if it hasn't been previously cleared.  The if statement (line 24)
checks to see if the \textbf{dp} member is not \textbf{NULL}.  If the mp\_int is a valid mp\_int then \textbf{dp} cannot be
\textbf{NULL} in which case the if statement will evaluate to true.

The digits of the mp\_int are cleared by the for loop (line 26) which assigns a zero to every digit.  Similar to mp\_init()
the digits are assigned zero instead of using block memory operations (such as memset()) since this is more portable.

The digits are deallocated off the heap via the XFREE macro.  Similar to XMALLOC the XFREE macro actually evaluates to
a standard C library function.  In this case the free() function.  Since free() only deallocates the memory the pointer
still has to be reset to \textbf{NULL} manually (line 34).

Now that the digits have been cleared and deallocated the other members are set to their final values (lines 35 and 36).

\section{Maintenance Algorithms}

The previous sections describes how to initialize and clear an mp\_int structure.  To further support operations
that are to be performed on mp\_int structures (such as addition and multiplication) the dependent algorithms must be
able to augment the precision of an mp\_int and
initialize mp\_ints with differing initial conditions.

These algorithms complete the set of low level algorithms required to work with mp\_int structures in the higher level
algorithms such as addition, multiplication and modular exponentiation.

\subsection{Augmenting an mp\_int's Precision}
When storing a value in an mp\_int structure, a sufficient number of digits must be available to accomodate the entire
result of an operation without loss of precision.  Quite often the size of the array given by the \textbf{alloc} member
is large enough to simply increase the \textbf{used} digit count.  However, when the size of the array is too small it
must be re-sized appropriately to accomodate the result.  The mp\_grow algorithm will provide this functionality.

\newpage\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_grow}. \\
\textbf{Input}.   An mp\_int $a$ and an integer $b$. \\
\textbf{Output}.  $a$ is expanded to accomodate $b$ digits. \\
\hline \\
1.  if $a.alloc \ge b$ then return(\textit{MP\_OKAY}) \\
2.  $u \leftarrow b\mbox{ (mod }MP\_PREC\mbox{)}$ \\
3.  $v \leftarrow b + 2 \cdot MP\_PREC - u$ \\
4.  Re-allocate the array of digits $a$ to size $v$ \\
5.  If the allocation failed then return(\textit{MP\_MEM}). \\
6.  for n from a.alloc to $v - 1$ do  \\
\hspace{+3mm}6.1  $a_n \leftarrow 0$ \\
7.  $a.alloc \leftarrow v$ \\
8.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_grow}
\end{figure}

\textbf{Algorithm mp\_grow.}
It is ideal to prevent re-allocations from being performed if they are not required (step one).  This is useful to
prevent mp\_ints from growing excessively in code that erroneously calls mp\_grow.

The requested digit count is padded up to next multiple of \textbf{MP\_PREC} plus an additional \textbf{MP\_PREC} (steps two and three).
This helps prevent many trivial reallocations that would grow an mp\_int by trivially small values.

It is assumed that the reallocation (step four) leaves the lower $a.alloc$ digits of the mp\_int intact.  This is much
akin to how the \textit{realloc} function from the standard C library works.  Since the newly allocated digits are
assumed to contain undefined values they are initially set to zero.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_grow.c
\vspace{-3mm}
\begin{alltt}
016
017   /* grow as required */
018   int mp_grow (mp_int * a, int size)
019   \{
020     int     i;
021     mp_digit *tmp;
022
023     /* if the alloc size is smaller alloc more ram */
024     if (a->alloc < size) \{
025       /* ensure there are always at least MP_PREC digits extra on top */
026       size += (MP_PREC * 2) - (size % MP_PREC);
027
028       /* reallocate the array a->dp
029        *
030        * We store the return in a temporary variable
031        * in case the operation failed we don't want
032        * to overwrite the dp member of a.
033        */
034       tmp = OPT_CAST(mp_digit) XREALLOC (a->dp, sizeof (mp_digit) * size);
035       if (tmp == NULL) \{
036         /* reallocation failed but "a" is still valid [can be freed] */
037         return MP_MEM;
038       \}
039
040       /* reallocation succeeded so set a->dp */
041       a->dp = tmp;
042
043       /* zero excess digits */
044       i        = a->alloc;
045       a->alloc = size;
046       for (; i < a->alloc; i++) \{
047         a->dp[i] = 0;
048       \}
049     \}
050     return MP_OKAY;
051   \}
052   #endif
053
\end{alltt}
\end{small}

A quick optimization is to first determine if a memory re-allocation is required at all.  The if statement (line 24) checks
if the \textbf{alloc} member of the mp\_int is smaller than the requested digit count.  If the count is not larger than \textbf{alloc}
the function skips the re-allocation part thus saving time.

When a re-allocation is performed it is turned into an optimal request to save time in the future.  The requested digit count is
padded upwards to 2nd multiple of \textbf{MP\_PREC} larger than \textbf{alloc} (line 26).  The XREALLOC function is used
to re-allocate the memory.  As per the other functions XREALLOC is actually a macro which evaluates to realloc by default.  The realloc
function leaves the base of the allocation intact which means the first \textbf{alloc} digits of the mp\_int are the same as before
the re-allocation.  All	that is left is to clear the newly allocated digits and return.

Note that the re-allocation result is actually stored in a temporary pointer $tmp$.  This is to allow this function to return
an error with a valid pointer.  Earlier releases of the library stored the result of XREALLOC into the mp\_int $a$.  That would
result in a memory leak if XREALLOC ever failed.

\subsection{Initializing Variable Precision mp\_ints}
Occasionally the number of digits required will be known in advance of an initialization, based on, for example, the size
of input mp\_ints to a given algorithm.  The purpose of algorithm mp\_init\_size is similar to mp\_init except that it
will allocate \textit{at least} a specified number of digits.

\begin{figure}[here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_init\_size}. \\
\textbf{Input}.   An mp\_int $a$ and the requested number of digits $b$. \\
\textbf{Output}.  $a$ is initialized to hold at least $b$ digits. \\
\hline \\
1.  $u \leftarrow b \mbox{ (mod }MP\_PREC\mbox{)}$ \\
2.  $v \leftarrow b + 2 \cdot MP\_PREC - u$ \\
3.  Allocate $v$ digits. \\
4.  for $n$ from $0$ to $v - 1$ do \\
\hspace{3mm}4.1  $a_n \leftarrow 0$ \\
5.  $a.sign \leftarrow MP\_ZPOS$\\
6.  $a.used \leftarrow 0$\\
7.  $a.alloc \leftarrow v$\\
8.  Return(\textit{MP\_OKAY})\\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_init\_size}
\end{figure}

\textbf{Algorithm mp\_init\_size.}
This algorithm will initialize an mp\_int structure $a$ like algorithm mp\_init with the exception that the number of
digits allocated can be controlled by the second input argument $b$.  The input size is padded upwards so it is a
multiple of \textbf{MP\_PREC} plus an additional \textbf{MP\_PREC} digits.  This padding is used to prevent trivial
allocations from becoming a bottleneck in the rest of the algorithms.

Like algorithm mp\_init, the mp\_int structure is initialized to a default state representing the integer zero.  This
particular algorithm is useful if it is known ahead of time the approximate size of the input.  If the approximation is
correct no further memory re-allocations are required to work with the mp\_int.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_init\_size.c
\vspace{-3mm}
\begin{alltt}
016
017   /* init an mp_init for a given size */
018   int mp_init_size (mp_int * a, int size)
019   \{
020     int x;
021
022     /* pad size so there are always extra digits */
023     size += (MP_PREC * 2) - (size % MP_PREC);
024
025     /* alloc mem */
026     a->dp = OPT_CAST(mp_digit) XMALLOC (sizeof (mp_digit) * size);
027     if (a->dp == NULL) \{
028       return MP_MEM;
029     \}
030
031     /* set the members */
032     a->used  = 0;
033     a->alloc = size;
034     a->sign  = MP_ZPOS;
035
036     /* zero the digits */
037     for (x = 0; x < size; x++) \{
038         a->dp[x] = 0;
039     \}
040
041     return MP_OKAY;
042   \}
043   #endif
044
\end{alltt}
\end{small}

The number of digits $b$ requested is padded (line 23) by first augmenting it to the next multiple of
\textbf{MP\_PREC} and then adding \textbf{MP\_PREC} to the result.  If the memory can be successfully allocated the
mp\_int is placed in a default state representing the integer zero.  Otherwise, the error code \textbf{MP\_MEM} will be
returned (line 28).

The digits are allocated with the malloc() function (line 26) and set to zero afterwards (line 37).  The
\textbf{used} count is set to zero, the \textbf{alloc} count set to the padded digit count and the \textbf{sign} flag set
to \textbf{MP\_ZPOS} to achieve a default valid mp\_int state (lines 32, 33 and 34).  If the function
returns succesfully then it is correct to assume that the mp\_int structure is in a valid state for the remainder of the
functions to work with.

\subsection{Multiple Integer Initializations and Clearings}
Occasionally a function will require a series of mp\_int data types to be made available simultaneously.
The purpose of algorithm mp\_init\_multi is to initialize a variable length array of mp\_int structures in a single
statement.  It is essentially a shortcut to multiple initializations.

\newpage\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_init\_multi}. \\
\textbf{Input}.   Variable length array $V_k$ of mp\_int variables of length $k$. \\
\textbf{Output}.  The array is initialized such that each mp\_int of $V_k$ is ready to use. \\
\hline \\
1.  for $n$ from 0 to $k - 1$ do \\
\hspace{+3mm}1.1.  Initialize the mp\_int $V_n$ (\textit{mp\_init}) \\
\hspace{+3mm}1.2.  If initialization failed then do \\
\hspace{+6mm}1.2.1.  for $j$ from $0$ to $n$ do \\
\hspace{+9mm}1.2.1.1.  Free the mp\_int $V_j$ (\textit{mp\_clear}) \\
\hspace{+6mm}1.2.2.   Return(\textit{MP\_MEM}) \\
2.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_init\_multi}
\end{figure}

\textbf{Algorithm mp\_init\_multi.}
The algorithm will initialize the array of mp\_int variables one at a time.  If a runtime error has been detected
(\textit{step 1.2}) all of the previously initialized variables are cleared.  The goal is an ``all or nothing''
initialization which allows for quick recovery from runtime errors.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_init\_multi.c
\vspace{-3mm}
\begin{alltt}
016   #include <stdarg.h>
017
018   int mp_init_multi(mp_int *mp, ...)
019   \{
020       mp_err res = MP_OKAY;      /* Assume ok until proven otherwise */
021       int n = 0;                 /* Number of ok inits */
022       mp_int* cur_arg = mp;
023       va_list args;
024
025       va_start(args, mp);        /* init args to next argument from caller */
026       while (cur_arg != NULL) \{
027           if (mp_init(cur_arg) != MP_OKAY) \{
028               /* Oops - error! Back-track and mp_clear what we already
029                  succeeded in init-ing, then return error.
030               */
031               va_list clean_args;
032
033               /* end the current list */
034               va_end(args);
035
036               /* now start cleaning up */
037               cur_arg = mp;
038               va_start(clean_args, mp);
039               while (n-- != 0) \{
040                   mp_clear(cur_arg);
041                   cur_arg = va_arg(clean_args, mp_int*);
042               \}
043               va_end(clean_args);
044               res = MP_MEM;
045               break;
046           \}
047           n++;
048           cur_arg = va_arg(args, mp_int*);
049       \}
050       va_end(args);
051       return res;                /* Assumed ok, if error flagged above. */
052   \}
053
054   #endif
055
\end{alltt}
\end{small}

This function intializes a variable length list of mp\_int structure pointers.  However, instead of having the mp\_int
structures in an actual C array they are simply passed as arguments to the function.  This function makes use of the
``...'' argument syntax of the C programming language.  The list is terminated with a final \textbf{NULL} argument
appended on the right.

The function uses the ``stdarg.h'' \textit{va} functions to step portably through the arguments to the function.  A count
$n$ of succesfully initialized mp\_int structures is maintained (line 47) such that if a failure does occur,
the algorithm can backtrack and free the previously initialized structures (lines 27 to 46).


\subsection{Clamping Excess Digits}
When a function anticipates a result will be $n$ digits it is simpler to assume this is true within the body of
the function instead of checking during the computation.  For example, a multiplication of a $i$ digit number by a
$j$ digit produces a result of at most $i + j$ digits.  It is entirely possible that the result is $i + j - 1$
though, with no final carry into the last position.  However, suppose the destination had to be first expanded
(\textit{via mp\_grow}) to accomodate $i + j - 1$ digits than further expanded to accomodate the final carry.
That would be a considerable waste of time since heap operations are relatively slow.

The ideal solution is to always assume the result is $i + j$ and fix up the \textbf{used} count after the function
terminates.  This way a single heap operation (\textit{at most}) is required.  However, if the result was not checked
there would be an excess high order zero digit.

For example, suppose the product of two integers was $x_n = (0x_{n-1}x_{n-2}...x_0)_{\beta}$.  The leading zero digit
will not contribute to the precision of the result.  In fact, through subsequent operations more leading zero digits would
accumulate to the point the size of the integer would be prohibitive.  As a result even though the precision is very
low the representation is excessively large.

The mp\_clamp algorithm is designed to solve this very problem.  It will trim high-order zeros by decrementing the
\textbf{used} count until a non-zero most significant digit is found.  Also in this system, zero is considered to be a
positive number which means that if the \textbf{used} count is decremented to zero, the sign must be set to
\textbf{MP\_ZPOS}.

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_clamp}. \\
\textbf{Input}.   An mp\_int $a$ \\
\textbf{Output}.  Any excess leading zero digits of $a$ are removed \\
\hline \\
1.  while $a.used > 0$ and $a_{a.used - 1} = 0$ do \\
\hspace{+3mm}1.1  $a.used \leftarrow a.used - 1$ \\
2.  if $a.used = 0$ then do \\
\hspace{+3mm}2.1  $a.sign \leftarrow MP\_ZPOS$ \\
\hline \\
\end{tabular}
\end{center}
\caption{Algorithm mp\_clamp}
\end{figure}

\textbf{Algorithm mp\_clamp.}
As can be expected this algorithm is very simple.  The loop on step one is expected to iterate only once or twice at
the most.  For example, this will happen in cases where there is not a carry to fill the last position.  Step two fixes the sign for
when all of the digits are zero to ensure that the mp\_int is valid at all times.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_clamp.c
\vspace{-3mm}
\begin{alltt}
016
017   /* trim unused digits
018    *
019    * This is used to ensure that leading zero digits are
020    * trimed and the leading "used" digit will be non-zero
021    * Typically very fast.  Also fixes the sign if there
022    * are no more leading digits
023    */
024   void
025   mp_clamp (mp_int * a)
026   \{
027     /* decrease used while the most significant digit is
028      * zero.
029      */
030     while ((a->used > 0) && (a->dp[a->used - 1] == 0)) \{
031       --(a->used);
032     \}
033
034     /* reset the sign flag if used == 0 */
035     if (a->used == 0) \{
036       a->sign = MP_ZPOS;
037     \}
038   \}
039   #endif
040
\end{alltt}
\end{small}

Note on line 27 how to test for the \textbf{used} count is made on the left of the \&\& operator.  In the C programming
language the terms to \&\& are evaluated left to right with a boolean short-circuit if any condition fails.  This is
important since if the \textbf{used} is zero the test on the right would fetch below the array.  That is obviously
undesirable.  The parenthesis on line 30 is used to make sure the \textbf{used} count is decremented and not
the pointer ``a''.

\section*{Exercises}
\begin{tabular}{cl}
$\left [ 1 \right ]$ & Discuss the relevance of the \textbf{used} member of the mp\_int structure. \\
                     & \\
$\left [ 1 \right ]$ & Discuss the consequences of not using padding when performing allocations.  \\
                     & \\
$\left [ 2 \right ]$ & Estimate an ideal value for \textbf{MP\_PREC} when performing 1024-bit RSA \\
                     & encryption when $\beta = 2^{28}$.  \\
                     & \\
$\left [ 1 \right ]$ & Discuss the relevance of the algorithm mp\_clamp.  What does it prevent? \\
                     & \\
$\left [ 1 \right ]$ & Give an example of when the algorithm  mp\_init\_copy might be useful. \\
                     & \\
\end{tabular}


%%%
% CHAPTER FOUR
%%%

\chapter{Basic Operations}

\section{Introduction}
In the previous chapter a series of low level algorithms were established that dealt with initializing and maintaining
mp\_int structures.  This chapter will discuss another set of seemingly non-algebraic algorithms which will form the low
level basis of the entire library.  While these algorithm are relatively trivial it is important to understand how they
work before proceeding since these algorithms will be used almost intrinsically in the following chapters.

The algorithms in this chapter deal primarily with more ``programmer'' related tasks such as creating copies of
mp\_int structures, assigning small values to mp\_int structures and comparisons of the values mp\_int structures
represent.

\section{Assigning Values to mp\_int Structures}
\subsection{Copying an mp\_int}
Assigning the value that a given mp\_int structure represents to another mp\_int structure shall be known as making
a copy for the purposes of this text.  The copy of the mp\_int will be a separate entity that represents the same
value as the mp\_int it was copied from.  The mp\_copy algorithm provides this functionality.

\newpage\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_copy}. \\
\textbf{Input}.  An mp\_int $a$ and $b$. \\
\textbf{Output}.  Store a copy of $a$ in $b$. \\
\hline \\
1.  If $b.alloc < a.used$ then grow $b$ to $a.used$ digits.  (\textit{mp\_grow}) \\
2.  for $n$ from 0 to $a.used - 1$ do \\
\hspace{3mm}2.1  $b_{n} \leftarrow a_{n}$ \\
3.  for $n$ from $a.used$ to $b.used - 1$ do \\
\hspace{3mm}3.1  $b_{n} \leftarrow 0$ \\
4.  $b.used \leftarrow a.used$ \\
5.  $b.sign \leftarrow a.sign$ \\
6.  return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_copy}
\end{figure}

\textbf{Algorithm mp\_copy.}
This algorithm copies the mp\_int $a$ such that upon succesful termination of the algorithm the mp\_int $b$ will
represent the same integer as the mp\_int $a$.  The mp\_int $b$ shall be a complete and distinct copy of the
mp\_int $a$ meaing that the mp\_int $a$ can be modified and it shall not affect the value of the mp\_int $b$.

If $b$ does not have enough room for the digits of $a$ it must first have its precision augmented via the mp\_grow
algorithm.  The digits of $a$ are copied over the digits of $b$ and any excess digits of $b$ are set to zero (step two
and three).  The \textbf{used} and \textbf{sign} members of $a$ are finally copied over the respective members of
$b$.

\textbf{Remark.}  This algorithm also introduces a new idiosyncrasy that will be used throughout the rest of the
text.  The error return codes of other algorithms are not explicitly checked in the pseudo-code presented.  For example, in
step one of the mp\_copy algorithm the return of mp\_grow is not explicitly checked to ensure it succeeded.  Text space is
limited so it is assumed that if a algorithm fails it will clear all temporarily allocated mp\_ints and return
the error code itself.  However, the C code presented will demonstrate all of the error handling logic required to
implement the pseudo-code.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_copy.c
\vspace{-3mm}
\begin{alltt}
016
017   /* copy, b = a */
018   int
019   mp_copy (mp_int * a, mp_int * b)
020   \{
021     int     res, n;
022
023     /* if dst == src do nothing */
024     if (a == b) \{
025       return MP_OKAY;
026     \}
027
028     /* grow dest */
029     if (b->alloc < a->used) \{
030        if ((res = mp_grow (b, a->used)) != MP_OKAY) \{
031           return res;
032        \}
033     \}
034
035     /* zero b and copy the parameters over */
036     \{
037       mp_digit *tmpa, *tmpb;
038
039       /* pointer aliases */
040
041       /* source */
042       tmpa = a->dp;
043
044       /* destination */
045       tmpb = b->dp;
046
047       /* copy all the digits */
048       for (n = 0; n < a->used; n++) \{
049         *tmpb++ = *tmpa++;
050       \}
051
052       /* clear high digits */
053       for (; n < b->used; n++) \{
054         *tmpb++ = 0;
055       \}
056     \}
057
058     /* copy used count and sign */
059     b->used = a->used;
060     b->sign = a->sign;
061     return MP_OKAY;
062   \}
063   #endif
064
\end{alltt}
\end{small}

Occasionally a dependent algorithm may copy an mp\_int effectively into itself such as when the input and output
mp\_int structures passed to a function are one and the same.  For this case it is optimal to return immediately without
copying digits (line 24).

The mp\_int $b$ must have enough digits to accomodate the used digits of the mp\_int $a$.  If $b.alloc$ is less than
$a.used$ the algorithm mp\_grow is used to augment the precision of $b$ (lines 29 to 33).  In order to
simplify the inner loop that copies the digits from $a$ to $b$, two aliases $tmpa$ and $tmpb$ point directly at the digits
of the mp\_ints $a$ and $b$ respectively.  These aliases (lines 42 and 45) allow the compiler to access the digits without first dereferencing the
mp\_int pointers and then subsequently the pointer to the digits.

After the aliases are established the digits from $a$ are copied into $b$ (lines 48 to 50) and then the excess
digits of $b$ are set to zero (lines 53 to 55).  Both ``for'' loops make use of the pointer aliases and in
fact the alias for $b$ is carried through into the second ``for'' loop to clear the excess digits.  This optimization
allows the alias to stay in a machine register fairly easy between the two loops.

\textbf{Remarks.}  The use of pointer aliases is an implementation methodology first introduced in this function that will
be used considerably in other functions.  Technically, a pointer alias is simply a short hand alias used to lower the
number of pointer dereferencing operations required to access data.  For example, a for loop may resemble

\begin{alltt}
for (x = 0; x < 100; x++) \{
    a->num[4]->dp[x] = 0;
\}
\end{alltt}

This could be re-written using aliases as

\begin{alltt}
mp_digit *tmpa;
a = a->num[4]->dp;
for (x = 0; x < 100; x++) \{
    *a++ = 0;
\}
\end{alltt}

In this case an alias is used to access the
array of digits within an mp\_int structure directly.  It may seem that a pointer alias is strictly not required
as a compiler may optimize out the redundant pointer operations.  However, there are two dominant reasons to use aliases.

The first reason is that most compilers will not effectively optimize pointer arithmetic.  For example, some optimizations
may work for the Microsoft Visual C++ compiler (MSVC) and not for the GNU C Compiler (GCC).  Also some optimizations may
work for GCC and not MSVC.  As such it is ideal to find a common ground for as many compilers as possible.  Pointer
aliases optimize the code considerably before the compiler even reads the source code which means the end compiled code
stands a better chance of being faster.

The second reason is that pointer aliases often can make an algorithm simpler to read.  Consider the first ``for''
loop of the function mp\_copy() re-written to not use pointer aliases.

\begin{alltt}
    /* copy all the digits */
    for (n = 0; n < a->used; n++) \{
      b->dp[n] = a->dp[n];
    \}
\end{alltt}

Whether this code is harder to read depends strongly on the individual.  However, it is quantifiably slightly more
complicated as there are four variables within the statement instead of just two.

\subsubsection{Nested Statements}
Another commonly used technique in the source routines is that certain sections of code are nested.  This is used in
particular with the pointer aliases to highlight code phases.  For example, a Comba multiplier (discussed in chapter six)
will typically have three different phases.  First the temporaries are initialized, then the columns calculated and
finally the carries are propagated.  In this example the middle column production phase will typically be nested as it
uses temporary variables and aliases the most.

The nesting also simplies the source code as variables that are nested are only valid for their scope.  As a result
the various temporary variables required do not propagate into other sections of code.


\subsection{Creating a Clone}
Another common operation is to make a local temporary copy of an mp\_int argument.  To initialize an mp\_int
and then copy another existing mp\_int into the newly intialized mp\_int will be known as creating a clone.  This is
useful within functions that need to modify an argument but do not wish to actually modify the original copy.  The
mp\_init\_copy algorithm has been designed to help perform this task.

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_init\_copy}. \\
\textbf{Input}.   An mp\_int $a$ and $b$\\
\textbf{Output}.  $a$ is initialized to be a copy of $b$. \\
\hline \\
1.  Init $a$.  (\textit{mp\_init}) \\
2.  Copy $b$ to $a$.  (\textit{mp\_copy}) \\
3.  Return the status of the copy operation. \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_init\_copy}
\end{figure}

\textbf{Algorithm mp\_init\_copy.}
This algorithm will initialize an mp\_int variable and copy another previously initialized mp\_int variable into it.  As
such this algorithm will perform two operations in one step.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_init\_copy.c
\vspace{-3mm}
\begin{alltt}
016
017   /* creates "a" then copies b into it */
018   int mp_init_copy (mp_int * a, mp_int * b)
019   \{
020     int     res;
021
022     if ((res = mp_init_size (a, b->used)) != MP_OKAY) \{
023       return res;
024     \}
025     return mp_copy (b, a);
026   \}
027   #endif
028
\end{alltt}
\end{small}

This will initialize \textbf{a} and make it a verbatim copy of the contents of \textbf{b}.  Note that
\textbf{a} will have its own memory allocated which means that \textbf{b} may be cleared after the call
and \textbf{a} will be left intact.

\section{Zeroing an Integer}
Reseting an mp\_int to the default state is a common step in many algorithms.  The mp\_zero algorithm will be the algorithm used to
perform this task.

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_zero}. \\
\textbf{Input}.   An mp\_int $a$ \\
\textbf{Output}.  Zero the contents of $a$ \\
\hline \\
1.  $a.used \leftarrow 0$ \\
2.  $a.sign \leftarrow$ MP\_ZPOS \\
3.  for $n$ from 0 to $a.alloc - 1$ do \\
\hspace{3mm}3.1  $a_n \leftarrow 0$ \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_zero}
\end{figure}

\textbf{Algorithm mp\_zero.}
This algorithm simply resets a mp\_int to the default state.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_zero.c
\vspace{-3mm}
\begin{alltt}
016
017   /* set to zero */
018   void mp_zero (mp_int * a)
019   \{
020     int       n;
021     mp_digit *tmp;
022
023     a->sign = MP_ZPOS;
024     a->used = 0;
025
026     tmp = a->dp;
027     for (n = 0; n < a->alloc; n++) \{
028        *tmp++ = 0;
029     \}
030   \}
031   #endif
032
\end{alltt}
\end{small}

After the function is completed, all of the digits are zeroed, the \textbf{used} count is zeroed and the
\textbf{sign} variable is set to \textbf{MP\_ZPOS}.

\section{Sign Manipulation}
\subsection{Absolute Value}
With the mp\_int representation of an integer, calculating the absolute value is trivial.  The mp\_abs algorithm will compute
the absolute value of an mp\_int.

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_abs}. \\
\textbf{Input}.   An mp\_int $a$ \\
\textbf{Output}.  Computes $b = \vert a \vert$ \\
\hline \\
1.  Copy $a$ to $b$.  (\textit{mp\_copy}) \\
2.  If the copy failed return(\textit{MP\_MEM}). \\
3.  $b.sign \leftarrow MP\_ZPOS$ \\
4.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_abs}
\end{figure}

\textbf{Algorithm mp\_abs.}
This algorithm computes the absolute of an mp\_int input.  First it copies $a$ over $b$.  This is an example of an
algorithm where the check in mp\_copy that determines if the source and destination are equal proves useful.  This allows,
for instance, the developer to pass the same mp\_int as the source and destination to this function without addition
logic to handle it.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_abs.c
\vspace{-3mm}
\begin{alltt}
016
017   /* b = |a|
018    *
019    * Simple function copies the input and fixes the sign to positive
020    */
021   int
022   mp_abs (mp_int * a, mp_int * b)
023   \{
024     int     res;
025
026     /* copy a to b */
027     if (a != b) \{
028        if ((res = mp_copy (a, b)) != MP_OKAY) \{
029          return res;
030        \}
031     \}
032
033     /* force the sign of b to positive */
034     b->sign = MP_ZPOS;
035
036     return MP_OKAY;
037   \}
038   #endif
039
\end{alltt}
\end{small}

This fairly trivial algorithm first eliminates non--required duplications (line 27) and then sets the
\textbf{sign} flag to \textbf{MP\_ZPOS}.

\subsection{Integer Negation}
With the mp\_int representation of an integer, calculating the negation is also trivial.  The mp\_neg algorithm will compute
the negative of an mp\_int input.

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_neg}. \\
\textbf{Input}.   An mp\_int $a$ \\
\textbf{Output}.  Computes $b = -a$ \\
\hline \\
1.  Copy $a$ to $b$.  (\textit{mp\_copy}) \\
2.  If the copy failed return(\textit{MP\_MEM}). \\
3.  If $a.used = 0$ then return(\textit{MP\_OKAY}). \\
4.  If $a.sign = MP\_ZPOS$ then do \\
\hspace{3mm}4.1  $b.sign = MP\_NEG$. \\
5.  else do \\
\hspace{3mm}5.1  $b.sign = MP\_ZPOS$. \\
6.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_neg}
\end{figure}

\textbf{Algorithm mp\_neg.}
This algorithm computes the negation of an input.  First it copies $a$ over $b$.  If $a$ has no used digits then
the algorithm returns immediately.  Otherwise it flips the sign flag and stores the result in $b$.  Note that if
$a$ had no digits then it must be positive by definition.  Had step three been omitted then the algorithm would return
zero as negative.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_neg.c
\vspace{-3mm}
\begin{alltt}
016
017   /* b = -a */
018   int mp_neg (mp_int * a, mp_int * b)
019   \{
020     int     res;
021     if (a != b) \{
022        if ((res = mp_copy (a, b)) != MP_OKAY) \{
023           return res;
024        \}
025     \}
026
027     if (mp_iszero(b) != MP_YES) \{
028        b->sign = (a->sign == MP_ZPOS) ? MP_NEG : MP_ZPOS;
029     \} else \{
030        b->sign = MP_ZPOS;
031     \}
032
033     return MP_OKAY;
034   \}
035   #endif
036
\end{alltt}
\end{small}

Like mp\_abs() this function avoids non--required duplications (line 21) and then sets the sign.  We
have to make sure that only non--zero values get a \textbf{sign} of \textbf{MP\_NEG}.  If the mp\_int is zero
than the \textbf{sign} is hard--coded to \textbf{MP\_ZPOS}.

\section{Small Constants}
\subsection{Setting Small Constants}
Often a mp\_int must be set to a relatively small value such as $1$ or $2$.  For these cases the mp\_set algorithm is useful.

\newpage\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_set}. \\
\textbf{Input}.   An mp\_int $a$ and a digit $b$ \\
\textbf{Output}.  Make $a$ equivalent to $b$ \\
\hline \\
1.  Zero $a$ (\textit{mp\_zero}). \\
2.  $a_0 \leftarrow b \mbox{ (mod }\beta\mbox{)}$ \\
3.  $a.used \leftarrow  \left \lbrace \begin{array}{ll}
                              1 &  \mbox{if }a_0 > 0 \\
                              0 &  \mbox{if }a_0 = 0
                              \end{array} \right .$ \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_set}
\end{figure}

\textbf{Algorithm mp\_set.}
This algorithm sets a mp\_int to a small single digit value.  Step number 1 ensures that the integer is reset to the default state.  The
single digit is set (\textit{modulo $\beta$}) and the \textbf{used} count is adjusted accordingly.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_set.c
\vspace{-3mm}
\begin{alltt}
016
017   /* set to a digit */
018   void mp_set (mp_int * a, mp_digit b)
019   \{
020     mp_zero (a);
021     a->dp[0] = b & MP_MASK;
022     a->used  = (a->dp[0] != 0) ? 1 : 0;
023   \}
024   #endif
025
\end{alltt}
\end{small}

First we zero (line 20) the mp\_int to make sure that the other members are initialized for a
small positive constant.  mp\_zero() ensures that the \textbf{sign} is positive and the \textbf{used} count
is zero.  Next we set the digit and reduce it modulo $\beta$ (line 21).  After this step we have to
check if the resulting digit is zero or not.  If it is not then we set the \textbf{used} count to one, otherwise
to zero.

We can quickly reduce modulo $\beta$ since it is of the form $2^k$ and a quick binary AND operation with
$2^k - 1$ will perform the same operation.

One important limitation of this function is that it will only set one digit.  The size of a digit is not fixed, meaning source that uses
this function should take that into account.  Only trivially small constants can be set using this function.

\subsection{Setting Large Constants}
To overcome the limitations of the mp\_set algorithm the mp\_set\_int algorithm is ideal.  It accepts a ``long''
data type as input and will always treat it as a 32-bit integer.

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_set\_int}. \\
\textbf{Input}.   An mp\_int $a$ and a ``long'' integer $b$ \\
\textbf{Output}.  Make $a$ equivalent to $b$ \\
\hline \\
1.  Zero $a$ (\textit{mp\_zero}) \\
2.  for $n$ from 0 to 7 do \\
\hspace{3mm}2.1  $a \leftarrow a \cdot 16$ (\textit{mp\_mul2d}) \\
\hspace{3mm}2.2  $u \leftarrow \lfloor b / 2^{4(7 - n)} \rfloor \mbox{ (mod }16\mbox{)}$\\
\hspace{3mm}2.3  $a_0 \leftarrow a_0 + u$ \\
\hspace{3mm}2.4  $a.used \leftarrow a.used + 1$ \\
3.  Clamp excess used digits (\textit{mp\_clamp}) \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_set\_int}
\end{figure}

\textbf{Algorithm mp\_set\_int.}
The algorithm performs eight iterations of a simple loop where in each iteration four bits from the source are added to the
mp\_int.  Step 2.1 will multiply the current result by sixteen making room for four more bits in the less significant positions.  In step 2.2 the
next four bits from the source are extracted and are added to the mp\_int. The \textbf{used} digit count is
incremented to reflect the addition.  The \textbf{used} digit counter is incremented since if any of the leading digits were zero the mp\_int would have
zero digits used and the newly added four bits would be ignored.

Excess zero digits are trimmed in steps 2.1 and 3 by using higher level algorithms mp\_mul2d and mp\_clamp.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_set\_int.c
\vspace{-3mm}
\begin{alltt}
016
017   /* set a 32-bit const */
018   int mp_set_int (mp_int * a, unsigned long b)
019   \{
020     int     x, res;
021
022     mp_zero (a);
023
024     /* set four bits at a time */
025     for (x = 0; x < 8; x++) \{
026       /* shift the number up four bits */
027       if ((res = mp_mul_2d (a, 4, a)) != MP_OKAY) \{
028         return res;
029       \}
030
031       /* OR in the top four bits of the source */
032       a->dp[0] |= (b >> 28) & 15;
033
034       /* shift the source up to the next four bits */
035       b <<= 4;
036
037       /* ensure that digits are not clamped off */
038       a->used += 1;
039     \}
040     mp_clamp (a);
041     return MP_OKAY;
042   \}
043   #endif
044
\end{alltt}
\end{small}

This function sets four bits of the number at a time to handle all practical \textbf{DIGIT\_BIT} sizes.  The weird
addition on line 38 ensures that the newly added in bits are added to the number of digits.  While it may not
seem obvious as to why the digit counter does not grow exceedingly large it is because of the shift on line 27
as well as the  call to mp\_clamp() on line 40.  Both functions will clamp excess leading digits which keeps
the number of used digits low.

\section{Comparisons}
\subsection{Unsigned Comparisions}
Comparing a multiple precision integer is performed with the exact same algorithm used to compare two decimal numbers.  For example,
to compare $1,234$ to $1,264$ the digits are extracted by their positions.  That is we compare $1 \cdot 10^3 + 2 \cdot 10^2 + 3 \cdot 10^1 + 4 \cdot 10^0$
to $1 \cdot 10^3 + 2 \cdot 10^2 + 6 \cdot 10^1 + 4 \cdot 10^0$ by comparing single digits at a time starting with the highest magnitude
positions.  If any leading digit of one integer is greater than a digit in the same position of another integer then obviously it must be greater.

The first comparision routine that will be developed is the unsigned magnitude compare which will perform a comparison based on the digits of two
mp\_int variables alone.  It will ignore the sign of the two inputs.  Such a function is useful when an absolute comparison is required or if the
signs are known to agree in advance.

To facilitate working with the results of the comparison functions three constants are required.

\begin{figure}[here]
\begin{center}
\begin{tabular}{|r|l|}
\hline \textbf{Constant} & \textbf{Meaning} \\
\hline \textbf{MP\_GT} & Greater Than \\
\hline \textbf{MP\_EQ} & Equal To \\
\hline \textbf{MP\_LT} & Less Than \\
\hline
\end{tabular}
\end{center}
\caption{Comparison Return Codes}
\end{figure}

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_cmp\_mag}. \\
\textbf{Input}.   Two mp\_ints $a$ and $b$.  \\
\textbf{Output}.  Unsigned comparison results ($a$ to the left of $b$). \\
\hline \\
1.  If $a.used > b.used$ then return(\textit{MP\_GT}) \\
2.  If $a.used < b.used$ then return(\textit{MP\_LT}) \\
3.  for n from $a.used - 1$ to 0 do \\
\hspace{+3mm}3.1  if $a_n > b_n$ then return(\textit{MP\_GT}) \\
\hspace{+3mm}3.2  if $a_n < b_n$ then return(\textit{MP\_LT}) \\
4.  Return(\textit{MP\_EQ}) \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_cmp\_mag}
\end{figure}

\textbf{Algorithm mp\_cmp\_mag.}
By saying ``$a$ to the left of $b$'' it is meant that the comparison is with respect to $a$, that is if $a$ is greater than $b$ it will return
\textbf{MP\_GT} and similar with respect to when $a = b$ and $a < b$.  The first two steps compare the number of digits used in both $a$ and $b$.
Obviously if the digit counts differ there would be an imaginary zero digit in the smaller number where the leading digit of the larger number is.
If both have the same number of digits than the actual digits themselves must be compared starting at the leading digit.

By step three both inputs must have the same number of digits so its safe to start from either $a.used - 1$ or $b.used - 1$ and count down to
the zero'th digit.  If after all of the digits have been compared, no difference is found, the algorithm returns \textbf{MP\_EQ}.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_cmp\_mag.c
\vspace{-3mm}
\begin{alltt}
016
017   /* compare maginitude of two ints (unsigned) */
018   int mp_cmp_mag (mp_int * a, mp_int * b)
019   \{
020     int     n;
021     mp_digit *tmpa, *tmpb;
022
023     /* compare based on # of non-zero digits */
024     if (a->used > b->used) \{
025       return MP_GT;
026     \}
027
028     if (a->used < b->used) \{
029       return MP_LT;
030     \}
031
032     /* alias for a */
033     tmpa = a->dp + (a->used - 1);
034
035     /* alias for b */
036     tmpb = b->dp + (a->used - 1);
037
038     /* compare based on digits  */
039     for (n = 0; n < a->used; ++n, --tmpa, --tmpb) \{
040       if (*tmpa > *tmpb) \{
041         return MP_GT;
042       \}
043
044       if (*tmpa < *tmpb) \{
045         return MP_LT;
046       \}
047     \}
048     return MP_EQ;
049   \}
050   #endif
051
\end{alltt}
\end{small}

The two if statements (lines 24 and 28) compare the number of digits in the two inputs.  These two are
performed before all of the digits are compared since it is a very cheap test to perform and can potentially save
considerable time.  The implementation given is also not valid without those two statements.  $b.alloc$ may be
smaller than $a.used$, meaning that undefined values will be read from $b$ past the end of the array of digits.



\subsection{Signed Comparisons}
Comparing with sign considerations is also fairly critical in several routines (\textit{division for example}).  Based on an unsigned magnitude
comparison a trivial signed comparison algorithm can be written.

\begin{figure}[here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_cmp}. \\
\textbf{Input}.   Two mp\_ints $a$ and $b$ \\
\textbf{Output}.  Signed Comparison Results ($a$ to the left of $b$) \\
\hline \\
1.  if $a.sign = MP\_NEG$ and $b.sign = MP\_ZPOS$ then return(\textit{MP\_LT}) \\
2.  if $a.sign = MP\_ZPOS$ and $b.sign = MP\_NEG$ then return(\textit{MP\_GT}) \\
3.  if $a.sign = MP\_NEG$ then \\
\hspace{+3mm}3.1  Return the unsigned comparison of $b$ and $a$ (\textit{mp\_cmp\_mag}) \\
4   Otherwise \\
\hspace{+3mm}4.1  Return the unsigned comparison of $a$ and $b$ \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_cmp}
\end{figure}

\textbf{Algorithm mp\_cmp.}
The first two steps compare the signs of the two inputs.  If the signs do not agree then it can return right away with the appropriate
comparison code.  When the signs are equal the digits of the inputs must be compared to determine the correct result.  In step
three the unsigned comparision flips the order of the arguments since they are both negative.  For instance, if $-a > -b$ then
$\vert a \vert < \vert b \vert$.  Step number four will compare the two when they are both positive.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_cmp.c
\vspace{-3mm}
\begin{alltt}
016
017   /* compare two ints (signed)*/
018   int
019   mp_cmp (mp_int * a, mp_int * b)
020   \{
021     /* compare based on sign */
022     if (a->sign != b->sign) \{
023        if (a->sign == MP_NEG) \{
024           return MP_LT;
025        \} else \{
026           return MP_GT;
027        \}
028     \}
029
030     /* compare digits */
031     if (a->sign == MP_NEG) \{
032        /* if negative compare opposite direction */
033        return mp_cmp_mag(b, a);
034     \} else \{
035        return mp_cmp_mag(a, b);
036     \}
037   \}
038   #endif
039
\end{alltt}
\end{small}

The two if statements (lines 22 and 23) perform the initial sign comparison.  If the signs are not the equal then which ever
has the positive sign is larger.   The inputs are compared (line 31) based on magnitudes.  If the signs were both
negative then the unsigned comparison is performed in the opposite direction (line 33).  Otherwise, the signs are assumed to
be both positive and a forward direction unsigned comparison is performed.

\section*{Exercises}
\begin{tabular}{cl}
$\left [ 2 \right ]$ & Modify algorithm mp\_set\_int to accept as input a variable length array of bits. \\
                     & \\
$\left [ 3 \right ]$ & Give the probability that algorithm mp\_cmp\_mag will have to compare $k$ digits  \\
                     & of two random digits (of equal magnitude) before a difference is found. \\
                     & \\
$\left [ 1 \right ]$ & Suggest a simple method to speed up the implementation of mp\_cmp\_mag based  \\
                     & on the observations made in the previous problem. \\
                     &
\end{tabular}

\chapter{Basic Arithmetic}
\section{Introduction}
At this point algorithms for initialization, clearing, zeroing, copying, comparing and setting small constants have been
established.  The next logical set of algorithms to develop are addition, subtraction and digit shifting algorithms.  These
algorithms make use of the lower level algorithms and are the cruicial building block for the multiplication algorithms.  It is very important
that these algorithms are highly optimized.  On their own they are simple $O(n)$ algorithms but they can be called from higher level algorithms
which easily places them at $O(n^2)$ or even $O(n^3)$ work levels.

All of the algorithms within this chapter make use of the logical bit shift operations denoted by $<<$ and $>>$ for left and right
logical shifts respectively.  A logical shift is analogous to sliding the decimal point of radix-10 representations.  For example, the real
number $0.9345$ is equivalent to $93.45\%$ which is found by sliding the the decimal two places to the right (\textit{multiplying by $\beta^2 = 10^2$}).
Algebraically a binary logical shift is equivalent to a division or multiplication by a power of two.
For example, $a << k = a \cdot 2^k$ while $a >> k = \lfloor a/2^k \rfloor$.

One significant difference between a logical shift and the way decimals are shifted is that digits below the zero'th position are removed
from the number.  For example, consider $1101_2 >> 1$ using decimal notation this would produce $110.1_2$.  However, with a logical shift the
result is $110_2$.

\section{Addition and Subtraction}
In common twos complement fixed precision arithmetic negative numbers are easily represented by subtraction from the modulus.  For example, with 32-bit integers
$a - b\mbox{ (mod }2^{32}\mbox{)}$ is the same as $a + (2^{32} - b) \mbox{ (mod }2^{32}\mbox{)}$  since $2^{32} \equiv 0 \mbox{ (mod }2^{32}\mbox{)}$.
As a result subtraction can be performed with a trivial series of logical operations and an addition.

However, in multiple precision arithmetic negative numbers are not represented in the same way.  Instead a sign flag is used to keep track of the
sign of the integer.  As a result signed addition and subtraction are actually implemented as conditional usage of lower level addition or
subtraction algorithms with the sign fixed up appropriately.

The lower level algorithms will add or subtract integers without regard to the sign flag.  That is they will add or subtract the magnitude of
the integers respectively.

\subsection{Low Level Addition}
An unsigned addition of multiple precision integers is performed with the same long-hand algorithm used to add decimal numbers.  That is to add the
trailing digits first and propagate the resulting carry upwards.  Since this is a lower level algorithm the name will have a ``s\_'' prefix.
Historically that convention stems from the MPI library where ``s\_'' stood for static functions that were hidden from the developer entirely.

\newpage
\begin{figure}[!here]
\begin{center}
\begin{small}
\begin{tabular}{l}
\hline Algorithm \textbf{s\_mp\_add}. \\
\textbf{Input}.   Two mp\_ints $a$ and $b$ \\
\textbf{Output}.  The unsigned addition $c = \vert a \vert + \vert b \vert$. \\
\hline \\
1.  if $a.used > b.used$ then \\
\hspace{+3mm}1.1  $min \leftarrow b.used$ \\
\hspace{+3mm}1.2  $max \leftarrow a.used$ \\
\hspace{+3mm}1.3  $x   \leftarrow a$ \\
2.  else  \\
\hspace{+3mm}2.1  $min \leftarrow a.used$ \\
\hspace{+3mm}2.2  $max \leftarrow b.used$ \\
\hspace{+3mm}2.3  $x   \leftarrow b$ \\
3.  If $c.alloc < max + 1$ then grow $c$ to hold at least $max + 1$ digits (\textit{mp\_grow}) \\
4.  $oldused \leftarrow c.used$ \\
5.  $c.used \leftarrow max + 1$ \\
6.  $u \leftarrow 0$ \\
7.  for $n$ from $0$ to $min - 1$ do \\
\hspace{+3mm}7.1  $c_n \leftarrow a_n + b_n + u$ \\
\hspace{+3mm}7.2  $u \leftarrow c_n >> lg(\beta)$ \\
\hspace{+3mm}7.3  $c_n \leftarrow c_n \mbox{ (mod }\beta\mbox{)}$ \\
8.  if $min \ne max$ then do \\
\hspace{+3mm}8.1  for $n$ from $min$ to $max - 1$ do \\
\hspace{+6mm}8.1.1  $c_n \leftarrow x_n + u$ \\
\hspace{+6mm}8.1.2  $u \leftarrow c_n >> lg(\beta)$ \\
\hspace{+6mm}8.1.3  $c_n \leftarrow c_n \mbox{ (mod }\beta\mbox{)}$ \\
9.  $c_{max} \leftarrow u$ \\
10.  if $olduse > max$ then \\
\hspace{+3mm}10.1  for $n$ from $max + 1$ to $oldused - 1$ do \\
\hspace{+6mm}10.1.1  $c_n \leftarrow 0$ \\
11.  Clamp excess digits in $c$.  (\textit{mp\_clamp}) \\
12.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{small}
\end{center}
\caption{Algorithm s\_mp\_add}
\end{figure}

\textbf{Algorithm s\_mp\_add.}
This algorithm is loosely based on algorithm 14.7 of HAC \cite[pp. 594]{HAC} but has been extended to allow the inputs to have different magnitudes.
Coincidentally the description of algorithm A in Knuth \cite[pp. 266]{TAOCPV2} shares the same deficiency as the algorithm from \cite{HAC}.  Even the
MIX pseudo  machine code presented by Knuth \cite[pp. 266-267]{TAOCPV2} is incapable of handling inputs which are of different magnitudes.

The first thing that has to be accomplished is to sort out which of the two inputs is the largest.  The addition logic
will simply add all of the smallest input to the largest input and store that first part of the result in the
destination.  Then it will apply a simpler addition loop to excess digits of the larger input.

The first two steps will handle sorting the inputs such that $min$ and $max$ hold the digit counts of the two
inputs.  The variable $x$ will be an mp\_int alias for the largest input or the second input $b$ if they have the
same number of digits.  After the inputs are sorted the destination $c$ is grown as required to accomodate the sum
of the two inputs.  The original \textbf{used} count of $c$ is copied and set to the new used count.

At this point the first addition loop will go through as many digit positions that both inputs have.  The carry
variable $\mu$ is set to zero outside the loop.  Inside the loop an ``addition'' step requires three statements to produce
one digit of the summand.  First
two digits from $a$ and $b$ are added together along with the carry $\mu$.  The carry of this step is extracted and stored
in $\mu$ and finally the digit of the result $c_n$ is truncated within the range $0 \le c_n < \beta$.

Now all of the digit positions that both inputs have in common have been exhausted.  If $min \ne max$ then $x$ is an alias
for one of the inputs that has more digits.  A simplified addition loop is then used to essentially copy the remaining digits
and the carry to the destination.

The final carry is stored in $c_{max}$ and digits above $max$ upto $oldused$ are zeroed which completes the addition.


\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_s\_mp\_add.c
\vspace{-3mm}
\begin{alltt}
016
017   /* low level addition, based on HAC pp.594, Algorithm 14.7 */
018   int
019   s_mp_add (mp_int * a, mp_int * b, mp_int * c)
020   \{
021     mp_int *x;
022     int     olduse, res, min, max;
023
024     /* find sizes, we let |a| <= |b| which means we have to sort
025      * them.  "x" will point to the input with the most digits
026      */
027     if (a->used > b->used) \{
028       min = b->used;
029       max = a->used;
030       x = a;
031     \} else \{
032       min = a->used;
033       max = b->used;
034       x = b;
035     \}
036
037     /* init result */
038     if (c->alloc < (max + 1)) \{
039       if ((res = mp_grow (c, max + 1)) != MP_OKAY) \{
040         return res;
041       \}
042     \}
043
044     /* get old used digit count and set new one */
045     olduse = c->used;
046     c->used = max + 1;
047
048     \{
049       mp_digit u, *tmpa, *tmpb, *tmpc;
050       int i;
051
052       /* alias for digit pointers */
053
054       /* first input */
055       tmpa = a->dp;
056
057       /* second input */
058       tmpb = b->dp;
059
060       /* destination */
061       tmpc = c->dp;
062
063       /* zero the carry */
064       u = 0;
065       for (i = 0; i < min; i++) \{
066         /* Compute the sum at one digit, T[i] = A[i] + B[i] + U */
067         *tmpc = *tmpa++ + *tmpb++ + u;
068
069         /* U = carry bit of T[i] */
070         u = *tmpc >> ((mp_digit)DIGIT_BIT);
071
072         /* take away carry bit from T[i] */
073         *tmpc++ &= MP_MASK;
074       \}
075
076       /* now copy higher words if any, that is in A+B
077        * if A or B has more digits add those in
078        */
079       if (min != max) \{
080         for (; i < max; i++) \{
081           /* T[i] = X[i] + U */
082           *tmpc = x->dp[i] + u;
083
084           /* U = carry bit of T[i] */
085           u = *tmpc >> ((mp_digit)DIGIT_BIT);
086
087           /* take away carry bit from T[i] */
088           *tmpc++ &= MP_MASK;
089         \}
090       \}
091
092       /* add carry */
093       *tmpc++ = u;
094
095       /* clear digits above oldused */
096       for (i = c->used; i < olduse; i++) \{
097         *tmpc++ = 0;
098       \}
099     \}
100
101     mp_clamp (c);
102     return MP_OKAY;
103   \}
104   #endif
105
\end{alltt}
\end{small}

We first sort (lines 27 to 35) the inputs based on magnitude and determine the $min$ and $max$ variables.
Note that $x$ is a pointer to an mp\_int assigned to the largest input, in effect it is a local alias.  Next we
grow the destination (37 to 42) ensure that it can accomodate the result of the addition.

Similar to the implementation of mp\_copy this function uses the braced code and local aliases coding style.  The three aliases that are on
lines 55, 58 and 61 represent the two inputs and destination variables respectively.  These aliases are used to ensure the
compiler does not have to dereference $a$, $b$ or $c$ (respectively) to access the digits of the respective mp\_int.

The initial carry $u$ will be cleared (line 64), note that $u$ is of type mp\_digit which ensures type
compatibility within the implementation.  The initial addition (line 65 to 74) adds digits from
both inputs until the smallest input runs out of digits.  Similarly the conditional addition loop
(line 80 to 90) adds the remaining digits from the larger of the two inputs.  The addition is finished
with the final carry being stored in $tmpc$ (line 93).  Note the ``++'' operator within the same expression.
After line 93, $tmpc$ will point to the $c.used$'th digit of the mp\_int $c$.  This is useful
for the next loop (line 96 to 99) which set any old upper digits to zero.

\subsection{Low Level Subtraction}
The low level unsigned subtraction algorithm is very similar to the low level unsigned addition algorithm.  The principle difference is that the
unsigned subtraction algorithm requires the result to be positive.  That is when computing $a - b$ the condition $\vert a \vert \ge \vert b\vert$ must
be met for this algorithm to function properly.  Keep in mind this low level algorithm is not meant to be used in higher level algorithms directly.
This algorithm as will be shown can be used to create functional signed addition and subtraction algorithms.


For this algorithm a new variable is required to make the description simpler.  Recall from section 1.3.1 that a mp\_digit must be able to represent
the range $0 \le x < 2\beta$ for the algorithms to work correctly.  However, it is allowable that a mp\_digit represent a larger range of values.  For
this algorithm we will assume that the variable $\gamma$ represents the number of bits available in a
mp\_digit (\textit{this implies $2^{\gamma} > \beta$}).

For example, the default for LibTomMath is to use a ``unsigned long'' for the mp\_digit ``type'' while $\beta = 2^{28}$.  In ISO C an ``unsigned long''
data type must be able to represent $0 \le x < 2^{32}$ meaning that in this case $\gamma \ge 32$.

\newpage\begin{figure}[!here]
\begin{center}
\begin{small}
\begin{tabular}{l}
\hline Algorithm \textbf{s\_mp\_sub}. \\
\textbf{Input}.   Two mp\_ints $a$ and $b$ ($\vert a \vert \ge \vert b \vert$) \\
\textbf{Output}.  The unsigned subtraction $c = \vert a \vert - \vert b \vert$. \\
\hline \\
1.  $min \leftarrow b.used$ \\
2.  $max \leftarrow a.used$ \\
3.  If $c.alloc < max$ then grow $c$ to hold at least $max$ digits.  (\textit{mp\_grow}) \\
4.  $oldused \leftarrow c.used$ \\
5.  $c.used \leftarrow max$ \\
6.  $u \leftarrow 0$ \\
7.  for $n$ from $0$ to $min - 1$ do \\
\hspace{3mm}7.1  $c_n \leftarrow a_n - b_n - u$ \\
\hspace{3mm}7.2  $u   \leftarrow c_n >> (\gamma - 1)$ \\
\hspace{3mm}7.3  $c_n \leftarrow c_n \mbox{ (mod }\beta\mbox{)}$ \\
8.  if $min < max$ then do \\
\hspace{3mm}8.1  for $n$ from $min$ to $max - 1$ do \\
\hspace{6mm}8.1.1  $c_n \leftarrow a_n - u$ \\
\hspace{6mm}8.1.2  $u   \leftarrow c_n >> (\gamma - 1)$ \\
\hspace{6mm}8.1.3  $c_n \leftarrow c_n \mbox{ (mod }\beta\mbox{)}$ \\
9. if $oldused > max$ then do \\
\hspace{3mm}9.1  for $n$ from $max$ to $oldused - 1$ do \\
\hspace{6mm}9.1.1  $c_n \leftarrow 0$ \\
10. Clamp excess digits of $c$.  (\textit{mp\_clamp}). \\
11. Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{small}
\end{center}
\caption{Algorithm s\_mp\_sub}
\end{figure}

\textbf{Algorithm s\_mp\_sub.}
This algorithm performs the unsigned subtraction of two mp\_int variables under the restriction that the result must be positive.  That is when
passing variables $a$ and $b$ the condition that $\vert a \vert \ge \vert b \vert$ must be met for the algorithm to function correctly.  This
algorithm is loosely based on algorithm 14.9 \cite[pp. 595]{HAC} and is similar to algorithm S in \cite[pp. 267]{TAOCPV2} as well.  As was the case
of the algorithm s\_mp\_add both other references lack discussion concerning various practical details such as when the inputs differ in magnitude.

The initial sorting of the inputs is trivial in this algorithm since $a$ is guaranteed to have at least the same magnitude of $b$.  Steps 1 and 2
set the $min$ and $max$ variables.  Unlike the addition routine there is guaranteed to be no carry which means that the final result can be at
most $max$ digits in length as opposed to $max + 1$.  Similar to the addition algorithm the \textbf{used} count of $c$ is copied locally and
set to the maximal count for the operation.

The subtraction loop that begins on step seven is essentially the same as the addition loop of algorithm s\_mp\_add except single precision
subtraction is used instead.  Note the use of the $\gamma$ variable to extract the carry (\textit{also known as the borrow}) within the subtraction
loops.  Under the assumption that two's complement single precision arithmetic is used this will successfully extract the desired carry.

For example, consider subtracting $0101_2$ from $0100_2$ where $\gamma = 4$ and $\beta = 2$.  The least significant bit will force a carry upwards to
the third bit which will be set to zero after the borrow.  After the very first bit has been subtracted $4 - 1 \equiv 0011_2$ will remain,  When the
third bit of $0101_2$ is subtracted from the result it will cause another carry.  In this case though the carry will be forced to propagate all the
way to the most significant bit.

Recall that $\beta < 2^{\gamma}$.  This means that if a carry does occur just before the $lg(\beta)$'th bit it will propagate all the way to the most
significant bit.  Thus, the high order bits of the mp\_digit that are not part of the actual digit will either be all zero, or all one. All that
is needed is a single zero or one bit for the carry.  Therefore a single logical shift right by $\gamma - 1$ positions is sufficient to extract the
carry.  This method of carry extraction may seem awkward but the reason for it becomes apparent when the implementation is discussed.

If $b$ has a smaller magnitude than $a$ then step 9 will force the carry and copy operation to propagate through the larger input $a$ into $c$.  Step
10 will ensure that any leading digits of $c$ above the $max$'th position are zeroed.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_s\_mp\_sub.c
\vspace{-3mm}
\begin{alltt}
016
017   /* low level subtraction (assumes |a| > |b|), HAC pp.595 Algorithm 14.9 */
018   int
019   s_mp_sub (mp_int * a, mp_int * b, mp_int * c)
020   \{
021     int     olduse, res, min, max;
022
023     /* find sizes */
024     min = b->used;
025     max = a->used;
026
027     /* init result */
028     if (c->alloc < max) \{
029       if ((res = mp_grow (c, max)) != MP_OKAY) \{
030         return res;
031       \}
032     \}
033     olduse = c->used;
034     c->used = max;
035
036     \{
037       mp_digit u, *tmpa, *tmpb, *tmpc;
038       int i;
039
040       /* alias for digit pointers */
041       tmpa = a->dp;
042       tmpb = b->dp;
043       tmpc = c->dp;
044
045       /* set carry to zero */
046       u = 0;
047       for (i = 0; i < min; i++) \{
048         /* T[i] = A[i] - B[i] - U */
049         *tmpc = (*tmpa++ - *tmpb++) - u;
050
051         /* U = carry bit of T[i]
052          * Note this saves performing an AND operation since
053          * if a carry does occur it will propagate all the way to the
054          * MSB.  As a result a single shift is enough to get the carry
055          */
056         u = *tmpc >> ((mp_digit)((CHAR_BIT * sizeof(mp_digit)) - 1));
057
058         /* Clear carry from T[i] */
059         *tmpc++ &= MP_MASK;
060       \}
061
062       /* now copy higher words if any, e.g. if A has more digits than B  */
063       for (; i < max; i++) \{
064         /* T[i] = A[i] - U */
065         *tmpc = *tmpa++ - u;
066
067         /* U = carry bit of T[i] */
068         u = *tmpc >> ((mp_digit)((CHAR_BIT * sizeof(mp_digit)) - 1));
069
070         /* Clear carry from T[i] */
071         *tmpc++ &= MP_MASK;
072       \}
073
074       /* clear digits above used (since we may not have grown result above) */

075       for (i = c->used; i < olduse; i++) \{
076         *tmpc++ = 0;
077       \}
078     \}
079
080     mp_clamp (c);
081     return MP_OKAY;
082   \}
083
084   #endif
085
\end{alltt}
\end{small}

Like low level addition we ``sort'' the inputs.  Except in this case the sorting is hardcoded
(lines 24 and 25).  In reality the $min$ and $max$ variables are only aliases and are only
used to make the source code easier to read.  Again the pointer alias optimization is used
within this algorithm.  The aliases $tmpa$, $tmpb$ and $tmpc$ are initialized
(lines 41, 42 and 43) for $a$, $b$ and $c$ respectively.

The first subtraction loop (lines 46 through 60) subtract digits from both inputs until the smaller of
the two inputs has been exhausted.  As remarked earlier there is an implementation reason for using the ``awkward''
method of extracting the carry (line 56).  The traditional method for extracting the carry would be to shift
by $lg(\beta)$ positions and logically AND the least significant bit.  The AND operation is required because all of
the bits above the $\lg(\beta)$'th bit will be set to one after a carry occurs from subtraction.  This carry
extraction requires two relatively cheap operations to extract the carry.  The other method is to simply shift the
most significant bit to the least significant bit thus extracting the carry with a single cheap operation.  This
optimization only works on twos compliment machines which is a safe assumption to make.

If $a$ has a larger magnitude than $b$ an additional loop (lines 63 through 72) is required to propagate
the carry through $a$ and copy the result to $c$.

\subsection{High Level Addition}
Now that both lower level addition and subtraction algorithms have been established an effective high level signed addition algorithm can be
established.  This high level addition algorithm will be what other algorithms and developers will use to perform addition of mp\_int data
types.

Recall from section 5.2 that an mp\_int represents an integer with an unsigned mantissa (\textit{the array of digits}) and a \textbf{sign}
flag.  A high level addition is actually performed as a series of eight separate cases which can be optimized down to three unique cases.

\begin{figure}[!here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_add}. \\
\textbf{Input}.   Two mp\_ints $a$ and $b$  \\
\textbf{Output}.  The signed addition $c = a + b$. \\
\hline \\
1.  if $a.sign = b.sign$ then do \\
\hspace{3mm}1.1  $c.sign \leftarrow a.sign$  \\
\hspace{3mm}1.2  $c \leftarrow \vert a \vert + \vert b \vert$ (\textit{s\_mp\_add})\\
2.  else do \\
\hspace{3mm}2.1  if $\vert a \vert < \vert b \vert$ then do (\textit{mp\_cmp\_mag})  \\
\hspace{6mm}2.1.1  $c.sign \leftarrow b.sign$ \\
\hspace{6mm}2.1.2  $c \leftarrow \vert b \vert - \vert a \vert$ (\textit{s\_mp\_sub}) \\
\hspace{3mm}2.2  else do \\
\hspace{6mm}2.2.1  $c.sign \leftarrow a.sign$ \\
\hspace{6mm}2.2.2  $c \leftarrow \vert a \vert - \vert b \vert$ \\
3.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_add}
\end{figure}

\textbf{Algorithm mp\_add.}
This algorithm performs the signed addition of two mp\_int variables.  There is no reference algorithm to draw upon from
either \cite{TAOCPV2} or \cite{HAC} since they both only provide unsigned operations.  The algorithm is fairly
straightforward but restricted since subtraction can only produce positive results.

\begin{figure}[here]
\begin{small}
\begin{center}
\begin{tabular}{|c|c|c|c|c|}
\hline \textbf{Sign of $a$} & \textbf{Sign of $b$} & \textbf{$\vert a \vert > \vert b \vert $} & \textbf{Unsigned Operation} & \textbf{Result Sign Flag} \\
\hline $+$ & $+$ & Yes & $c = a + b$ & $a.sign$ \\
\hline $+$ & $+$ & No  & $c = a + b$ & $a.sign$ \\
\hline $-$ & $-$ & Yes & $c = a + b$ & $a.sign$ \\
\hline $-$ & $-$ & No  & $c = a + b$ & $a.sign$ \\
\hline &&&&\\

\hline $+$ & $-$ & No  & $c = b - a$ & $b.sign$ \\
\hline $-$ & $+$ & No  & $c = b - a$ & $b.sign$ \\

\hline &&&&\\

\hline $+$ & $-$ & Yes & $c = a - b$ & $a.sign$ \\
\hline $-$ & $+$ & Yes & $c = a - b$ & $a.sign$ \\

\hline
\end{tabular}
\end{center}
\end{small}
\caption{Addition Guide Chart}
\label{fig:AddChart}
\end{figure}

Figure~\ref{fig:AddChart} lists all of the eight possible input combinations and is sorted to show that only three
specific cases need to be handled.  The return code of the unsigned operations at step 1.2, 2.1.2 and 2.2.2 are
forwarded to step three to check for errors.  This simplifies the description of the algorithm considerably and best
follows how the implementation actually was achieved.

Also note how the \textbf{sign} is set before the unsigned addition or subtraction is performed.  Recall from the descriptions of algorithms
s\_mp\_add and s\_mp\_sub that the mp\_clamp function is used at the end to trim excess digits.  The mp\_clamp algorithm will set the \textbf{sign}
to \textbf{MP\_ZPOS} when the \textbf{used} digit count reaches zero.

For example, consider performing $-a + a$ with algorithm mp\_add.  By the description of the algorithm the sign is set to \textbf{MP\_NEG} which would
produce a result of $-0$.  However, since the sign is set first then the unsigned addition is performed the subsequent usage of algorithm mp\_clamp
within algorithm s\_mp\_add will force $-0$ to become $0$.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_add.c
\vspace{-3mm}
\begin{alltt}
016
017   /* high level addition (handles signs) */
018   int mp_add (mp_int * a, mp_int * b, mp_int * c)
019   \{
020     int     sa, sb, res;
021
022     /* get sign of both inputs */
023     sa = a->sign;
024     sb = b->sign;
025
026     /* handle two cases, not four */
027     if (sa == sb) \{
028       /* both positive or both negative */
029       /* add their magnitudes, copy the sign */
030       c->sign = sa;
031       res = s_mp_add (a, b, c);
032     \} else \{
033       /* one positive, the other negative */
034       /* subtract the one with the greater magnitude from */
035       /* the one of the lesser magnitude.  The result gets */
036       /* the sign of the one with the greater magnitude. */
037       if (mp_cmp_mag (a, b) == MP_LT) \{
038         c->sign = sb;
039         res = s_mp_sub (b, a, c);
040       \} else \{
041         c->sign = sa;
042         res = s_mp_sub (a, b, c);
043       \}
044     \}
045     return res;
046   \}
047
048   #endif
049
\end{alltt}
\end{small}

The source code follows the algorithm fairly closely.  The most notable new source code addition is the usage of the $res$ integer variable which
is used to pass result of the unsigned operations forward.  Unlike in the algorithm, the variable $res$ is merely returned as is without
explicitly checking it and returning the constant \textbf{MP\_OKAY}.  The observation is this algorithm will succeed or fail only if the lower
level functions do so.  Returning their return code is sufficient.

\subsection{High Level Subtraction}
The high level signed subtraction algorithm is essentially the same as the high level signed addition algorithm.

\newpage\begin{figure}[!here]
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_sub}. \\
\textbf{Input}.   Two mp\_ints $a$ and $b$  \\
\textbf{Output}.  The signed subtraction $c = a - b$. \\
\hline \\
1.  if $a.sign \ne b.sign$ then do \\
\hspace{3mm}1.1  $c.sign \leftarrow a.sign$ \\
\hspace{3mm}1.2  $c \leftarrow \vert a \vert + \vert b \vert$ (\textit{s\_mp\_add}) \\
2.  else do \\
\hspace{3mm}2.1  if $\vert a \vert \ge \vert b \vert$ then do (\textit{mp\_cmp\_mag}) \\
\hspace{6mm}2.1.1  $c.sign \leftarrow a.sign$ \\
\hspace{6mm}2.1.2  $c \leftarrow \vert a \vert  - \vert b \vert$ (\textit{s\_mp\_sub}) \\
\hspace{3mm}2.2  else do \\
\hspace{6mm}2.2.1  $c.sign \leftarrow  \left \lbrace \begin{array}{ll}
                              MP\_ZPOS &  \mbox{if }a.sign = MP\_NEG \\
                              MP\_NEG  &  \mbox{otherwise} \\
                              \end{array} \right .$ \\
\hspace{6mm}2.2.2  $c \leftarrow \vert b \vert  - \vert a \vert$ \\
3.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\caption{Algorithm mp\_sub}
\end{figure}

\textbf{Algorithm mp\_sub.}
This algorithm performs the signed subtraction of two inputs.  Similar to algorithm mp\_add there is no reference in either \cite{TAOCPV2} or
\cite{HAC}.  Also this algorithm is restricted by algorithm s\_mp\_sub.  Chart \ref{fig:SubChart} lists the eight possible inputs and
the operations required.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{|c|c|c|c|c|}
\hline \textbf{Sign of $a$} & \textbf{Sign of $b$} & \textbf{$\vert a \vert \ge \vert b \vert $} & \textbf{Unsigned Operation} & \textbf{Result Sign Flag} \\
\hline $+$ & $-$ & Yes & $c = a + b$ & $a.sign$ \\
\hline $+$ & $-$ & No  & $c = a + b$ & $a.sign$ \\
\hline $-$ & $+$ & Yes & $c = a + b$ & $a.sign$ \\
\hline $-$ & $+$ & No  & $c = a + b$ & $a.sign$ \\
\hline &&&& \\
\hline $+$ & $+$ & Yes & $c = a - b$ & $a.sign$ \\
\hline $-$ & $-$ & Yes & $c = a - b$ & $a.sign$ \\
\hline &&&& \\
\hline $+$ & $+$ & No  & $c = b - a$ & $\mbox{opposite of }a.sign$ \\
\hline $-$ & $-$ & No  & $c = b - a$ & $\mbox{opposite of }a.sign$ \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Subtraction Guide Chart}
\label{fig:SubChart}
\end{figure}

Similar to the case of algorithm mp\_add the \textbf{sign} is set first before the unsigned addition or subtraction.  That is to prevent the
algorithm from producing $-a - -a = -0$ as a result.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_sub.c
\vspace{-3mm}
\begin{alltt}
016
017   /* high level subtraction (handles signs) */
018   int
019   mp_sub (mp_int * a, mp_int * b, mp_int * c)
020   \{
021     int     sa, sb, res;
022
023     sa = a->sign;
024     sb = b->sign;
025
026     if (sa != sb) \{
027       /* subtract a negative from a positive, OR */
028       /* subtract a positive from a negative. */
029       /* In either case, ADD their magnitudes, */
030       /* and use the sign of the first number. */
031       c->sign = sa;
032       res = s_mp_add (a, b, c);
033     \} else \{
034       /* subtract a positive from a positive, OR */
035       /* subtract a negative from a negative. */
036       /* First, take the difference between their */
037       /* magnitudes, then... */
038       if (mp_cmp_mag (a, b) != MP_LT) \{
039         /* Copy the sign from the first */
040         c->sign = sa;
041         /* The first has a larger or equal magnitude */
042         res = s_mp_sub (a, b, c);
043       \} else \{
044         /* The result has the *opposite* sign from */
045         /* the first number. */
046         c->sign = (sa == MP_ZPOS) ? MP_NEG : MP_ZPOS;
047         /* The second has a larger magnitude */
048         res = s_mp_sub (b, a, c);
049       \}
050     \}
051     return res;
052   \}
053
054   #endif
055
\end{alltt}
\end{small}

Much like the implementation of algorithm mp\_add the variable $res$ is used to catch the return code of the unsigned addition or subtraction operations
and forward it to the end of the function.  On line 38 the ``not equal to'' \textbf{MP\_LT} expression is used to emulate a
``greater than or equal to'' comparison.

\section{Bit and Digit Shifting}
It is quite common to think of a multiple precision integer as a polynomial in $x$, that is $y = f(\beta)$ where $f(x) = \sum_{i=0}^{n-1} a_i x^i$.
This notation arises within discussion of Montgomery and Diminished Radix Reduction as well as Karatsuba multiplication and squaring.

In order to facilitate operations on polynomials in $x$ as above a series of simple ``digit'' algorithms have to be established.  That is to shift
the digits left or right as well to shift individual bits of the digits left and right.  It is important to note that not all ``shift'' operations
are on radix-$\beta$ digits.

\subsection{Multiplication by Two}

In a binary system where the radix is a power of two multiplication by two not only arises often in other algorithms it is a fairly efficient
operation to perform.  A single precision logical shift left is sufficient to multiply a single digit by two.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_mul\_2}. \\
\textbf{Input}.   One mp\_int $a$ \\
\textbf{Output}.  $b = 2a$. \\
\hline \\
1.  If $b.alloc < a.used + 1$ then grow $b$ to hold $a.used + 1$ digits.  (\textit{mp\_grow}) \\
2.  $oldused \leftarrow b.used$ \\
3.  $b.used \leftarrow a.used$ \\
4.  $r \leftarrow 0$ \\
5.  for $n$ from 0 to $a.used - 1$ do \\
\hspace{3mm}5.1  $rr \leftarrow a_n >> (lg(\beta) - 1)$ \\
\hspace{3mm}5.2  $b_n \leftarrow (a_n << 1) + r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{3mm}5.3  $r \leftarrow rr$ \\
6.  If $r \ne 0$ then do \\
\hspace{3mm}6.1  $b_{n + 1} \leftarrow r$ \\
\hspace{3mm}6.2  $b.used \leftarrow b.used + 1$ \\
7.  If $b.used < oldused - 1$ then do \\
\hspace{3mm}7.1  for $n$ from $b.used$ to $oldused - 1$ do \\
\hspace{6mm}7.1.1  $b_n \leftarrow 0$ \\
8.  $b.sign \leftarrow a.sign$ \\
9.  Return(\textit{MP\_OKAY}).\\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_mul\_2}
\end{figure}

\textbf{Algorithm mp\_mul\_2.}
This algorithm will quickly multiply a mp\_int by two provided $\beta$ is a power of two.  Neither \cite{TAOCPV2} nor \cite{HAC} describe such
an algorithm despite the fact it arises often in other algorithms.  The algorithm is setup much like the lower level algorithm s\_mp\_add since
it is for all intents and purposes equivalent to the operation $b = \vert a \vert + \vert a \vert$.

Step 1 and 2 grow the input as required to accomodate the maximum number of \textbf{used} digits in the result.  The initial \textbf{used} count
is set to $a.used$ at step 4.  Only if there is a final carry will the \textbf{used} count require adjustment.

Step 6 is an optimization implementation of the addition loop for this specific case.  That is since the two values being added together
are the same there is no need to perform two reads from the digits of $a$.  Step 6.1 performs a single precision shift on the current digit $a_n$ to
obtain what will be the carry for the next iteration.  Step 6.2 calculates the $n$'th digit of the result as single precision shift of $a_n$ plus
the previous carry.  Recall from section 4.1 that $a_n << 1$ is equivalent to $a_n \cdot 2$.  An iteration of the addition loop is finished with
forwarding the carry to the next iteration.

Step 7 takes care of any final carry by setting the $a.used$'th digit of the result to the carry and augmenting the \textbf{used} count of $b$.
Step 8 clears any leading digits of $b$ in case it originally had a larger magnitude than $a$.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_mul\_2.c
\vspace{-3mm}
\begin{alltt}
016
017   /* b = a*2 */
018   int mp_mul_2(mp_int * a, mp_int * b)
019   \{
020     int     x, res, oldused;
021
022     /* grow to accomodate result */
023     if (b->alloc < (a->used + 1)) \{
024       if ((res = mp_grow (b, a->used + 1)) != MP_OKAY) \{
025         return res;
026       \}
027     \}
028
029     oldused = b->used;
030     b->used = a->used;
031
032     \{
033       mp_digit r, rr, *tmpa, *tmpb;
034
035       /* alias for source */
036       tmpa = a->dp;
037
038       /* alias for dest */
039       tmpb = b->dp;
040
041       /* carry */
042       r = 0;
043       for (x = 0; x < a->used; x++) \{
044
045         /* get what will be the *next* carry bit from the
046          * MSB of the current digit
047          */
048         rr = *tmpa >> ((mp_digit)(DIGIT_BIT - 1));
049
050         /* now shift up this digit, add in the carry [from the previous] */
051         *tmpb++ = ((*tmpa++ << ((mp_digit)1)) | r) & MP_MASK;
052
053         /* copy the carry that would be from the source
054          * digit into the next iteration
055          */
056         r = rr;
057       \}
058
059       /* new leading digit? */
060       if (r != 0) \{
061         /* add a MSB which is always 1 at this point */
062         *tmpb = 1;
063         ++(b->used);
064       \}
065
066       /* now zero any excess digits on the destination
067        * that we didn't write to
068        */
069       tmpb = b->dp + b->used;
070       for (x = b->used; x < oldused; x++) \{
071         *tmpb++ = 0;
072       \}
073     \}
074     b->sign = a->sign;
075     return MP_OKAY;
076   \}
077   #endif
078
\end{alltt}
\end{small}

This implementation is essentially an optimized implementation of s\_mp\_add for the case of doubling an input.  The only noteworthy difference
is the use of the logical shift operator on line 51 to perform a single precision doubling.

\subsection{Division by Two}
A division by two can just as easily be accomplished with a logical shift right as multiplication by two can be with a logical shift left.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_div\_2}. \\
\textbf{Input}.   One mp\_int $a$ \\
\textbf{Output}.  $b = a/2$. \\
\hline \\
1.  If $b.alloc < a.used$ then grow $b$ to hold $a.used$ digits.  (\textit{mp\_grow}) \\
2.  If the reallocation failed return(\textit{MP\_MEM}). \\
3.  $oldused \leftarrow b.used$ \\
4.  $b.used \leftarrow a.used$ \\
5.  $r \leftarrow 0$ \\
6.  for $n$ from $b.used - 1$ to $0$ do \\
\hspace{3mm}6.1  $rr \leftarrow a_n \mbox{ (mod }2\mbox{)}$\\
\hspace{3mm}6.2  $b_n \leftarrow (a_n >> 1) + (r << (lg(\beta) - 1)) \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{3mm}6.3  $r \leftarrow rr$ \\
7.  If $b.used < oldused - 1$ then do \\
\hspace{3mm}7.1  for $n$ from $b.used$ to $oldused - 1$ do \\
\hspace{6mm}7.1.1  $b_n \leftarrow 0$ \\
8.  $b.sign \leftarrow a.sign$ \\
9.  Clamp excess digits of $b$.  (\textit{mp\_clamp}) \\
10.  Return(\textit{MP\_OKAY}).\\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_div\_2}
\end{figure}

\textbf{Algorithm mp\_div\_2.}
This algorithm will divide an mp\_int by two using logical shifts to the right.  Like mp\_mul\_2 it uses a modified low level addition
core as the basis of the algorithm.  Unlike mp\_mul\_2 the shift operations work from the leading digit to the trailing digit.  The algorithm
could be written to work from the trailing digit to the leading digit however, it would have to stop one short of $a.used - 1$ digits to prevent
reading past the end of the array of digits.

Essentially the loop at step 6 is similar to that of mp\_mul\_2 except the logical shifts go in the opposite direction and the carry is at the
least significant bit not the most significant bit.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_div\_2.c
\vspace{-3mm}
\begin{alltt}
016
017   /* b = a/2 */
018   int mp_div_2(mp_int * a, mp_int * b)
019   \{
020     int     x, res, oldused;
021
022     /* copy */
023     if (b->alloc < a->used) \{
024       if ((res = mp_grow (b, a->used)) != MP_OKAY) \{
025         return res;
026       \}
027     \}
028
029     oldused = b->used;
030     b->used = a->used;
031     \{
032       mp_digit r, rr, *tmpa, *tmpb;
033
034       /* source alias */
035       tmpa = a->dp + b->used - 1;
036
037       /* dest alias */
038       tmpb = b->dp + b->used - 1;
039
040       /* carry */
041       r = 0;
042       for (x = b->used - 1; x >= 0; x--) \{
043         /* get the carry for the next iteration */
044         rr = *tmpa & 1;
045
046         /* shift the current digit, add in carry and store */
047         *tmpb-- = (*tmpa-- >> 1) | (r << (DIGIT_BIT - 1));
048
049         /* forward carry to next iteration */
050         r = rr;
051       \}
052
053       /* zero excess digits */
054       tmpb = b->dp + b->used;
055       for (x = b->used; x < oldused; x++) \{
056         *tmpb++ = 0;
057       \}
058     \}
059     b->sign = a->sign;
060     mp_clamp (b);
061     return MP_OKAY;
062   \}
063   #endif
064
\end{alltt}
\end{small}

\section{Polynomial Basis Operations}
Recall from section 4.3 that any integer can be represented as a polynomial in $x$ as $y = f(\beta)$.  Such a representation is also known as
the polynomial basis \cite[pp. 48]{ROSE}. Given such a notation a multiplication or division by $x$ amounts to shifting whole digits a single
place.  The need for such operations arises in several other higher level algorithms such as Barrett and Montgomery reduction, integer
division and Karatsuba multiplication.

Converting from an array of digits to polynomial basis is very simple.  Consider the integer $y \equiv (a_2, a_1, a_0)_{\beta}$ and recall that
$y = \sum_{i=0}^{2} a_i \beta^i$.  Simply replace $\beta$ with $x$ and the expression is in polynomial basis.  For example, $f(x) = 8x + 9$ is the
polynomial basis representation for $89$ using radix ten.  That is, $f(10) = 8(10) + 9 = 89$.

\subsection{Multiplication by $x$}

Given a polynomial in $x$ such as $f(x) = a_n x^n + a_{n-1} x^{n-1} + ... + a_0$ multiplying by $x$ amounts to shifting the coefficients up one
degree.  In this case $f(x) \cdot x = a_n x^{n+1} + a_{n-1} x^n + ... + a_0 x$.  From a scalar basis point of view multiplying by $x$ is equivalent to
multiplying by the integer $\beta$.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_lshd}. \\
\textbf{Input}.   One mp\_int $a$ and an integer $b$ \\
\textbf{Output}.  $a \leftarrow a \cdot \beta^b$ (equivalent to multiplication by $x^b$). \\
\hline \\
1.  If $b \le 0$ then return(\textit{MP\_OKAY}). \\
2.  If $a.alloc < a.used + b$ then grow $a$ to at least $a.used + b$ digits.  (\textit{mp\_grow}). \\
3.  If the reallocation failed return(\textit{MP\_MEM}). \\
4.  $a.used \leftarrow a.used + b$ \\
5.  $i \leftarrow a.used - 1$ \\
6.  $j \leftarrow a.used - 1 - b$ \\
7.  for $n$ from $a.used - 1$ to $b$ do \\
\hspace{3mm}7.1  $a_{i} \leftarrow a_{j}$ \\
\hspace{3mm}7.2  $i \leftarrow i - 1$ \\
\hspace{3mm}7.3  $j \leftarrow j - 1$ \\
8.  for $n$ from 0 to $b - 1$ do \\
\hspace{3mm}8.1  $a_n \leftarrow 0$ \\
9.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_lshd}
\end{figure}

\textbf{Algorithm mp\_lshd.}
This algorithm multiplies an mp\_int by the $b$'th power of $x$.  This is equivalent to multiplying by $\beta^b$.  The algorithm differs
from the other algorithms presented so far as it performs the operation in place instead storing the result in a separate location.  The
motivation behind this change is due to the way this function is typically used.  Algorithms such as mp\_add store the result in an optionally
different third mp\_int because the original inputs are often still required.  Algorithm mp\_lshd (\textit{and similarly algorithm mp\_rshd}) is
typically used on values where the original value is no longer required.  The algorithm will return success immediately if
$b \le 0$ since the rest of algorithm is only valid when $b > 0$.

First the destination $a$ is grown as required to accomodate the result.  The counters $i$ and $j$ are used to form a \textit{sliding window} over
the digits of $a$ of length $b$.  The head of the sliding window is at $i$ (\textit{the leading digit}) and the tail at $j$ (\textit{the trailing digit}).
The loop on step 7 copies the digit from the tail to the head.  In each iteration the window is moved down one digit.   The last loop on
step 8 sets the lower $b$ digits to zero.

\newpage
\begin{center}
\begin{figure}[here]
\includegraphics{pics/sliding_window.ps}
\caption{Sliding Window Movement}
\label{pic:sliding_window}
\end{figure}
\end{center}

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_lshd.c
\vspace{-3mm}
\begin{alltt}
016
017   /* shift left a certain amount of digits */
018   int mp_lshd (mp_int * a, int b)
019   \{
020     int     x, res;
021
022     /* if its less than zero return */
023     if (b <= 0) \{
024       return MP_OKAY;
025     \}
026
027     /* grow to fit the new digits */
028     if (a->alloc < (a->used + b)) \{
029        if ((res = mp_grow (a, a->used + b)) != MP_OKAY) \{
030          return res;
031        \}
032     \}
033
034     \{
035       mp_digit *top, *bottom;
036
037       /* increment the used by the shift amount then copy upwards */
038       a->used += b;
039
040       /* top */
041       top = a->dp + a->used - 1;
042
043       /* base */
044       bottom = (a->dp + a->used - 1) - b;
045
046       /* much like mp_rshd this is implemented using a sliding window
047        * except the window goes the otherway around.  Copying from
048        * the bottom to the top.  see bn_mp_rshd.c for more info.
049        */
050       for (x = a->used - 1; x >= b; x--) \{
051         *top-- = *bottom--;
052       \}
053
054       /* zero the lower digits */
055       top = a->dp;
056       for (x = 0; x < b; x++) \{
057         *top++ = 0;
058       \}
059     \}
060     return MP_OKAY;
061   \}
062   #endif
063
\end{alltt}
\end{small}

The if statement (line 23) ensures that the $b$ variable is greater than zero since we do not interpret negative
shift counts properly.  The \textbf{used} count is incremented by $b$ before the copy loop begins.  This elminates
the need for an additional variable in the for loop.  The variable $top$ (line 41) is an alias
for the leading digit while $bottom$ (line 44) is an alias for the trailing edge.  The aliases form a
window of exactly $b$ digits over the input.

\subsection{Division by $x$}

Division by powers of $x$ is easily achieved by shifting the digits right and removing any that will end up to the right of the zero'th digit.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_rshd}. \\
\textbf{Input}.   One mp\_int $a$ and an integer $b$ \\
\textbf{Output}.  $a \leftarrow a / \beta^b$ (Divide by $x^b$). \\
\hline \\
1.  If $b \le 0$ then return. \\
2.  If $a.used \le b$ then do \\
\hspace{3mm}2.1  Zero $a$.  (\textit{mp\_zero}). \\
\hspace{3mm}2.2  Return. \\
3.  $i \leftarrow 0$ \\
4.  $j \leftarrow b$ \\
5.  for $n$ from 0 to $a.used - b - 1$ do \\
\hspace{3mm}5.1  $a_i \leftarrow a_j$ \\
\hspace{3mm}5.2  $i \leftarrow i + 1$ \\
\hspace{3mm}5.3  $j \leftarrow j + 1$ \\
6.  for $n$ from $a.used - b$ to $a.used - 1$ do \\
\hspace{3mm}6.1  $a_n \leftarrow 0$ \\
7.  $a.used \leftarrow a.used - b$ \\
8.  Return. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_rshd}
\end{figure}

\textbf{Algorithm mp\_rshd.}
This algorithm divides the input in place by the $b$'th power of $x$.  It is analogous to dividing by a $\beta^b$ but much quicker since
it does not require single precision division.  This algorithm does not actually return an error code as it cannot fail.

If the input $b$ is less than one the algorithm quickly returns without performing any work.  If the \textbf{used} count is less than or equal
to the shift count $b$ then it will simply zero the input and return.

After the trivial cases of inputs have been handled the sliding window is setup.  Much like the case of algorithm mp\_lshd a sliding window that
is $b$ digits wide is used to copy the digits.  Unlike mp\_lshd the window slides in the opposite direction from the trailing to the leading digit.
Also the digits are copied from the leading to the trailing edge.

Once the window copy is complete the upper digits must be zeroed and the \textbf{used} count decremented.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_rshd.c
\vspace{-3mm}
\begin{alltt}
016
017   /* shift right a certain amount of digits */
018   void mp_rshd (mp_int * a, int b)
019   \{
020     int     x;
021
022     /* if b <= 0 then ignore it */
023     if (b <= 0) \{
024       return;
025     \}
026
027     /* if b > used then simply zero it and return */
028     if (a->used <= b) \{
029       mp_zero (a);
030       return;
031     \}
032
033     \{
034       mp_digit *bottom, *top;
035
036       /* shift the digits down */
037
038       /* bottom */
039       bottom = a->dp;
040
041       /* top [offset into digits] */
042       top = a->dp + b;
043
044       /* this is implemented as a sliding window where
045        * the window is b-digits long and digits from
046        * the top of the window are copied to the bottom
047        *
048        * e.g.
049
050        b-2 | b-1 | b0 | b1 | b2 | ... | bb |   ---->
051                    /\symbol{92}                   |      ---->
052                     \symbol{92}-------------------/      ---->
053        */
054       for (x = 0; x < (a->used - b); x++) \{
055         *bottom++ = *top++;
056       \}
057
058       /* zero the top digits */
059       for (; x < a->used; x++) \{
060         *bottom++ = 0;
061       \}
062     \}
063
064     /* remove excess digits */
065     a->used -= b;
066   \}
067   #endif
068
\end{alltt}
\end{small}

The only noteworthy element of this routine is the lack of a return type since it cannot fail.  Like mp\_lshd() we
form a sliding window except we copy in the other direction.  After the window (line 59) we then zero
the upper digits of the input to make sure the result is correct.

\section{Powers of Two}

Now that algorithms for moving single bits as well as whole digits exist algorithms for moving the ``in between'' distances are required.  For
example, to quickly multiply by $2^k$ for any $k$ without using a full multiplier algorithm would prove useful.  Instead of performing single
shifts $k$ times to achieve a multiplication by $2^{\pm k}$ a mixture of whole digit shifting and partial digit shifting is employed.

\subsection{Multiplication by Power of Two}

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_mul\_2d}. \\
\textbf{Input}.   One mp\_int $a$ and an integer $b$ \\
\textbf{Output}.  $c \leftarrow a \cdot 2^b$. \\
\hline \\
1.  $c \leftarrow a$.  (\textit{mp\_copy}) \\
2.  If $c.alloc < c.used + \lfloor b / lg(\beta) \rfloor + 2$ then grow $c$ accordingly. \\
3.  If the reallocation failed return(\textit{MP\_MEM}). \\
4.  If $b \ge lg(\beta)$ then \\
\hspace{3mm}4.1  $c \leftarrow c \cdot \beta^{\lfloor b / lg(\beta) \rfloor}$ (\textit{mp\_lshd}). \\
\hspace{3mm}4.2  If step 4.1 failed return(\textit{MP\_MEM}). \\
5.  $d \leftarrow b \mbox{ (mod }lg(\beta)\mbox{)}$ \\
6.  If $d \ne 0$ then do \\
\hspace{3mm}6.1  $mask \leftarrow 2^d$ \\
\hspace{3mm}6.2  $r \leftarrow 0$ \\
\hspace{3mm}6.3  for $n$ from $0$ to $c.used - 1$ do \\
\hspace{6mm}6.3.1  $rr \leftarrow c_n >> (lg(\beta) - d) \mbox{ (mod }mask\mbox{)}$ \\
\hspace{6mm}6.3.2  $c_n \leftarrow (c_n << d) + r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{6mm}6.3.3  $r \leftarrow rr$ \\
\hspace{3mm}6.4  If $r > 0$ then do \\
\hspace{6mm}6.4.1  $c_{c.used} \leftarrow r$ \\
\hspace{6mm}6.4.2  $c.used \leftarrow c.used + 1$ \\
7.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_mul\_2d}
\end{figure}

\textbf{Algorithm mp\_mul\_2d.}
This algorithm multiplies $a$ by $2^b$ and stores the result in $c$.  The algorithm uses algorithm mp\_lshd and a derivative of algorithm mp\_mul\_2 to
quickly compute the product.

First the algorithm will multiply $a$ by $x^{\lfloor b / lg(\beta) \rfloor}$ which will ensure that the remainder multiplicand is less than
$\beta$.  For example, if $b = 37$ and $\beta = 2^{28}$ then this step will multiply by $x$ leaving a multiplication by $2^{37 - 28} = 2^{9}$
left.

After the digits have been shifted appropriately at most $lg(\beta) - 1$ shifts are left to perform.  Step 5 calculates the number of remaining shifts
required.  If it is non-zero a modified shift loop is used to calculate the remaining product.
Essentially the loop is a generic version of algorithm mp\_mul\_2 designed to handle any shift count in the range $1 \le x < lg(\beta)$.  The $mask$
variable is used to extract the upper $d$ bits to form the carry for the next iteration.

This algorithm is loosely measured as a $O(2n)$ algorithm which means that if the input is $n$-digits that it takes $2n$ ``time'' to
complete.  It is possible to optimize this algorithm down to a $O(n)$ algorithm at a cost of making the algorithm slightly harder to follow.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_mul\_2d.c
\vspace{-3mm}
\begin{alltt}
016
017   /* shift left by a certain bit count */
018   int mp_mul_2d (mp_int * a, int b, mp_int * c)
019   \{
020     mp_digit d;
021     int      res;
022
023     /* copy */
024     if (a != c) \{
025        if ((res = mp_copy (a, c)) != MP_OKAY) \{
026          return res;
027        \}
028     \}
029
030     if (c->alloc < (int)(c->used + (b / DIGIT_BIT) + 1)) \{
031        if ((res = mp_grow (c, c->used + (b / DIGIT_BIT) + 1)) != MP_OKAY) \{
032          return res;
033        \}
034     \}
035
036     /* shift by as many digits in the bit count */
037     if (b >= (int)DIGIT_BIT) \{
038       if ((res = mp_lshd (c, b / DIGIT_BIT)) != MP_OKAY) \{
039         return res;
040       \}
041     \}
042
043     /* shift any bit count < DIGIT_BIT */
044     d = (mp_digit) (b % DIGIT_BIT);
045     if (d != 0) \{
046       mp_digit *tmpc, shift, mask, r, rr;
047       int x;
048
049       /* bitmask for carries */
050       mask = (((mp_digit)1) << d) - 1;
051
052       /* shift for msbs */
053       shift = DIGIT_BIT - d;
054
055       /* alias */
056       tmpc = c->dp;
057
058       /* carry */
059       r    = 0;
060       for (x = 0; x < c->used; x++) \{
061         /* get the higher bits of the current word */
062         rr = (*tmpc >> shift) & mask;
063
064         /* shift the current word and OR in the carry */
065         *tmpc = ((*tmpc << d) | r) & MP_MASK;
066         ++tmpc;
067
068         /* set the carry to the carry bits of the current word */
069         r = rr;
070       \}
071
072       /* set final carry */
073       if (r != 0) \{
074          c->dp[(c->used)++] = r;
075       \}
076     \}
077     mp_clamp (c);
078     return MP_OKAY;
079   \}
080   #endif
081
\end{alltt}
\end{small}

The shifting is performed in--place which means the first step (line 24) is to copy the input to the
destination.  We avoid calling mp\_copy() by making sure the mp\_ints are different.  The destination then
has to be grown (line 31) to accomodate the result.

If the shift count $b$ is larger than $lg(\beta)$ then a call to mp\_lshd() is used to handle all of the multiples
of $lg(\beta)$.  Leaving only a remaining shift of $lg(\beta) - 1$ or fewer bits left.  Inside the actual shift
loop (lines 45 to 76) we make use of pre--computed values $shift$ and $mask$.   These are used to
extract the carry bit(s) to pass into the next iteration of the loop.  The $r$ and $rr$ variables form a
chain between consecutive iterations to propagate the carry.

\subsection{Division by Power of Two}

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_div\_2d}. \\
\textbf{Input}.   One mp\_int $a$ and an integer $b$ \\
\textbf{Output}.  $c \leftarrow \lfloor a / 2^b \rfloor, d \leftarrow a \mbox{ (mod }2^b\mbox{)}$. \\
\hline \\
1.  If $b \le 0$ then do \\
\hspace{3mm}1.1  $c \leftarrow a$ (\textit{mp\_copy}) \\
\hspace{3mm}1.2  $d \leftarrow 0$ (\textit{mp\_zero}) \\
\hspace{3mm}1.3  Return(\textit{MP\_OKAY}). \\
2.  $c \leftarrow a$ \\
3.  $d \leftarrow a \mbox{ (mod }2^b\mbox{)}$ (\textit{mp\_mod\_2d}) \\
4.  If $b \ge lg(\beta)$ then do \\
\hspace{3mm}4.1  $c \leftarrow \lfloor c/\beta^{\lfloor b/lg(\beta) \rfloor} \rfloor$ (\textit{mp\_rshd}). \\
5.  $k \leftarrow b \mbox{ (mod }lg(\beta)\mbox{)}$ \\
6.  If $k \ne 0$ then do \\
\hspace{3mm}6.1  $mask \leftarrow 2^k$ \\
\hspace{3mm}6.2  $r \leftarrow 0$ \\
\hspace{3mm}6.3  for $n$ from $c.used - 1$ to $0$ do \\
\hspace{6mm}6.3.1  $rr \leftarrow c_n \mbox{ (mod }mask\mbox{)}$ \\
\hspace{6mm}6.3.2  $c_n \leftarrow (c_n >> k) + (r << (lg(\beta) - k))$ \\
\hspace{6mm}6.3.3  $r \leftarrow rr$ \\
7.  Clamp excess digits of $c$.  (\textit{mp\_clamp}) \\
8.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_div\_2d}
\end{figure}

\textbf{Algorithm mp\_div\_2d.}
This algorithm will divide an input $a$ by $2^b$ and produce the quotient and remainder.  The algorithm is designed much like algorithm
mp\_mul\_2d by first using whole digit shifts then single precision shifts.  This algorithm will also produce the remainder of the division
by using algorithm mp\_mod\_2d.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_div\_2d.c
\vspace{-3mm}
\begin{alltt}
016
017   /* shift right by a certain bit count (store quotient in c, optional remaind
      er in d) */
018   int mp_div_2d (mp_int * a, int b, mp_int * c, mp_int * d)
019   \{
020     mp_digit D, r, rr;
021     int     x, res;
022     mp_int  t;
023
024
025     /* if the shift count is <= 0 then we do no work */
026     if (b <= 0) \{
027       res = mp_copy (a, c);
028       if (d != NULL) \{
029         mp_zero (d);
030       \}
031       return res;
032     \}
033
034     if ((res = mp_init (&t)) != MP_OKAY) \{
035       return res;
036     \}
037
038     /* get the remainder */
039     if (d != NULL) \{
040       if ((res = mp_mod_2d (a, b, &t)) != MP_OKAY) \{
041         mp_clear (&t);
042         return res;
043       \}
044     \}
045
046     /* copy */
047     if ((res = mp_copy (a, c)) != MP_OKAY) \{
048       mp_clear (&t);
049       return res;
050     \}
051
052     /* shift by as many digits in the bit count */
053     if (b >= (int)DIGIT_BIT) \{
054       mp_rshd (c, b / DIGIT_BIT);
055     \}
056
057     /* shift any bit count < DIGIT_BIT */
058     D = (mp_digit) (b % DIGIT_BIT);
059     if (D != 0) \{
060       mp_digit *tmpc, mask, shift;
061
062       /* mask */
063       mask = (((mp_digit)1) << D) - 1;
064
065       /* shift for lsb */
066       shift = DIGIT_BIT - D;
067
068       /* alias */
069       tmpc = c->dp + (c->used - 1);
070
071       /* carry */
072       r = 0;
073       for (x = c->used - 1; x >= 0; x--) \{
074         /* get the lower  bits of this word in a temp */
075         rr = *tmpc & mask;
076
077         /* shift the current word and mix in the carry bits from the previous
      word */
078         *tmpc = (*tmpc >> D) | (r << shift);
079         --tmpc;
080
081         /* set the carry to the carry bits of the current word found above */
082         r = rr;
083       \}
084     \}
085     mp_clamp (c);
086     if (d != NULL) \{
087       mp_exch (&t, d);
088     \}
089     mp_clear (&t);
090     return MP_OKAY;
091   \}
092   #endif
093
\end{alltt}
\end{small}

The implementation of algorithm mp\_div\_2d is slightly different than the algorithm specifies.  The remainder $d$ may be optionally
ignored by passing \textbf{NULL} as the pointer to the mp\_int variable.    The temporary mp\_int variable $t$ is used to hold the
result of the remainder operation until the end.  This allows $d$ and $a$ to represent the same mp\_int without modifying $a$ before
the quotient is obtained.

The remainder of the source code is essentially the same as the source code for mp\_mul\_2d.  The only significant difference is
the direction of the shifts.

\subsection{Remainder of Division by Power of Two}

The last algorithm in the series of polynomial basis power of two algorithms is calculating the remainder of division by $2^b$.  This
algorithm benefits from the fact that in twos complement arithmetic $a \mbox{ (mod }2^b\mbox{)}$ is the same as $a$ AND $2^b - 1$.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_mod\_2d}. \\
\textbf{Input}.   One mp\_int $a$ and an integer $b$ \\
\textbf{Output}.  $c \leftarrow a \mbox{ (mod }2^b\mbox{)}$. \\
\hline \\
1.  If $b \le 0$ then do \\
\hspace{3mm}1.1  $c \leftarrow 0$ (\textit{mp\_zero}) \\
\hspace{3mm}1.2  Return(\textit{MP\_OKAY}). \\
2.  If $b > a.used \cdot lg(\beta)$ then do \\
\hspace{3mm}2.1  $c \leftarrow a$ (\textit{mp\_copy}) \\
\hspace{3mm}2.2  Return the result of step 2.1. \\
3.  $c \leftarrow a$ \\
4.  If step 3 failed return(\textit{MP\_MEM}). \\
5.  for $n$ from $\lceil b / lg(\beta) \rceil$ to $c.used$ do \\
\hspace{3mm}5.1  $c_n \leftarrow 0$ \\
6.  $k \leftarrow b \mbox{ (mod }lg(\beta)\mbox{)}$ \\
7.  $c_{\lfloor b / lg(\beta) \rfloor} \leftarrow c_{\lfloor b / lg(\beta) \rfloor} \mbox{ (mod }2^{k}\mbox{)}$. \\
8.  Clamp excess digits of $c$.  (\textit{mp\_clamp}) \\
9.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_mod\_2d}
\end{figure}

\textbf{Algorithm mp\_mod\_2d.}
This algorithm will quickly calculate the value of $a \mbox{ (mod }2^b\mbox{)}$.  First if $b$ is less than or equal to zero the
result is set to zero.  If $b$ is greater than the number of bits in $a$ then it simply copies $a$ to $c$ and returns.  Otherwise, $a$
is copied to $b$, leading digits are removed and the remaining leading digit is trimed to the exact bit count.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_mod\_2d.c
\vspace{-3mm}
\begin{alltt}
016
017   /* calc a value mod 2**b */
018   int
019   mp_mod_2d (mp_int * a, int b, mp_int * c)
020   \{
021     int     x, res;
022
023     /* if b is <= 0 then zero the int */
024     if (b <= 0) \{
025       mp_zero (c);
026       return MP_OKAY;
027     \}
028
029     /* if the modulus is larger than the value than return */
030     if (b >= (int) (a->used * DIGIT_BIT)) \{
031       res = mp_copy (a, c);
032       return res;
033     \}
034
035     /* copy */
036     if ((res = mp_copy (a, c)) != MP_OKAY) \{
037       return res;
038     \}
039
040     /* zero digits above the last digit of the modulus */
041     for (x = (b / DIGIT_BIT) + (((b % DIGIT_BIT) == 0) ? 0 : 1); x < c->used;
      x++) \{
042       c->dp[x] = 0;
043     \}
044     /* clear the digit that is not completely outside/inside the modulus */
045     c->dp[b / DIGIT_BIT] &=
046       (mp_digit) ((((mp_digit) 1) << (((mp_digit) b) % DIGIT_BIT)) - ((mp_digi
      t) 1));
047     mp_clamp (c);
048     return MP_OKAY;
049   \}
050   #endif
051
\end{alltt}
\end{small}

We first avoid cases of $b \le 0$ by simply mp\_zero()'ing the destination in such cases.  Next if $2^b$ is larger
than the input we just mp\_copy() the input and return right away.  After this point we know we must actually
perform some work to produce the remainder.

Recalling that reducing modulo $2^k$ and a binary ``and'' with $2^k - 1$ are numerically equivalent we can quickly reduce
the number.  First we zero any digits above the last digit in $2^b$ (line 41).  Next we reduce the
leading digit of both (line 45) and then mp\_clamp().

\section*{Exercises}
\begin{tabular}{cl}
$\left [ 3 \right ] $ & Devise an algorithm that performs $a \cdot 2^b$ for generic values of $b$ \\
                      & in $O(n)$ time. \\
                      &\\
$\left [ 3 \right ] $ & Devise an efficient algorithm to multiply by small low hamming  \\
                      & weight values such as $3$, $5$ and $9$.  Extend it to handle all values \\
                      & upto $64$ with a hamming weight less than three. \\
                      &\\
$\left [ 2 \right ] $ & Modify the preceding algorithm to handle values of the form \\
                      & $2^k - 1$ as well. \\
                      &\\
$\left [ 3 \right ] $ & Using only algorithms mp\_mul\_2, mp\_div\_2 and mp\_add create an \\
                      & algorithm to multiply two integers in roughly $O(2n^2)$ time for \\
                      & any $n$-bit input.  Note that the time of addition is ignored in the \\
                      & calculation.  \\
                      & \\
$\left [ 5 \right ] $ & Improve the previous algorithm to have a working time of at most \\
                      & $O \left (2^{(k-1)}n + \left ({2n^2 \over k} \right ) \right )$ for an appropriate choice of $k$.  Again ignore \\
                      & the cost of addition. \\
                      & \\
$\left [ 2 \right ] $ & Devise a chart to find optimal values of $k$ for the previous problem \\
                      & for $n = 64 \ldots 1024$ in steps of $64$. \\
                      & \\
$\left [ 2 \right ] $ & Using only algorithms mp\_abs and mp\_sub devise another method for \\
                      & calculating the result of a signed comparison. \\
                      &
\end{tabular}

\chapter{Multiplication and Squaring}
\section{The Multipliers}
For most number theoretic problems including certain public key cryptographic algorithms, the ``multipliers'' form the most important subset of
algorithms of any multiple precision integer package.  The set of multiplier algorithms include integer multiplication, squaring and modular reduction
where in each of the algorithms single precision multiplication is the dominant operation performed.  This chapter will discuss integer multiplication
and squaring, leaving modular reductions for the subsequent chapter.

The importance of the multiplier algorithms is for the most part driven by the fact that certain popular public key algorithms are based on modular
exponentiation, that is computing $d \equiv a^b \mbox{ (mod }c\mbox{)}$ for some arbitrary choice of $a$, $b$, $c$ and $d$.  During a modular
exponentiation the majority\footnote{Roughly speaking a modular exponentiation will spend about 40\% of the time performing modular reductions,
35\% of the time performing squaring and 25\% of the time performing multiplications.} of the processor time is spent performing single precision
multiplications.

For centuries general purpose multiplication has required a lengthly $O(n^2)$ process, whereby each digit of one multiplicand has to be multiplied
against every digit of the other multiplicand.  Traditional long-hand multiplication is based on this process;  while the techniques can differ the
overall algorithm used is essentially the same.  Only ``recently'' have faster algorithms been studied.  First Karatsuba multiplication was discovered in
1962.  This algorithm can multiply two numbers with considerably fewer single precision multiplications when compared to the long-hand approach.
This technique led to the discovery of polynomial basis algorithms (\textit{good reference?}) and subquently Fourier Transform based solutions.

\section{Multiplication}
\subsection{The Baseline Multiplication}
\label{sec:basemult}
\index{baseline multiplication}
Computing the product of two integers in software can be achieved using a trivial adaptation of the standard $O(n^2)$ long-hand multiplication
algorithm that school children are taught.  The algorithm is considered an $O(n^2)$ algorithm since for two $n$-digit inputs $n^2$ single precision
multiplications are required.  More specifically for a $m$ and $n$ digit input $m \cdot n$ single precision multiplications are required.  To
simplify most discussions, it will be assumed that the inputs have comparable number of digits.

The ``baseline multiplication'' algorithm is designed to act as the ``catch-all'' algorithm, only to be used when the faster algorithms cannot be
used.  This algorithm does not use any particularly interesting optimizations and should ideally be avoided if possible.    One important
facet of this algorithm, is that it has been modified to only produce a certain amount of output digits as resolution.  The importance of this
modification will become evident during the discussion of Barrett modular reduction.  Recall that for a $n$ and $m$ digit input the product
will be at most $n + m$ digits.  Therefore, this algorithm can be reduced to a full multiplier by having it produce $n + m$ digits of the product.

Recall from sub-section 4.2.2 the definition of $\gamma$ as the number of bits in the type \textbf{mp\_digit}.  We shall now extend the variable set to
include $\alpha$ which shall represent the number of bits in the type \textbf{mp\_word}.  This implies that $2^{\alpha} > 2 \cdot \beta^2$.  The
constant $\delta = 2^{\alpha - 2lg(\beta)}$ will represent the maximal weight of any column in a product (\textit{see sub-section 5.2.2 for more information}).

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{s\_mp\_mul\_digs}. \\
\textbf{Input}.   mp\_int $a$, mp\_int $b$ and an integer $digs$ \\
\textbf{Output}.  $c \leftarrow \vert a \vert \cdot \vert b \vert \mbox{ (mod }\beta^{digs}\mbox{)}$. \\
\hline \\
1.  If min$(a.used, b.used) < \delta$ then do \\
\hspace{3mm}1.1  Calculate $c = \vert a \vert \cdot \vert b \vert$ by the Comba method (\textit{see algorithm~\ref{fig:COMBAMULT}}).  \\
\hspace{3mm}1.2  Return the result of step 1.1 \\
\\
Allocate and initialize a temporary mp\_int. \\
2.  Init $t$ to be of size $digs$ \\
3.  If step 2 failed return(\textit{MP\_MEM}). \\
4.  $t.used \leftarrow digs$ \\
\\
Compute the product. \\
5.  for $ix$ from $0$ to $a.used - 1$ do \\
\hspace{3mm}5.1  $u \leftarrow 0$ \\
\hspace{3mm}5.2  $pb \leftarrow \mbox{min}(b.used, digs - ix)$ \\
\hspace{3mm}5.3  If $pb < 1$ then goto step 6. \\
\hspace{3mm}5.4  for $iy$ from $0$ to $pb - 1$ do \\
\hspace{6mm}5.4.1  $\hat r \leftarrow t_{iy + ix} + a_{ix} \cdot b_{iy} + u$ \\
\hspace{6mm}5.4.2  $t_{iy + ix} \leftarrow \hat r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{6mm}5.4.3  $u \leftarrow \lfloor \hat r / \beta \rfloor$ \\
\hspace{3mm}5.5  if $ix + pb < digs$ then do \\
\hspace{6mm}5.5.1  $t_{ix + pb} \leftarrow u$ \\
6.  Clamp excess digits of $t$. \\
7.  Swap $c$ with $t$ \\
8.  Clear $t$ \\
9.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm s\_mp\_mul\_digs}
\end{figure}

\textbf{Algorithm s\_mp\_mul\_digs.}
This algorithm computes the unsigned product of two inputs $a$ and $b$, limited to an output precision of $digs$ digits.  While it may seem
a bit awkward to modify the function from its simple $O(n^2)$ description, the usefulness of partial multipliers will arise in a subsequent
algorithm.  The algorithm is loosely based on algorithm 14.12 from \cite[pp. 595]{HAC} and is similar to Algorithm M of Knuth \cite[pp. 268]{TAOCPV2}.
Algorithm s\_mp\_mul\_digs differs from these cited references since it can produce a variable output precision regardless of the precision of the
inputs.

The first thing this algorithm checks for is whether a Comba multiplier can be used instead.   If the minimum digit count of either
input is less than $\delta$, then the Comba method may be used instead.    After the Comba method is ruled out, the baseline algorithm begins.  A
temporary mp\_int variable $t$ is used to hold the intermediate result of the product.  This allows the algorithm to be used to
compute products when either $a = c$ or $b = c$ without overwriting the inputs.

All of step 5 is the infamous $O(n^2)$ multiplication loop slightly modified to only produce upto $digs$ digits of output.  The $pb$ variable
is given the count of digits to read from $b$ inside the nested loop.  If $pb \le 1$ then no more output digits can be produced and the algorithm
will exit the loop.  The best way to think of the loops are as a series of $pb \times 1$ multiplications.    That is, in each pass of the
innermost loop $a_{ix}$ is multiplied against $b$ and the result is added (\textit{with an appropriate shift}) to $t$.

For example, consider multiplying $576$ by $241$.  That is equivalent to computing $10^0(1)(576) + 10^1(4)(576) + 10^2(2)(576)$ which is best
visualized in the following table.

\begin{figure}[here]
\begin{center}
\begin{tabular}{|c|c|c|c|c|c|l|}
\hline   &&          & 5 & 7 & 6 & \\
\hline   $\times$&&  & 2 & 4 & 1 & \\
\hline &&&&&&\\
  &&          & 5 & 7 & 6 & $10^0(1)(576)$ \\
  &2 &   3    & 6 & 1 & 6 & $10^1(4)(576) + 10^0(1)(576)$ \\
  1 & 3 & 8 & 8 & 1 & 6 &   $10^2(2)(576) + 10^1(4)(576) + 10^0(1)(576)$ \\
\hline
\end{tabular}
\end{center}
\caption{Long-Hand Multiplication Diagram}
\end{figure}

Each row of the product is added to the result after being shifted to the left (\textit{multiplied by a power of the radix}) by the appropriate
count.  That is in pass $ix$ of the inner loop the product is added starting at the $ix$'th digit of the reult.

Step 5.4.1 introduces the hat symbol (\textit{e.g. $\hat r$}) which represents a double precision variable.  The multiplication on that step
is assumed to be a double wide output single precision multiplication.  That is, two single precision variables are multiplied to produce a
double precision result.  The step is somewhat optimized from a long-hand multiplication algorithm because the carry from the addition in step
5.4.1 is propagated through the nested loop.  If the carry was not propagated immediately it would overflow the single precision digit
$t_{ix+iy}$ and the result would be lost.

At step 5.5 the nested loop is finished and any carry that was left over should be forwarded.  The carry does not have to be added to the $ix+pb$'th
digit since that digit is assumed to be zero at this point.  However, if $ix + pb \ge digs$ the carry is not set as it would make the result
exceed the precision requested.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_s\_mp\_mul\_digs.c
\vspace{-3mm}
\begin{alltt}
016
017   /* multiplies |a| * |b| and only computes upto digs digits of result
018    * HAC pp. 595, Algorithm 14.12  Modified so you can control how
019    * many digits of output are created.
020    */
021   int s_mp_mul_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
022   \{
023     mp_int  t;
024     int     res, pa, pb, ix, iy;
025     mp_digit u;
026     mp_word r;
027     mp_digit tmpx, *tmpt, *tmpy;
028
029     /* can we use the fast multiplier? */
030     if (((digs) < MP_WARRAY) &&
031         (MIN (a->used, b->used) <
032             (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) \{
033       return fast_s_mp_mul_digs (a, b, c, digs);
034     \}
035
036     if ((res = mp_init_size (&t, digs)) != MP_OKAY) \{
037       return res;
038     \}
039     t.used = digs;
040
041     /* compute the digits of the product directly */
042     pa = a->used;
043     for (ix = 0; ix < pa; ix++) \{
044       /* set the carry to zero */
045       u = 0;
046
047       /* limit ourselves to making digs digits of output */
048       pb = MIN (b->used, digs - ix);
049
050       /* setup some aliases */
051       /* copy of the digit from a used within the nested loop */
052       tmpx = a->dp[ix];
053
054       /* an alias for the destination shifted ix places */
055       tmpt = t.dp + ix;
056
057       /* an alias for the digits of b */
058       tmpy = b->dp;
059
060       /* compute the columns of the output and propagate the carry */
061       for (iy = 0; iy < pb; iy++) \{
062         /* compute the column as a mp_word */
063         r       = (mp_word)*tmpt +
064                   ((mp_word)tmpx * (mp_word)*tmpy++) +
065                   (mp_word)u;
066
067         /* the new column is the lower part of the result */
068         *tmpt++ = (mp_digit) (r & ((mp_word) MP_MASK));
069
070         /* get the carry word from the result */
071         u       = (mp_digit) (r >> ((mp_word) DIGIT_BIT));
072       \}
073       /* set carry if it is placed below digs */
074       if ((ix + iy) < digs) \{
075         *tmpt = u;
076       \}
077     \}
078
079     mp_clamp (&t);
080     mp_exch (&t, c);
081
082     mp_clear (&t);
083     return MP_OKAY;
084   \}
085   #endif
086
\end{alltt}
\end{small}

First we determine (line 30) if the Comba method can be used first since it's faster.  The conditions for
sing the Comba routine are that min$(a.used, b.used) < \delta$ and the number of digits of output is less than
\textbf{MP\_WARRAY}.  This new constant is used to control the stack usage in the Comba routines.  By default it is
set to $\delta$ but can be reduced when memory is at a premium.

If we cannot use the Comba method we proceed to setup the baseline routine.  We allocate the the destination mp\_int
$t$ (line 36) to the exact size of the output to avoid further re--allocations.  At this point we now
begin the $O(n^2)$ loop.

This implementation of multiplication has the caveat that it can be trimmed to only produce a variable number of
digits as output.  In each iteration of the outer loop the $pb$ variable is set (line 48) to the maximum
number of inner loop iterations.

Inside the inner loop we calculate $\hat r$ as the mp\_word product of the two mp\_digits and the addition of the
carry from the previous iteration.  A particularly important observation is that most modern optimizing
C compilers (GCC for instance) can recognize that a $N \times N \rightarrow 2N$ multiplication is all that
is required for the product.  In x86 terms for example, this means using the MUL instruction.

Each digit of the product is stored in turn (line 68) and the carry propagated (line 71) to the
next iteration.

\subsection{Faster Multiplication by the ``Comba'' Method}

One of the huge drawbacks of the ``baseline'' algorithms is that at the $O(n^2)$ level the carry must be
computed and propagated upwards.  This makes the nested loop very sequential and hard to unroll and implement
in parallel.  The ``Comba'' \cite{COMBA} method is named after little known (\textit{in cryptographic venues}) Paul G.
Comba who described a method of implementing fast multipliers that do not require nested carry fixup operations.  As an
interesting aside it seems that Paul Barrett describes a similar technique in his 1986 paper \cite{BARRETT} written
five years before.

At the heart of the Comba technique is once again the long-hand algorithm.  Except in this case a slight
twist is placed on how the columns of the result are produced.  In the standard long-hand algorithm rows of products
are produced then added together to form the final result.  In the baseline algorithm the columns are added together
after each iteration to get the result instantaneously.

In the Comba algorithm the columns of the result are produced entirely independently of each other.  That is at
the $O(n^2)$ level a simple multiplication and addition step is performed.  The carries of the columns are propagated
after the nested loop to reduce the amount of work requiored. Succintly the first step of the algorithm is to compute
the product vector $\vec x$ as follows.

\begin{equation}
\vec x_n = \sum_{i+j = n} a_ib_j, \forall n \in \lbrace 0, 1, 2, \ldots, i + j \rbrace
\end{equation}

Where $\vec x_n$ is the $n'th$ column of the output vector.  Consider the following example which computes the vector $\vec x$ for the multiplication
of $576$ and $241$.

\newpage\begin{figure}[here]
\begin{small}
\begin{center}
\begin{tabular}{|c|c|c|c|c|c|}
  \hline &          & 5 & 7 & 6 & First Input\\
  \hline $\times$ & & 2 & 4 & 1 & Second Input\\
\hline            &                        & $1 \cdot 5 = 5$   & $1 \cdot 7 = 7$   & $1 \cdot 6 = 6$ & First pass \\
                  &  $4 \cdot 5 = 20$      & $4 \cdot 7+5=33$  & $4 \cdot 6+7=31$  & 6               & Second pass \\
   $2 \cdot 5 = 10$ &  $2 \cdot 7 + 20 = 34$ & $2 \cdot 6+33=45$ & 31                & 6             & Third pass \\
\hline 10 & 34 & 45 & 31 & 6 & Final Result \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Comba Multiplication Diagram}
\end{figure}

At this point the vector $x = \left < 10, 34, 45, 31, 6 \right >$ is the result of the first step of the Comba multipler.
Now the columns must be fixed by propagating the carry upwards.  The resultant vector will have one extra dimension over the input vector which is
congruent to adding a leading zero digit.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Comba Fixup}. \\
\textbf{Input}.   Vector $\vec x$ of dimension $k$ \\
\textbf{Output}.  Vector $\vec x$ such that the carries have been propagated. \\
\hline \\
1.  for $n$ from $0$ to $k - 1$ do \\
\hspace{3mm}1.1 $\vec x_{n+1} \leftarrow \vec x_{n+1} + \lfloor \vec x_{n}/\beta \rfloor$ \\
\hspace{3mm}1.2 $\vec x_{n} \leftarrow \vec x_{n} \mbox{ (mod }\beta\mbox{)}$ \\
2.  Return($\vec x$). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Comba Fixup}
\end{figure}

With that algorithm and $k = 5$ and $\beta = 10$ the following vector is produced $\vec x= \left < 1, 3, 8, 8, 1, 6 \right >$.  In this case
$241 \cdot 576$ is in fact $138816$ and the procedure succeeded.  If the algorithm is correct and as will be demonstrated shortly more
efficient than the baseline algorithm why not simply always use this algorithm?

\subsubsection{Column Weight.}
At the nested $O(n^2)$ level the Comba method adds the product of two single precision variables to each column of the output
independently.  A serious obstacle is if the carry is lost, due to lack of precision before the algorithm has a chance to fix
the carries.  For example, in the multiplication of two three-digit numbers the third column of output will be the sum of
three single precision multiplications.  If the precision of the accumulator for the output digits is less then $3 \cdot (\beta - 1)^2$ then
an overflow can occur and the carry information will be lost.  For any $m$ and $n$ digit inputs the maximum weight of any column is
min$(m, n)$ which is fairly obvious.

The maximum number of terms in any column of a product is known as the ``column weight'' and strictly governs when the algorithm can be used.  Recall
from earlier that a double precision type has $\alpha$ bits of resolution and a single precision digit has $lg(\beta)$ bits of precision.  Given these
two quantities we must not violate the following

\begin{equation}
k \cdot \left (\beta - 1 \right )^2 < 2^{\alpha}
\end{equation}

Which reduces to

\begin{equation}
k \cdot \left ( \beta^2 - 2\beta + 1 \right ) < 2^{\alpha}
\end{equation}

Let $\rho = lg(\beta)$ represent the number of bits in a single precision digit.  By further re-arrangement of the equation the final solution is
found.

\begin{equation}
k  < {{2^{\alpha}} \over {\left (2^{2\rho} - 2^{\rho + 1} + 1 \right )}}
\end{equation}

The defaults for LibTomMath are $\beta = 2^{28}$ and $\alpha = 2^{64}$ which means that $k$ is bounded by $k < 257$.  In this configuration
the smaller input may not have more than $256$ digits if the Comba method is to be used.  This is quite satisfactory for most applications since
$256$ digits would allow for numbers in the range of $0 \le x < 2^{7168}$ which, is much larger than most public key cryptographic algorithms require.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{fast\_s\_mp\_mul\_digs}. \\
\textbf{Input}.   mp\_int $a$, mp\_int $b$ and an integer $digs$ \\
\textbf{Output}.  $c \leftarrow \vert a \vert \cdot \vert b \vert \mbox{ (mod }\beta^{digs}\mbox{)}$. \\
\hline \\
Place an array of \textbf{MP\_WARRAY} single precision digits named $W$ on the stack. \\
1.  If $c.alloc < digs$ then grow $c$ to $digs$ digits. (\textit{mp\_grow}) \\
2.  If step 1 failed return(\textit{MP\_MEM}).\\
\\
3.  $pa \leftarrow \mbox{MIN}(digs, a.used + b.used)$ \\
\\
4.  $\_ \hat W \leftarrow 0$ \\
5.  for $ix$ from 0 to $pa - 1$ do \\
\hspace{3mm}5.1  $ty \leftarrow \mbox{MIN}(b.used - 1, ix)$ \\
\hspace{3mm}5.2  $tx \leftarrow ix - ty$ \\
\hspace{3mm}5.3  $iy \leftarrow \mbox{MIN}(a.used - tx, ty + 1)$ \\
\hspace{3mm}5.4  for $iz$ from 0 to $iy - 1$ do \\
\hspace{6mm}5.4.1  $\_ \hat W \leftarrow \_ \hat W + a_{tx+iy}b_{ty-iy}$ \\
\hspace{3mm}5.5  $W_{ix} \leftarrow \_ \hat W (\mbox{mod }\beta)$\\
\hspace{3mm}5.6  $\_ \hat W \leftarrow \lfloor \_ \hat W / \beta \rfloor$ \\
\\
6.  $oldused \leftarrow c.used$ \\
7.  $c.used \leftarrow digs$ \\
8.  for $ix$ from $0$ to $pa$ do \\
\hspace{3mm}8.1  $c_{ix} \leftarrow W_{ix}$ \\
9.  for $ix$ from $pa + 1$ to $oldused - 1$ do \\
\hspace{3mm}9.1 $c_{ix} \leftarrow 0$ \\
\\
10.  Clamp $c$. \\
11.  Return MP\_OKAY. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm fast\_s\_mp\_mul\_digs}
\label{fig:COMBAMULT}
\end{figure}

\textbf{Algorithm fast\_s\_mp\_mul\_digs.}
This algorithm performs the unsigned multiplication of $a$ and $b$ using the Comba method limited to $digs$ digits of precision.

The outer loop of this algorithm is more complicated than that of the baseline multiplier.  This is because on the inside of the
loop we want to produce one column per pass.  This allows the accumulator $\_ \hat W$ to be placed in CPU registers and
reduce the memory bandwidth to two \textbf{mp\_digit} reads per iteration.

The $ty$ variable is set to the minimum count of $ix$ or the number of digits in $b$.  That way if $a$ has more digits than
$b$ this will be limited to $b.used - 1$.  The $tx$ variable is set to the to the distance past $b.used$ the variable
$ix$ is.  This is used for the immediately subsequent statement where we find $iy$.

The variable $iy$ is the minimum digits we can read from either $a$ or $b$ before running out.  Computing one column at a time
means we have to scan one integer upwards and the other downwards.  $a$ starts at $tx$ and $b$ starts at $ty$.  In each
pass we are producing the $ix$'th output column and we note that $tx + ty = ix$.  As we move $tx$ upwards we have to
move $ty$ downards so the equality remains valid.  The $iy$ variable is the number of iterations until
$tx \ge a.used$ or $ty < 0$ occurs.

After every inner pass we store the lower half of the accumulator into $W_{ix}$ and then propagate the carry of the accumulator
into the next round by dividing $\_ \hat W$ by $\beta$.

To measure the benefits of the Comba method over the baseline method consider the number of operations that are required.  If the
cost in terms of time of a multiply and addition is $p$ and the cost of a carry propagation is $q$ then a baseline multiplication would require
$O \left ((p + q)n^2 \right )$ time to multiply two $n$-digit numbers.  The Comba method requires only $O(pn^2 + qn)$ time, however in practice,
the speed increase is actually much more.  With $O(n)$ space the algorithm can be reduced to $O(pn + qn)$ time by implementing the $n$ multiply
and addition operations in the nested loop in parallel.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_fast\_s\_mp\_mul\_digs.c
\vspace{-3mm}
\begin{alltt}
016
017   /* Fast (comba) multiplier
018    *
019    * This is the fast column-array [comba] multiplier.  It is
020    * designed to compute the columns of the product first
021    * then handle the carries afterwards.  This has the effect
022    * of making the nested loops that compute the columns very
023    * simple and schedulable on super-scalar processors.
024    *
025    * This has been modified to produce a variable number of
026    * digits of output so if say only a half-product is required
027    * you don't have to compute the upper half (a feature
028    * required for fast Barrett reduction).
029    *
030    * Based on Algorithm 14.12 on pp.595 of HAC.
031    *
032    */
033   int fast_s_mp_mul_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
034   \{
035     int     olduse, res, pa, ix, iz;
036     mp_digit W[MP_WARRAY];
037     mp_word  _W;
038
039     /* grow the destination as required */
040     if (c->alloc < digs) \{
041       if ((res = mp_grow (c, digs)) != MP_OKAY) \{
042         return res;
043       \}
044     \}
045
046     /* number of output digits to produce */
047     pa = MIN(digs, a->used + b->used);
048
049     /* clear the carry */
050     _W = 0;
051     for (ix = 0; ix < pa; ix++) \{
052         int      tx, ty;
053         int      iy;
054         mp_digit *tmpx, *tmpy;
055
056         /* get offsets into the two bignums */
057         ty = MIN(b->used-1, ix);
058         tx = ix - ty;
059
060         /* setup temp aliases */
061         tmpx = a->dp + tx;
062         tmpy = b->dp + ty;
063
064         /* this is the number of times the loop will iterrate, essentially
065            while (tx++ < a->used && ty-- >= 0) \{ ... \}
066          */
067         iy = MIN(a->used-tx, ty+1);
068
069         /* execute loop */
070         for (iz = 0; iz < iy; ++iz) \{
071            _W += ((mp_word)*tmpx++)*((mp_word)*tmpy--);
072
073         \}
074
075         /* store term */
076         W[ix] = ((mp_digit)_W) & MP_MASK;
077
078         /* make next carry */
079         _W = _W >> ((mp_word)DIGIT_BIT);
080     \}
081
082     /* setup dest */
083     olduse  = c->used;
084     c->used = pa;
085
086     \{
087       mp_digit *tmpc;
088       tmpc = c->dp;
089       for (ix = 0; ix < (pa + 1); ix++) \{
090         /* now extract the previous digit [below the carry] */
091         *tmpc++ = W[ix];
092       \}
093
094       /* clear unused digits [that existed in the old copy of c] */
095       for (; ix < olduse; ix++) \{
096         *tmpc++ = 0;
097       \}
098     \}
099     mp_clamp (c);
100     return MP_OKAY;
101   \}
102   #endif
103
\end{alltt}
\end{small}

As per the pseudo--code we first calculate $pa$ (line 47) as the number of digits to output.  Next we begin the outer loop
to produce the individual columns of the product.  We use the two aliases $tmpx$ and $tmpy$ (lines 61, 62) to point
inside the two multiplicands quickly.

The inner loop (lines 70 to 73) of this implementation is where the tradeoff come into play.  Originally this comba
implementation was ``row--major'' which means it adds to each of the columns in each pass.  After the outer loop it would then fix
the carries.  This was very fast except it had an annoying drawback.  You had to read a mp\_word and two mp\_digits and write
one mp\_word per iteration.  On processors such as the Athlon XP and P4 this did not matter much since the cache bandwidth
is very high and it can keep the ALU fed with data.  It did, however, matter on older and embedded cpus where cache is often
slower and also often doesn't exist.  This new algorithm only performs two reads per iteration under the assumption that the
compiler has aliased $\_ \hat W$ to a CPU register.

After the inner loop we store the current accumulator in $W$ and shift $\_ \hat W$ (lines 76, 79) to forward it as
a carry for the next pass.  After the outer loop we use the final carry (line 76) as the last digit of the product.

\subsection{Polynomial Basis Multiplication}
To break the $O(n^2)$ barrier in multiplication requires a completely different look at integer multiplication.  In the following algorithms
the use of polynomial basis representation for two integers $a$ and $b$ as $f(x) = \sum_{i=0}^{n} a_i x^i$ and
$g(x) = \sum_{i=0}^{n} b_i x^i$ respectively, is required.  In this system both $f(x)$ and $g(x)$ have $n + 1$ terms and are of the $n$'th degree.

The product $a \cdot b \equiv f(x)g(x)$ is the polynomial $W(x) = \sum_{i=0}^{2n} w_i x^i$.  The coefficients $w_i$ will
directly yield the desired product when $\beta$ is substituted for $x$.  The direct solution to solve for the $2n + 1$ coefficients
requires $O(n^2)$ time and would in practice be slower than the Comba technique.

However, numerical analysis theory indicates that only $2n + 1$ distinct points in $W(x)$ are required to determine the values of the $2n + 1$ unknown
coefficients.   This means by finding $\zeta_y = W(y)$ for $2n + 1$ small values of $y$ the coefficients of $W(x)$ can be found with
Gaussian elimination.  This technique is also occasionally refered to as the \textit{interpolation technique} (\textit{references please...}) since in
effect an interpolation based on $2n + 1$ points will yield a polynomial equivalent to $W(x)$.

The coefficients of the polynomial $W(x)$ are unknown which makes finding $W(y)$ for any value of $y$ impossible.  However, since
$W(x) = f(x)g(x)$ the equivalent $\zeta_y = f(y) g(y)$ can be used in its place.  The benefit of this technique stems from the
fact that $f(y)$ and $g(y)$ are much smaller than either $a$ or $b$ respectively.  As a result finding the $2n + 1$ relations required
by multiplying $f(y)g(y)$ involves multiplying integers that are much smaller than either of the inputs.

When picking points to gather relations there are always three obvious points to choose, $y = 0, 1$ and $ \infty$.  The $\zeta_0$ term
is simply the product $W(0) = w_0 = a_0 \cdot b_0$.  The $\zeta_1$ term is the product
$W(1) = \left (\sum_{i = 0}^{n} a_i \right ) \left (\sum_{i = 0}^{n} b_i \right )$.  The third point $\zeta_{\infty}$ is less obvious but rather
simple to explain.  The $2n + 1$'th coefficient of $W(x)$ is numerically equivalent to the most significant column in an integer multiplication.
The point at $\infty$ is used symbolically to represent the most significant column, that is $W(\infty) = w_{2n} = a_nb_n$.  Note that the
points at $y = 0$ and $\infty$ yield the coefficients $w_0$ and $w_{2n}$ directly.

If more points are required they should be of small values and powers of two such as $2^q$ and the related \textit{mirror points}
$\left (2^q \right )^{2n}  \cdot \zeta_{2^{-q}}$ for small values of $q$.  The term ``mirror point'' stems from the fact that
$\left (2^q \right )^{2n}  \cdot \zeta_{2^{-q}}$ can be calculated in the exact opposite fashion as $\zeta_{2^q}$.  For
example, when $n = 2$ and $q = 1$ then following two equations are equivalent to the point $\zeta_{2}$ and its mirror.

\begin{eqnarray}
\zeta_{2}                  = f(2)g(2) = (4a_2 + 2a_1 + a_0)(4b_2 + 2b_1 + b_0) \nonumber \\
16 \cdot \zeta_{1 \over 2} = 4f({1\over 2}) \cdot 4g({1 \over 2}) = (a_2 + 2a_1 + 4a_0)(b_2 + 2b_1 + 4b_0)
\end{eqnarray}

Using such points will allow the values of $f(y)$ and $g(y)$ to be independently calculated using only left shifts.  For example, when $n = 2$ the
polynomial $f(2^q)$ is equal to $2^q((2^qa_2) + a_1) + a_0$.  This technique of polynomial representation is known as Horner's method.

As a general rule of the algorithm when the inputs are split into $n$ parts each there are $2n - 1$ multiplications.  Each multiplication is of
multiplicands that have $n$ times fewer digits than the inputs.  The asymptotic running time of this algorithm is
$O \left ( k^{lg_n(2n - 1)} \right )$ for $k$ digit inputs (\textit{assuming they have the same number of digits}).  Figure~\ref{fig:exponent}
summarizes the exponents for various values of $n$.

\begin{figure}
\begin{center}
\begin{tabular}{|c|c|c|}
\hline \textbf{Split into $n$ Parts} & \textbf{Exponent}  & \textbf{Notes}\\
\hline $2$ & $1.584962501$ & This is Karatsuba Multiplication. \\
\hline $3$ & $1.464973520$ & This is Toom-Cook Multiplication. \\
\hline $4$ & $1.403677461$ &\\
\hline $5$ & $1.365212389$ &\\
\hline $10$ & $1.278753601$ &\\
\hline $100$ & $1.149426538$ &\\
\hline $1000$ & $1.100270931$ &\\
\hline $10000$ & $1.075252070$ &\\
\hline
\end{tabular}
\end{center}
\caption{Asymptotic Running Time of Polynomial Basis Multiplication}
\label{fig:exponent}
\end{figure}

At first it may seem like a good idea to choose $n = 1000$ since the exponent is approximately $1.1$.  However, the overhead
of solving for the 2001 terms of $W(x)$ will certainly consume any savings the algorithm could offer for all but exceedingly large
numbers.

\subsubsection{Cutoff Point}
The polynomial basis multiplication algorithms all require fewer single precision multiplications than a straight Comba approach.  However,
the algorithms incur an overhead (\textit{at the $O(n)$ work level}) since they require a system of equations to be solved.  This makes the
polynomial basis approach more costly to use with small inputs.

Let $m$ represent the number of digits in the multiplicands (\textit{assume both multiplicands have the same number of digits}).  There exists a
point $y$ such that when $m < y$ the polynomial basis algorithms are more costly than Comba, when $m = y$ they are roughly the same cost and
when $m > y$ the Comba methods are slower than the polynomial basis algorithms.

The exact location of $y$ depends on several key architectural elements of the computer platform in question.

\begin{enumerate}
\item  The ratio of clock cycles for single precision multiplication versus other simpler operations such as addition, shifting, etc.  For example
on the AMD Athlon the ratio is roughly $17 : 1$ while on the Intel P4 it is $29 : 1$.  The higher the ratio in favour of multiplication the lower
the cutoff point $y$ will be.

\item  The complexity of the linear system of equations (\textit{for the coefficients of $W(x)$}) is.  Generally speaking as the number of splits
grows the complexity grows substantially.  Ideally solving the system will only involve addition, subtraction and shifting of integers.  This
directly reflects on the ratio previous mentioned.

\item  To a lesser extent memory bandwidth and function call overheads.  Provided the values are in the processor cache this is less of an
influence over the cutoff point.

\end{enumerate}

A clean cutoff point separation occurs when a point $y$ is found such that all of the cutoff point conditions are met.  For example, if the point
is too low then there will be values of $m$ such that $m > y$ and the Comba method is still faster.  Finding the cutoff points is fairly simple when
a high resolution timer is available.

\subsection{Karatsuba Multiplication}
Karatsuba \cite{KARA} multiplication when originally proposed in 1962 was among the first set of algorithms to break the $O(n^2)$ barrier for
general purpose multiplication.  Given two polynomial basis representations $f(x) = ax + b$ and $g(x) = cx + d$, Karatsuba proved with
light algebra \cite{KARAP} that the following polynomial is equivalent to multiplication of the two integers the polynomials represent.

\begin{equation}
f(x) \cdot g(x) = acx^2 + ((a + b)(c + d) - (ac + bd))x + bd
\end{equation}

Using the observation that $ac$ and $bd$ could be re-used only three half sized multiplications would be required to produce the product.  Applying
this algorithm recursively, the work factor becomes $O(n^{lg(3)})$ which is substantially better than the work factor $O(n^2)$ of the Comba technique.  It turns
out what Karatsuba did not know or at least did not publish was that this is simply polynomial basis multiplication with the points
$\zeta_0$, $\zeta_{\infty}$ and $\zeta_{1}$.  Consider the resultant system of equations.

\begin{center}
\begin{tabular}{rcrcrcrc}
$\zeta_{0}$ &      $=$ &  &  &  & & $w_0$ \\
$\zeta_{1}$ &      $=$ & $w_2$ & $+$ & $w_1$ & $+$ & $w_0$ \\
$\zeta_{\infty}$ & $=$ & $w_2$ &  & &  & \\
\end{tabular}
\end{center}

By adding the first and last equation to the equation in the middle the term $w_1$ can be isolated and all three coefficients solved for.  The simplicity
of this system of equations has made Karatsuba fairly popular.  In fact the cutoff point is often fairly low\footnote{With LibTomMath 0.18 it is 70 and 109 digits for the Intel P4 and AMD Athlon respectively.}
making it an ideal algorithm to speed up certain public key cryptosystems such as RSA and Diffie-Hellman.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_karatsuba\_mul}. \\
\textbf{Input}.   mp\_int $a$ and mp\_int $b$ \\
\textbf{Output}.  $c \leftarrow \vert a \vert \cdot \vert b \vert$ \\
\hline \\
1.  Init the following mp\_int variables: $x0$, $x1$, $y0$, $y1$, $t1$, $x0y0$, $x1y1$.\\
2.  If step 2 failed then return(\textit{MP\_MEM}). \\
\\
Split the input.  e.g. $a = x1 \cdot \beta^B + x0$ \\
3.  $B \leftarrow \mbox{min}(a.used, b.used)/2$ \\
4.  $x0 \leftarrow a \mbox{ (mod }\beta^B\mbox{)}$ (\textit{mp\_mod\_2d}) \\
5.  $y0 \leftarrow b \mbox{ (mod }\beta^B\mbox{)}$ \\
6.  $x1 \leftarrow \lfloor a / \beta^B \rfloor$ (\textit{mp\_rshd}) \\
7.  $y1 \leftarrow \lfloor b / \beta^B \rfloor$ \\
\\
Calculate the three products. \\
8.  $x0y0 \leftarrow x0 \cdot y0$ (\textit{mp\_mul}) \\
9.  $x1y1 \leftarrow x1 \cdot y1$ \\
10.  $t1 \leftarrow x1 + x0$ (\textit{mp\_add}) \\
11.  $x0 \leftarrow y1 + y0$ \\
12.  $t1 \leftarrow t1 \cdot x0$ \\
\\
Calculate the middle term. \\
13.  $x0 \leftarrow x0y0 + x1y1$ \\
14.  $t1 \leftarrow t1 - x0$ (\textit{s\_mp\_sub}) \\
\\
Calculate the final product. \\
15.  $t1 \leftarrow t1 \cdot \beta^B$ (\textit{mp\_lshd}) \\
16.  $x1y1 \leftarrow x1y1 \cdot \beta^{2B}$ \\
17.  $t1 \leftarrow x0y0 + t1$ \\
18.  $c \leftarrow t1 + x1y1$ \\
19.  Clear all of the temporary variables. \\
20.  Return(\textit{MP\_OKAY}).\\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_karatsuba\_mul}
\end{figure}

\textbf{Algorithm mp\_karatsuba\_mul.}
This algorithm computes the unsigned product of two inputs using the Karatsuba multiplication algorithm.  It is loosely based on the description
from Knuth \cite[pp. 294-295]{TAOCPV2}.

\index{radix point}
In order to split the two inputs into their respective halves, a suitable \textit{radix point} must be chosen.  The radix point chosen must
be used for both of the inputs meaning that it must be smaller than the smallest input.  Step 3 chooses the radix point $B$ as half of the
smallest input \textbf{used} count.  After the radix point is chosen the inputs are split into lower and upper halves.  Step 4 and 5
compute the lower halves.  Step 6 and 7 computer the upper halves.

After the halves have been computed the three intermediate half-size products must be computed.  Step 8 and 9 compute the trivial products
$x0 \cdot y0$ and $x1 \cdot y1$.  The mp\_int $x0$ is used as a temporary variable after $x1 + x0$ has been computed.  By using $x0$ instead
of an additional temporary variable, the algorithm can avoid an addition memory allocation operation.

The remaining steps 13 through 18 compute the Karatsuba polynomial through a variety of digit shifting and addition operations.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_karatsuba\_mul.c
\vspace{-3mm}
\begin{alltt}
016
017   /* c = |a| * |b| using Karatsuba Multiplication using
018    * three half size multiplications
019    *
020    * Let B represent the radix [e.g. 2**DIGIT_BIT] and
021    * let n represent half of the number of digits in
022    * the min(a,b)
023    *
024    * a = a1 * B**n + a0
025    * b = b1 * B**n + b0
026    *
027    * Then, a * b =>
028      a1b1 * B**2n + ((a1 + a0)(b1 + b0) - (a0b0 + a1b1)) * B + a0b0
029    *
030    * Note that a1b1 and a0b0 are used twice and only need to be
031    * computed once.  So in total three half size (half # of
032    * digit) multiplications are performed, a0b0, a1b1 and
033    * (a1+b1)(a0+b0)
034    *
035    * Note that a multiplication of half the digits requires
036    * 1/4th the number of single precision multiplications so in
037    * total after one call 25% of the single precision multiplications
038    * are saved.  Note also that the call to mp_mul can end up back
039    * in this function if the a0, a1, b0, or b1 are above the threshold.
040    * This is known as divide-and-conquer and leads to the famous
041    * O(N**lg(3)) or O(N**1.584) work which is asymptopically lower than
042    * the standard O(N**2) that the baseline/comba methods use.
043    * Generally though the overhead of this method doesn't pay off
044    * until a certain size (N ~ 80) is reached.
045    */
046   int mp_karatsuba_mul (mp_int * a, mp_int * b, mp_int * c)
047   \{
048     mp_int  x0, x1, y0, y1, t1, x0y0, x1y1;
049     int     B, err;
050
051     /* default the return code to an error */
052     err = MP_MEM;
053
054     /* min # of digits */
055     B = MIN (a->used, b->used);
056
057     /* now divide in two */
058     B = B >> 1;
059
060     /* init copy all the temps */
061     if (mp_init_size (&x0, B) != MP_OKAY)
062       goto ERR;
063     if (mp_init_size (&x1, a->used - B) != MP_OKAY)
064       goto X0;
065     if (mp_init_size (&y0, B) != MP_OKAY)
066       goto X1;
067     if (mp_init_size (&y1, b->used - B) != MP_OKAY)
068       goto Y0;
069
070     /* init temps */
071     if (mp_init_size (&t1, B * 2) != MP_OKAY)
072       goto Y1;
073     if (mp_init_size (&x0y0, B * 2) != MP_OKAY)
074       goto T1;
075     if (mp_init_size (&x1y1, B * 2) != MP_OKAY)
076       goto X0Y0;
077
078     /* now shift the digits */
079     x0.used = y0.used = B;
080     x1.used = a->used - B;
081     y1.used = b->used - B;
082
083     \{
084       int x;
085       mp_digit *tmpa, *tmpb, *tmpx, *tmpy;
086
087       /* we copy the digits directly instead of using higher level functions
088        * since we also need to shift the digits
089        */
090       tmpa = a->dp;
091       tmpb = b->dp;
092
093       tmpx = x0.dp;
094       tmpy = y0.dp;
095       for (x = 0; x < B; x++) \{
096         *tmpx++ = *tmpa++;
097         *tmpy++ = *tmpb++;
098       \}
099
100       tmpx = x1.dp;
101       for (x = B; x < a->used; x++) \{
102         *tmpx++ = *tmpa++;
103       \}
104
105       tmpy = y1.dp;
106       for (x = B; x < b->used; x++) \{
107         *tmpy++ = *tmpb++;
108       \}
109     \}
110
111     /* only need to clamp the lower words since by definition the
112      * upper words x1/y1 must have a known number of digits
113      */
114     mp_clamp (&x0);
115     mp_clamp (&y0);
116
117     /* now calc the products x0y0 and x1y1 */
118     /* after this x0 is no longer required, free temp [x0==t2]! */
119     if (mp_mul (&x0, &y0, &x0y0) != MP_OKAY)
120       goto X1Y1;          /* x0y0 = x0*y0 */
121     if (mp_mul (&x1, &y1, &x1y1) != MP_OKAY)
122       goto X1Y1;          /* x1y1 = x1*y1 */
123
124     /* now calc x1+x0 and y1+y0 */
125     if (s_mp_add (&x1, &x0, &t1) != MP_OKAY)
126       goto X1Y1;          /* t1 = x1 - x0 */
127     if (s_mp_add (&y1, &y0, &x0) != MP_OKAY)
128       goto X1Y1;          /* t2 = y1 - y0 */
129     if (mp_mul (&t1, &x0, &t1) != MP_OKAY)
130       goto X1Y1;          /* t1 = (x1 + x0) * (y1 + y0) */
131
132     /* add x0y0 */
133     if (mp_add (&x0y0, &x1y1, &x0) != MP_OKAY)
134       goto X1Y1;          /* t2 = x0y0 + x1y1 */
135     if (s_mp_sub (&t1, &x0, &t1) != MP_OKAY)
136       goto X1Y1;          /* t1 = (x1+x0)*(y1+y0) - (x1y1 + x0y0) */
137
138     /* shift by B */
139     if (mp_lshd (&t1, B) != MP_OKAY)
140       goto X1Y1;          /* t1 = (x0y0 + x1y1 - (x1-x0)*(y1-y0))<<B */
141     if (mp_lshd (&x1y1, B * 2) != MP_OKAY)
142       goto X1Y1;          /* x1y1 = x1y1 << 2*B */
143
144     if (mp_add (&x0y0, &t1, &t1) != MP_OKAY)
145       goto X1Y1;          /* t1 = x0y0 + t1 */
146     if (mp_add (&t1, &x1y1, c) != MP_OKAY)
147       goto X1Y1;          /* t1 = x0y0 + t1 + x1y1 */
148
149     /* Algorithm succeeded set the return code to MP_OKAY */
150     err = MP_OKAY;
151
152   X1Y1:mp_clear (&x1y1);
153   X0Y0:mp_clear (&x0y0);
154   T1:mp_clear (&t1);
155   Y1:mp_clear (&y1);
156   Y0:mp_clear (&y0);
157   X1:mp_clear (&x1);
158   X0:mp_clear (&x0);
159   ERR:
160     return err;
161   \}
162   #endif
163
\end{alltt}
\end{small}

The new coding element in this routine, not  seen in previous routines, is the usage of goto statements.  The conventional
wisdom is that goto statements should be avoided.  This is generally true, however when every single function call can fail, it makes sense
to handle error recovery with a single piece of code.  Lines 61 to 75 handle initializing all of the temporary variables
required.  Note how each of the if statements goes to a different label in case of failure.  This allows the routine to correctly free only
the temporaries that have been successfully allocated so far.

The temporary variables are all initialized using the mp\_init\_size routine since they are expected to be large.  This saves the
additional reallocation that would have been necessary.  Also $x0$, $x1$, $y0$ and $y1$ have to be able to hold at least their respective
number of digits for the next section of code.

The first algebraic portion of the algorithm is to split the two inputs into their halves.  However, instead of using mp\_mod\_2d and mp\_rshd
to extract the halves, the respective code has been placed inline within the body of the function.  To initialize the halves, the \textbf{used} and
\textbf{sign} members are copied first.  The first for loop on line 101 copies the lower halves.  Since they are both the same magnitude it
is simpler to calculate both lower halves in a single loop.  The for loop on lines 106 and 106 calculate the upper halves $x1$ and
$y1$ respectively.

By inlining the calculation of the halves, the Karatsuba multiplier has a slightly lower overhead and can be used for smaller magnitude inputs.

When line 150 is reached, the algorithm has completed succesfully.  The ``error status'' variable $err$ is set to \textbf{MP\_OKAY} so that
the same code that handles errors can be used to clear the temporary variables and return.

\subsection{Toom-Cook $3$-Way Multiplication}
Toom-Cook $3$-Way \cite{TOOM} multiplication is essentially the polynomial basis algorithm for $n = 2$ except that the points  are
chosen such that $\zeta$ is easy to compute and the resulting system of equations easy to reduce.  Here, the points $\zeta_{0}$,
$16 \cdot \zeta_{1 \over 2}$, $\zeta_1$, $\zeta_2$ and $\zeta_{\infty}$ make up the five required points to solve for the coefficients
of the $W(x)$.

With the five relations that Toom-Cook specifies, the following system of equations is formed.

\begin{center}
\begin{tabular}{rcrcrcrcrcr}
$\zeta_0$                    & $=$ & $0w_4$ & $+$ & $0w_3$ & $+$ & $0w_2$ & $+$ & $0w_1$ & $+$ & $1w_0$  \\
$16 \cdot \zeta_{1 \over 2}$ & $=$ & $1w_4$ & $+$ & $2w_3$ & $+$ & $4w_2$ & $+$ & $8w_1$ & $+$ & $16w_0$  \\
$\zeta_1$                    & $=$ & $1w_4$ & $+$ & $1w_3$ & $+$ & $1w_2$ & $+$ & $1w_1$ & $+$ & $1w_0$  \\
$\zeta_2$                    & $=$ & $16w_4$ & $+$ & $8w_3$ & $+$ & $4w_2$ & $+$ & $2w_1$ & $+$ & $1w_0$  \\
$\zeta_{\infty}$             & $=$ & $1w_4$ & $+$ & $0w_3$ & $+$ & $0w_2$ & $+$ & $0w_1$ & $+$ & $0w_0$  \\
\end{tabular}
\end{center}

A trivial solution to this matrix requires $12$ subtractions, two multiplications by a small power of two, two divisions by a small power
of two, two divisions by three and one multiplication by three.  All of these $19$ sub-operations require less than quadratic time, meaning that
the algorithm can be faster than a baseline multiplication.  However, the greater complexity of this algorithm places the cutoff point
(\textbf{TOOM\_MUL\_CUTOFF}) where Toom-Cook becomes more efficient much higher than the Karatsuba cutoff point.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_toom\_mul}. \\
\textbf{Input}.   mp\_int $a$ and mp\_int $b$ \\
\textbf{Output}.  $c \leftarrow  a  \cdot  b $ \\
\hline \\
Split $a$ and $b$ into three pieces.  E.g. $a = a_2 \beta^{2k} + a_1 \beta^{k} + a_0$ \\
1.  $k \leftarrow \lfloor \mbox{min}(a.used, b.used) / 3 \rfloor$ \\
2.  $a_0 \leftarrow a \mbox{ (mod }\beta^{k}\mbox{)}$ \\
3.  $a_1 \leftarrow \lfloor a / \beta^k \rfloor$, $a_1 \leftarrow a_1 \mbox{ (mod }\beta^{k}\mbox{)}$ \\
4.  $a_2 \leftarrow \lfloor a / \beta^{2k} \rfloor$, $a_2 \leftarrow a_2 \mbox{ (mod }\beta^{k}\mbox{)}$ \\
5.  $b_0 \leftarrow a \mbox{ (mod }\beta^{k}\mbox{)}$ \\
6.  $b_1 \leftarrow \lfloor a / \beta^k \rfloor$, $b_1 \leftarrow b_1 \mbox{ (mod }\beta^{k}\mbox{)}$ \\
7.  $b_2 \leftarrow \lfloor a / \beta^{2k} \rfloor$, $b_2 \leftarrow b_2 \mbox{ (mod }\beta^{k}\mbox{)}$ \\
\\
Find the five equations for $w_0, w_1, ..., w_4$. \\
8.  $w_0 \leftarrow a_0 \cdot b_0$ \\
9.  $w_4 \leftarrow a_2 \cdot b_2$ \\
10. $tmp_1 \leftarrow 2 \cdot a_0$, $tmp_1 \leftarrow a_1 + tmp_1$, $tmp_1 \leftarrow 2 \cdot tmp_1$, $tmp_1 \leftarrow tmp_1 + a_2$ \\
11. $tmp_2 \leftarrow 2 \cdot b_0$, $tmp_2 \leftarrow b_1 + tmp_2$, $tmp_2 \leftarrow 2 \cdot tmp_2$, $tmp_2 \leftarrow tmp_2 + b_2$ \\
12. $w_1 \leftarrow tmp_1 \cdot tmp_2$ \\
13. $tmp_1 \leftarrow 2 \cdot a_2$, $tmp_1 \leftarrow a_1 + tmp_1$, $tmp_1 \leftarrow 2 \cdot tmp_1$, $tmp_1 \leftarrow tmp_1 + a_0$ \\
14. $tmp_2 \leftarrow 2 \cdot b_2$, $tmp_2 \leftarrow b_1 + tmp_2$, $tmp_2 \leftarrow 2 \cdot tmp_2$, $tmp_2 \leftarrow tmp_2 + b_0$ \\
15. $w_3 \leftarrow tmp_1 \cdot tmp_2$ \\
16. $tmp_1 \leftarrow a_0 + a_1$, $tmp_1 \leftarrow tmp_1 + a_2$, $tmp_2 \leftarrow b_0 + b_1$, $tmp_2 \leftarrow tmp_2 + b_2$ \\
17. $w_2 \leftarrow tmp_1 \cdot tmp_2$ \\
\\
Continued on the next page.\\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_toom\_mul}
\end{figure}

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_toom\_mul} (continued). \\
\textbf{Input}.   mp\_int $a$ and mp\_int $b$ \\
\textbf{Output}.  $c \leftarrow a \cdot  b $ \\
\hline \\
Now solve the system of equations. \\
18. $w_1 \leftarrow w_4 - w_1$, $w_3 \leftarrow w_3 - w_0$ \\
19. $w_1 \leftarrow \lfloor w_1 / 2 \rfloor$, $w_3 \leftarrow \lfloor w_3 / 2 \rfloor$ \\
20. $w_2 \leftarrow w_2 - w_0$, $w_2 \leftarrow w_2 - w_4$ \\
21. $w_1 \leftarrow w_1 - w_2$, $w_3 \leftarrow w_3 - w_2$ \\
22. $tmp_1 \leftarrow 8 \cdot w_0$, $w_1 \leftarrow w_1 - tmp_1$, $tmp_1 \leftarrow 8 \cdot w_4$, $w_3 \leftarrow w_3 - tmp_1$ \\
23. $w_2 \leftarrow 3 \cdot w_2$, $w_2 \leftarrow w_2 - w_1$, $w_2 \leftarrow w_2 - w_3$ \\
24. $w_1 \leftarrow w_1 - w_2$, $w_3 \leftarrow w_3 - w_2$ \\
25. $w_1 \leftarrow \lfloor w_1 / 3 \rfloor, w_3 \leftarrow \lfloor w_3 / 3 \rfloor$ \\
\\
Now substitute $\beta^k$ for $x$ by shifting $w_0, w_1, ..., w_4$. \\
26. for $n$ from $1$ to $4$ do \\
\hspace{3mm}26.1  $w_n \leftarrow w_n \cdot \beta^{nk}$ \\
27. $c \leftarrow w_0 + w_1$, $c \leftarrow c + w_2$, $c \leftarrow c + w_3$, $c \leftarrow c + w_4$ \\
28. Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_toom\_mul (continued)}
\end{figure}

\textbf{Algorithm mp\_toom\_mul.}
This algorithm computes the product of two mp\_int variables $a$ and $b$ using the Toom-Cook approach.  Compared to the Karatsuba multiplication, this
algorithm has a lower asymptotic running time of approximately $O(n^{1.464})$ but at an obvious cost in overhead.  In this
description, several statements have been compounded to save space.  The intention is that the statements are executed from left to right across
any given step.

The two inputs $a$ and $b$ are first split into three $k$-digit integers $a_0, a_1, a_2$ and $b_0, b_1, b_2$ respectively.  From these smaller
integers the coefficients of the polynomial basis representations $f(x)$ and $g(x)$ are known and can be used to find the relations required.

The first two relations $w_0$ and $w_4$ are the points $\zeta_{0}$ and $\zeta_{\infty}$ respectively.  The relation $w_1, w_2$ and $w_3$ correspond
to the points $16 \cdot \zeta_{1 \over 2}, \zeta_{2}$ and $\zeta_{1}$ respectively.  These are found using logical shifts to independently find
$f(y)$ and $g(y)$ which significantly speeds up the algorithm.

After the five relations $w_0, w_1, \ldots, w_4$ have been computed, the system they represent must be solved in order for the unknown coefficients
$w_1, w_2$ and $w_3$ to be isolated.  The steps 18 through 25 perform the system reduction required as previously described.  Each step of
the reduction represents the comparable matrix operation that would be performed had this been performed by pencil.  For example, step 18 indicates
that row $1$ must be subtracted from row $4$ and simultaneously row $0$ subtracted from row $3$.

Once the coeffients have been isolated, the polynomial $W(x) = \sum_{i=0}^{2n} w_i x^i$ is known.  By substituting $\beta^{k}$ for $x$, the integer
result $a \cdot b$ is produced.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_toom\_mul.c
\vspace{-3mm}
\begin{alltt}
016
017   /* multiplication using the Toom-Cook 3-way algorithm
018    *
019    * Much more complicated than Karatsuba but has a lower
020    * asymptotic running time of O(N**1.464).  This algorithm is
021    * only particularly useful on VERY large inputs
022    * (we're talking 1000s of digits here...).
023   */
024   int mp_toom_mul(mp_int *a, mp_int *b, mp_int *c)
025   \{
026       mp_int w0, w1, w2, w3, w4, tmp1, tmp2, a0, a1, a2, b0, b1, b2;
027       int res, B;
028
029       /* init temps */
030       if ((res = mp_init_multi(&w0, &w1, &w2, &w3, &w4,
031                                &a0, &a1, &a2, &b0, &b1,
032                                &b2, &tmp1, &tmp2, NULL)) != MP_OKAY) \{
033          return res;
034       \}
035
036       /* B */
037       B = MIN(a->used, b->used) / 3;
038
039       /* a = a2 * B**2 + a1 * B + a0 */
040       if ((res = mp_mod_2d(a, DIGIT_BIT * B, &a0)) != MP_OKAY) \{
041          goto ERR;
042       \}
043
044       if ((res = mp_copy(a, &a1)) != MP_OKAY) \{
045          goto ERR;
046       \}
047       mp_rshd(&a1, B);
048       if ((res = mp_mod_2d(&a1, DIGIT_BIT * B, &a1)) != MP_OKAY) \{
049          goto ERR;
050       \}
051
052       if ((res = mp_copy(a, &a2)) != MP_OKAY) \{
053          goto ERR;
054       \}
055       mp_rshd(&a2, B*2);
056
057       /* b = b2 * B**2 + b1 * B + b0 */
058       if ((res = mp_mod_2d(b, DIGIT_BIT * B, &b0)) != MP_OKAY) \{
059          goto ERR;
060       \}
061
062       if ((res = mp_copy(b, &b1)) != MP_OKAY) \{
063          goto ERR;
064       \}
065       mp_rshd(&b1, B);
066       (void)mp_mod_2d(&b1, DIGIT_BIT * B, &b1);
067
068       if ((res = mp_copy(b, &b2)) != MP_OKAY) \{
069          goto ERR;
070       \}
071       mp_rshd(&b2, B*2);
072
073       /* w0 = a0*b0 */
074       if ((res = mp_mul(&a0, &b0, &w0)) != MP_OKAY) \{
075          goto ERR;
076       \}
077
078       /* w4 = a2 * b2 */
079       if ((res = mp_mul(&a2, &b2, &w4)) != MP_OKAY) \{
080          goto ERR;
081       \}
082
083       /* w1 = (a2 + 2(a1 + 2a0))(b2 + 2(b1 + 2b0)) */
084       if ((res = mp_mul_2(&a0, &tmp1)) != MP_OKAY) \{
085          goto ERR;
086       \}
087       if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) \{
088          goto ERR;
089       \}
090       if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) \{
091          goto ERR;
092       \}
093       if ((res = mp_add(&tmp1, &a2, &tmp1)) != MP_OKAY) \{
094          goto ERR;
095       \}
096
097       if ((res = mp_mul_2(&b0, &tmp2)) != MP_OKAY) \{
098          goto ERR;
099       \}
100       if ((res = mp_add(&tmp2, &b1, &tmp2)) != MP_OKAY) \{
101          goto ERR;
102       \}
103       if ((res = mp_mul_2(&tmp2, &tmp2)) != MP_OKAY) \{
104          goto ERR;
105       \}
106       if ((res = mp_add(&tmp2, &b2, &tmp2)) != MP_OKAY) \{
107          goto ERR;
108       \}
109
110       if ((res = mp_mul(&tmp1, &tmp2, &w1)) != MP_OKAY) \{
111          goto ERR;
112       \}
113
114       /* w3 = (a0 + 2(a1 + 2a2))(b0 + 2(b1 + 2b2)) */
115       if ((res = mp_mul_2(&a2, &tmp1)) != MP_OKAY) \{
116          goto ERR;
117       \}
118       if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) \{
119          goto ERR;
120       \}
121       if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) \{
122          goto ERR;
123       \}
124       if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) \{
125          goto ERR;
126       \}
127
128       if ((res = mp_mul_2(&b2, &tmp2)) != MP_OKAY) \{
129          goto ERR;
130       \}
131       if ((res = mp_add(&tmp2, &b1, &tmp2)) != MP_OKAY) \{
132          goto ERR;
133       \}
134       if ((res = mp_mul_2(&tmp2, &tmp2)) != MP_OKAY) \{
135          goto ERR;
136       \}
137       if ((res = mp_add(&tmp2, &b0, &tmp2)) != MP_OKAY) \{
138          goto ERR;
139       \}
140
141       if ((res = mp_mul(&tmp1, &tmp2, &w3)) != MP_OKAY) \{
142          goto ERR;
143       \}
144
145
146       /* w2 = (a2 + a1 + a0)(b2 + b1 + b0) */
147       if ((res = mp_add(&a2, &a1, &tmp1)) != MP_OKAY) \{
148          goto ERR;
149       \}
150       if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) \{
151          goto ERR;
152       \}
153       if ((res = mp_add(&b2, &b1, &tmp2)) != MP_OKAY) \{
154          goto ERR;
155       \}
156       if ((res = mp_add(&tmp2, &b0, &tmp2)) != MP_OKAY) \{
157          goto ERR;
158       \}
159       if ((res = mp_mul(&tmp1, &tmp2, &w2)) != MP_OKAY) \{
160          goto ERR;
161       \}
162
163       /* now solve the matrix
164
165          0  0  0  0  1
166          1  2  4  8  16
167          1  1  1  1  1
168          16 8  4  2  1
169          1  0  0  0  0
170
171          using 12 subtractions, 4 shifts,
172                 2 small divisions and 1 small multiplication
173        */
174
175       /* r1 - r4 */
176       if ((res = mp_sub(&w1, &w4, &w1)) != MP_OKAY) \{
177          goto ERR;
178       \}
179       /* r3 - r0 */
180       if ((res = mp_sub(&w3, &w0, &w3)) != MP_OKAY) \{
181          goto ERR;
182       \}
183       /* r1/2 */
184       if ((res = mp_div_2(&w1, &w1)) != MP_OKAY) \{
185          goto ERR;
186       \}
187       /* r3/2 */
188       if ((res = mp_div_2(&w3, &w3)) != MP_OKAY) \{
189          goto ERR;
190       \}
191       /* r2 - r0 - r4 */
192       if ((res = mp_sub(&w2, &w0, &w2)) != MP_OKAY) \{
193          goto ERR;
194       \}
195       if ((res = mp_sub(&w2, &w4, &w2)) != MP_OKAY) \{
196          goto ERR;
197       \}
198       /* r1 - r2 */
199       if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) \{
200          goto ERR;
201       \}
202       /* r3 - r2 */
203       if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) \{
204          goto ERR;
205       \}
206       /* r1 - 8r0 */
207       if ((res = mp_mul_2d(&w0, 3, &tmp1)) != MP_OKAY) \{
208          goto ERR;
209       \}
210       if ((res = mp_sub(&w1, &tmp1, &w1)) != MP_OKAY) \{
211          goto ERR;
212       \}
213       /* r3 - 8r4 */
214       if ((res = mp_mul_2d(&w4, 3, &tmp1)) != MP_OKAY) \{
215          goto ERR;
216       \}
217       if ((res = mp_sub(&w3, &tmp1, &w3)) != MP_OKAY) \{
218          goto ERR;
219       \}
220       /* 3r2 - r1 - r3 */
221       if ((res = mp_mul_d(&w2, 3, &w2)) != MP_OKAY) \{
222          goto ERR;
223       \}
224       if ((res = mp_sub(&w2, &w1, &w2)) != MP_OKAY) \{
225          goto ERR;
226       \}
227       if ((res = mp_sub(&w2, &w3, &w2)) != MP_OKAY) \{
228          goto ERR;
229       \}
230       /* r1 - r2 */
231       if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) \{
232          goto ERR;
233       \}
234       /* r3 - r2 */
235       if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) \{
236          goto ERR;
237       \}
238       /* r1/3 */
239       if ((res = mp_div_3(&w1, &w1, NULL)) != MP_OKAY) \{
240          goto ERR;
241       \}
242       /* r3/3 */
243       if ((res = mp_div_3(&w3, &w3, NULL)) != MP_OKAY) \{
244          goto ERR;
245       \}
246
247       /* at this point shift W[n] by B*n */
248       if ((res = mp_lshd(&w1, 1*B)) != MP_OKAY) \{
249          goto ERR;
250       \}
251       if ((res = mp_lshd(&w2, 2*B)) != MP_OKAY) \{
252          goto ERR;
253       \}
254       if ((res = mp_lshd(&w3, 3*B)) != MP_OKAY) \{
255          goto ERR;
256       \}
257       if ((res = mp_lshd(&w4, 4*B)) != MP_OKAY) \{
258          goto ERR;
259       \}
260
261       if ((res = mp_add(&w0, &w1, c)) != MP_OKAY) \{
262          goto ERR;
263       \}
264       if ((res = mp_add(&w2, &w3, &tmp1)) != MP_OKAY) \{
265          goto ERR;
266       \}
267       if ((res = mp_add(&w4, &tmp1, &tmp1)) != MP_OKAY) \{
268          goto ERR;
269       \}
270       if ((res = mp_add(&tmp1, c, c)) != MP_OKAY) \{
271          goto ERR;
272       \}
273
274   ERR:
275       mp_clear_multi(&w0, &w1, &w2, &w3, &w4,
276                      &a0, &a1, &a2, &b0, &b1,
277                      &b2, &tmp1, &tmp2, NULL);
278       return res;
279   \}
280
281   #endif
282
\end{alltt}
\end{small}

The first obvious thing to note is that this algorithm is complicated.  The complexity is worth it if you are multiplying very
large numbers.  For example, a 10,000 digit multiplication takes approximaly 99,282,205 fewer single precision multiplications with
Toom--Cook than a Comba or baseline approach (this is a savings of more than 99$\%$).  For most ``crypto'' sized numbers this
algorithm is not practical as Karatsuba has a much lower cutoff point.

First we split $a$ and $b$ into three roughly equal portions.  This has been accomplished (lines 40 to 71) with
combinations of mp\_rshd() and mp\_mod\_2d() function calls.  At this point $a = a2 \cdot \beta^2 + a1 \cdot \beta + a0$ and similiarly
for $b$.

Next we compute the five points $w0, w1, w2, w3$ and $w4$.  Recall that $w0$ and $w4$ can be computed directly from the portions so
we get those out of the way first (lines 74 and 79).  Next we compute $w1, w2$ and $w3$ using Horners method.

After this point we solve for the actual values of $w1, w2$ and $w3$ by reducing the $5 \times 5$ system which is relatively
straight forward.

\subsection{Signed Multiplication}
Now that algorithms to handle multiplications of every useful dimensions have been developed, a rather simple finishing touch is required.  So far all
of the multiplication algorithms have been unsigned multiplications which leaves only a signed multiplication algorithm to be established.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_mul}. \\
\textbf{Input}.   mp\_int $a$ and mp\_int $b$ \\
\textbf{Output}.  $c \leftarrow a \cdot b$ \\
\hline \\
1.  If $a.sign = b.sign$ then \\
\hspace{3mm}1.1  $sign = MP\_ZPOS$ \\
2.  else \\
\hspace{3mm}2.1  $sign = MP\_ZNEG$ \\
3.  If min$(a.used, b.used) \ge TOOM\_MUL\_CUTOFF$ then  \\
\hspace{3mm}3.1  $c \leftarrow a \cdot b$ using algorithm mp\_toom\_mul \\
4.  else if min$(a.used, b.used) \ge KARATSUBA\_MUL\_CUTOFF$ then \\
\hspace{3mm}4.1  $c \leftarrow a \cdot b$ using algorithm mp\_karatsuba\_mul \\
5.  else \\
\hspace{3mm}5.1  $digs \leftarrow a.used + b.used + 1$ \\
\hspace{3mm}5.2  If $digs < MP\_ARRAY$ and min$(a.used, b.used) \le \delta$ then \\
\hspace{6mm}5.2.1  $c \leftarrow a \cdot b \mbox{ (mod }\beta^{digs}\mbox{)}$ using algorithm fast\_s\_mp\_mul\_digs.  \\
\hspace{3mm}5.3  else \\
\hspace{6mm}5.3.1  $c \leftarrow a \cdot b \mbox{ (mod }\beta^{digs}\mbox{)}$ using algorithm s\_mp\_mul\_digs.  \\
6.  $c.sign \leftarrow sign$ \\
7.  Return the result of the unsigned multiplication performed. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_mul}
\end{figure}

\textbf{Algorithm mp\_mul.}
This algorithm performs the signed multiplication of two inputs.  It will make use of any of the three unsigned multiplication algorithms
available when the input is of appropriate size.  The \textbf{sign} of the result is not set until the end of the algorithm since algorithm
s\_mp\_mul\_digs will clear it.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_mul.c
\vspace{-3mm}
\begin{alltt}
016
017   /* high level multiplication (handles sign) */
018   int mp_mul (mp_int * a, mp_int * b, mp_int * c)
019   \{
020     int     res, neg;
021     neg = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
022
023     /* use Toom-Cook? */
024   #ifdef BN_MP_TOOM_MUL_C
025     if (MIN (a->used, b->used) >= TOOM_MUL_CUTOFF) \{
026       res = mp_toom_mul(a, b, c);
027     \} else
028   #endif
029   #ifdef BN_MP_KARATSUBA_MUL_C
030     /* use Karatsuba? */
031     if (MIN (a->used, b->used) >= KARATSUBA_MUL_CUTOFF) \{
032       res = mp_karatsuba_mul (a, b, c);
033     \} else
034   #endif
035     \{
036       /* can we use the fast multiplier?
037        *
038        * The fast multiplier can be used if the output will
039        * have less than MP_WARRAY digits and the number of
040        * digits won't affect carry propagation
041        */
042       int     digs = a->used + b->used + 1;
043
044   #ifdef BN_FAST_S_MP_MUL_DIGS_C
045       if ((digs < MP_WARRAY) &&
046           (MIN(a->used, b->used) <=
047            (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) \{
048         res = fast_s_mp_mul_digs (a, b, c, digs);
049       \} else
050   #endif
051       \{
052   #ifdef BN_S_MP_MUL_DIGS_C
053         res = s_mp_mul (a, b, c); /* uses s_mp_mul_digs */
054   #else
055         res = MP_VAL;
056   #endif
057       \}
058     \}
059     c->sign = (c->used > 0) ? neg : MP_ZPOS;
060     return res;
061   \}
062   #endif
063
\end{alltt}
\end{small}

The implementation is rather simplistic and is not particularly noteworthy.  Line 23 computes the sign of the result using the ``?''
operator from the C programming language.  Line 47 computes $\delta$ using the fact that $1 << k$ is equal to $2^k$.

\section{Squaring}
\label{sec:basesquare}

Squaring is a special case of multiplication where both multiplicands are equal.  At first it may seem like there is no significant optimization
available but in fact there is.  Consider the multiplication of $576$ against $241$.  In total there will be nine single precision multiplications
performed which are $1\cdot 6$, $1 \cdot 7$, $1 \cdot 5$, $4 \cdot 6$, $4 \cdot 7$, $4 \cdot 5$, $2 \cdot  6$, $2 \cdot 7$ and $2 \cdot 5$.  Now consider
the multiplication of $123$ against $123$.  The nine products are $3 \cdot 3$, $3 \cdot 2$, $3 \cdot 1$, $2 \cdot 3$, $2 \cdot 2$, $2 \cdot 1$,
$1 \cdot 3$, $1 \cdot 2$ and $1 \cdot 1$.  On closer inspection some of the products are equivalent.  For example, $3 \cdot 2 = 2 \cdot 3$
and $3 \cdot 1 = 1 \cdot 3$.

For any $n$-digit input, there are ${{\left (n^2 + n \right)}\over 2}$ possible unique single precision multiplications required compared to the $n^2$
required for multiplication.  The following diagram gives an example of the operations required.

\begin{figure}[here]
\begin{center}
\begin{tabular}{ccccc|c}
&&1&2&3&\\
$\times$ &&1&2&3&\\
\hline && $3 \cdot 1$ & $3 \cdot 2$ & $3 \cdot 3$ & Row 0\\
       & $2 \cdot 1$  & $2 \cdot 2$ & $2 \cdot 3$ && Row 1 \\
         $1 \cdot 1$  & $1 \cdot 2$ & $1 \cdot 3$ &&& Row 2 \\
\end{tabular}
\end{center}
\caption{Squaring Optimization Diagram}
\end{figure}

Starting from zero and numbering the columns from right to left a very simple pattern becomes obvious.  For the purposes of this discussion let $x$
represent the number being squared.  The first observation is that in row $k$ the $2k$'th column of the product has a $\left (x_k \right)^2$ term in it.

The second observation is that every column $j$ in row $k$ where $j \ne 2k$ is part of a double product.  Every non-square term of a column will
appear twice hence the name ``double product''.  Every odd column is made up entirely of double products.  In fact every column is made up of double
products and at most one square (\textit{see the exercise section}).

The third and final observation is that for row $k$ the first unique non-square term, that is, one that hasn't already appeared in an earlier row,
occurs at column $2k + 1$.  For example, on row $1$ of the previous squaring, column one is part of the double product with column one from row zero.
Column two of row one is a square and column three is the first unique column.

\subsection{The Baseline Squaring Algorithm}
The baseline squaring algorithm is meant to be a catch-all squaring algorithm.  It will handle any of the input sizes that the faster routines
will not handle.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{s\_mp\_sqr}. \\
\textbf{Input}.   mp\_int $a$ \\
\textbf{Output}.  $b \leftarrow a^2$ \\
\hline \\
1.  Init a temporary mp\_int of at least $2 \cdot a.used +1$ digits.  (\textit{mp\_init\_size}) \\
2.  If step 1 failed return(\textit{MP\_MEM}) \\
3.  $t.used \leftarrow 2 \cdot a.used + 1$ \\
4.  For $ix$ from 0 to $a.used - 1$ do \\
\hspace{3mm}Calculate the square. \\
\hspace{3mm}4.1  $\hat r \leftarrow t_{2ix} + \left (a_{ix} \right )^2$ \\
\hspace{3mm}4.2  $t_{2ix} \leftarrow \hat r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{3mm}Calculate the double products after the square. \\
\hspace{3mm}4.3  $u \leftarrow \lfloor \hat r / \beta \rfloor$ \\
\hspace{3mm}4.4  For $iy$ from $ix + 1$ to $a.used - 1$ do \\
\hspace{6mm}4.4.1  $\hat r \leftarrow 2 \cdot a_{ix}a_{iy} + t_{ix + iy} + u$ \\
\hspace{6mm}4.4.2  $t_{ix + iy} \leftarrow \hat r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{6mm}4.4.3  $u \leftarrow \lfloor \hat r / \beta \rfloor$ \\
\hspace{3mm}Set the last carry. \\
\hspace{3mm}4.5  While $u > 0$ do \\
\hspace{6mm}4.5.1  $iy \leftarrow iy + 1$ \\
\hspace{6mm}4.5.2  $\hat r \leftarrow t_{ix + iy} + u$ \\
\hspace{6mm}4.5.3  $t_{ix + iy} \leftarrow \hat r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{6mm}4.5.4  $u \leftarrow \lfloor \hat r / \beta \rfloor$ \\
5.  Clamp excess digits of $t$.  (\textit{mp\_clamp}) \\
6.  Exchange $b$ and $t$. \\
7.  Clear $t$ (\textit{mp\_clear}) \\
8.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm s\_mp\_sqr}
\end{figure}

\textbf{Algorithm s\_mp\_sqr.}
This algorithm computes the square of an input using the three observations on squaring.  It is based fairly faithfully on  algorithm 14.16 of HAC
\cite[pp.596-597]{HAC}.  Similar to algorithm s\_mp\_mul\_digs, a temporary mp\_int is allocated to hold the result of the squaring.  This allows the
destination mp\_int to be the same as the source mp\_int.

The outer loop of this algorithm begins on step 4. It is best to think of the outer loop as walking down the rows of the partial results, while
the inner loop computes the columns of the partial result.  Step 4.1 and 4.2 compute the square term for each row, and step 4.3 and 4.4 propagate
the carry and compute the double products.

The requirement that a mp\_word be able to represent the range $0 \le x < 2 \beta^2$ arises from this
very algorithm.  The product $a_{ix}a_{iy}$ will lie in the range $0 \le x \le \beta^2 - 2\beta + 1$ which is obviously less than $\beta^2$ meaning that
when it is multiplied by two, it can be properly represented by a mp\_word.

Similar to algorithm s\_mp\_mul\_digs, after every pass of the inner loop, the destination is correctly set to the sum of all of the partial
results calculated so far.  This involves expensive carry propagation which will be eliminated in the next algorithm.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_s\_mp\_sqr.c
\vspace{-3mm}
\begin{alltt}
016
017   /* low level squaring, b = a*a, HAC pp.596-597, Algorithm 14.16 */
018   int s_mp_sqr (mp_int * a, mp_int * b)
019   \{
020     mp_int  t;
021     int     res, ix, iy, pa;
022     mp_word r;
023     mp_digit u, tmpx, *tmpt;
024
025     pa = a->used;
026     if ((res = mp_init_size (&t, (2 * pa) + 1)) != MP_OKAY) \{
027       return res;
028     \}
029
030     /* default used is maximum possible size */
031     t.used = (2 * pa) + 1;
032
033     for (ix = 0; ix < pa; ix++) \{
034       /* first calculate the digit at 2*ix */
035       /* calculate double precision result */
036       r = (mp_word)t.dp[2*ix] +
037           ((mp_word)a->dp[ix] * (mp_word)a->dp[ix]);
038
039       /* store lower part in result */
040       t.dp[ix+ix] = (mp_digit) (r & ((mp_word) MP_MASK));
041
042       /* get the carry */
043       u           = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
044
045       /* left hand side of A[ix] * A[iy] */
046       tmpx        = a->dp[ix];
047
048       /* alias for where to store the results */
049       tmpt        = t.dp + ((2 * ix) + 1);
050
051       for (iy = ix + 1; iy < pa; iy++) \{
052         /* first calculate the product */
053         r       = ((mp_word)tmpx) * ((mp_word)a->dp[iy]);
054
055         /* now calculate the double precision result, note we use
056          * addition instead of *2 since it's easier to optimize
057          */
058         r       = ((mp_word) *tmpt) + r + r + ((mp_word) u);
059
060         /* store lower part */
061         *tmpt++ = (mp_digit) (r & ((mp_word) MP_MASK));
062
063         /* get carry */
064         u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
065       \}
066       /* propagate upwards */
067       while (u != ((mp_digit) 0)) \{
068         r       = ((mp_word) *tmpt) + ((mp_word) u);
069         *tmpt++ = (mp_digit) (r & ((mp_word) MP_MASK));
070         u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
071       \}
072     \}
073
074     mp_clamp (&t);
075     mp_exch (&t, b);
076     mp_clear (&t);
077     return MP_OKAY;
078   \}
079   #endif
080
\end{alltt}
\end{small}

Inside the outer loop (line 33) the square term is calculated on line 36.  The carry (line 43) has been
extracted from the mp\_word accumulator using a right shift.  Aliases for $a_{ix}$ and $t_{ix+iy}$ are initialized
(lines 46 and 49) to simplify the inner loop.  The doubling is performed using two
additions (line 58) since it is usually faster than shifting, if not at least as fast.

The important observation is that the inner loop does not begin at $iy = 0$ like for multiplication.  As such the inner loops
get progressively shorter as the algorithm proceeds.  This is what leads to the savings compared to using a multiplication to
square a number.

\subsection{Faster Squaring by the ``Comba'' Method}
A major drawback to the baseline method is the requirement for single precision shifting inside the $O(n^2)$ nested loop.  Squaring has an additional
drawback that it must double the product inside the inner loop as well.  As for multiplication, the Comba technique can be used to eliminate these
performance hazards.

The first obvious solution is to make an array of mp\_words which will hold all of the columns.  This will indeed eliminate all of the carry
propagation operations from the inner loop.  However, the inner product must still be doubled $O(n^2)$ times.  The solution stems from the simple fact
that $2a + 2b + 2c = 2(a + b + c)$.  That is the sum of all of the double products is equal to double the sum of all the products.  For example,
$ab + ba + ac + ca = 2ab + 2ac = 2(ab + ac)$.

However, we cannot simply double all of the columns, since the squares appear only once per row.  The most practical solution is to have two
mp\_word arrays.  One array will hold the squares and the other array will hold the double products.  With both arrays the doubling and
carry propagation can be moved to a $O(n)$ work level outside the $O(n^2)$ level.  In this case, we have an even simpler solution in mind.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{fast\_s\_mp\_sqr}. \\
\textbf{Input}.   mp\_int $a$ \\
\textbf{Output}.  $b \leftarrow a^2$ \\
\hline \\
Place an array of \textbf{MP\_WARRAY} mp\_digits named $W$ on the stack. \\
1.  If $b.alloc < 2a.used + 1$ then grow $b$ to $2a.used + 1$ digits.  (\textit{mp\_grow}). \\
2.  If step 1 failed return(\textit{MP\_MEM}). \\
\\
3.  $pa \leftarrow 2 \cdot a.used$ \\
4.  $\hat W1 \leftarrow 0$ \\
5.  for $ix$ from $0$ to $pa - 1$ do \\
\hspace{3mm}5.1  $\_ \hat W \leftarrow 0$ \\
\hspace{3mm}5.2  $ty \leftarrow \mbox{MIN}(a.used - 1, ix)$ \\
\hspace{3mm}5.3  $tx \leftarrow ix - ty$ \\
\hspace{3mm}5.4  $iy \leftarrow \mbox{MIN}(a.used - tx, ty + 1)$ \\
\hspace{3mm}5.5  $iy \leftarrow \mbox{MIN}(iy, \lfloor \left (ty - tx + 1 \right )/2 \rfloor)$ \\
\hspace{3mm}5.6  for $iz$ from $0$ to $iz - 1$ do \\
\hspace{6mm}5.6.1  $\_ \hat W \leftarrow \_ \hat W + a_{tx + iz}a_{ty - iz}$ \\
\hspace{3mm}5.7  $\_ \hat W \leftarrow 2 \cdot \_ \hat W  + \hat W1$ \\
\hspace{3mm}5.8  if $ix$ is even then \\
\hspace{6mm}5.8.1  $\_ \hat W \leftarrow \_ \hat W + \left ( a_{\lfloor ix/2 \rfloor}\right )^2$ \\
\hspace{3mm}5.9  $W_{ix} \leftarrow \_ \hat W (\mbox{mod }\beta)$ \\
\hspace{3mm}5.10  $\hat W1 \leftarrow \lfloor \_ \hat W / \beta \rfloor$ \\
\\
6.  $oldused \leftarrow b.used$ \\
7.  $b.used \leftarrow 2 \cdot a.used$ \\
8.  for $ix$ from $0$ to $pa - 1$ do \\
\hspace{3mm}8.1  $b_{ix} \leftarrow W_{ix}$ \\
9.  for $ix$ from $pa$ to $oldused - 1$ do \\
\hspace{3mm}9.1  $b_{ix} \leftarrow 0$ \\
10.  Clamp excess digits from $b$.  (\textit{mp\_clamp}) \\
11.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm fast\_s\_mp\_sqr}
\end{figure}

\textbf{Algorithm fast\_s\_mp\_sqr.}
This algorithm computes the square of an input using the Comba technique.  It is designed to be a replacement for algorithm
s\_mp\_sqr when the number of input digits is less than \textbf{MP\_WARRAY} and less than $\delta \over 2$.
This algorithm is very similar to the Comba multiplier except with a few key differences we shall make note of.

First, we have an accumulator and carry variables $\_ \hat W$ and $\hat W1$ respectively.  This is because the inner loop
products are to be doubled.  If we had added the previous carry in we would be doubling too much.  Next we perform an
addition MIN condition on $iy$ (step 5.5) to prevent overlapping digits.  For example, $a_3 \cdot a_5$ is equal
$a_5 \cdot a_3$.  Whereas in the multiplication case we would have $5 < a.used$ and $3 \ge 0$ is maintained since we double the sum
of the products just outside the inner loop we have to avoid doing this.  This is also a good thing since we perform
fewer multiplications and the routine ends up being faster.

Finally the last difference is the addition of the ``square'' term outside the inner loop (step 5.8).  We add in the square
only to even outputs and it is the square of the term at the $\lfloor ix / 2 \rfloor$ position.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_fast\_s\_mp\_sqr.c
\vspace{-3mm}
\begin{alltt}
016
017   /* the jist of squaring...
018    * you do like mult except the offset of the tmpx [one that
019    * starts closer to zero] can't equal the offset of tmpy.
020    * So basically you set up iy like before then you min it with
021    * (ty-tx) so that it never happens.  You double all those
022    * you add in the inner loop
023
024   After that loop you do the squares and add them in.
025   */
026
027   int fast_s_mp_sqr (mp_int * a, mp_int * b)
028   \{
029     int       olduse, res, pa, ix, iz;
030     mp_digit   W[MP_WARRAY], *tmpx;
031     mp_word   W1;
032
033     /* grow the destination as required */
034     pa = a->used + a->used;
035     if (b->alloc < pa) \{
036       if ((res = mp_grow (b, pa)) != MP_OKAY) \{
037         return res;
038       \}
039     \}
040
041     /* number of output digits to produce */
042     W1 = 0;
043     for (ix = 0; ix < pa; ix++) \{
044         int      tx, ty, iy;
045         mp_word  _W;
046         mp_digit *tmpy;
047
048         /* clear counter */
049         _W = 0;
050
051         /* get offsets into the two bignums */
052         ty = MIN(a->used-1, ix);
053         tx = ix - ty;
054
055         /* setup temp aliases */
056         tmpx = a->dp + tx;
057         tmpy = a->dp + ty;
058
059         /* this is the number of times the loop will iterrate, essentially
060            while (tx++ < a->used && ty-- >= 0) \{ ... \}
061          */
062         iy = MIN(a->used-tx, ty+1);
063
064         /* now for squaring tx can never equal ty
065          * we halve the distance since they approach at a rate of 2x
066          * and we have to round because odd cases need to be executed
067          */
068         iy = MIN(iy, ((ty-tx)+1)>>1);
069
070         /* execute loop */
071         for (iz = 0; iz < iy; iz++) \{
072            _W += ((mp_word)*tmpx++)*((mp_word)*tmpy--);
073         \}
074
075         /* double the inner product and add carry */
076         _W = _W + _W + W1;
077
078         /* even columns have the square term in them */
079         if ((ix&1) == 0) \{
080            _W += ((mp_word)a->dp[ix>>1])*((mp_word)a->dp[ix>>1]);
081         \}
082
083         /* store it */
084         W[ix] = (mp_digit)(_W & MP_MASK);
085
086         /* make next carry */
087         W1 = _W >> ((mp_word)DIGIT_BIT);
088     \}
089
090     /* setup dest */
091     olduse  = b->used;
092     b->used = a->used+a->used;
093
094     \{
095       mp_digit *tmpb;
096       tmpb = b->dp;
097       for (ix = 0; ix < pa; ix++) \{
098         *tmpb++ = W[ix] & MP_MASK;
099       \}
100
101       /* clear unused digits [that existed in the old copy of c] */
102       for (; ix < olduse; ix++) \{
103         *tmpb++ = 0;
104       \}
105     \}
106     mp_clamp (b);
107     return MP_OKAY;
108   \}
109   #endif
110
\end{alltt}
\end{small}

This implementation is essentially a copy of Comba multiplication with the appropriate changes added to make it faster for
the special case of squaring.

\subsection{Polynomial Basis Squaring}
The same algorithm that performs optimal polynomial basis multiplication can be used to perform polynomial basis squaring.  The minor exception
is that $\zeta_y = f(y)g(y)$ is actually equivalent to $\zeta_y = f(y)^2$ since $f(y) = g(y)$.  Instead of performing $2n + 1$
multiplications to find the $\zeta$ relations, squaring operations are performed instead.

\subsection{Karatsuba Squaring}
Let $f(x) = ax + b$ represent the polynomial basis representation of a number to square.
Let $h(x) = \left ( f(x) \right )^2$ represent the square of the polynomial.  The Karatsuba equation can be modified to square a
number with the following equation.

\begin{equation}
h(x) = a^2x^2 + \left ((a + b)^2 - (a^2 + b^2) \right )x + b^2
\end{equation}

Upon closer inspection this equation only requires the calculation of three half-sized squares: $a^2$, $b^2$ and $(a + b)^2$.  As in
Karatsuba multiplication, this algorithm can be applied recursively on the input and will achieve an asymptotic running time of
$O \left ( n^{lg(3)} \right )$.

If the asymptotic times of Karatsuba squaring and multiplication are the same, why not simply use the multiplication algorithm
instead?  The answer to this arises from the cutoff point for squaring.  As in multiplication there exists a cutoff point, at which the
time required for a Comba based squaring and a Karatsuba based squaring meet.  Due to the overhead inherent in the Karatsuba method, the cutoff
point is fairly high.  For example, on an AMD Athlon XP processor with $\beta = 2^{28}$, the cutoff point is around 127 digits.

Consider squaring a 200 digit number with this technique.  It will be split into two 100 digit halves which are subsequently squared.
The 100 digit halves will not be squared using Karatsuba, but instead using the faster Comba based squaring algorithm.  If Karatsuba multiplication
were used instead, the 100 digit numbers would be squared with a slower Comba based multiplication.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_karatsuba\_sqr}. \\
\textbf{Input}.   mp\_int $a$ \\
\textbf{Output}.  $b \leftarrow a^2$ \\
\hline \\
1.  Initialize the following temporary mp\_ints:  $x0$, $x1$, $t1$, $t2$, $x0x0$ and $x1x1$. \\
2.  If any of the initializations on step 1 failed return(\textit{MP\_MEM}). \\
\\
Split the input.  e.g. $a = x1\beta^B + x0$ \\
3.  $B \leftarrow \lfloor a.used / 2 \rfloor$ \\
4.  $x0 \leftarrow a \mbox{ (mod }\beta^B\mbox{)}$ (\textit{mp\_mod\_2d}) \\
5.  $x1 \leftarrow \lfloor a / \beta^B \rfloor$ (\textit{mp\_lshd}) \\
\\
Calculate the three squares. \\
6.  $x0x0 \leftarrow x0^2$ (\textit{mp\_sqr}) \\
7.  $x1x1 \leftarrow x1^2$ \\
8.  $t1 \leftarrow x1 + x0$ (\textit{s\_mp\_add}) \\
9.  $t1 \leftarrow t1^2$ \\
\\
Compute the middle term. \\
10.  $t2 \leftarrow x0x0 + x1x1$ (\textit{s\_mp\_add}) \\
11.  $t1 \leftarrow t1 - t2$ \\
\\
Compute final product. \\
12.  $t1 \leftarrow t1\beta^B$ (\textit{mp\_lshd}) \\
13.  $x1x1 \leftarrow x1x1\beta^{2B}$ \\
14.  $t1 \leftarrow t1 + x0x0$ \\
15.  $b \leftarrow t1 + x1x1$ \\
16.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_karatsuba\_sqr}
\end{figure}

\textbf{Algorithm mp\_karatsuba\_sqr.}
This algorithm computes the square of an input $a$ using the Karatsuba technique.  This algorithm is very similar to the Karatsuba based
multiplication algorithm with the exception that the three half-size multiplications have been replaced with three half-size squarings.

The radix point for squaring is simply placed exactly in the middle of the digits when the input has an odd number of digits, otherwise it is
placed just below the middle.  Step 3, 4 and 5 compute the two halves required using $B$
as the radix point.  The first two squares in steps 6 and 7 are rather straightforward while the last square is of a more compact form.

By expanding $\left (x1 + x0 \right )^2$, the $x1^2$ and $x0^2$ terms in the middle disappear, that is $(x0 - x1)^2 - (x1^2 + x0^2)  = 2 \cdot x0 \cdot x1$.
Now if $5n$ single precision additions and a squaring of $n$-digits is faster than multiplying two $n$-digit numbers and doubling then
this method is faster.  Assuming no further recursions occur, the difference can be estimated with the following inequality.

Let $p$ represent the cost of a single precision addition and $q$ the cost of a single precision multiplication both in terms of time\footnote{Or
machine clock cycles.}.

\begin{equation}
5pn +{{q(n^2 + n)} \over 2} \le pn + qn^2
\end{equation}

For example, on an AMD Athlon XP processor $p = {1 \over 3}$ and $q = 6$.  This implies that the following inequality should hold.
\begin{center}
\begin{tabular}{rcl}
${5n \over 3} + 3n^2 + 3n$     & $<$ & ${n \over 3} + 6n^2$ \\
${5 \over 3} + 3n + 3$     & $<$ & ${1 \over 3} + 6n$ \\
${13 \over 9}$     & $<$ & $n$ \\
\end{tabular}
\end{center}

This results in a cutoff point around $n = 2$.  As a consequence it is actually faster to compute the middle term the ``long way'' on processors
where multiplication is substantially slower\footnote{On the Athlon there is a 1:17 ratio between clock cycles for addition and multiplication.  On
the Intel P4 processor this ratio is 1:29 making this method even more beneficial.  The only common exception is the ARMv4 processor which has a
ratio of 1:7.  } than simpler operations such as addition.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_karatsuba\_sqr.c
\vspace{-3mm}
\begin{alltt}
016
017   /* Karatsuba squaring, computes b = a*a using three
018    * half size squarings
019    *
020    * See comments of karatsuba_mul for details.  It
021    * is essentially the same algorithm but merely
022    * tuned to perform recursive squarings.
023    */
024   int mp_karatsuba_sqr (mp_int * a, mp_int * b)
025   \{
026     mp_int  x0, x1, t1, t2, x0x0, x1x1;
027     int     B, err;
028
029     err = MP_MEM;
030
031     /* min # of digits */
032     B = a->used;
033
034     /* now divide in two */
035     B = B >> 1;
036
037     /* init copy all the temps */
038     if (mp_init_size (&x0, B) != MP_OKAY)
039       goto ERR;
040     if (mp_init_size (&x1, a->used - B) != MP_OKAY)
041       goto X0;
042
043     /* init temps */
044     if (mp_init_size (&t1, a->used * 2) != MP_OKAY)
045       goto X1;
046     if (mp_init_size (&t2, a->used * 2) != MP_OKAY)
047       goto T1;
048     if (mp_init_size (&x0x0, B * 2) != MP_OKAY)
049       goto T2;
050     if (mp_init_size (&x1x1, (a->used - B) * 2) != MP_OKAY)
051       goto X0X0;
052
053     \{
054       int x;
055       mp_digit *dst, *src;
056
057       src = a->dp;
058
059       /* now shift the digits */
060       dst = x0.dp;
061       for (x = 0; x < B; x++) \{
062         *dst++ = *src++;
063       \}
064
065       dst = x1.dp;
066       for (x = B; x < a->used; x++) \{
067         *dst++ = *src++;
068       \}
069     \}
070
071     x0.used = B;
072     x1.used = a->used - B;
073
074     mp_clamp (&x0);
075
076     /* now calc the products x0*x0 and x1*x1 */
077     if (mp_sqr (&x0, &x0x0) != MP_OKAY)
078       goto X1X1;           /* x0x0 = x0*x0 */
079     if (mp_sqr (&x1, &x1x1) != MP_OKAY)
080       goto X1X1;           /* x1x1 = x1*x1 */
081
082     /* now calc (x1+x0)**2 */
083     if (s_mp_add (&x1, &x0, &t1) != MP_OKAY)
084       goto X1X1;           /* t1 = x1 - x0 */
085     if (mp_sqr (&t1, &t1) != MP_OKAY)
086       goto X1X1;           /* t1 = (x1 - x0) * (x1 - x0) */
087
088     /* add x0y0 */
089     if (s_mp_add (&x0x0, &x1x1, &t2) != MP_OKAY)
090       goto X1X1;           /* t2 = x0x0 + x1x1 */
091     if (s_mp_sub (&t1, &t2, &t1) != MP_OKAY)
092       goto X1X1;           /* t1 = (x1+x0)**2 - (x0x0 + x1x1) */
093
094     /* shift by B */
095     if (mp_lshd (&t1, B) != MP_OKAY)
096       goto X1X1;           /* t1 = (x0x0 + x1x1 - (x1-x0)*(x1-x0))<<B */
097     if (mp_lshd (&x1x1, B * 2) != MP_OKAY)
098       goto X1X1;           /* x1x1 = x1x1 << 2*B */
099
100     if (mp_add (&x0x0, &t1, &t1) != MP_OKAY)
101       goto X1X1;           /* t1 = x0x0 + t1 */
102     if (mp_add (&t1, &x1x1, b) != MP_OKAY)
103       goto X1X1;           /* t1 = x0x0 + t1 + x1x1 */
104
105     err = MP_OKAY;
106
107   X1X1:mp_clear (&x1x1);
108   X0X0:mp_clear (&x0x0);
109   T2:mp_clear (&t2);
110   T1:mp_clear (&t1);
111   X1:mp_clear (&x1);
112   X0:mp_clear (&x0);
113   ERR:
114     return err;
115   \}
116   #endif
117
\end{alltt}
\end{small}

This implementation is largely based on the implementation of algorithm mp\_karatsuba\_mul.  It uses the same inline style to copy and
shift the input into the two halves.  The loop from line 53 to line 69 has been modified since only one input exists.  The \textbf{used}
count of both $x0$ and $x1$ is fixed up and $x0$ is clamped before the calculations begin.  At this point $x1$ and $x0$ are valid equivalents
to the respective halves as if mp\_rshd and mp\_mod\_2d had been used.

By inlining the copy and shift operations the cutoff point for Karatsuba multiplication can be lowered.  On the Athlon the cutoff point
is exactly at the point where Comba squaring can no longer be used (\textit{128 digits}).  On slower processors such as the Intel P4
it is actually below the Comba limit (\textit{at 110 digits}).

This routine uses the same error trap coding style as mp\_karatsuba\_sqr.  As the temporary variables are initialized errors are
redirected to the error trap higher up.  If the algorithm completes without error the error code is set to \textbf{MP\_OKAY} and
mp\_clears are executed normally.

\subsection{Toom-Cook Squaring}
The Toom-Cook squaring algorithm mp\_toom\_sqr is heavily based on the algorithm mp\_toom\_mul with the exception that squarings are used
instead of multiplication to find the five relations.  The reader is encouraged to read the description of the latter algorithm and try to
derive their own Toom-Cook squaring algorithm.

\subsection{High Level Squaring}
\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_sqr}. \\
\textbf{Input}.   mp\_int $a$ \\
\textbf{Output}.  $b \leftarrow a^2$ \\
\hline \\
1.  If $a.used \ge TOOM\_SQR\_CUTOFF$ then  \\
\hspace{3mm}1.1  $b \leftarrow a^2$ using algorithm mp\_toom\_sqr \\
2.  else if $a.used \ge KARATSUBA\_SQR\_CUTOFF$ then \\
\hspace{3mm}2.1  $b \leftarrow a^2$ using algorithm mp\_karatsuba\_sqr \\
3.  else \\
\hspace{3mm}3.1  $digs \leftarrow a.used + b.used + 1$ \\
\hspace{3mm}3.2  If $digs < MP\_ARRAY$ and $a.used \le \delta$ then \\
\hspace{6mm}3.2.1  $b \leftarrow a^2$ using algorithm fast\_s\_mp\_sqr.  \\
\hspace{3mm}3.3  else \\
\hspace{6mm}3.3.1  $b \leftarrow a^2$ using algorithm s\_mp\_sqr.  \\
4.  $b.sign \leftarrow MP\_ZPOS$ \\
5.  Return the result of the unsigned squaring performed. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_sqr}
\end{figure}

\textbf{Algorithm mp\_sqr.}
This algorithm computes the square of the input using one of four different algorithms.  If the input is very large and has at least
\textbf{TOOM\_SQR\_CUTOFF} or \textbf{KARATSUBA\_SQR\_CUTOFF} digits then either the Toom-Cook or the Karatsuba Squaring algorithm is used.  If
neither of the polynomial basis algorithms should be used then either the Comba or baseline algorithm is used.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_sqr.c
\vspace{-3mm}
\begin{alltt}
016
017   /* computes b = a*a */
018   int
019   mp_sqr (mp_int * a, mp_int * b)
020   \{
021     int     res;
022
023   #ifdef BN_MP_TOOM_SQR_C
024     /* use Toom-Cook? */
025     if (a->used >= TOOM_SQR_CUTOFF) \{
026       res = mp_toom_sqr(a, b);
027     /* Karatsuba? */
028     \} else
029   #endif
030   #ifdef BN_MP_KARATSUBA_SQR_C
031     if (a->used >= KARATSUBA_SQR_CUTOFF) \{
032       res = mp_karatsuba_sqr (a, b);
033     \} else
034   #endif
035     \{
036   #ifdef BN_FAST_S_MP_SQR_C
037       /* can we use the fast comba multiplier? */
038       if ((((a->used * 2) + 1) < MP_WARRAY) &&
039            (a->used <
040            (1 << (((sizeof(mp_word) * CHAR_BIT) - (2 * DIGIT_BIT)) - 1)))) \{
041         res = fast_s_mp_sqr (a, b);
042       \} else
043   #endif
044       \{
045   #ifdef BN_S_MP_SQR_C
046         res = s_mp_sqr (a, b);
047   #else
048         res = MP_VAL;
049   #endif
050       \}
051     \}
052     b->sign = MP_ZPOS;
053     return res;
054   \}
055   #endif
056
\end{alltt}
\end{small}

\section*{Exercises}
\begin{tabular}{cl}
$\left [ 3 \right ] $ & Devise an efficient algorithm for selection of the radix point to handle inputs \\
                      & that have different number of digits in Karatsuba multiplication. \\
                      & \\
$\left [ 2 \right ] $ & In section 5.3 the fact that every column of a squaring is made up \\
                      & of double products and at most one square is stated.  Prove this statement. \\
                      & \\
$\left [ 3 \right ] $ & Prove the equation for Karatsuba squaring. \\
                      & \\
$\left [ 1 \right ] $ & Prove that Karatsuba squaring requires $O \left (n^{lg(3)} \right )$ time. \\
                      & \\
$\left [ 2 \right ] $ & Determine the minimal ratio between addition and multiplication clock cycles \\
                      & required for equation $6.7$ to be true.  \\
                      & \\
$\left [ 3 \right ] $ & Implement a threaded version of Comba multiplication (and squaring) where you \\
                      & compute subsets of the columns in each thread.  Determine a cutoff point where \\
                      & it is effective and add the logic to mp\_mul() and mp\_sqr(). \\
                      &\\
$\left [ 4 \right ] $ & Same as the previous but also modify the Karatsuba and Toom-Cook.  You must \\
                      & increase the throughput of mp\_exptmod() for random odd moduli in the range \\
                      & $512 \ldots 4096$ bits significantly ($> 2x$) to complete this challenge. \\
                      & \\
\end{tabular}

\chapter{Modular Reduction}
\section{Basics of Modular Reduction}
\index{modular residue}
Modular reduction is an operation that arises quite often within public key cryptography algorithms and various number theoretic algorithms,
such as factoring.  Modular reduction algorithms are the third class of algorithms of the ``multipliers'' set.  A number $a$ is said to be \textit{reduced}
modulo another number $b$ by finding the remainder of the division $a/b$.  Full integer division with remainder is a topic to be covered
in~\ref{sec:division}.

Modular reduction is equivalent to solving for $r$ in the following equation.  $a = bq + r$ where $q = \lfloor a/b \rfloor$.  The result
$r$ is said to be ``congruent to $a$ modulo $b$'' which is also written as $r \equiv a \mbox{ (mod }b\mbox{)}$.  In other vernacular $r$ is known as the
``modular residue'' which leads to ``quadratic residue''\footnote{That's fancy talk for $b \equiv a^2 \mbox{ (mod }p\mbox{)}$.} and
other forms of residues.

Modular reductions are normally used to create either finite groups, rings or fields.  The most common usage for performance driven modular reductions
is in modular exponentiation algorithms.  That is to compute $d = a^b \mbox{ (mod }c\mbox{)}$ as fast as possible.  This operation is used in the
RSA and Diffie-Hellman public key algorithms, for example.  Modular multiplication and squaring also appears as a fundamental operation in
elliptic curve cryptographic algorithms.  As will be discussed in the subsequent chapter there exist fast algorithms for computing modular
exponentiations without having to perform (\textit{in this example}) $b - 1$ multiplications.  These algorithms will produce partial results in the
range $0 \le x < c^2$ which can be taken advantage of to create several efficient algorithms.   They have also been used to create redundancy check
algorithms known as CRCs, error correction codes such as Reed-Solomon and solve a variety of number theoeretic problems.

\section{The Barrett Reduction}
The Barrett reduction algorithm \cite{BARRETT} was inspired by fast division algorithms which multiply by the reciprocal to emulate
division.  Barretts observation was that the residue $c$ of $a$ modulo $b$ is equal to

\begin{equation}
c = a - b \cdot \lfloor a/b \rfloor
\end{equation}

Since algorithms such as modular exponentiation would be using the same modulus extensively, typical DSP\footnote{It is worth noting that Barrett's paper
targeted the DSP56K processor.}  intuition would indicate the next step would be to replace $a/b$ by a multiplication by the reciprocal.  However,
DSP intuition on its own will not work as these numbers are considerably larger than the precision of common DSP floating point data types.
It would take another common optimization to optimize the algorithm.

\subsection{Fixed Point Arithmetic}
The trick used to optimize the above equation is based on a technique of emulating floating point data types with fixed precision integers.  Fixed
point arithmetic would become very popular as it greatly optimize the ``3d-shooter'' genre of games in the mid 1990s when floating point units were
fairly slow if not unavailable.   The idea behind fixed point arithmetic is to take a normal $k$-bit integer data type and break it into $p$-bit
integer and a $q$-bit fraction part (\textit{where $p+q = k$}).

In this system a $k$-bit integer $n$ would actually represent $n/2^q$.  For example, with $q = 4$ the integer $n = 37$ would actually represent the
value $2.3125$.  To multiply two fixed point numbers the integers are multiplied using traditional arithmetic and subsequently normalized by
moving the implied decimal point back to where it should be.  For example, with $q = 4$ to multiply the integers $9$ and $5$ they must be converted
to fixed point first by multiplying by $2^q$.  Let $a = 9(2^q)$ represent the fixed point representation of $9$ and $b = 5(2^q)$ represent the
fixed point representation of $5$.  The product $ab$ is equal to $45(2^{2q})$ which when normalized by dividing by $2^q$ produces $45(2^q)$.

This technique became popular since a normal integer multiplication and logical shift right are the only required operations to perform a multiplication
of two fixed point numbers.  Using fixed point arithmetic, division can be easily approximated by multiplying by the reciprocal.  If $2^q$ is
equivalent to one than $2^q/b$ is equivalent to the fixed point approximation of $1/b$ using real arithmetic.  Using this fact dividing an integer
$a$ by another integer $b$ can be achieved with the following expression.

\begin{equation}
\lfloor a / b \rfloor \mbox{ }\approx\mbox{ } \lfloor (a \cdot \lfloor 2^q / b \rfloor)/2^q \rfloor
\end{equation}

The precision of the division is proportional to the value of $q$.  If the divisor $b$ is used frequently as is the case with
modular exponentiation pre-computing $2^q/b$ will allow a division to be performed with a multiplication and a right shift.  Both operations
are considerably faster than division on most processors.

Consider dividing $19$ by $5$.  The correct result is $\lfloor 19/5 \rfloor = 3$.  With $q = 3$ the reciprocal is $\lfloor 2^q/5 \rfloor = 1$ which
leads to a product of $19$ which when divided by $2^q$ produces $2$.  However, with $q = 4$ the reciprocal is $\lfloor 2^q/5 \rfloor = 3$ and
the result of the emulated division is $\lfloor 3 \cdot 19 / 2^q \rfloor = 3$ which is correct.  The value of $2^q$ must be close to or ideally
larger than the dividend.  In effect if $a$ is the dividend then $q$ should allow $0 \le \lfloor a/2^q \rfloor \le 1$ in order for this approach
to work correctly.  Plugging this form of divison into the original equation the following modular residue equation arises.

\begin{equation}
c = a - b \cdot \lfloor (a \cdot \lfloor 2^q / b \rfloor)/2^q \rfloor
\end{equation}

Using the notation from \cite{BARRETT} the value of $\lfloor 2^q / b \rfloor$ will be represented by the $\mu$ symbol.  Using the $\mu$
variable also helps re-inforce the idea that it is meant to be computed once and re-used.

\begin{equation}
c = a - b \cdot \lfloor (a \cdot \mu)/2^q \rfloor
\end{equation}

Provided that $2^q \ge a$ this algorithm will produce a quotient that is either exactly correct or off by a value of one.  In the context of Barrett
reduction the value of $a$ is bound by $0 \le a \le (b - 1)^2$ meaning that $2^q \ge b^2$ is sufficient to ensure the reciprocal will have enough
precision.

Let $n$ represent the number of digits in $b$.  This algorithm requires approximately $2n^2$ single precision multiplications to produce the quotient and
another $n^2$ single precision multiplications to find the residue.  In total $3n^2$ single precision multiplications are required to
reduce the number.

For example, if $b = 1179677$ and $q = 41$ ($2^q > b^2$), then the reciprocal $\mu$ is equal to $\lfloor 2^q / b \rfloor = 1864089$.  Consider reducing
$a = 180388626447$ modulo $b$ using the above reduction equation.  The quotient using the new formula is $\lfloor (a \cdot \mu) / 2^q \rfloor = 152913$.
By subtracting $152913b$ from $a$ the correct residue $a \equiv 677346 \mbox{ (mod }b\mbox{)}$ is found.

\subsection{Choosing a Radix Point}
Using the fixed point representation a modular reduction can be performed with $3n^2$ single precision multiplications.  If that were the best
that could be achieved a full division\footnote{A division requires approximately $O(2cn^2)$ single precision multiplications for a small value of $c$.
See~\ref{sec:division} for further details.} might as well be used in its place.  The key to optimizing the reduction is to reduce the precision of
the initial multiplication that finds the quotient.

Let $a$ represent the number of which the residue is sought.  Let $b$ represent the modulus used to find the residue.  Let $m$ represent
the number of digits in $b$.  For the purposes of this discussion we will assume that the number of digits in $a$ is $2m$, which is generally true if
two $m$-digit numbers have been multiplied.  Dividing $a$ by $b$ is the same as dividing a $2m$ digit integer by a $m$ digit integer.  Digits below the
$m - 1$'th digit of $a$ will contribute at most a value of $1$ to the quotient because $\beta^k < b$ for any $0 \le k \le m - 1$.  Another way to
express this is by re-writing $a$ as two parts.  If $a' \equiv a \mbox{ (mod }b^m\mbox{)}$ and $a'' = a - a'$ then
${a \over b} \equiv {{a' + a''} \over b}$ which is equivalent to ${a' \over b} + {a'' \over b}$.  Since $a'$ is bound to be less than $b$ the quotient
is bound by $0 \le {a' \over b} < 1$.

Since the digits of $a'$ do not contribute much to the quotient the observation is that they might as well be zero.  However, if the digits
``might as well be zero'' they might as well not be there in the first place.  Let $q_0 = \lfloor a/\beta^{m-1} \rfloor$ represent the input
with the irrelevant digits trimmed.  Now the modular reduction is trimmed to the almost equivalent equation

\begin{equation}
c = a - b \cdot \lfloor (q_0 \cdot \mu) / \beta^{m+1} \rfloor
\end{equation}

Note that the original divisor $2^q$ has been replaced with $\beta^{m+1}$ where in this case $q$ is a multiple of $lg(\beta)$. Also note that the
exponent on the divisor when added to the amount $q_0$ was shifted by equals $2m$.  If the optimization had not been performed the divisor
would have the exponent $2m$ so in the end the exponents do ``add up''. Using the above equation the quotient
$\lfloor (q_0 \cdot \mu) / \beta^{m+1} \rfloor$ can be off from the true quotient by at most two.  The original fixed point quotient can be off
by as much as one (\textit{provided the radix point is chosen suitably}) and now that the lower irrelevent digits have been trimmed the quotient
can be off by an additional value of one for a total of at most two.  This implies that
$0 \le a - b \cdot \lfloor (q_0 \cdot \mu) / \beta^{m+1} \rfloor < 3b$.  By first subtracting $b$ times the quotient and then conditionally subtracting
$b$ once or twice the residue is found.

The quotient is now found using $(m + 1)(m) = m^2 + m$ single precision multiplications and the residue with an additional $m^2$ single
precision multiplications, ignoring the subtractions required.  In total $2m^2 + m$ single precision multiplications are required to find the residue.
This is considerably faster than the original attempt.

For example, let $\beta = 10$ represent the radix of the digits.  Let $b = 9999$ represent the modulus which implies $m = 4$. Let $a = 99929878$
represent the value of which the residue is desired.  In this case $q = 8$ since $10^7 < 9999^2$ meaning that $\mu = \lfloor \beta^{q}/b \rfloor = 10001$.
With the new observation the multiplicand for the quotient is equal to $q_0 = \lfloor a / \beta^{m - 1} \rfloor = 99929$.  The quotient is then
$\lfloor (q_0 \cdot \mu) / \beta^{m+1} \rfloor = 9993$.  Subtracting $9993b$ from $a$ and the correct residue $a \equiv 9871 \mbox{ (mod }b\mbox{)}$
is found.

\subsection{Trimming the Quotient}
So far the reduction algorithm has been optimized from $3m^2$ single precision multiplications down to $2m^2 + m$ single precision multiplications.  As
it stands now the algorithm is already fairly fast compared to a full integer division algorithm.  However, there is still room for
optimization.

After the first multiplication inside the quotient ($q_0 \cdot \mu$) the value is shifted right by $m + 1$ places effectively nullifying the lower
half of the product.  It would be nice to be able to remove those digits from the product to effectively cut down the number of single precision
multiplications.  If the number of digits in the modulus $m$ is far less than $\beta$ a full product is not required for the algorithm to work properly.
In fact the lower $m - 2$ digits will not affect the upper half of the product at all and do not need to be computed.

The value of $\mu$ is a $m$-digit number and $q_0$ is a $m + 1$ digit number.  Using a full multiplier $(m + 1)(m) = m^2 + m$ single precision
multiplications would be required.  Using a multiplier that will only produce digits at and above the $m - 1$'th digit reduces the number
of single precision multiplications to ${m^2 + m} \over 2$ single precision multiplications.

\subsection{Trimming the Residue}
After the quotient has been calculated it is used to reduce the input.  As previously noted the algorithm is not exact and it can be off by a small
multiple of the modulus, that is $0 \le a - b \cdot \lfloor (q_0 \cdot \mu) / \beta^{m+1} \rfloor < 3b$.  If $b$ is $m$ digits than the
result of reduction equation is a value of at most $m + 1$ digits (\textit{provided $3 < \beta$}) implying that the upper $m - 1$ digits are
implicitly zero.

The next optimization arises from this very fact.  Instead of computing $b \cdot \lfloor (q_0 \cdot \mu) / \beta^{m+1} \rfloor$ using a full
$O(m^2)$ multiplication algorithm only the lower $m+1$ digits of the product have to be computed.  Similarly the value of $a$ can
be reduced modulo $\beta^{m+1}$ before the multiple of $b$ is subtracted which simplifes the subtraction as well.  A multiplication that produces
only the lower $m+1$ digits requires ${m^2 + 3m - 2} \over 2$ single precision multiplications.

With both optimizations in place the algorithm is the algorithm Barrett proposed.  It requires $m^2 + 2m - 1$ single precision multiplications which
is considerably faster than the straightforward $3m^2$ method.

\subsection{The Barrett Algorithm}
\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_reduce}. \\
\textbf{Input}.   mp\_int $a$, mp\_int $b$ and $\mu = \lfloor \beta^{2m}/b \rfloor, m = \lceil lg_{\beta}(b) \rceil, (0 \le a < b^2, b > 1)$ \\
\textbf{Output}.  $a \mbox{ (mod }b\mbox{)}$ \\
\hline \\
Let $m$ represent the number of digits in $b$.  \\
1.  Make a copy of $a$ and store it in $q$.  (\textit{mp\_init\_copy}) \\
2.  $q \leftarrow \lfloor q / \beta^{m - 1} \rfloor$ (\textit{mp\_rshd}) \\
\\
Produce the quotient. \\
3.  $q \leftarrow q \cdot \mu$  (\textit{note: only produce digits at or above $m-1$}) \\
4.  $q \leftarrow \lfloor q / \beta^{m + 1} \rfloor$ \\
\\
Subtract the multiple of modulus from the input. \\
5.  $a \leftarrow a \mbox{ (mod }\beta^{m+1}\mbox{)}$ (\textit{mp\_mod\_2d}) \\
6.  $q \leftarrow q \cdot b \mbox{ (mod }\beta^{m+1}\mbox{)}$ (\textit{s\_mp\_mul\_digs}) \\
7.  $a \leftarrow a - q$ (\textit{mp\_sub}) \\
\\
Add $\beta^{m+1}$ if a carry occured. \\
8.  If $a < 0$ then (\textit{mp\_cmp\_d}) \\
\hspace{3mm}8.1  $q \leftarrow 1$ (\textit{mp\_set}) \\
\hspace{3mm}8.2  $q \leftarrow q \cdot \beta^{m+1}$ (\textit{mp\_lshd}) \\
\hspace{3mm}8.3  $a \leftarrow a + q$ \\
\\
Now subtract the modulus if the residue is too large (e.g. quotient too small). \\
9.  While $a \ge b$ do (\textit{mp\_cmp}) \\
\hspace{3mm}9.1  $c \leftarrow a - b$ \\
10.  Clear $q$. \\
11.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_reduce}
\end{figure}

\textbf{Algorithm mp\_reduce.}
This algorithm will reduce the input $a$ modulo $b$ in place using the Barrett algorithm.  It is loosely based on algorithm 14.42 of HAC
\cite[pp.  602]{HAC} which is based on the paper from Paul Barrett \cite{BARRETT}.  The algorithm has several restrictions and assumptions which must
be adhered to for the algorithm to work.

First the modulus $b$ is assumed to be positive and greater than one.  If the modulus were less than or equal to one than subtracting
a multiple of it would either accomplish nothing or actually enlarge the input.  The input $a$ must be in the range $0 \le a < b^2$ in order
for the quotient to have enough precision.  If $a$ is the product of two numbers that were already reduced modulo $b$, this will not be a problem.
Technically the algorithm will still work if $a \ge b^2$ but it will take much longer to finish.  The value of $\mu$ is passed as an argument to this
algorithm and is assumed to be calculated and stored before the algorithm is used.

Recall that the multiplication for the quotient on step 3 must only produce digits at or above the $m-1$'th position.  An algorithm called
$s\_mp\_mul\_high\_digs$ which has not been presented is used to accomplish this task.  The algorithm is based on $s\_mp\_mul\_digs$ except that
instead of stopping at a given level of precision it starts at a given level of precision.  This optimal algorithm can only be used if the number
of digits in $b$ is very much smaller than $\beta$.

While it is known that
$a \ge b \cdot \lfloor (q_0 \cdot \mu) / \beta^{m+1} \rfloor$ only the lower $m+1$ digits are being used to compute the residue, so an implied
``borrow'' from the higher digits might leave a negative result.  After the multiple of the modulus has been subtracted from $a$ the residue must be
fixed up in case it is negative.  The invariant $\beta^{m+1}$ must be added to the residue to make it positive again.

The while loop at step 9 will subtract $b$ until the residue is less than $b$.  If the algorithm is performed correctly this step is
performed at most twice, and on average once. However, if $a \ge b^2$ than it will iterate substantially more times than it should.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_reduce.c
\vspace{-3mm}
\begin{alltt}
016
017   /* reduces x mod m, assumes 0 < x < m**2, mu is
018    * precomputed via mp_reduce_setup.
019    * From HAC pp.604 Algorithm 14.42
020    */
021   int mp_reduce (mp_int * x, mp_int * m, mp_int * mu)
022   \{
023     mp_int  q;
024     int     res, um = m->used;
025
026     /* q = x */
027     if ((res = mp_init_copy (&q, x)) != MP_OKAY) \{
028       return res;
029     \}
030
031     /* q1 = x / b**(k-1)  */
032     mp_rshd (&q, um - 1);
033
034     /* according to HAC this optimization is ok */
035     if (((mp_digit) um) > (((mp_digit)1) << (DIGIT_BIT - 1))) \{
036       if ((res = mp_mul (&q, mu, &q)) != MP_OKAY) \{
037         goto CLEANUP;
038       \}
039     \} else \{
040   #ifdef BN_S_MP_MUL_HIGH_DIGS_C
041       if ((res = s_mp_mul_high_digs (&q, mu, &q, um)) != MP_OKAY) \{
042         goto CLEANUP;
043       \}
044   #elif defined(BN_FAST_S_MP_MUL_HIGH_DIGS_C)
045       if ((res = fast_s_mp_mul_high_digs (&q, mu, &q, um)) != MP_OKAY) \{
046         goto CLEANUP;
047       \}
048   #else
049       \{
050         res = MP_VAL;
051         goto CLEANUP;
052       \}
053   #endif
054     \}
055
056     /* q3 = q2 / b**(k+1) */
057     mp_rshd (&q, um + 1);
058
059     /* x = x mod b**(k+1), quick (no division) */
060     if ((res = mp_mod_2d (x, DIGIT_BIT * (um + 1), x)) != MP_OKAY) \{
061       goto CLEANUP;
062     \}
063
064     /* q = q * m mod b**(k+1), quick (no division) */
065     if ((res = s_mp_mul_digs (&q, m, &q, um + 1)) != MP_OKAY) \{
066       goto CLEANUP;
067     \}
068
069     /* x = x - q */
070     if ((res = mp_sub (x, &q, x)) != MP_OKAY) \{
071       goto CLEANUP;
072     \}
073
074     /* If x < 0, add b**(k+1) to it */
075     if (mp_cmp_d (x, 0) == MP_LT) \{
076       mp_set (&q, 1);
077       if ((res = mp_lshd (&q, um + 1)) != MP_OKAY)
078         goto CLEANUP;
079       if ((res = mp_add (x, &q, x)) != MP_OKAY)
080         goto CLEANUP;
081     \}
082
083     /* Back off if it's too big */
084     while (mp_cmp (x, m) != MP_LT) \{
085       if ((res = s_mp_sub (x, m, x)) != MP_OKAY) \{
086         goto CLEANUP;
087       \}
088     \}
089
090   CLEANUP:
091     mp_clear (&q);
092
093     return res;
094   \}
095   #endif
096
\end{alltt}
\end{small}

The first multiplication that determines the quotient can be performed by only producing the digits from $m - 1$ and up.  This essentially halves
the number of single precision multiplications required.  However, the optimization is only safe if $\beta$ is much larger than the number of digits
in the modulus.  In the source code this is evaluated on lines 36 to 43 where algorithm s\_mp\_mul\_high\_digs is used when it is
safe to do so.

\subsection{The Barrett Setup Algorithm}
In order to use algorithm mp\_reduce the value of $\mu$ must be calculated in advance.  Ideally this value should be computed once and stored for
future use so that the Barrett algorithm can be used without delay.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_reduce\_setup}. \\
\textbf{Input}.   mp\_int $a$ ($a > 1$)  \\
\textbf{Output}.  $\mu \leftarrow \lfloor \beta^{2m}/a \rfloor$ \\
\hline \\
1.  $\mu \leftarrow 2^{2 \cdot lg(\beta) \cdot  m}$ (\textit{mp\_2expt}) \\
2.  $\mu \leftarrow \lfloor \mu / b \rfloor$ (\textit{mp\_div}) \\
3.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_reduce\_setup}
\end{figure}

\textbf{Algorithm mp\_reduce\_setup.}
This algorithm computes the reciprocal $\mu$ required for Barrett reduction.  First $\beta^{2m}$ is calculated as $2^{2 \cdot lg(\beta) \cdot  m}$ which
is equivalent and much faster.  The final value is computed by taking the integer quotient of $\lfloor \mu / b \rfloor$.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_reduce\_setup.c
\vspace{-3mm}
\begin{alltt}
016
017   /* pre-calculate the value required for Barrett reduction
018    * For a given modulus "b" it calulates the value required in "a"
019    */
020   int mp_reduce_setup (mp_int * a, mp_int * b)
021   \{
022     int     res;
023
024     if ((res = mp_2expt (a, b->used * 2 * DIGIT_BIT)) != MP_OKAY) \{
025       return res;
026     \}
027     return mp_div (a, b, a, NULL);
028   \}
029   #endif
030
\end{alltt}
\end{small}

This simple routine calculates the reciprocal $\mu$ required by Barrett reduction.  Note the extended usage of algorithm mp\_div where the variable
which would received the remainder is passed as NULL.  As will be discussed in~\ref{sec:division} the division routine allows both the quotient and the
remainder to be passed as NULL meaning to ignore the value.

\section{The Montgomery Reduction}
Montgomery reduction\footnote{Thanks to Niels Ferguson for his insightful explanation of the algorithm.} \cite{MONT} is by far the most interesting
form of reduction in common use.  It computes a modular residue which is not actually equal to the residue of the input yet instead equal to a
residue times a constant.  However, as perplexing as this may sound the algorithm is relatively simple and very efficient.

Throughout this entire section the variable $n$ will represent the modulus used to form the residue.  As will be discussed shortly the value of
$n$ must be odd.  The variable $x$ will represent the quantity of which the residue is sought.  Similar to the Barrett algorithm the input
is restricted to $0 \le x < n^2$.  To begin the description some simple number theory facts must be established.

\textbf{Fact 1.}  Adding $n$ to $x$ does not change the residue since in effect it adds one to the quotient $\lfloor x / n \rfloor$.  Another way
to explain this is that $n$ is (\textit{or multiples of $n$ are}) congruent to zero modulo $n$.  Adding zero will not change the value of the residue.

\textbf{Fact 2.}  If $x$ is even then performing a division by two in $\Z$ is congruent to $x \cdot 2^{-1} \mbox{ (mod }n\mbox{)}$.  Actually
this is an application of the fact that if $x$ is evenly divisible by any $k \in \Z$ then division in $\Z$ will be congruent to
multiplication by $k^{-1}$ modulo $n$.

From these two simple facts the following simple algorithm can be derived.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Montgomery Reduction}. \\
\textbf{Input}.   Integer $x$, $n$ and $k$ \\
\textbf{Output}.  $2^{-k}x \mbox{ (mod }n\mbox{)}$ \\
\hline \\
1.  for $t$ from $1$ to $k$ do \\
\hspace{3mm}1.1  If $x$ is odd then \\
\hspace{6mm}1.1.1  $x \leftarrow x + n$ \\
\hspace{3mm}1.2  $x \leftarrow x/2$ \\
2.  Return $x$. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Montgomery Reduction}
\end{figure}

The algorithm reduces the input one bit at a time using the two congruencies stated previously.  Inside the loop $n$, which is odd, is
added to $x$ if $x$ is odd.  This forces $x$ to be even which allows the division by two in $\Z$ to be congruent to a modular division by two.  Since
$x$ is assumed to be initially much larger than $n$ the addition of $n$ will contribute an insignificant magnitude to $x$.  Let $r$ represent the
final result of the Montgomery algorithm.  If $k > lg(n)$ and $0 \le x < n^2$ then the final result is limited to
$0 \le r < \lfloor x/2^k \rfloor + n$.  As a result at most a single subtraction is required to get the residue desired.

\begin{figure}[here]
\begin{small}
\begin{center}
\begin{tabular}{|c|l|}
\hline \textbf{Step number ($t$)} & \textbf{Result ($x$)} \\
\hline $1$ & $x + n = 5812$, $x/2 = 2906$ \\
\hline $2$ & $x/2 = 1453$ \\
\hline $3$ & $x + n = 1710$, $x/2 = 855$ \\
\hline $4$ & $x + n = 1112$, $x/2 = 556$ \\
\hline $5$ & $x/2 = 278$ \\
\hline $6$ & $x/2 = 139$ \\
\hline $7$ & $x + n = 396$, $x/2 = 198$ \\
\hline $8$ & $x/2 = 99$ \\
\hline $9$ & $x + n = 356$, $x/2 = 178$ \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Example of Montgomery Reduction (I)}
\label{fig:MONT1}
\end{figure}

Consider the example in figure~\ref{fig:MONT1} which reduces $x = 5555$ modulo $n = 257$ when $k = 9$ (note $\beta^k = 512$ which is larger than $n$).  The result of
the algorithm $r = 178$ is congruent to the value of $2^{-9} \cdot 5555 \mbox{ (mod }257\mbox{)}$.  When $r$ is multiplied by $2^9$ modulo $257$ the correct residue
$r \equiv 158$ is produced.

Let $k = \lfloor lg(n) \rfloor + 1$ represent the number of bits in $n$.  The current algorithm requires $2k^2$ single precision shifts
and $k^2$ single precision additions.  At this rate the algorithm is most certainly slower than Barrett reduction and not terribly useful.
Fortunately there exists an alternative representation of the algorithm.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Montgomery Reduction} (modified I). \\
\textbf{Input}.   Integer $x$, $n$ and $k$ ($2^k > n$) \\
\textbf{Output}.  $2^{-k}x \mbox{ (mod }n\mbox{)}$ \\
\hline \\
1.  for $t$ from $1$ to $k$ do \\
\hspace{3mm}1.1  If the $t$'th bit of $x$ is one then \\
\hspace{6mm}1.1.1  $x \leftarrow x + 2^tn$ \\
2.  Return $x/2^k$. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Montgomery Reduction (modified I)}
\end{figure}

This algorithm is equivalent since $2^tn$ is a multiple of $n$ and the lower $k$ bits of $x$ are zero by step 2.  The number of single
precision shifts has now been reduced from $2k^2$ to $k^2 + k$ which is only a small improvement.

\begin{figure}[here]
\begin{small}
\begin{center}
\begin{tabular}{|c|l|r|}
\hline \textbf{Step number ($t$)} & \textbf{Result ($x$)} & \textbf{Result ($x$) in Binary} \\
\hline -- & $5555$ & $1010110110011$ \\
\hline $1$ & $x + 2^{0}n = 5812$ &  $1011010110100$ \\
\hline $2$ & $5812$ & $1011010110100$ \\
\hline $3$ & $x + 2^{2}n = 6840$ & $1101010111000$ \\
\hline $4$ & $x + 2^{3}n = 8896$ & $10001011000000$ \\
\hline $5$ & $8896$ & $10001011000000$ \\
\hline $6$ & $8896$ & $10001011000000$ \\
\hline $7$ & $x + 2^{6}n = 25344$ & $110001100000000$ \\
\hline $8$ & $25344$ & $110001100000000$ \\
\hline $9$ & $x + 2^{7}n = 91136$ & $10110010000000000$ \\
\hline -- & $x/2^k = 178$ & \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Example of Montgomery Reduction (II)}
\label{fig:MONT2}
\end{figure}

Figure~\ref{fig:MONT2} demonstrates the modified algorithm reducing $x = 5555$ modulo $n = 257$ with $k = 9$.
With this algorithm a single shift right at the end is the only right shift required to reduce the input instead of $k$ right shifts inside the
loop.  Note that for the iterations $t = 2, 5, 6$ and $8$ where the result $x$ is not changed.  In those iterations the $t$'th bit of $x$ is
zero and the appropriate multiple of $n$ does not need to be added to force the $t$'th bit of the result to zero.

\subsection{Digit Based Montgomery Reduction}
Instead of computing the reduction on a bit-by-bit basis it is actually much faster to compute it on digit-by-digit basis.  Consider the
previous algorithm re-written to compute the Montgomery reduction in this new fashion.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Montgomery Reduction} (modified II). \\
\textbf{Input}.   Integer $x$, $n$ and $k$ ($\beta^k > n$) \\
\textbf{Output}.  $\beta^{-k}x \mbox{ (mod }n\mbox{)}$ \\
\hline \\
1.  for $t$ from $0$ to $k - 1$ do \\
\hspace{3mm}1.1  $x \leftarrow x + \mu n \beta^t$ \\
2.  Return $x/\beta^k$. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Montgomery Reduction (modified II)}
\end{figure}

The value $\mu n \beta^t$ is a multiple of the modulus $n$ meaning that it will not change the residue.  If the first digit of
the value $\mu n \beta^t$ equals the negative (modulo $\beta$) of the $t$'th digit of $x$ then the addition will result in a zero digit.  This
problem breaks down to solving the following congruency.

\begin{center}
\begin{tabular}{rcl}
$x_t + \mu n_0$ & $\equiv$ & $0 \mbox{ (mod }\beta\mbox{)}$ \\
$\mu n_0$ & $\equiv$ & $-x_t \mbox{ (mod }\beta\mbox{)}$ \\
$\mu$ & $\equiv$ & $-x_t/n_0 \mbox{ (mod }\beta\mbox{)}$ \\
\end{tabular}
\end{center}

In each iteration of the loop on step 1 a new value of $\mu$ must be calculated.  The value of $-1/n_0 \mbox{ (mod }\beta\mbox{)}$ is used
extensively in this algorithm and should be precomputed.  Let $\rho$ represent the negative of the modular inverse of $n_0$ modulo $\beta$.

For example, let $\beta = 10$ represent the radix.  Let $n = 17$ represent the modulus which implies $k = 2$ and $\rho \equiv 7$.  Let $x = 33$
represent the value to reduce.

\newpage\begin{figure}
\begin{center}
\begin{tabular}{|c|c|c|}
\hline \textbf{Step ($t$)} & \textbf{Value of $x$} & \textbf{Value of $\mu$} \\
\hline --                 & $33$ & --\\
\hline $0$                 & $33 + \mu n = 50$ & $1$ \\
\hline $1$                 & $50 + \mu n \beta = 900$ & $5$ \\
\hline
\end{tabular}
\end{center}
\caption{Example of Montgomery Reduction}
\end{figure}

The final result $900$ is then divided by $\beta^k$ to produce the final result $9$.  The first observation is that $9 \nequiv x \mbox{ (mod }n\mbox{)}$
which implies the result is not the modular residue of $x$ modulo $n$.  However, recall that the residue is actually multiplied by $\beta^{-k}$ in
the algorithm.  To get the true residue the value must be multiplied by $\beta^k$.  In this case $\beta^k \equiv 15 \mbox{ (mod }n\mbox{)}$ and
the correct residue is $9 \cdot 15 \equiv 16 \mbox{ (mod }n\mbox{)}$.

\subsection{Baseline Montgomery Reduction}
The baseline Montgomery reduction algorithm will produce the residue for any size input.  It is designed to be a catch-all algororithm for
Montgomery reductions.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_montgomery\_reduce}. \\
\textbf{Input}.   mp\_int $x$, mp\_int $n$ and a digit $\rho \equiv -1/n_0 \mbox{ (mod }n\mbox{)}$. \\
\hspace{11.5mm}($0 \le x < n^2, n > 1, (n, \beta) = 1, \beta^k > n$) \\
\textbf{Output}.  $\beta^{-k}x \mbox{ (mod }n\mbox{)}$ \\
\hline \\
1.  $digs \leftarrow 2n.used + 1$ \\
2.  If $digs < MP\_ARRAY$ and $m.used < \delta$ then \\
\hspace{3mm}2.1  Use algorithm fast\_mp\_montgomery\_reduce instead. \\
\\
Setup $x$ for the reduction. \\
3.  If $x.alloc < digs$ then grow $x$ to $digs$ digits. \\
4.  $x.used \leftarrow digs$ \\
\\
Eliminate the lower $k$ digits. \\
5.  For $ix$ from $0$ to $k - 1$ do \\
\hspace{3mm}5.1  $\mu \leftarrow x_{ix} \cdot \rho \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{3mm}5.2  $u \leftarrow 0$ \\
\hspace{3mm}5.3  For $iy$ from $0$ to $k - 1$ do \\
\hspace{6mm}5.3.1  $\hat r \leftarrow \mu n_{iy} + x_{ix + iy} + u$ \\
\hspace{6mm}5.3.2  $x_{ix + iy} \leftarrow \hat r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{6mm}5.3.3  $u \leftarrow \lfloor \hat r / \beta \rfloor$ \\
\hspace{3mm}5.4  While $u > 0$ do \\
\hspace{6mm}5.4.1  $iy \leftarrow iy + 1$ \\
\hspace{6mm}5.4.2  $x_{ix + iy} \leftarrow x_{ix + iy} + u$ \\
\hspace{6mm}5.4.3  $u \leftarrow \lfloor x_{ix+iy} / \beta \rfloor$ \\
\hspace{6mm}5.4.4  $x_{ix + iy} \leftarrow x_{ix+iy} \mbox{ (mod }\beta\mbox{)}$ \\
\\
Divide by $\beta^k$ and fix up as required. \\
6.  $x \leftarrow \lfloor x / \beta^k \rfloor$ \\
7.  If $x \ge n$ then \\
\hspace{3mm}7.1  $x \leftarrow x - n$ \\
8.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_montgomery\_reduce}
\end{figure}

\textbf{Algorithm mp\_montgomery\_reduce.}
This algorithm reduces the input $x$ modulo $n$ in place using the Montgomery reduction algorithm.  The algorithm is loosely based
on algorithm 14.32 of \cite[pp.601]{HAC} except it merges the multiplication of $\mu n \beta^t$ with the addition in the inner loop.  The
restrictions on this algorithm are fairly easy to adapt to.  First $0 \le x < n^2$ bounds the input to numbers in the same range as
for the Barrett algorithm.  Additionally if $n > 1$ and $n$ is odd there will exist a modular inverse $\rho$.  $\rho$ must be calculated in
advance of this algorithm.  Finally the variable $k$ is fixed and a pseudonym for $n.used$.

Step 2 decides whether a faster Montgomery algorithm can be used.  It is based on the Comba technique meaning that there are limits on
the size of the input.  This algorithm is discussed in sub-section 6.3.3.

Step 5 is the main reduction loop of the algorithm.  The value of $\mu$ is calculated once per iteration in the outer loop.  The inner loop
calculates $x + \mu n \beta^{ix}$ by multiplying $\mu n$ and adding the result to $x$ shifted by $ix$ digits.  Both the addition and
multiplication are performed in the same loop to save time and memory.  Step 5.4 will handle any additional carries that escape the inner loop.

Using a quick inspection this algorithm requires $n$ single precision multiplications for the outer loop and $n^2$ single precision multiplications
in the inner loop.  In total $n^2 + n$ single precision multiplications which compares favourably to Barrett at $n^2 + 2n - 1$ single precision
multiplications.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_montgomery\_reduce.c
\vspace{-3mm}
\begin{alltt}
016
017   /* computes xR**-1 == x (mod N) via Montgomery Reduction */
018   int
019   mp_montgomery_reduce (mp_int * x, mp_int * n, mp_digit rho)
020   \{
021     int     ix, res, digs;
022     mp_digit mu;
023
024     /* can the fast reduction [comba] method be used?
025      *
026      * Note that unlike in mul you're safely allowed *less*
027      * than the available columns [255 per default] since carries
028      * are fixed up in the inner loop.
029      */
030     digs = (n->used * 2) + 1;
031     if ((digs < MP_WARRAY) &&
032         (n->used <
033         (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) \{
034       return fast_mp_montgomery_reduce (x, n, rho);
035     \}
036
037     /* grow the input as required */
038     if (x->alloc < digs) \{
039       if ((res = mp_grow (x, digs)) != MP_OKAY) \{
040         return res;
041       \}
042     \}
043     x->used = digs;
044
045     for (ix = 0; ix < n->used; ix++) \{
046       /* mu = ai * rho mod b
047        *
048        * The value of rho must be precalculated via
049        * montgomery_setup() such that
050        * it equals -1/n0 mod b this allows the
051        * following inner loop to reduce the
052        * input one digit at a time
053        */
054       mu = (mp_digit) (((mp_word)x->dp[ix] * (mp_word)rho) & MP_MASK);
055
056       /* a = a + mu * m * b**i */
057       \{
058         int iy;
059         mp_digit *tmpn, *tmpx, u;
060         mp_word r;
061
062         /* alias for digits of the modulus */
063         tmpn = n->dp;
064
065         /* alias for the digits of x [the input] */
066         tmpx = x->dp + ix;
067
068         /* set the carry to zero */
069         u = 0;
070
071         /* Multiply and add in place */
072         for (iy = 0; iy < n->used; iy++) \{
073           /* compute product and sum */
074           r       = ((mp_word)mu * (mp_word)*tmpn++) +
075                      (mp_word) u + (mp_word) *tmpx;
076
077           /* get carry */
078           u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
079
080           /* fix digit */
081           *tmpx++ = (mp_digit)(r & ((mp_word) MP_MASK));
082         \}
083         /* At this point the ix'th digit of x should be zero */
084
085
086         /* propagate carries upwards as required*/
087         while (u != 0) \{
088           *tmpx   += u;
089           u        = *tmpx >> DIGIT_BIT;
090           *tmpx++ &= MP_MASK;
091         \}
092       \}
093     \}
094
095     /* at this point the n.used'th least
096      * significant digits of x are all zero
097      * which means we can shift x to the
098      * right by n.used digits and the
099      * residue is unchanged.
100      */
101
102     /* x = x/b**n.used */
103     mp_clamp(x);
104     mp_rshd (x, n->used);
105
106     /* if x >= n then x = x - n */
107     if (mp_cmp_mag (x, n) != MP_LT) \{
108       return s_mp_sub (x, n, x);
109     \}
110
111     return MP_OKAY;
112   \}
113   #endif
114
\end{alltt}
\end{small}

This is the baseline implementation of the Montgomery reduction algorithm.  Lines 30 to 35 determine if the Comba based
routine can be used instead.  Line 48 computes the value of $\mu$ for that particular iteration of the outer loop.

The multiplication $\mu n \beta^{ix}$ is performed in one step in the inner loop.  The alias $tmpx$ refers to the $ix$'th digit of $x$ and
the alias $tmpn$ refers to the modulus $n$.

\subsection{Faster ``Comba'' Montgomery Reduction}

The Montgomery reduction requires fewer single precision multiplications than a Barrett reduction, however it is much slower due to the serial
nature of the inner loop.  The Barrett reduction algorithm requires two slightly modified multipliers which can be implemented with the Comba
technique.  The Montgomery reduction algorithm cannot directly use the Comba technique to any significant advantage since the inner loop calculates
a $k \times 1$ product $k$ times.

The biggest obstacle is that at the $ix$'th iteration of the outer loop the value of $x_{ix}$ is required to calculate $\mu$.  This means the
carries from $0$ to $ix - 1$ must have been propagated upwards to form a valid $ix$'th digit.  The solution as it turns out is very simple.
Perform a Comba like multiplier and inside the outer loop just after the inner loop fix up the $ix + 1$'th digit by forwarding the carry.

With this change in place the Montgomery reduction algorithm can be performed with a Comba style multiplication loop which substantially increases
the speed of the algorithm.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{fast\_mp\_montgomery\_reduce}. \\
\textbf{Input}.   mp\_int $x$, mp\_int $n$ and a digit $\rho \equiv -1/n_0 \mbox{ (mod }n\mbox{)}$. \\
\hspace{11.5mm}($0 \le x < n^2, n > 1, (n, \beta) = 1, \beta^k > n$) \\
\textbf{Output}.  $\beta^{-k}x \mbox{ (mod }n\mbox{)}$ \\
\hline \\
Place an array of \textbf{MP\_WARRAY} mp\_word variables called $\hat W$ on the stack. \\
1.  if $x.alloc < n.used + 1$ then grow $x$ to $n.used + 1$ digits. \\
Copy the digits of $x$ into the array $\hat W$ \\
2.  For $ix$ from $0$ to $x.used - 1$ do \\
\hspace{3mm}2.1  $\hat W_{ix} \leftarrow x_{ix}$ \\
3.  For $ix$ from $x.used$ to $2n.used - 1$ do \\
\hspace{3mm}3.1  $\hat W_{ix} \leftarrow 0$ \\
Elimiate the lower $k$ digits. \\
4.  for $ix$ from $0$ to $n.used - 1$ do \\
\hspace{3mm}4.1  $\mu \leftarrow \hat W_{ix} \cdot \rho \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{3mm}4.2  For $iy$ from $0$ to $n.used - 1$ do \\
\hspace{6mm}4.2.1  $\hat W_{iy + ix} \leftarrow \hat W_{iy + ix} + \mu \cdot n_{iy}$ \\
\hspace{3mm}4.3  $\hat W_{ix + 1} \leftarrow \hat W_{ix + 1} + \lfloor \hat W_{ix} / \beta \rfloor$ \\
Propagate carries upwards. \\
5.  for $ix$ from $n.used$ to $2n.used + 1$ do \\
\hspace{3mm}5.1  $\hat W_{ix + 1} \leftarrow \hat W_{ix + 1} + \lfloor \hat W_{ix} / \beta \rfloor$ \\
Shift right and reduce modulo $\beta$ simultaneously. \\
6.  for $ix$ from $0$ to $n.used + 1$ do \\
\hspace{3mm}6.1  $x_{ix} \leftarrow \hat W_{ix + n.used} \mbox{ (mod }\beta\mbox{)}$ \\
Zero excess digits and fixup $x$. \\
7.  if $x.used > n.used + 1$ then do \\
\hspace{3mm}7.1  for $ix$ from $n.used + 1$ to $x.used - 1$ do \\
\hspace{6mm}7.1.1  $x_{ix} \leftarrow 0$ \\
8.  $x.used \leftarrow n.used + 1$ \\
9.  Clamp excessive digits of $x$. \\
10.  If $x \ge n$ then \\
\hspace{3mm}10.1  $x \leftarrow x - n$ \\
11.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm fast\_mp\_montgomery\_reduce}
\end{figure}

\textbf{Algorithm fast\_mp\_montgomery\_reduce.}
This algorithm will compute the Montgomery reduction of $x$ modulo $n$ using the Comba technique.  It is on most computer platforms significantly
faster than algorithm mp\_montgomery\_reduce and algorithm mp\_reduce (\textit{Barrett reduction}).  The algorithm has the same restrictions
on the input as the baseline reduction algorithm.  An additional two restrictions are imposed on this algorithm.  The number of digits $k$ in the
the modulus $n$ must not violate $MP\_WARRAY > 2k +1$ and $n < \delta$.   When $\beta = 2^{28}$ this algorithm can be used to reduce modulo
a modulus of at most $3,556$ bits in length.

As in the other Comba reduction algorithms there is a $\hat W$ array which stores the columns of the product.  It is initially filled with the
contents of $x$ with the excess digits zeroed.  The reduction loop is very similar the to the baseline loop at heart.  The multiplication on step
4.1 can be single precision only since $ab \mbox{ (mod }\beta\mbox{)} \equiv (a \mbox{ mod }\beta)(b \mbox{ mod }\beta)$.  Some multipliers such
as those on the ARM processors take a variable length time to complete depending on the number of bytes of result it must produce.  By performing
a single precision multiplication instead half the amount of time is spent.

Also note that digit $\hat W_{ix}$ must have the carry from the $ix - 1$'th digit propagated upwards in order for this to work.  That is what step
4.3 will do.  In effect over the $n.used$ iterations of the outer loop the $n.used$'th lower columns all have the their carries propagated forwards.  Note
how the upper bits of those same words are not reduced modulo $\beta$.  This is because those values will be discarded shortly and there is no
point.

Step 5 will propagate the remainder of the carries upwards.  On step 6 the columns are reduced modulo $\beta$ and shifted simultaneously as they are
stored in the destination $x$.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_fast\_mp\_montgomery\_reduce.c
\vspace{-3mm}
\begin{alltt}
016
017   /* computes xR**-1 == x (mod N) via Montgomery Reduction
018    *
019    * This is an optimized implementation of montgomery_reduce
020    * which uses the comba method to quickly calculate the columns of the
021    * reduction.
022    *
023    * Based on Algorithm 14.32 on pp.601 of HAC.
024   */
025   int fast_mp_montgomery_reduce (mp_int * x, mp_int * n, mp_digit rho)
026   \{
027     int     ix, res, olduse;
028     mp_word W[MP_WARRAY];
029
030     /* get old used count */
031     olduse = x->used;
032
033     /* grow a as required */
034     if (x->alloc < (n->used + 1)) \{
035       if ((res = mp_grow (x, n->used + 1)) != MP_OKAY) \{
036         return res;
037       \}
038     \}
039
040     /* first we have to get the digits of the input into
041      * an array of double precision words W[...]
042      */
043     \{
044       mp_word *_W;
045       mp_digit *tmpx;
046
047       /* alias for the W[] array */
048       _W   = W;
049
050       /* alias for the digits of  x*/
051       tmpx = x->dp;
052
053       /* copy the digits of a into W[0..a->used-1] */
054       for (ix = 0; ix < x->used; ix++) \{
055         *_W++ = *tmpx++;
056       \}
057
058       /* zero the high words of W[a->used..m->used*2] */
059       for (; ix < ((n->used * 2) + 1); ix++) \{
060         *_W++ = 0;
061       \}
062     \}
063
064     /* now we proceed to zero successive digits
065      * from the least significant upwards
066      */
067     for (ix = 0; ix < n->used; ix++) \{
068       /* mu = ai * m' mod b
069        *
070        * We avoid a double precision multiplication (which isn't required)
071        * by casting the value down to a mp_digit.  Note this requires
072        * that W[ix-1] have  the carry cleared (see after the inner loop)
073        */
074       mp_digit mu;
075       mu = (mp_digit) (((W[ix] & MP_MASK) * rho) & MP_MASK);
076
077       /* a = a + mu * m * b**i
078        *
079        * This is computed in place and on the fly.  The multiplication
080        * by b**i is handled by offseting which columns the results
081        * are added to.
082        *
083        * Note the comba method normally doesn't handle carries in the
084        * inner loop In this case we fix the carry from the previous
085        * column since the Montgomery reduction requires digits of the
086        * result (so far) [see above] to work.  This is
087        * handled by fixing up one carry after the inner loop.  The
088        * carry fixups are done in order so after these loops the
089        * first m->used words of W[] have the carries fixed
090        */
091       \{
092         int iy;
093         mp_digit *tmpn;
094         mp_word *_W;
095
096         /* alias for the digits of the modulus */
097         tmpn = n->dp;
098
099         /* Alias for the columns set by an offset of ix */
100         _W = W + ix;
101
102         /* inner loop */
103         for (iy = 0; iy < n->used; iy++) \{
104             *_W++ += ((mp_word)mu) * ((mp_word)*tmpn++);
105         \}
106       \}
107
108       /* now fix carry for next digit, W[ix+1] */
109       W[ix + 1] += W[ix] >> ((mp_word) DIGIT_BIT);
110     \}
111
112     /* now we have to propagate the carries and
113      * shift the words downward [all those least
114      * significant digits we zeroed].
115      */
116     \{
117       mp_digit *tmpx;
118       mp_word *_W, *_W1;
119
120       /* nox fix rest of carries */
121
122       /* alias for current word */
123       _W1 = W + ix;
124
125       /* alias for next word, where the carry goes */
126       _W = W + ++ix;
127
128       for (; ix <= ((n->used * 2) + 1); ix++) \{
129         *_W++ += *_W1++ >> ((mp_word) DIGIT_BIT);
130       \}
131
132       /* copy out, A = A/b**n
133        *
134        * The result is A/b**n but instead of converting from an
135        * array of mp_word to mp_digit than calling mp_rshd
136        * we just copy them in the right order
137        */
138
139       /* alias for destination word */
140       tmpx = x->dp;
141
142       /* alias for shifted double precision result */
143       _W = W + n->used;
144
145       for (ix = 0; ix < (n->used + 1); ix++) \{
146         *tmpx++ = (mp_digit)(*_W++ & ((mp_word) MP_MASK));
147       \}
148
149       /* zero oldused digits, if the input a was larger than
150        * m->used+1 we'll have to clear the digits
151        */
152       for (; ix < olduse; ix++) \{
153         *tmpx++ = 0;
154       \}
155     \}
156
157     /* set the max used and clamp */
158     x->used = n->used + 1;
159     mp_clamp (x);
160
161     /* if A >= m then A = A - m */
162     if (mp_cmp_mag (x, n) != MP_LT) \{
163       return s_mp_sub (x, n, x);
164     \}
165     return MP_OKAY;
166   \}
167   #endif
168
\end{alltt}
\end{small}

The $\hat W$ array is first filled with digits of $x$ on line 50 then the rest of the digits are zeroed on line 54.  Both loops share
the same alias variables to make the code easier to read.

The value of $\mu$ is calculated in an interesting fashion.  First the value $\hat W_{ix}$ is reduced modulo $\beta$ and cast to a mp\_digit.  This
forces the compiler to use a single precision multiplication and prevents any concerns about loss of precision.   Line 109 fixes the carry
for the next iteration of the loop by propagating the carry from $\hat W_{ix}$ to $\hat W_{ix+1}$.

The for loop on line 108 propagates the rest of the carries upwards through the columns.  The for loop on line 125 reduces the columns
modulo $\beta$ and shifts them $k$ places at the same time.  The alias $\_ \hat W$ actually refers to the array $\hat W$ starting at the $n.used$'th
digit, that is $\_ \hat W_{t} = \hat W_{n.used + t}$.

\subsection{Montgomery Setup}
To calculate the variable $\rho$ a relatively simple algorithm will be required.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_montgomery\_setup}. \\
\textbf{Input}.   mp\_int $n$ ($n > 1$ and $(n, 2) = 1$) \\
\textbf{Output}.  $\rho \equiv -1/n_0 \mbox{ (mod }\beta\mbox{)}$ \\
\hline \\
1.  $b \leftarrow n_0$ \\
2.  If $b$ is even return(\textit{MP\_VAL}) \\
3.  $x \leftarrow (((b + 2) \mbox{ AND } 4) << 1) + b$ \\
4.  for $k$ from 0 to $\lceil lg(lg(\beta)) \rceil - 2$ do \\
\hspace{3mm}4.1  $x \leftarrow x \cdot (2 - bx)$ \\
5.  $\rho \leftarrow \beta - x \mbox{ (mod }\beta\mbox{)}$ \\
6.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_montgomery\_setup}
\end{figure}

\textbf{Algorithm mp\_montgomery\_setup.}
This algorithm will calculate the value of $\rho$ required within the Montgomery reduction algorithms.  It uses a very interesting trick
to calculate $1/n_0$ when $\beta$ is a power of two.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_montgomery\_setup.c
\vspace{-3mm}
\begin{alltt}
016
017   /* setups the montgomery reduction stuff */
018   int
019   mp_montgomery_setup (mp_int * n, mp_digit * rho)
020   \{
021     mp_digit x, b;
022
023   /* fast inversion mod 2**k
024    *
025    * Based on the fact that
026    *
027    * XA = 1 (mod 2**n)  =>  (X(2-XA)) A = 1 (mod 2**2n)
028    *                    =>  2*X*A - X*X*A*A = 1
029    *                    =>  2*(1) - (1)     = 1
030    */
031     b = n->dp[0];
032
033     if ((b & 1) == 0) \{
034       return MP_VAL;
035     \}
036
037     x = (((b + 2) & 4) << 1) + b; /* here x*a==1 mod 2**4 */
038     x *= 2 - (b * x);             /* here x*a==1 mod 2**8 */
039   #if !defined(MP_8BIT)
040     x *= 2 - (b * x);             /* here x*a==1 mod 2**16 */
041   #endif
042   #if defined(MP_64BIT) || !(defined(MP_8BIT) || defined(MP_16BIT))
043     x *= 2 - (b * x);             /* here x*a==1 mod 2**32 */
044   #endif
045   #ifdef MP_64BIT
046     x *= 2 - (b * x);             /* here x*a==1 mod 2**64 */
047   #endif
048
049     /* rho = -1/m mod b */
050     *rho = (mp_digit)(((mp_word)1 << ((mp_word) DIGIT_BIT)) - x) & MP_MASK;
051
052     return MP_OKAY;
053   \}
054   #endif
055
\end{alltt}
\end{small}

This source code computes the value of $\rho$ required to perform Montgomery reduction.  It has been modified to avoid performing excess
multiplications when $\beta$ is not the default 28-bits.

\section{The Diminished Radix Algorithm}
The Diminished Radix method of modular reduction \cite{DRMET} is a fairly clever technique which can be more efficient than either the Barrett
or Montgomery methods for certain forms of moduli.  The technique is based on the following simple congruence.

\begin{equation}
(x \mbox{ mod } n) + k \lfloor x / n \rfloor \equiv x \mbox{ (mod }(n - k)\mbox{)}
\end{equation}

This observation was used in the MMB \cite{MMB} block cipher to create a diffusion primitive.  It used the fact that if $n = 2^{31}$ and $k=1$ that
then a x86 multiplier could produce the 62-bit product and use  the ``shrd'' instruction to perform a double-precision right shift.  The proof
of the above equation is very simple.  First write $x$ in the product form.

\begin{equation}
x = qn + r
\end{equation}

Now reduce both sides modulo $(n - k)$.

\begin{equation}
x \equiv qk + r  \mbox{ (mod }(n-k)\mbox{)}
\end{equation}

The variable $n$ reduces modulo $n - k$ to $k$.  By putting $q = \lfloor x/n \rfloor$ and $r = x \mbox{ mod } n$
into the equation the original congruence is reproduced, thus concluding the proof.  The following algorithm is based on this observation.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Diminished Radix Reduction}. \\
\textbf{Input}.   Integer $x$, $n$, $k$ \\
\textbf{Output}.  $x \mbox{ mod } (n - k)$ \\
\hline \\
1.  $q \leftarrow \lfloor x / n \rfloor$ \\
2.  $q \leftarrow k \cdot q$ \\
3.  $x \leftarrow x \mbox{ (mod }n\mbox{)}$ \\
4.  $x \leftarrow x + q$ \\
5.  If $x \ge (n - k)$ then \\
\hspace{3mm}5.1  $x \leftarrow x - (n - k)$ \\
\hspace{3mm}5.2  Goto step 1. \\
6.  Return $x$ \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Diminished Radix Reduction}
\label{fig:DR}
\end{figure}

This algorithm will reduce $x$ modulo $n - k$ and return the residue.  If $0 \le x < (n - k)^2$ then the algorithm will loop almost always
once or twice and occasionally three times.  For simplicity sake the value of $x$ is bounded by the following simple polynomial.

\begin{equation}
0 \le x < n^2 + k^2 - 2nk
\end{equation}

The true bound is  $0 \le x < (n - k - 1)^2$ but this has quite a few more terms.  The value of $q$ after step 1 is bounded by the following.

\begin{equation}
q < n - 2k - k^2/n
\end{equation}

Since $k^2$ is going to be considerably smaller than $n$ that term will always be zero.  The value of $x$ after step 3 is bounded trivially as
$0 \le x < n$.  By step four the sum $x + q$ is bounded by

\begin{equation}
0 \le q + x < (k + 1)n - 2k^2 - 1
\end{equation}

With a second pass $q$ will be loosely bounded by $0 \le q < k^2$ after step 2 while $x$ will still be loosely bounded by $0 \le x < n$ after step 3.  After the second pass it is highly unlike that the
sum in step 4 will exceed $n - k$.  In practice fewer than three passes of the algorithm are required to reduce virtually every input in the
range $0 \le x < (n - k - 1)^2$.

\begin{figure}
\begin{small}
\begin{center}
\begin{tabular}{|l|}
\hline
$x = 123456789, n = 256, k = 3$ \\
\hline $q \leftarrow \lfloor x/n \rfloor = 482253$ \\
$q \leftarrow q*k = 1446759$ \\
$x \leftarrow x \mbox{ mod } n = 21$ \\
$x \leftarrow x + q = 1446780$ \\
$x \leftarrow x - (n - k) = 1446527$ \\
\hline
$q \leftarrow \lfloor x/n \rfloor = 5650$ \\
$q \leftarrow q*k = 16950$ \\
$x \leftarrow x \mbox{ mod } n = 127$ \\
$x \leftarrow x + q = 17077$ \\
$x \leftarrow x - (n - k) = 16824$ \\
\hline
$q \leftarrow \lfloor x/n \rfloor = 65$ \\
$q \leftarrow q*k = 195$ \\
$x \leftarrow x \mbox{ mod } n = 184$ \\
$x \leftarrow x + q = 379$ \\
$x \leftarrow x - (n - k) = 126$ \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Example Diminished Radix Reduction}
\label{fig:EXDR}
\end{figure}

Figure~\ref{fig:EXDR} demonstrates the reduction of $x = 123456789$ modulo $n - k = 253$ when $n = 256$ and $k = 3$.  Note that even while $x$
is considerably larger than $(n - k - 1)^2 = 63504$ the algorithm still converges on the modular residue exceedingly fast.  In this case only
three passes were required to find the residue $x \equiv 126$.


\subsection{Choice of Moduli}
On the surface this algorithm looks like a very expensive algorithm.  It requires a couple of subtractions followed by multiplication and other
modular reductions.  The usefulness of this algorithm becomes exceedingly clear when an appropriate modulus is chosen.

Division in general is a very expensive operation to perform.  The one exception is when the division is by a power of the radix of representation used.
Division by ten for example is simple for pencil and paper mathematics since it amounts to shifting the decimal place to the right.  Similarly division
by two (\textit{or powers of two}) is very simple for binary computers to perform.  It would therefore seem logical to choose $n$ of the form $2^p$
which would imply that $\lfloor x / n \rfloor$ is a simple shift of $x$ right $p$ bits.

However, there is one operation related to division of power of twos that is even faster than this.  If $n = \beta^p$ then the division may be
performed by moving whole digits to the right $p$ places.  In practice division by $\beta^p$ is much faster than division by $2^p$ for any $p$.
Also with the choice of $n = \beta^p$ reducing $x$ modulo $n$ merely requires zeroing the digits above the $p-1$'th digit of $x$.

Throughout the next section the term ``restricted modulus'' will refer to a modulus of the form $\beta^p - k$ whereas the term ``unrestricted
modulus'' will refer to a modulus of the form $2^p - k$.  The word ``restricted'' in this case refers to the fact that it is based on the
$2^p$ logic except $p$ must be a multiple of $lg(\beta)$.

\subsection{Choice of $k$}
Now that division and reduction (\textit{step 1 and 3 of figure~\ref{fig:DR}}) have been optimized to simple digit operations the multiplication by $k$
in step 2 is the most expensive operation.  Fortunately the choice of $k$ is not terribly limited.  For all intents and purposes it might
as well be a single digit.  The smaller the value of $k$ is the faster the algorithm will be.

\subsection{Restricted Diminished Radix Reduction}
The restricted Diminished Radix algorithm can quickly reduce an input modulo a modulus of the form $n = \beta^p - k$.  This algorithm can reduce
an input $x$ within the range $0 \le x < n^2$ using only a couple passes of the algorithm demonstrated in figure~\ref{fig:DR}.  The implementation
of this algorithm has been optimized to avoid additional overhead associated with a division by $\beta^p$, the multiplication by $k$ or the addition
of $x$ and $q$.  The resulting algorithm is very efficient and can lead to substantial improvements over Barrett and Montgomery reduction when modular
exponentiations are performed.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_dr\_reduce}. \\
\textbf{Input}.   mp\_int $x$, $n$ and a mp\_digit $k = \beta - n_0$ \\
\hspace{11.5mm}($0 \le x < n^2$, $n > 1$, $0 < k < \beta$) \\
\textbf{Output}.  $x \mbox{ mod } n$ \\
\hline \\
1.  $m \leftarrow n.used$ \\
2.  If $x.alloc < 2m$ then grow $x$ to $2m$ digits. \\
3.  $\mu \leftarrow 0$ \\
4.  for $i$ from $0$ to $m - 1$ do \\
\hspace{3mm}4.1  $\hat r \leftarrow k \cdot x_{m+i} + x_{i} + \mu$ \\
\hspace{3mm}4.2  $x_{i} \leftarrow \hat r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{3mm}4.3  $\mu \leftarrow \lfloor \hat r / \beta \rfloor$ \\
5.  $x_{m} \leftarrow \mu$ \\
6.  for $i$ from $m + 1$ to $x.used - 1$ do \\
\hspace{3mm}6.1  $x_{i} \leftarrow 0$ \\
7.  Clamp excess digits of $x$. \\
8.  If $x \ge n$ then \\
\hspace{3mm}8.1  $x \leftarrow x - n$ \\
\hspace{3mm}8.2  Goto step 3. \\
9.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_dr\_reduce}
\end{figure}

\textbf{Algorithm mp\_dr\_reduce.}
This algorithm will perform the Dimished Radix reduction of $x$ modulo $n$.  It has similar restrictions to that of the Barrett reduction
with the addition that $n$ must be of the form $n = \beta^m - k$ where $0 < k <\beta$.

This algorithm essentially implements the pseudo-code in figure~\ref{fig:DR} except with a slight optimization.  The division by $\beta^m$, multiplication by $k$
and addition of $x \mbox{ mod }\beta^m$ are all performed simultaneously inside the loop on step 4.  The division by $\beta^m$ is emulated by accessing
the term at the $m+i$'th position which is subsequently multiplied by $k$ and added to the term at the $i$'th position.  After the loop the $m$'th
digit is set to the carry and the upper digits are zeroed.  Steps 5 and 6 emulate the reduction modulo $\beta^m$ that should have happend to
$x$ before the addition of the multiple of the upper half.

At step 8 if $x$ is still larger than $n$ another pass of the algorithm is required.  First $n$ is subtracted from $x$ and then the algorithm resumes
at step 3.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_dr\_reduce.c
\vspace{-3mm}
\begin{alltt}
016
017   /* reduce "x" in place modulo "n" using the Diminished Radix algorithm.
018    *
019    * Based on algorithm from the paper
020    *
021    * "Generating Efficient Primes for Discrete Log Cryptosystems"
022    *                 Chae Hoon Lim, Pil Joong Lee,
023    *          POSTECH Information Research Laboratories
024    *
025    * The modulus must be of a special format [see manual]
026    *
027    * Has been modified to use algorithm 7.10 from the LTM book instead
028    *
029    * Input x must be in the range 0 <= x <= (n-1)**2
030    */
031   int
032   mp_dr_reduce (mp_int * x, mp_int * n, mp_digit k)
033   \{
034     int      err, i, m;
035     mp_word  r;
036     mp_digit mu, *tmpx1, *tmpx2;
037
038     /* m = digits in modulus */
039     m = n->used;
040
041     /* ensure that "x" has at least 2m digits */
042     if (x->alloc < (m + m)) \{
043       if ((err = mp_grow (x, m + m)) != MP_OKAY) \{
044         return err;
045       \}
046     \}
047
048   /* top of loop, this is where the code resumes if
049    * another reduction pass is required.
050    */
051   top:
052     /* aliases for digits */
053     /* alias for lower half of x */
054     tmpx1 = x->dp;
055
056     /* alias for upper half of x, or x/B**m */
057     tmpx2 = x->dp + m;
058
059     /* set carry to zero */
060     mu = 0;
061
062     /* compute (x mod B**m) + k * [x/B**m] inline and inplace */
063     for (i = 0; i < m; i++) \{
064         r         = (((mp_word)*tmpx2++) * (mp_word)k) + *tmpx1 + mu;
065         *tmpx1++  = (mp_digit)(r & MP_MASK);
066         mu        = (mp_digit)(r >> ((mp_word)DIGIT_BIT));
067     \}
068
069     /* set final carry */
070     *tmpx1++ = mu;
071
072     /* zero words above m */
073     for (i = m + 1; i < x->used; i++) \{
074         *tmpx1++ = 0;
075     \}
076
077     /* clamp, sub and return */
078     mp_clamp (x);
079
080     /* if x >= n then subtract and reduce again
081      * Each successive "recursion" makes the input smaller and smaller.
082      */
083     if (mp_cmp_mag (x, n) != MP_LT) \{
084       if ((err = s_mp_sub(x, n, x)) != MP_OKAY) \{
085         return err;
086       \}
087       goto top;
088     \}
089     return MP_OKAY;
090   \}
091   #endif
092
\end{alltt}
\end{small}

The first step is to grow $x$ as required to $2m$ digits since the reduction is performed in place on $x$.  The label on line 51 is where
the algorithm will resume if further reduction passes are required.  In theory it could be placed at the top of the function however, the size of
the modulus and question of whether $x$ is large enough are invariant after the first pass meaning that it would be a waste of time.

The aliases $tmpx1$ and $tmpx2$ refer to the digits of $x$ where the latter is offset by $m$ digits.  By reading digits from $x$ offset by $m$ digits
a division by $\beta^m$ can be simulated virtually for free.  The loop on line 63 performs the bulk of the work (\textit{corresponds to step 4 of algorithm 7.11})
in this algorithm.

By line 70 the pointer $tmpx1$ points to the $m$'th digit of $x$ which is where the final carry will be placed.  Similarly by line 73 the
same pointer will point to the $m+1$'th digit where the zeroes will be placed.

Since the algorithm is only valid if both $x$ and $n$ are greater than zero an unsigned comparison suffices to determine if another pass is required.
With the same logic at line 84 the value of $x$ is known to be greater than or equal to $n$ meaning that an unsigned subtraction can be used
as well.  Since the destination of the subtraction is the larger of the inputs the call to algorithm s\_mp\_sub cannot fail and the return code
does not need to be checked.

\subsubsection{Setup}
To setup the restricted Diminished Radix algorithm the value $k = \beta - n_0$ is required.  This algorithm is not really complicated but provided for
completeness.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_dr\_setup}. \\
\textbf{Input}.   mp\_int $n$ \\
\textbf{Output}.  $k = \beta - n_0$ \\
\hline \\
1.  $k \leftarrow \beta - n_0$ \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_dr\_setup}
\end{figure}

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_dr\_setup.c
\vspace{-3mm}
\begin{alltt}
016
017   /* determines the setup value */
018   void mp_dr_setup(mp_int *a, mp_digit *d)
019   \{
020      /* the casts are required if DIGIT_BIT is one less than
021       * the number of bits in a mp_digit [e.g. DIGIT_BIT==31]
022       */
023      *d = (mp_digit)((((mp_word)1) << ((mp_word)DIGIT_BIT)) -
024           ((mp_word)a->dp[0]));
025   \}
026
027   #endif
028
\end{alltt}
\end{small}

\subsubsection{Modulus Detection}
Another algorithm which will be useful is the ability to detect a restricted Diminished Radix modulus.  An integer is said to be
of restricted Diminished Radix form if all of the digits are equal to $\beta - 1$ except the trailing digit which may be any value.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_dr\_is\_modulus}. \\
\textbf{Input}.   mp\_int $n$ \\
\textbf{Output}.  $1$ if $n$ is in D.R form, $0$ otherwise \\
\hline
1.  If $n.used < 2$ then return($0$). \\
2.  for $ix$ from $1$ to $n.used - 1$ do \\
\hspace{3mm}2.1  If $n_{ix} \ne \beta - 1$ return($0$). \\
3.  Return($1$). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_dr\_is\_modulus}
\end{figure}

\textbf{Algorithm mp\_dr\_is\_modulus.}
This algorithm determines if a value is in Diminished Radix form.  Step 1 rejects obvious cases where fewer than two digits are
in the mp\_int.  Step 2 tests all but the first digit to see if they are equal to $\beta - 1$.  If the algorithm manages to get to
step 3 then $n$ must be of Diminished Radix form.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_dr\_is\_modulus.c
\vspace{-3mm}
\begin{alltt}
016
017   /* determines if a number is a valid DR modulus */
018   int mp_dr_is_modulus(mp_int *a)
019   \{
020      int ix;
021
022      /* must be at least two digits */
023      if (a->used < 2) \{
024         return 0;
025      \}
026
027      /* must be of the form b**k - a [a <= b] so all
028       * but the first digit must be equal to -1 (mod b).
029       */
030      for (ix = 1; ix < a->used; ix++) \{
031          if (a->dp[ix] != MP_MASK) \{
032             return 0;
033          \}
034      \}
035      return 1;
036   \}
037
038   #endif
039
\end{alltt}
\end{small}

\subsection{Unrestricted Diminished Radix Reduction}
The unrestricted Diminished Radix algorithm allows modular reductions to be performed when the modulus is of the form $2^p - k$.  This algorithm
is a straightforward adaptation of algorithm~\ref{fig:DR}.

In general the restricted Diminished Radix reduction algorithm is much faster since it has considerably lower overhead.  However, this new
algorithm is much faster than either Montgomery or Barrett reduction when the moduli are of the appropriate form.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_reduce\_2k}. \\
\textbf{Input}.   mp\_int $a$ and $n$.  mp\_digit $k$  \\
\hspace{11.5mm}($a \ge 0$, $n > 1$, $0 < k < \beta$, $n + k$ is a power of two) \\
\textbf{Output}.  $a \mbox{ (mod }n\mbox{)}$ \\
\hline
1.  $p \leftarrow \lceil lg(n) \rceil$  (\textit{mp\_count\_bits}) \\
2.  While $a \ge n$ do \\
\hspace{3mm}2.1  $q \leftarrow \lfloor a / 2^p \rfloor$ (\textit{mp\_div\_2d}) \\
\hspace{3mm}2.2  $a \leftarrow a \mbox{ (mod }2^p\mbox{)}$ (\textit{mp\_mod\_2d}) \\
\hspace{3mm}2.3  $q \leftarrow q \cdot k$ (\textit{mp\_mul\_d}) \\
\hspace{3mm}2.4  $a \leftarrow a - q$ (\textit{s\_mp\_sub}) \\
\hspace{3mm}2.5  If $a \ge n$ then do \\
\hspace{6mm}2.5.1  $a \leftarrow a - n$ \\
3.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_reduce\_2k}
\end{figure}

\textbf{Algorithm mp\_reduce\_2k.}
This algorithm quickly reduces an input $a$ modulo an unrestricted Diminished Radix modulus $n$.  Division by $2^p$ is emulated with a right
shift which makes the algorithm fairly inexpensive to use.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_reduce\_2k.c
\vspace{-3mm}
\begin{alltt}
016
017   /* reduces a modulo n where n is of the form 2**p - d */
018   int mp_reduce_2k(mp_int *a, mp_int *n, mp_digit d)
019   \{
020      mp_int q;
021      int    p, res;
022
023      if ((res = mp_init(&q)) != MP_OKAY) \{
024         return res;
025      \}
026
027      p = mp_count_bits(n);
028   top:
029      /* q = a/2**p, a = a mod 2**p */
030      if ((res = mp_div_2d(a, p, &q, a)) != MP_OKAY) \{
031         goto ERR;
032      \}
033
034      if (d != 1) \{
035         /* q = q * d */
036         if ((res = mp_mul_d(&q, d, &q)) != MP_OKAY) \{
037            goto ERR;
038         \}
039      \}
040
041      /* a = a + q */
042      if ((res = s_mp_add(a, &q, a)) != MP_OKAY) \{
043         goto ERR;
044      \}
045
046      if (mp_cmp_mag(a, n) != MP_LT) \{
047         if ((res = s_mp_sub(a, n, a)) != MP_OKAY) \{
048            goto ERR;
049         \}
050         goto top;
051      \}
052
053   ERR:
054      mp_clear(&q);
055      return res;
056   \}
057
058   #endif
059
\end{alltt}
\end{small}

The algorithm mp\_count\_bits calculates the number of bits in an mp\_int which is used to find the initial value of $p$.  The call to mp\_div\_2d
on line 30 calculates both the quotient $q$ and the remainder $a$ required.  By doing both in a single function call the code size
is kept fairly small.  The multiplication by $k$ is only performed if $k > 1$. This allows reductions modulo $2^p - 1$ to be performed without
any multiplications.

The unsigned s\_mp\_add, mp\_cmp\_mag and s\_mp\_sub are used in place of their full sign counterparts since the inputs are only valid if they are
positive.  By using the unsigned versions the overhead is kept to a minimum.

\subsubsection{Unrestricted Setup}
To setup this reduction algorithm the value of $k = 2^p - n$ is required.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_reduce\_2k\_setup}. \\
\textbf{Input}.   mp\_int $n$   \\
\textbf{Output}.  $k = 2^p - n$ \\
\hline
1.  $p \leftarrow \lceil lg(n) \rceil$  (\textit{mp\_count\_bits}) \\
2.  $x \leftarrow 2^p$ (\textit{mp\_2expt}) \\
3.  $x \leftarrow x - n$ (\textit{mp\_sub}) \\
4.  $k \leftarrow x_0$ \\
5.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_reduce\_2k\_setup}
\end{figure}

\textbf{Algorithm mp\_reduce\_2k\_setup.}
This algorithm computes the value of $k$ required for the algorithm mp\_reduce\_2k.  By making a temporary variable $x$ equal to $2^p$ a subtraction
is sufficient to solve for $k$.  Alternatively if $n$ has more than one digit the value of $k$ is simply $\beta - n_0$.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_reduce\_2k\_setup.c
\vspace{-3mm}
\begin{alltt}
016
017   /* determines the setup value */
018   int mp_reduce_2k_setup(mp_int *a, mp_digit *d)
019   \{
020      int res, p;
021      mp_int tmp;
022
023      if ((res = mp_init(&tmp)) != MP_OKAY) \{
024         return res;
025      \}
026
027      p = mp_count_bits(a);
028      if ((res = mp_2expt(&tmp, p)) != MP_OKAY) \{
029         mp_clear(&tmp);
030         return res;
031      \}
032
033      if ((res = s_mp_sub(&tmp, a, &tmp)) != MP_OKAY) \{
034         mp_clear(&tmp);
035         return res;
036      \}
037
038      *d = tmp.dp[0];
039      mp_clear(&tmp);
040      return MP_OKAY;
041   \}
042   #endif
043
\end{alltt}
\end{small}

\subsubsection{Unrestricted Detection}
An integer $n$ is a valid unrestricted Diminished Radix modulus if either of the following are true.

\begin{enumerate}
\item  The number has only one digit.
\item  The number has more than one digit and every bit from the $\beta$'th to the most significant is one.
\end{enumerate}

If either condition is true than there is a power of two $2^p$ such that $0 < 2^p - n < \beta$.   If the input is only
one digit than it will always be of the correct form.  Otherwise all of the bits above the first digit must be one.  This arises from the fact
that there will be value of $k$ that when added to the modulus causes a carry in the first digit which propagates all the way to the most
significant bit.  The resulting sum will be a power of two.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_reduce\_is\_2k}. \\
\textbf{Input}.   mp\_int $n$   \\
\textbf{Output}.  $1$ if of proper form, $0$ otherwise \\
\hline
1.  If $n.used = 0$ then return($0$). \\
2.  If $n.used = 1$ then return($1$). \\
3.  $p \leftarrow \lceil lg(n) \rceil$  (\textit{mp\_count\_bits}) \\
4.  for $x$ from $lg(\beta)$ to $p$ do \\
\hspace{3mm}4.1  If the ($x \mbox{ mod }lg(\beta)$)'th bit of the $\lfloor x / lg(\beta) \rfloor$ of $n$ is zero then return($0$). \\
5.  Return($1$). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_reduce\_is\_2k}
\end{figure}

\textbf{Algorithm mp\_reduce\_is\_2k.}
This algorithm quickly determines if a modulus is of the form required for algorithm mp\_reduce\_2k to function properly.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_reduce\_is\_2k.c
\vspace{-3mm}
\begin{alltt}
016
017   /* determines if mp_reduce_2k can be used */
018   int mp_reduce_is_2k(mp_int *a)
019   \{
020      int ix, iy, iw;
021      mp_digit iz;
022
023      if (a->used == 0) \{
024         return MP_NO;
025      \} else if (a->used == 1) \{
026         return MP_YES;
027      \} else if (a->used > 1) \{
028         iy = mp_count_bits(a);
029         iz = 1;
030         iw = 1;
031
032         /* Test every bit from the second digit up, must be 1 */
033         for (ix = DIGIT_BIT; ix < iy; ix++) \{
034             if ((a->dp[iw] & iz) == 0) \{
035                return MP_NO;
036             \}
037             iz <<= 1;
038             if (iz > (mp_digit)MP_MASK) \{
039                ++iw;
040                iz = 1;
041             \}
042         \}
043      \}
044      return MP_YES;
045   \}
046
047   #endif
048
\end{alltt}
\end{small}



\section{Algorithm Comparison}
So far three very different algorithms for modular reduction have been discussed.  Each of the algorithms have their own strengths and weaknesses
that makes having such a selection very useful.  The following table sumarizes the three algorithms along with comparisons of work factors.  Since
all three algorithms have the restriction that $0 \le x < n^2$ and $n > 1$ those limitations are not included in the table.

\begin{center}
\begin{small}
\begin{tabular}{|c|c|c|c|c|c|}
\hline \textbf{Method} & \textbf{Work Required} & \textbf{Limitations} & \textbf{$m = 8$} & \textbf{$m = 32$} & \textbf{$m = 64$} \\
\hline Barrett    & $m^2 + 2m - 1$ & None              & $79$ & $1087$ & $4223$ \\
\hline Montgomery & $m^2 + m$      & $n$ must be odd   & $72$ & $1056$ & $4160$ \\
\hline D.R.       & $2m$           & $n = \beta^m - k$ & $16$ & $64$   & $128$  \\
\hline
\end{tabular}
\end{small}
\end{center}

In theory Montgomery and Barrett reductions would require roughly the same amount of time to complete.  However, in practice since Montgomery
reduction can be written as a single function with the Comba technique it is much faster.  Barrett reduction suffers from the overhead of
calling the half precision multipliers, addition and division by $\beta$ algorithms.

For almost every cryptographic algorithm Montgomery reduction is the algorithm of choice.  The one set of algorithms where Diminished Radix reduction truly
shines are based on the discrete logarithm problem such as Diffie-Hellman \cite{DH} and ElGamal \cite{ELGAMAL}.  In these algorithms
primes of the form $\beta^m - k$ can be found and shared amongst users.  These primes will allow the Diminished Radix algorithm to be used in
modular exponentiation to greatly speed up the operation.



\section*{Exercises}
\begin{tabular}{cl}
$\left [ 3 \right ]$ & Prove that the ``trick'' in algorithm mp\_montgomery\_setup actually \\
                     & calculates the correct value of $\rho$. \\
                     & \\
$\left [ 2 \right ]$ & Devise an algorithm to reduce modulo $n + k$ for small $k$ quickly.  \\
                     & \\
$\left [ 4 \right ]$ & Prove that the pseudo-code algorithm ``Diminished Radix Reduction'' \\
                     & (\textit{figure~\ref{fig:DR}}) terminates.  Also prove the probability that it will \\
                     & terminate within $1 \le k \le 10$ iterations. \\
                     & \\
\end{tabular}


\chapter{Exponentiation}
Exponentiation is the operation of raising one variable to the power of another, for example, $a^b$.  A variant of exponentiation, computed
in a finite field or ring, is called modular exponentiation.  This latter style of operation is typically used in public key
cryptosystems such as RSA and Diffie-Hellman.  The ability to quickly compute modular exponentiations is of great benefit to any
such cryptosystem and many methods have been sought to speed it up.

\section{Exponentiation Basics}
A trivial algorithm would simply multiply $a$ against itself $b - 1$ times to compute the exponentiation desired.  However, as $b$ grows in size
the number of multiplications becomes prohibitive.  Imagine what would happen if $b$ $\approx$ $2^{1024}$ as is the case when computing an RSA signature
with a $1024$-bit key.  Such a calculation could never be completed as it would take simply far too long.

Fortunately there is a very simple algorithm based on the laws of exponents.  Recall that $lg_a(a^b) = b$ and that $lg_a(a^ba^c) = b + c$ which
are two trivial relationships between the base and the exponent.  Let $b_i$ represent the $i$'th bit of $b$ starting from the least
significant bit.  If $b$ is a $k$-bit integer than the following equation is true.

\begin{equation}
a^b = \prod_{i=0}^{k-1} a^{2^i \cdot b_i}
\end{equation}

By taking the base $a$ logarithm of both sides of the equation the following equation is the result.

\begin{equation}
b = \sum_{i=0}^{k-1}2^i \cdot b_i
\end{equation}

The term $a^{2^i}$ can be found from the $i - 1$'th term by squaring the term since $\left ( a^{2^i} \right )^2$ is equal to
$a^{2^{i+1}}$.  This observation forms the basis of essentially all fast exponentiation algorithms.  It requires $k$ squarings and on average
$k \over 2$ multiplications to compute the result.  This is indeed quite an improvement over simply multiplying by $a$ a total of $b-1$ times.

While this current method is a considerable speed up there are further improvements to be made.  For example, the $a^{2^i}$ term does not need to
be computed in an auxilary variable.  Consider the following equivalent algorithm.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Left to Right Exponentiation}. \\
\textbf{Input}.   Integer $a$, $b$ and $k$ \\
\textbf{Output}.  $c = a^b$ \\
\hline \\
1.  $c \leftarrow 1$ \\
2.  for $i$ from $k - 1$ to $0$ do \\
\hspace{3mm}2.1  $c \leftarrow c^2$ \\
\hspace{3mm}2.2  $c \leftarrow c \cdot a^{b_i}$ \\
3.  Return $c$. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Left to Right Exponentiation}
\label{fig:LTOR}
\end{figure}

This algorithm starts from the most significant bit and works towards the least significant bit.  When the $i$'th bit of $b$ is set $a$ is
multiplied against the current product.  In each iteration the product is squared which doubles the exponent of the individual terms of the
product.

For example, let $b = 101100_2 \equiv 44_{10}$.  The following chart demonstrates the actions of the algorithm.

\newpage\begin{figure}
\begin{center}
\begin{tabular}{|c|c|}
\hline \textbf{Value of $i$} & \textbf{Value of $c$} \\
\hline - & $1$ \\
\hline $5$ & $a$ \\
\hline $4$ & $a^2$ \\
\hline $3$ & $a^4 \cdot a$ \\
\hline $2$ & $a^8 \cdot a^2 \cdot a$ \\
\hline $1$ & $a^{16} \cdot a^4 \cdot a^2$ \\
\hline $0$ & $a^{32} \cdot a^8 \cdot a^4$ \\
\hline
\end{tabular}
\end{center}
\caption{Example of Left to Right Exponentiation}
\end{figure}

When the product $a^{32} \cdot a^8 \cdot a^4$ is simplified it is equal $a^{44}$ which is the desired exponentiation.  This particular algorithm is
called ``Left to Right'' because it reads the exponent in that order.  All of the exponentiation algorithms that will be presented are of this nature.

\subsection{Single Digit Exponentiation}
The first algorithm in the series of exponentiation algorithms will be an unbounded algorithm where the exponent is a single digit.  It is intended
to be used when a small power of an input is required (\textit{e.g. $a^5$}).  It is faster than simply multiplying $b - 1$ times for all values of
$b$ that are greater than three.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_expt\_d}. \\
\textbf{Input}.   mp\_int $a$ and mp\_digit $b$ \\
\textbf{Output}.  $c = a^b$ \\
\hline \\
1.  $g \leftarrow a$ (\textit{mp\_init\_copy}) \\
2.  $c \leftarrow 1$ (\textit{mp\_set}) \\
3.  for $x$ from 1 to $lg(\beta)$ do \\
\hspace{3mm}3.1  $c \leftarrow c^2$ (\textit{mp\_sqr}) \\
\hspace{3mm}3.2  If $b$ AND $2^{lg(\beta) - 1} \ne 0$ then \\
\hspace{6mm}3.2.1  $c \leftarrow c \cdot g$ (\textit{mp\_mul}) \\
\hspace{3mm}3.3  $b \leftarrow b << 1$ \\
4.  Clear $g$. \\
5.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_expt\_d}
\end{figure}

\textbf{Algorithm mp\_expt\_d.}
This algorithm computes the value of $a$ raised to the power of a single digit $b$.  It uses the left to right exponentiation algorithm to
quickly compute the exponentiation.  It is loosely based on algorithm 14.79 of HAC \cite[pp. 615]{HAC} with the difference that the
exponent is a fixed width.

A copy of $a$ is made first to allow destination variable $c$ be the same as the source variable $a$.  The result is set to the initial value of
$1$ in the subsequent step.

Inside the loop the exponent is read from the most significant bit first down to the least significant bit.  First $c$ is invariably squared
on step 3.1.  In the following step if the most significant bit of $b$ is one the copy of $a$ is multiplied against $c$.  The value
of $b$ is shifted left one bit to make the next bit down from the most signficant bit the new most significant bit.  In effect each
iteration of the loop moves the bits of the exponent $b$ upwards to the most significant location.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_expt\_d\_ex.c
\vspace{-3mm}
\begin{alltt}
016
017   /* calculate c = a**b  using a square-multiply algorithm */
018   int mp_expt_d_ex (mp_int * a, mp_digit b, mp_int * c, int fast)
019   \{
020     int     res;
021     unsigned int x;
022
023     mp_int  g;
024
025     if ((res = mp_init_copy (&g, a)) != MP_OKAY) \{
026       return res;
027     \}
028
029     /* set initial result */
030     mp_set (c, 1);
031
032     if (fast != 0) \{
033       while (b > 0) \{
034         /* if the bit is set multiply */
035         if ((b & 1) != 0) \{
036           if ((res = mp_mul (c, &g, c)) != MP_OKAY) \{
037             mp_clear (&g);
038             return res;
039           \}
040         \}
041
042         /* square */
043         if (b > 1) \{
044           if ((res = mp_sqr (&g, &g)) != MP_OKAY) \{
045             mp_clear (&g);
046             return res;
047           \}
048         \}
049
050         /* shift to next bit */
051         b >>= 1;
052       \}
053     \}
054     else \{
055       for (x = 0; x < DIGIT_BIT; x++) \{
056         /* square */
057         if ((res = mp_sqr (c, c)) != MP_OKAY) \{
058           mp_clear (&g);
059           return res;
060         \}
061
062         /* if the bit is set multiply */
063         if ((b & (mp_digit) (((mp_digit)1) << (DIGIT_BIT - 1))) != 0) \{
064           if ((res = mp_mul (c, &g, c)) != MP_OKAY) \{
065              mp_clear (&g);
066              return res;
067           \}
068         \}
069
070         /* shift to next bit */
071         b <<= 1;
072       \}
073     \} /* if ... else */
074
075     mp_clear (&g);
076     return MP_OKAY;
077   \}
078   #endif
079
\end{alltt}
\end{small}

This describes only the algorithm that is used when the parameter $fast$ is $0$.  Line 30 sets the initial value of the result to $1$.  Next the loop on line 55 steps through each bit of the exponent starting from
the most significant down towards the least significant. The invariant squaring operation placed on line 57 is performed first.  After
the squaring the result $c$ is multiplied by the base $g$ if and only if the most significant bit of the exponent is set.  The shift on line
71 moves all of the bits of the exponent upwards towards the most significant location.

\section{$k$-ary Exponentiation}
When calculating an exponentiation the most time consuming bottleneck is the multiplications which are in general a small factor
slower than squaring.  Recall from the previous algorithm that $b_{i}$ refers to the $i$'th bit of the exponent $b$.  Suppose instead it referred to
the $i$'th $k$-bit digit of the exponent of $b$.  For $k = 1$ the definitions are synonymous and for $k > 1$ algorithm~\ref{fig:KARY}
computes the same exponentiation.  A group of $k$ bits from the exponent is called a \textit{window}.  That is it is a small window on only a
portion of the entire exponent.  Consider the following modification to the basic left to right exponentiation algorithm.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{$k$-ary Exponentiation}. \\
\textbf{Input}.   Integer $a$, $b$, $k$ and $t$ \\
\textbf{Output}.  $c = a^b$ \\
\hline \\
1.  $c \leftarrow 1$ \\
2.  for $i$ from $t - 1$ to $0$ do \\
\hspace{3mm}2.1  $c \leftarrow c^{2^k} $ \\
\hspace{3mm}2.2  Extract the $i$'th $k$-bit word from $b$ and store it in $g$. \\
\hspace{3mm}2.3  $c \leftarrow c \cdot a^g$ \\
3.  Return $c$. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{$k$-ary Exponentiation}
\label{fig:KARY}
\end{figure}

The squaring on step 2.1 can be calculated by squaring the value $c$ successively $k$ times.  If the values of $a^g$ for $0 < g < 2^k$ have been
precomputed this algorithm requires only $t$ multiplications and $tk$ squarings.  The table can be generated with $2^{k - 1} - 1$ squarings and
$2^{k - 1} + 1$ multiplications.  This algorithm assumes that the number of bits in the exponent is evenly divisible by $k$.
However, when it is not the remaining $0 < x \le k - 1$ bits can be handled with algorithm~\ref{fig:LTOR}.

Suppose $k = 4$ and $t = 100$.  This modified algorithm will require $109$ multiplications and $408$ squarings to compute the exponentiation.  The
original algorithm would on average have required $200$ multiplications and $400$ squrings to compute the same value.  The total number of squarings
has increased slightly but the number of multiplications has nearly halved.

\subsection{Optimal Values of $k$}
An optimal value of $k$ will minimize $2^{k} + \lceil n / k \rceil + n - 1$ for a fixed number of bits in the exponent $n$.  The simplest
approach is to brute force search amongst the values $k = 2, 3, \ldots, 8$ for the lowest result.  Table~\ref{fig:OPTK} lists optimal values of $k$
for various exponent sizes and compares the number of multiplication and squarings required against algorithm~\ref{fig:LTOR}.

\begin{figure}[here]
\begin{center}
\begin{small}
\begin{tabular}{|c|c|c|c|c|c|}
\hline \textbf{Exponent (bits)} & \textbf{Optimal $k$} & \textbf{Work at $k$} & \textbf{Work with ~\ref{fig:LTOR}} \\
\hline $16$ & $2$ & $27$ & $24$ \\
\hline $32$ & $3$ & $49$ & $48$ \\
\hline $64$ & $3$ & $92$ & $96$ \\
\hline $128$ & $4$ & $175$ & $192$ \\
\hline $256$ & $4$ & $335$ & $384$ \\
\hline $512$ & $5$ & $645$ & $768$ \\
\hline $1024$ & $6$ & $1257$ & $1536$ \\
\hline $2048$ & $6$ & $2452$ & $3072$ \\
\hline $4096$ & $7$ & $4808$ & $6144$ \\
\hline
\end{tabular}
\end{small}
\end{center}
\caption{Optimal Values of $k$ for $k$-ary Exponentiation}
\label{fig:OPTK}
\end{figure}

\subsection{Sliding-Window Exponentiation}
A simple modification to the previous algorithm is only generate the upper half of the table in the range $2^{k-1} \le g < 2^k$.  Essentially
this is a table for all values of $g$ where the most significant bit of $g$ is a one.  However, in order for this to be allowed in the
algorithm values of $g$ in the range $0 \le g < 2^{k-1}$ must be avoided.

Table~\ref{fig:OPTK2} lists optimal values of $k$ for various exponent sizes and compares the work required against algorithm {\ref{fig:KARY}}.

\begin{figure}[here]
\begin{center}
\begin{small}
\begin{tabular}{|c|c|c|c|c|c|}
\hline \textbf{Exponent (bits)} & \textbf{Optimal $k$} & \textbf{Work at $k$} & \textbf{Work with ~\ref{fig:KARY}} \\
\hline $16$ & $3$ & $24$ & $27$ \\
\hline $32$ & $3$ & $45$ & $49$ \\
\hline $64$ & $4$ & $87$ & $92$ \\
\hline $128$ & $4$ & $167$ & $175$ \\
\hline $256$ & $5$ & $322$ & $335$ \\
\hline $512$ & $6$ & $628$ & $645$ \\
\hline $1024$ & $6$ & $1225$ & $1257$ \\
\hline $2048$ & $7$ & $2403$ & $2452$ \\
\hline $4096$ & $8$ & $4735$ & $4808$ \\
\hline
\end{tabular}
\end{small}
\end{center}
\caption{Optimal Values of $k$ for Sliding Window Exponentiation}
\label{fig:OPTK2}
\end{figure}

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Sliding Window $k$-ary Exponentiation}. \\
\textbf{Input}.   Integer $a$, $b$, $k$ and $t$ \\
\textbf{Output}.  $c = a^b$ \\
\hline \\
1.  $c \leftarrow 1$ \\
2.  for $i$ from $t - 1$ to $0$ do \\
\hspace{3mm}2.1  If the $i$'th bit of $b$ is a zero then \\
\hspace{6mm}2.1.1   $c \leftarrow c^2$ \\
\hspace{3mm}2.2  else do \\
\hspace{6mm}2.2.1  $c \leftarrow c^{2^k}$ \\
\hspace{6mm}2.2.2  Extract the $k$ bits from $(b_{i}b_{i-1}\ldots b_{i-(k-1)})$ and store it in $g$. \\
\hspace{6mm}2.2.3  $c \leftarrow c \cdot a^g$ \\
\hspace{6mm}2.2.4  $i \leftarrow i - k$ \\
3.  Return $c$. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Sliding Window $k$-ary Exponentiation}
\end{figure}

Similar to the previous algorithm this algorithm must have a special handler when fewer than $k$ bits are left in the exponent.  While this
algorithm requires the same number of squarings it can potentially have fewer multiplications.  The pre-computed table $a^g$ is also half
the size as the previous table.

Consider the exponent $b = 111101011001000_2 \equiv 31432_{10}$ with $k = 3$ using both algorithms.  The first algorithm will divide the exponent up as
the following five $3$-bit words $b \equiv \left ( 111, 101, 011, 001, 000 \right )_{2}$.  The second algorithm will break the
exponent as $b \equiv \left ( 111, 101, 0, 110, 0, 100, 0 \right )_{2}$.  The single digit $0$ in the second representation are where
a single squaring took place instead of a squaring and multiplication.  In total the first method requires $10$ multiplications and $18$
squarings.  The second method requires $8$ multiplications and $18$ squarings.

In general the sliding window method is never slower than the generic $k$-ary method and often it is slightly faster.

\section{Modular Exponentiation}

Modular exponentiation is essentially computing the power of a base within a finite field or ring.  For example, computing
$d \equiv a^b \mbox{ (mod }c\mbox{)}$ is a modular exponentiation.  Instead of first computing $a^b$ and then reducing it
modulo $c$ the intermediate result is reduced modulo $c$ after every squaring or multiplication operation.

This guarantees that any intermediate result is bounded by $0 \le d \le c^2 - 2c + 1$ and can be reduced modulo $c$ quickly using
one of the algorithms presented in chapter six.

Before the actual modular exponentiation algorithm can be written a wrapper algorithm must be written first.  This algorithm
will allow the exponent $b$ to be negative which is computed as $c \equiv \left (1 / a \right )^{\vert b \vert} \mbox{(mod }d\mbox{)}$. The
value of $(1/a) \mbox{ mod }c$ is computed using the modular inverse (\textit{see \ref{sec;modinv}}).  If no inverse exists the algorithm
terminates with an error.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_exptmod}. \\
\textbf{Input}.   mp\_int $a$, $b$ and $c$ \\
\textbf{Output}.  $y \equiv g^x \mbox{ (mod }p\mbox{)}$ \\
\hline \\
1.  If $c.sign = MP\_NEG$ return(\textit{MP\_VAL}). \\
2.  If $b.sign = MP\_NEG$ then \\
\hspace{3mm}2.1  $g' \leftarrow g^{-1} \mbox{ (mod }c\mbox{)}$ \\
\hspace{3mm}2.2  $x' \leftarrow \vert x \vert$ \\
\hspace{3mm}2.3  Compute $d \equiv g'^{x'} \mbox{ (mod }c\mbox{)}$ via recursion. \\
3.  if $p$ is odd \textbf{OR} $p$ is a D.R. modulus then \\
\hspace{3mm}3.1  Compute $y \equiv g^{x} \mbox{ (mod }p\mbox{)}$ via algorithm mp\_exptmod\_fast. \\
4.  else \\
\hspace{3mm}4.1  Compute $y \equiv g^{x} \mbox{ (mod }p\mbox{)}$ via algorithm s\_mp\_exptmod. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_exptmod}
\end{figure}

\textbf{Algorithm mp\_exptmod.}
The first algorithm which actually performs modular exponentiation is algorithm s\_mp\_exptmod.  It is a sliding window $k$-ary algorithm
which uses Barrett reduction to reduce the product modulo $p$.  The second algorithm mp\_exptmod\_fast performs the same operation
except it uses either Montgomery or Diminished Radix reduction.  The two latter reduction algorithms are clumped in the same exponentiation
algorithm since their arguments are essentially the same (\textit{two mp\_ints and one mp\_digit}).

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_exptmod.c
\vspace{-3mm}
\begin{alltt}
016
017
018   /* this is a shell function that calls either the normal or Montgomery
019    * exptmod functions.  Originally the call to the montgomery code was
020    * embedded in the normal function but that wasted alot of stack space
021    * for nothing (since 99% of the time the Montgomery code would be called)
022    */
023   int mp_exptmod (mp_int * G, mp_int * X, mp_int * P, mp_int * Y)
024   \{
025     int dr;
026
027     /* modulus P must be positive */
028     if (P->sign == MP_NEG) \{
029        return MP_VAL;
030     \}
031
032     /* if exponent X is negative we have to recurse */
033     if (X->sign == MP_NEG) \{
034   #ifdef BN_MP_INVMOD_C
035        mp_int tmpG, tmpX;
036        int err;
037
038        /* first compute 1/G mod P */
039        if ((err = mp_init(&tmpG)) != MP_OKAY) \{
040           return err;
041        \}
042        if ((err = mp_invmod(G, P, &tmpG)) != MP_OKAY) \{
043           mp_clear(&tmpG);
044           return err;
045        \}
046
047        /* now get |X| */
048        if ((err = mp_init(&tmpX)) != MP_OKAY) \{
049           mp_clear(&tmpG);
050           return err;
051        \}
052        if ((err = mp_abs(X, &tmpX)) != MP_OKAY) \{
053           mp_clear_multi(&tmpG, &tmpX, NULL);
054           return err;
055        \}
056
057        /* and now compute (1/G)**|X| instead of G**X [X < 0] */
058        err = mp_exptmod(&tmpG, &tmpX, P, Y);
059        mp_clear_multi(&tmpG, &tmpX, NULL);
060        return err;
061   #else
062        /* no invmod */
063        return MP_VAL;
064   #endif
065     \}
066
067   /* modified diminished radix reduction */
068   #if defined(BN_MP_REDUCE_IS_2K_L_C) && defined(BN_MP_REDUCE_2K_L_C) && defin
      ed(BN_S_MP_EXPTMOD_C)
069     if (mp_reduce_is_2k_l(P) == MP_YES) \{
070        return s_mp_exptmod(G, X, P, Y, 1);
071     \}
072   #endif
073
074   #ifdef BN_MP_DR_IS_MODULUS_C
075     /* is it a DR modulus? */
076     dr = mp_dr_is_modulus(P);
077   #else
078     /* default to no */
079     dr = 0;
080   #endif
081
082   #ifdef BN_MP_REDUCE_IS_2K_C
083     /* if not, is it a unrestricted DR modulus? */
084     if (dr == 0) \{
085        dr = mp_reduce_is_2k(P) << 1;
086     \}
087   #endif
088
089     /* if the modulus is odd or dr != 0 use the montgomery method */
090   #ifdef BN_MP_EXPTMOD_FAST_C
091     if ((mp_isodd (P) == MP_YES) || (dr !=  0)) \{
092       return mp_exptmod_fast (G, X, P, Y, dr);
093     \} else \{
094   #endif
095   #ifdef BN_S_MP_EXPTMOD_C
096       /* otherwise use the generic Barrett reduction technique */
097       return s_mp_exptmod (G, X, P, Y, 0);
098   #else
099       /* no exptmod for evens */
100       return MP_VAL;
101   #endif
102   #ifdef BN_MP_EXPTMOD_FAST_C
103     \}
104   #endif
105   \}
106
107   #endif
108
\end{alltt}
\end{small}

In order to keep the algorithms in a known state the first step on line 28 is to reject any negative modulus as input.  If the exponent is
negative the algorithm tries to perform a modular exponentiation with the modular inverse of the base $G$.  The temporary variable $tmpG$ is assigned
the modular inverse of $G$ and $tmpX$ is assigned the absolute value of $X$.  The algorithm will recuse with these new values with a positive
exponent.

If the exponent is positive the algorithm resumes the exponentiation.  Line 76 determines if the modulus is of the restricted Diminished Radix
form.  If it is not line 69 attempts to determine if it is of a unrestricted Diminished Radix form.  The integer $dr$ will take on one
of three values.

\begin{enumerate}
\item $dr = 0$ means that the modulus is not of either restricted or unrestricted Diminished Radix form.
\item $dr = 1$ means that the modulus is of restricted Diminished Radix form.
\item $dr = 2$ means that the modulus is of unrestricted Diminished Radix form.
\end{enumerate}

Line 69 determines if the fast modular exponentiation algorithm can be used.  It is allowed if $dr \ne 0$ or if the modulus is odd.  Otherwise,
the slower s\_mp\_exptmod algorithm is used which uses Barrett reduction.

\subsection{Barrett Modular Exponentiation}

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{s\_mp\_exptmod}. \\
\textbf{Input}.   mp\_int $a$, $b$ and $c$ \\
\textbf{Output}.  $y \equiv g^x \mbox{ (mod }p\mbox{)}$ \\
\hline \\
1.  $k \leftarrow lg(x)$ \\
2.  $winsize \leftarrow  \left \lbrace \begin{array}{ll}
                              2 &  \mbox{if }k \le 7 \\
                              3 &  \mbox{if }7 < k \le 36 \\
                              4 &  \mbox{if }36 < k \le 140 \\
                              5 &  \mbox{if }140 < k \le 450 \\
                              6 &  \mbox{if }450 < k \le 1303 \\
                              7 &  \mbox{if }1303 < k \le 3529 \\
                              8 &  \mbox{if }3529 < k \\
                              \end{array} \right .$ \\
3.  Initialize $2^{winsize}$ mp\_ints in an array named $M$ and one mp\_int named $\mu$ \\
4.  Calculate the $\mu$ required for Barrett Reduction (\textit{mp\_reduce\_setup}). \\
5.  $M_1 \leftarrow g \mbox{ (mod }p\mbox{)}$ \\
\\
Setup the table of small powers of $g$.  First find $g^{2^{winsize}}$ and then all multiples of it. \\
6.  $k \leftarrow 2^{winsize - 1}$ \\
7.  $M_{k} \leftarrow M_1$ \\
8.  for $ix$ from 0 to $winsize - 2$ do \\
\hspace{3mm}8.1  $M_k \leftarrow \left ( M_k \right )^2$ (\textit{mp\_sqr})  \\
\hspace{3mm}8.2  $M_k \leftarrow M_k \mbox{ (mod }p\mbox{)}$ (\textit{mp\_reduce}) \\
9.  for $ix$ from $2^{winsize - 1} + 1$ to $2^{winsize} - 1$ do \\
\hspace{3mm}9.1  $M_{ix} \leftarrow M_{ix - 1} \cdot M_{1}$ (\textit{mp\_mul}) \\
\hspace{3mm}9.2  $M_{ix} \leftarrow M_{ix} \mbox{ (mod }p\mbox{)}$ (\textit{mp\_reduce}) \\
10.  $res \leftarrow 1$ \\
\\
Start Sliding Window. \\
11.  $mode \leftarrow 0, bitcnt \leftarrow 1, buf \leftarrow 0, digidx \leftarrow x.used - 1, bitcpy \leftarrow 0, bitbuf \leftarrow 0$ \\
12.  Loop \\
\hspace{3mm}12.1  $bitcnt \leftarrow bitcnt - 1$ \\
\hspace{3mm}12.2  If $bitcnt = 0$ then do \\
\hspace{6mm}12.2.1  If $digidx = -1$ goto step 13. \\
\hspace{6mm}12.2.2  $buf \leftarrow x_{digidx}$ \\
\hspace{6mm}12.2.3  $digidx \leftarrow digidx - 1$ \\
\hspace{6mm}12.2.4  $bitcnt \leftarrow lg(\beta)$ \\
Continued on next page. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm s\_mp\_exptmod}
\end{figure}

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{s\_mp\_exptmod} (\textit{continued}). \\
\textbf{Input}.   mp\_int $a$, $b$ and $c$ \\
\textbf{Output}.  $y \equiv g^x \mbox{ (mod }p\mbox{)}$ \\
\hline \\
\hspace{3mm}12.3  $y \leftarrow (buf >> (lg(\beta) - 1))$ AND $1$ \\
\hspace{3mm}12.4  $buf \leftarrow buf << 1$ \\
\hspace{3mm}12.5  if $mode = 0$ and $y = 0$ then goto step 12. \\
\hspace{3mm}12.6  if $mode = 1$ and $y = 0$ then do \\
\hspace{6mm}12.6.1  $res \leftarrow res^2$ \\
\hspace{6mm}12.6.2  $res \leftarrow res \mbox{ (mod }p\mbox{)}$ \\
\hspace{6mm}12.6.3  Goto step 12. \\
\hspace{3mm}12.7  $bitcpy \leftarrow bitcpy + 1$ \\
\hspace{3mm}12.8  $bitbuf \leftarrow bitbuf + (y << (winsize - bitcpy))$ \\
\hspace{3mm}12.9  $mode \leftarrow 2$ \\
\hspace{3mm}12.10  If $bitcpy = winsize$ then do \\
\hspace{6mm}Window is full so perform the squarings and single multiplication. \\
\hspace{6mm}12.10.1  for $ix$ from $0$ to $winsize -1$ do \\
\hspace{9mm}12.10.1.1  $res \leftarrow res^2$ \\
\hspace{9mm}12.10.1.2  $res \leftarrow res \mbox{ (mod }p\mbox{)}$ \\
\hspace{6mm}12.10.2  $res \leftarrow res \cdot M_{bitbuf}$ \\
\hspace{6mm}12.10.3  $res \leftarrow res \mbox{ (mod }p\mbox{)}$ \\
\hspace{6mm}Reset the window. \\
\hspace{6mm}12.10.4  $bitcpy \leftarrow 0, bitbuf \leftarrow 0, mode \leftarrow 1$ \\
\\
No more windows left.  Check for residual bits of exponent. \\
13.  If $mode = 2$ and $bitcpy > 0$ then do \\
\hspace{3mm}13.1  for $ix$ form $0$ to $bitcpy - 1$ do \\
\hspace{6mm}13.1.1  $res \leftarrow res^2$ \\
\hspace{6mm}13.1.2  $res \leftarrow res \mbox{ (mod }p\mbox{)}$ \\
\hspace{6mm}13.1.3  $bitbuf \leftarrow bitbuf << 1$ \\
\hspace{6mm}13.1.4  If $bitbuf$ AND $2^{winsize} \ne 0$ then do \\
\hspace{9mm}13.1.4.1  $res \leftarrow res \cdot M_{1}$ \\
\hspace{9mm}13.1.4.2  $res \leftarrow res \mbox{ (mod }p\mbox{)}$ \\
14.  $y \leftarrow res$ \\
15.  Clear $res$, $mu$ and the $M$ array. \\
16.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm s\_mp\_exptmod (continued)}
\end{figure}

\textbf{Algorithm s\_mp\_exptmod.}
This algorithm computes the $x$'th power of $g$ modulo $p$ and stores the result in $y$.  It takes advantage of the Barrett reduction
algorithm to keep the product small throughout the algorithm.

The first two steps determine the optimal window size based on the number of bits in the exponent.  The larger the exponent the
larger the window size becomes.  After a window size $winsize$ has been chosen an array of $2^{winsize}$ mp\_int variables is allocated.  This
table will hold the values of $g^x \mbox{ (mod }p\mbox{)}$ for $2^{winsize - 1} \le x < 2^{winsize}$.

After the table is allocated the first power of $g$ is found.  Since $g \ge p$ is allowed it must be first reduced modulo $p$ to make
the rest of the algorithm more efficient.  The first element of the table at $2^{winsize - 1}$ is found by squaring $M_1$ successively $winsize - 2$
times.  The rest of the table elements are found by multiplying the previous element by $M_1$ modulo $p$.

Now that the table is available the sliding window may begin.  The following list describes the functions of all the variables in the window.
\begin{enumerate}
\item The variable $mode$ dictates how the bits of the exponent are interpreted.
\begin{enumerate}
   \item When $mode = 0$ the bits are ignored since no non-zero bit of the exponent has been seen yet.  For example, if the exponent were simply
         $1$ then there would be $lg(\beta) - 1$ zero bits before the first non-zero bit.  In this case bits are ignored until a non-zero bit is found.
   \item When $mode = 1$ a non-zero bit has been seen before and a new $winsize$-bit window has not been formed yet.  In this mode leading $0$ bits
         are read and a single squaring is performed.  If a non-zero bit is read a new window is created.
   \item When $mode = 2$ the algorithm is in the middle of forming a window and new bits are appended to the window from the most significant bit
         downwards.
\end{enumerate}
\item The variable $bitcnt$ indicates how many bits are left in the current digit of the exponent left to be read.  When it reaches zero a new digit
      is fetched from the exponent.
\item The variable $buf$ holds the currently read digit of the exponent.
\item The variable $digidx$ is an index into the exponents digits.  It starts at the leading digit $x.used - 1$ and moves towards the trailing digit.
\item The variable $bitcpy$ indicates how many bits are in the currently formed window.  When it reaches $winsize$ the window is flushed and
      the appropriate operations performed.
\item The variable $bitbuf$ holds the current bits of the window being formed.
\end{enumerate}

All of step 12 is the window processing loop.  It will iterate while there are digits available form the exponent to read.  The first step
inside this loop is to extract a new digit if no more bits are available in the current digit.  If there are no bits left a new digit is
read and if there are no digits left than the loop terminates.

After a digit is made available step 12.3 will extract the most significant bit of the current digit and move all other bits in the digit
upwards.  In effect the digit is read from most significant bit to least significant bit and since the digits are read from leading to
trailing edges the entire exponent is read from most significant bit to least significant bit.

At step 12.5 if the $mode$ and currently extracted bit $y$ are both zero the bit is ignored and the next bit is read.  This prevents the
algorithm from having to perform trivial squaring and reduction operations before the first non-zero bit is read.  Step 12.6 and 12.7-10 handle
the two cases of $mode = 1$ and $mode = 2$ respectively.

\begin{center}
\begin{figure}[here]
\includegraphics{pics/expt_state.ps}
\caption{Sliding Window State Diagram}
\label{pic:expt_state}
\end{figure}
\end{center}

By step 13 there are no more digits left in the exponent.  However, there may be partial bits in the window left.  If $mode = 2$ then
a Left-to-Right algorithm is used to process the remaining few bits.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_s\_mp\_exptmod.c
\vspace{-3mm}
\begin{alltt}
016   #ifdef MP_LOW_MEM
017      #define TAB_SIZE 32
018   #else
019      #define TAB_SIZE 256
020   #endif
021
022   int s_mp_exptmod (mp_int * G, mp_int * X, mp_int * P, mp_int * Y, int redmod
      e)
023   \{
024     mp_int  M[TAB_SIZE], res, mu;
025     mp_digit buf;
026     int     err, bitbuf, bitcpy, bitcnt, mode, digidx, x, y, winsize;
027     int (*redux)(mp_int*,mp_int*,mp_int*);
028
029     /* find window size */
030     x = mp_count_bits (X);
031     if (x <= 7) \{
032       winsize = 2;
033     \} else if (x <= 36) \{
034       winsize = 3;
035     \} else if (x <= 140) \{
036       winsize = 4;
037     \} else if (x <= 450) \{
038       winsize = 5;
039     \} else if (x <= 1303) \{
040       winsize = 6;
041     \} else if (x <= 3529) \{
042       winsize = 7;
043     \} else \{
044       winsize = 8;
045     \}
046
047   #ifdef MP_LOW_MEM
048       if (winsize > 5) \{
049          winsize = 5;
050       \}
051   #endif
052
053     /* init M array */
054     /* init first cell */
055     if ((err = mp_init(&M[1])) != MP_OKAY) \{
056        return err;
057     \}
058
059     /* now init the second half of the array */
060     for (x = 1<<(winsize-1); x < (1 << winsize); x++) \{
061       if ((err = mp_init(&M[x])) != MP_OKAY) \{
062         for (y = 1<<(winsize-1); y < x; y++) \{
063           mp_clear (&M[y]);
064         \}
065         mp_clear(&M[1]);
066         return err;
067       \}
068     \}
069
070     /* create mu, used for Barrett reduction */
071     if ((err = mp_init (&mu)) != MP_OKAY) \{
072       goto LBL_M;
073     \}
074
075     if (redmode == 0) \{
076        if ((err = mp_reduce_setup (&mu, P)) != MP_OKAY) \{
077           goto LBL_MU;
078        \}
079        redux = mp_reduce;
080     \} else \{
081        if ((err = mp_reduce_2k_setup_l (P, &mu)) != MP_OKAY) \{
082           goto LBL_MU;
083        \}
084        redux = mp_reduce_2k_l;
085     \}
086
087     /* create M table
088      *
089      * The M table contains powers of the base,
090      * e.g. M[x] = G**x mod P
091      *
092      * The first half of the table is not
093      * computed though accept for M[0] and M[1]
094      */
095     if ((err = mp_mod (G, P, &M[1])) != MP_OKAY) \{
096       goto LBL_MU;
097     \}
098
099     /* compute the value at M[1<<(winsize-1)] by squaring
100      * M[1] (winsize-1) times
101      */
102     if ((err = mp_copy (&M[1], &M[1 << (winsize - 1)])) != MP_OKAY) \{
103       goto LBL_MU;
104     \}
105
106     for (x = 0; x < (winsize - 1); x++) \{
107       /* square it */
108       if ((err = mp_sqr (&M[1 << (winsize - 1)],
109                          &M[1 << (winsize - 1)])) != MP_OKAY) \{
110         goto LBL_MU;
111       \}
112
113       /* reduce modulo P */
114       if ((err = redux (&M[1 << (winsize - 1)], P, &mu)) != MP_OKAY) \{
115         goto LBL_MU;
116       \}
117     \}
118
119     /* create upper table, that is M[x] = M[x-1] * M[1] (mod P)
120      * for x = (2**(winsize - 1) + 1) to (2**winsize - 1)
121      */
122     for (x = (1 << (winsize - 1)) + 1; x < (1 << winsize); x++) \{
123       if ((err = mp_mul (&M[x - 1], &M[1], &M[x])) != MP_OKAY) \{
124         goto LBL_MU;
125       \}
126       if ((err = redux (&M[x], P, &mu)) != MP_OKAY) \{
127         goto LBL_MU;
128       \}
129     \}
130
131     /* setup result */
132     if ((err = mp_init (&res)) != MP_OKAY) \{
133       goto LBL_MU;
134     \}
135     mp_set (&res, 1);
136
137     /* set initial mode and bit cnt */
138     mode   = 0;
139     bitcnt = 1;
140     buf    = 0;
141     digidx = X->used - 1;
142     bitcpy = 0;
143     bitbuf = 0;
144
145     for (;;) \{
146       /* grab next digit as required */
147       if (--bitcnt == 0) \{
148         /* if digidx == -1 we are out of digits */
149         if (digidx == -1) \{
150           break;
151         \}
152         /* read next digit and reset the bitcnt */
153         buf    = X->dp[digidx--];
154         bitcnt = (int) DIGIT_BIT;
155       \}
156
157       /* grab the next msb from the exponent */
158       y     = (buf >> (mp_digit)(DIGIT_BIT - 1)) & 1;
159       buf <<= (mp_digit)1;
160
161       /* if the bit is zero and mode == 0 then we ignore it
162        * These represent the leading zero bits before the first 1 bit
163        * in the exponent.  Technically this opt is not required but it
164        * does lower the # of trivial squaring/reductions used
165        */
166       if ((mode == 0) && (y == 0)) \{
167         continue;
168       \}
169
170       /* if the bit is zero and mode == 1 then we square */
171       if ((mode == 1) && (y == 0)) \{
172         if ((err = mp_sqr (&res, &res)) != MP_OKAY) \{
173           goto LBL_RES;
174         \}
175         if ((err = redux (&res, P, &mu)) != MP_OKAY) \{
176           goto LBL_RES;
177         \}
178         continue;
179       \}
180
181       /* else we add it to the window */
182       bitbuf |= (y << (winsize - ++bitcpy));
183       mode    = 2;
184
185       if (bitcpy == winsize) \{
186         /* ok window is filled so square as required and multiply  */
187         /* square first */
188         for (x = 0; x < winsize; x++) \{
189           if ((err = mp_sqr (&res, &res)) != MP_OKAY) \{
190             goto LBL_RES;
191           \}
192           if ((err = redux (&res, P, &mu)) != MP_OKAY) \{
193             goto LBL_RES;
194           \}
195         \}
196
197         /* then multiply */
198         if ((err = mp_mul (&res, &M[bitbuf], &res)) != MP_OKAY) \{
199           goto LBL_RES;
200         \}
201         if ((err = redux (&res, P, &mu)) != MP_OKAY) \{
202           goto LBL_RES;
203         \}
204
205         /* empty window and reset */
206         bitcpy = 0;
207         bitbuf = 0;
208         mode   = 1;
209       \}
210     \}
211
212     /* if bits remain then square/multiply */
213     if ((mode == 2) && (bitcpy > 0)) \{
214       /* square then multiply if the bit is set */
215       for (x = 0; x < bitcpy; x++) \{
216         if ((err = mp_sqr (&res, &res)) != MP_OKAY) \{
217           goto LBL_RES;
218         \}
219         if ((err = redux (&res, P, &mu)) != MP_OKAY) \{
220           goto LBL_RES;
221         \}
222
223         bitbuf <<= 1;
224         if ((bitbuf & (1 << winsize)) != 0) \{
225           /* then multiply */
226           if ((err = mp_mul (&res, &M[1], &res)) != MP_OKAY) \{
227             goto LBL_RES;
228           \}
229           if ((err = redux (&res, P, &mu)) != MP_OKAY) \{
230             goto LBL_RES;
231           \}
232         \}
233       \}
234     \}
235
236     mp_exch (&res, Y);
237     err = MP_OKAY;
238   LBL_RES:mp_clear (&res);
239   LBL_MU:mp_clear (&mu);
240   LBL_M:
241     mp_clear(&M[1]);
242     for (x = 1<<(winsize-1); x < (1 << winsize); x++) \{
243       mp_clear (&M[x]);
244     \}
245     return err;
246   \}
247   #endif
248
\end{alltt}
\end{small}

Lines 31 through 45 determine the optimal window size based on the length of the exponent in bits.  The window divisions are sorted
from smallest to greatest so that in each \textbf{if} statement only one condition must be tested.  For example, by the \textbf{if} statement
on line 37 the value of $x$ is already known to be greater than $140$.

The conditional piece of code beginning on line 47 allows the window size to be restricted to five bits.  This logic is used to ensure
the table of precomputed powers of $G$ remains relatively small.

The for loop on line 60 initializes the $M$ array while lines 71 and 76 through 85 initialize the reduction
function that will be used for this modulus.

-- More later.

\section{Quick Power of Two}
Calculating $b = 2^a$ can be performed much quicker than with any of the previous algorithms.  Recall that a logical shift left $m << k$ is
equivalent to $m \cdot 2^k$.  By this logic when $m = 1$ a quick power of two can be achieved.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_2expt}. \\
\textbf{Input}.   integer $b$ \\
\textbf{Output}.  $a \leftarrow 2^b$ \\
\hline \\
1.  $a \leftarrow 0$ \\
2.  If $a.alloc < \lfloor b / lg(\beta) \rfloor + 1$ then grow $a$ appropriately. \\
3.  $a.used \leftarrow \lfloor b / lg(\beta) \rfloor + 1$ \\
4.  $a_{\lfloor b / lg(\beta) \rfloor} \leftarrow 1 << (b \mbox{ mod } lg(\beta))$ \\
5.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_2expt}
\end{figure}

\textbf{Algorithm mp\_2expt.}

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_2expt.c
\vspace{-3mm}
\begin{alltt}
016
017   /* computes a = 2**b
018    *
019    * Simple algorithm which zeroes the int, grows it then just sets one bit
020    * as required.
021    */
022   int
023   mp_2expt (mp_int * a, int b)
024   \{
025     int     res;
026
027     /* zero a as per default */
028     mp_zero (a);
029
030     /* grow a to accomodate the single bit */
031     if ((res = mp_grow (a, (b / DIGIT_BIT) + 1)) != MP_OKAY) \{
032       return res;
033     \}
034
035     /* set the used count of where the bit will go */
036     a->used = (b / DIGIT_BIT) + 1;
037
038     /* put the single bit in its place */
039     a->dp[b / DIGIT_BIT] = ((mp_digit)1) << (b % DIGIT_BIT);
040
041     return MP_OKAY;
042   \}
043   #endif
044
\end{alltt}
\end{small}

\chapter{Higher Level Algorithms}

This chapter discusses the various higher level algorithms that are required to complete a well rounded multiple precision integer package.  These
routines are less performance oriented than the algorithms of chapters five, six and seven but are no less important.

The first section describes a method of integer division with remainder that is universally well known.  It provides the signed division logic
for the package.  The subsequent section discusses a set of algorithms which allow a single digit to be the 2nd operand for a variety of operations.
These algorithms serve mostly to simplify other algorithms where small constants are required.  The last two sections discuss how to manipulate
various representations of integers.  For example, converting from an mp\_int to a string of character.

\section{Integer Division with Remainder}
\label{sec:division}

Integer division aside from modular exponentiation is the most intensive algorithm to compute.  Like addition, subtraction and multiplication
the basis of this algorithm is the long-hand division algorithm taught to school children.  Throughout this discussion several common variables
will be used.  Let $x$ represent the divisor and $y$ represent the dividend.  Let $q$ represent the integer quotient $\lfloor y / x \rfloor$ and
let $r$ represent the remainder $r = y - x \lfloor y / x \rfloor$.  The following simple algorithm will be used to start the discussion.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Radix-$\beta$ Integer Division}. \\
\textbf{Input}.   integer $x$ and $y$ \\
\textbf{Output}.  $q = \lfloor y/x\rfloor, r = y - xq$ \\
\hline \\
1.  $q \leftarrow 0$ \\
2.  $n \leftarrow \vert \vert y \vert \vert - \vert \vert x \vert \vert$ \\
3.  for $t$ from $n$ down to $0$ do \\
\hspace{3mm}3.1  Maximize $k$ such that $kx\beta^t$ is less than or equal to $y$ and $(k + 1)x\beta^t$ is greater. \\
\hspace{3mm}3.2  $q \leftarrow q + k\beta^t$ \\
\hspace{3mm}3.3  $y \leftarrow y - kx\beta^t$ \\
4.  $r \leftarrow y$ \\
5.  Return($q, r$) \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Radix-$\beta$ Integer Division}
\label{fig:raddiv}
\end{figure}

As children we are taught this very simple algorithm for the case of $\beta = 10$.  Almost instinctively several optimizations are taught for which
their reason of existing are never explained.  For this example let $y = 5471$ represent the dividend and $x = 23$ represent the divisor.

To find the first digit of the quotient the value of $k$ must be maximized such that $kx\beta^t$ is less than or equal to $y$ and
simultaneously $(k + 1)x\beta^t$ is greater than $y$.  Implicitly $k$ is the maximum value the $t$'th digit of the quotient may have.  The habitual method
used to find the maximum is to ``eyeball'' the two numbers, typically only the leading digits and quickly estimate a quotient.  By only using leading
digits a much simpler division may be used to form an educated guess at what the value must be.  In this case $k = \lfloor 54/23\rfloor = 2$ quickly
arises as a possible  solution.  Indeed $2x\beta^2 = 4600$ is less than $y = 5471$ and simultaneously $(k + 1)x\beta^2 = 6900$ is larger than $y$.
As a  result $k\beta^2$ is added to the quotient which now equals $q = 200$ and $4600$ is subtracted from $y$ to give a remainder of $y = 841$.

Again this process is repeated to produce the quotient digit $k = 3$ which makes the quotient $q = 200 + 3\beta = 230$ and the remainder
$y = 841 - 3x\beta = 181$.  Finally the last iteration of the loop produces $k = 7$ which leads to the quotient $q = 230 + 7 = 237$ and the
remainder $y = 181 - 7x = 20$.  The final quotient and remainder found are $q = 237$ and $r = y = 20$ which are indeed correct since
$237 \cdot 23 + 20 = 5471$ is true.

\subsection{Quotient Estimation}
\label{sec:divest}
As alluded to earlier the quotient digit $k$ can be estimated from only the leading digits of both the divisor and dividend.  When $p$ leading
digits are used from both the divisor and dividend to form an estimation the accuracy of the estimation rises as $p$ grows.  Technically
speaking the estimation is based on assuming the lower $\vert \vert y \vert \vert - p$ and $\vert \vert x \vert \vert - p$ lower digits of the
dividend and divisor are zero.

The value of the estimation may off by a few values in either direction and in general is fairly correct.  A simplification \cite[pp. 271]{TAOCPV2}
of the estimation technique is to use $t + 1$ digits of the dividend and $t$ digits of the divisor, in particularly when $t = 1$.  The estimate
using this technique is never too small.  For the following proof let $t = \vert \vert y \vert \vert - 1$ and $s = \vert \vert x \vert \vert - 1$
represent the most significant digits of the dividend and divisor respectively.

\textbf{Proof.}\textit{  The quotient $\hat k = \lfloor (y_t\beta + y_{t-1}) / x_s \rfloor$ is greater than or equal to
$k = \lfloor y / (x \cdot \beta^{\vert \vert y \vert \vert - \vert \vert x \vert \vert - 1}) \rfloor$. }
The first obvious case is when $\hat k = \beta - 1$ in which case the proof is concluded since the real quotient cannot be larger.  For all other
cases $\hat k = \lfloor (y_t\beta + y_{t-1}) / x_s \rfloor$ and $\hat k x_s \ge y_t\beta + y_{t-1} - x_s + 1$.  The latter portion of the inequalility
$-x_s + 1$ arises from the fact that a truncated integer division will give the same quotient for at most $x_s - 1$ values.  Next a series of
inequalities will prove the hypothesis.

\begin{equation}
y - \hat k x \le y - \hat k x_s\beta^s
\end{equation}

This is trivially true since $x \ge x_s\beta^s$.  Next we replace $\hat kx_s\beta^s$ by the previous inequality for $\hat kx_s$.

\begin{equation}
y - \hat k x \le y_t\beta^t + \ldots + y_0 - (y_t\beta^t + y_{t-1}\beta^{t-1} - x_s\beta^t + \beta^s)
\end{equation}

By simplifying the previous inequality the following inequality is formed.

\begin{equation}
y - \hat k x \le y_{t-2}\beta^{t-2} + \ldots + y_0 + x_s\beta^s - \beta^s
\end{equation}

Subsequently,

\begin{equation}
y_{t-2}\beta^{t-2} + \ldots +  y_0  + x_s\beta^s - \beta^s < x_s\beta^s \le x
\end{equation}

Which proves that $y - \hat kx \le x$ and by consequence $\hat k \ge k$ which concludes the proof.  \textbf{QED}


\subsection{Normalized Integers}
For the purposes of division a normalized input is when the divisors leading digit $x_n$ is greater than or equal to $\beta / 2$.  By multiplying both
$x$ and $y$ by $j = \lfloor (\beta / 2) / x_n \rfloor$ the quotient remains unchanged and the remainder is simply $j$ times the original
remainder.  The purpose of normalization is to ensure the leading digit of the divisor is sufficiently large such that the estimated quotient will
lie in the domain of a single digit.  Consider the maximum dividend $(\beta - 1) \cdot \beta + (\beta - 1)$ and the minimum divisor $\beta / 2$.

\begin{equation}
{{\beta^2 - 1} \over { \beta / 2}} \le 2\beta - {2 \over \beta}
\end{equation}

At most the quotient approaches $2\beta$, however, in practice this will not occur since that would imply the previous quotient digit was too small.

\subsection{Radix-$\beta$ Division with Remainder}
\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_div}. \\
\textbf{Input}.   mp\_int $a, b$ \\
\textbf{Output}.  $c = \lfloor a/b \rfloor$, $d = a - bc$ \\
\hline \\
1.  If $b = 0$ return(\textit{MP\_VAL}). \\
2.  If $\vert a \vert < \vert b \vert$ then do \\
\hspace{3mm}2.1  $d \leftarrow a$ \\
\hspace{3mm}2.2  $c \leftarrow 0$ \\
\hspace{3mm}2.3  Return(\textit{MP\_OKAY}). \\
\\
Setup the quotient to receive the digits. \\
3.  Grow $q$ to $a.used + 2$ digits. \\
4.  $q \leftarrow 0$ \\
5.  $x \leftarrow \vert a \vert , y \leftarrow \vert b \vert$ \\
6.  $sign \leftarrow  \left \lbrace \begin{array}{ll}
                              MP\_ZPOS &  \mbox{if }a.sign = b.sign \\
                              MP\_NEG  &  \mbox{otherwise} \\
                              \end{array} \right .$ \\
\\
Normalize the inputs such that the leading digit of $y$ is greater than or equal to $\beta / 2$. \\
7.  $norm \leftarrow (lg(\beta) - 1) - (\lceil lg(y) \rceil \mbox{ (mod }lg(\beta)\mbox{)})$ \\
8.  $x \leftarrow x \cdot 2^{norm}, y \leftarrow y \cdot 2^{norm}$ \\
\\
Find the leading digit of the quotient. \\
9.  $n \leftarrow x.used - 1, t \leftarrow y.used - 1$ \\
10.  $y \leftarrow y \cdot \beta^{n - t}$ \\
11.  While ($x \ge y$) do \\
\hspace{3mm}11.1  $q_{n - t} \leftarrow q_{n - t} + 1$ \\
\hspace{3mm}11.2  $x \leftarrow x - y$ \\
12.  $y \leftarrow \lfloor y / \beta^{n-t} \rfloor$ \\
\\
Continued on the next page. \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_div}
\end{figure}

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_div} (continued). \\
\textbf{Input}.   mp\_int $a, b$ \\
\textbf{Output}.  $c = \lfloor a/b \rfloor$, $d = a - bc$ \\
\hline \\
Now find the remainder fo the digits. \\
13.  for $i$ from $n$ down to $(t + 1)$ do \\
\hspace{3mm}13.1  If $i > x.used$ then jump to the next iteration of this loop. \\
\hspace{3mm}13.2  If $x_{i} = y_{t}$ then \\
\hspace{6mm}13.2.1  $q_{i - t - 1} \leftarrow \beta - 1$ \\
\hspace{3mm}13.3  else \\
\hspace{6mm}13.3.1  $\hat r \leftarrow x_{i} \cdot \beta + x_{i - 1}$ \\
\hspace{6mm}13.3.2  $\hat r \leftarrow \lfloor \hat r / y_{t} \rfloor$ \\
\hspace{6mm}13.3.3  $q_{i - t - 1} \leftarrow \hat r$ \\
\hspace{3mm}13.4  $q_{i - t - 1} \leftarrow q_{i - t - 1} + 1$ \\
\\
Fixup quotient estimation. \\
\hspace{3mm}13.5  Loop \\
\hspace{6mm}13.5.1  $q_{i - t - 1} \leftarrow q_{i - t - 1} - 1$ \\
\hspace{6mm}13.5.2  t$1 \leftarrow 0$ \\
\hspace{6mm}13.5.3  t$1_0 \leftarrow y_{t - 1}, $ t$1_1 \leftarrow y_t,$ t$1.used \leftarrow 2$ \\
\hspace{6mm}13.5.4  $t1 \leftarrow t1 \cdot q_{i - t - 1}$ \\
\hspace{6mm}13.5.5  t$2_0 \leftarrow x_{i - 2}, $ t$2_1 \leftarrow x_{i - 1}, $ t$2_2 \leftarrow x_i, $ t$2.used \leftarrow 3$ \\
\hspace{6mm}13.5.6  If $\vert t1 \vert > \vert t2 \vert$ then goto step 13.5. \\
\hspace{3mm}13.6  t$1 \leftarrow y \cdot q_{i - t - 1}$ \\
\hspace{3mm}13.7  t$1 \leftarrow $ t$1 \cdot \beta^{i - t - 1}$ \\
\hspace{3mm}13.8  $x \leftarrow x - $ t$1$ \\
\hspace{3mm}13.9  If $x.sign = MP\_NEG$ then \\
\hspace{6mm}13.10  t$1 \leftarrow y$ \\
\hspace{6mm}13.11  t$1 \leftarrow $ t$1 \cdot \beta^{i - t - 1}$ \\
\hspace{6mm}13.12  $x \leftarrow x + $ t$1$ \\
\hspace{6mm}13.13  $q_{i - t - 1} \leftarrow q_{i - t - 1} - 1$ \\
\\
Finalize the result. \\
14.  Clamp excess digits of $q$ \\
15.  $c \leftarrow q, c.sign \leftarrow sign$ \\
16.  $x.sign \leftarrow a.sign$ \\
17.  $d \leftarrow \lfloor x / 2^{norm} \rfloor$ \\
18.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_div (continued)}
\end{figure}
\textbf{Algorithm mp\_div.}
This algorithm will calculate quotient and remainder from an integer division given a dividend and divisor.  The algorithm is a signed
division and will produce a fully qualified quotient and remainder.

First the divisor $b$ must be non-zero which is enforced in step one.  If the divisor is larger than the dividend than the quotient is implicitly
zero and the remainder is the dividend.

After the first two trivial cases of inputs are handled the variable $q$ is setup to receive the digits of the quotient.  Two unsigned copies of the
divisor $y$ and dividend $x$ are made as well.  The core of the division algorithm is an unsigned division and will only work if the values are
positive.  Now the two values $x$ and $y$ must be normalized such that the leading digit of $y$ is greater than or equal to $\beta / 2$.
This is performed by shifting both to the left by enough bits to get the desired normalization.

At this point the division algorithm can begin producing digits of the quotient.  Recall that maximum value of the estimation used is
$2\beta - {2 \over \beta}$ which means that a digit of the quotient must be first produced by another means.  In this case $y$ is shifted
to the left (\textit{step ten}) so that it has the same number of digits as $x$.  The loop on step eleven will subtract multiples of the
shifted copy of $y$ until $x$ is smaller.  Since the leading digit of $y$ is greater than or equal to $\beta/2$ this loop will iterate at most two
times to produce the desired leading digit of the quotient.

Now the remainder of the digits can be produced.  The equation $\hat q = \lfloor {{x_i \beta + x_{i-1}}\over y_t} \rfloor$ is used to fairly
accurately approximate the true quotient digit.  The estimation can in theory produce an estimation as high as $2\beta - {2 \over \beta}$ but by
induction the upper quotient digit is correct (\textit{as established on step eleven}) and the estimate must be less than $\beta$.

Recall from section~\ref{sec:divest} that the estimation is never too low but may be too high.  The next step of the estimation process is
to refine the estimation.  The loop on step 13.5 uses $x_i\beta^2 + x_{i-1}\beta + x_{i-2}$ and $q_{i - t - 1}(y_t\beta + y_{t-1})$ as a higher
order approximation to adjust the quotient digit.

After both phases of estimation the quotient digit may still be off by a value of one\footnote{This is similar to the error introduced
by optimizing Barrett reduction.}.  Steps 13.6 and 13.7 subtract the multiple of the divisor from the dividend (\textit{Similar to step 3.3 of
algorithm~\ref{fig:raddiv}} and then subsequently add a multiple of the divisor if the quotient was too large.

Now that the quotient has been determine finializing the result is a matter of clamping the quotient, fixing the sizes and de-normalizing the
remainder.  An important aspect of this algorithm seemingly overlooked in other descriptions such as that of Algorithm 14.20 HAC \cite[pp. 598]{HAC}
is that when the estimations are being made (\textit{inside the loop on step 13.5}) that the digits $y_{t-1}$, $x_{i-2}$ and $x_{i-1}$ may lie
outside their respective boundaries.  For example, if $t = 0$ or $i \le 1$ then the digits would be undefined.  In those cases the digits should
respectively be replaced with a zero.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_div.c
\vspace{-3mm}
\begin{alltt}
016
017   #ifdef BN_MP_DIV_SMALL
018
019   /* slower bit-bang division... also smaller */
020   int mp_div(mp_int * a, mp_int * b, mp_int * c, mp_int * d)
021   \{
022      mp_int ta, tb, tq, q;
023      int    res, n, n2;
024
025     /* is divisor zero ? */
026     if (mp_iszero (b) == MP_YES) \{
027       return MP_VAL;
028     \}
029
030     /* if a < b then q=0, r = a */
031     if (mp_cmp_mag (a, b) == MP_LT) \{
032       if (d != NULL) \{
033         res = mp_copy (a, d);
034       \} else \{
035         res = MP_OKAY;
036       \}
037       if (c != NULL) \{
038         mp_zero (c);
039       \}
040       return res;
041     \}
042
043     /* init our temps */
044     if ((res = mp_init_multi(&ta, &tb, &tq, &q, NULL)) != MP_OKAY) \{
045        return res;
046     \}
047
048
049     mp_set(&tq, 1);
050     n = mp_count_bits(a) - mp_count_bits(b);
051     if (((res = mp_abs(a, &ta)) != MP_OKAY) ||
052         ((res = mp_abs(b, &tb)) != MP_OKAY) ||
053         ((res = mp_mul_2d(&tb, n, &tb)) != MP_OKAY) ||
054         ((res = mp_mul_2d(&tq, n, &tq)) != MP_OKAY)) \{
055         goto LBL_ERR;
056     \}
057
058     while (n-- >= 0) \{
059        if (mp_cmp(&tb, &ta) != MP_GT) \{
060           if (((res = mp_sub(&ta, &tb, &ta)) != MP_OKAY) ||
061               ((res = mp_add(&q, &tq, &q)) != MP_OKAY)) \{
062              goto LBL_ERR;
063           \}
064        \}
065        if (((res = mp_div_2d(&tb, 1, &tb, NULL)) != MP_OKAY) ||
066            ((res = mp_div_2d(&tq, 1, &tq, NULL)) != MP_OKAY)) \{
067              goto LBL_ERR;
068        \}
069     \}
070
071     /* now q == quotient and ta == remainder */
072     n  = a->sign;
073     n2 = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
074     if (c != NULL) \{
075        mp_exch(c, &q);
076        c->sign  = (mp_iszero(c) == MP_YES) ? MP_ZPOS : n2;
077     \}
078     if (d != NULL) \{
079        mp_exch(d, &ta);
080        d->sign = (mp_iszero(d) == MP_YES) ? MP_ZPOS : n;
081     \}
082   LBL_ERR:
083      mp_clear_multi(&ta, &tb, &tq, &q, NULL);
084      return res;
085   \}
086
087   #else
088
089   /* integer signed division.
090    * c*b + d == a [e.g. a/b, c=quotient, d=remainder]
091    * HAC pp.598 Algorithm 14.20
092    *
093    * Note that the description in HAC is horribly
094    * incomplete.  For example, it doesn't consider
095    * the case where digits are removed from 'x' in
096    * the inner loop.  It also doesn't consider the
097    * case that y has fewer than three digits, etc..
098    *
099    * The overall algorithm is as described as
100    * 14.20 from HAC but fixed to treat these cases.
101   */
102   int mp_div (mp_int * a, mp_int * b, mp_int * c, mp_int * d)
103   \{
104     mp_int  q, x, y, t1, t2;
105     int     res, n, t, i, norm, neg;
106
107     /* is divisor zero ? */
108     if (mp_iszero (b) == MP_YES) \{
109       return MP_VAL;
110     \}
111
112     /* if a < b then q=0, r = a */
113     if (mp_cmp_mag (a, b) == MP_LT) \{
114       if (d != NULL) \{
115         res = mp_copy (a, d);
116       \} else \{
117         res = MP_OKAY;
118       \}
119       if (c != NULL) \{
120         mp_zero (c);
121       \}
122       return res;
123     \}
124
125     if ((res = mp_init_size (&q, a->used + 2)) != MP_OKAY) \{
126       return res;
127     \}
128     q.used = a->used + 2;
129
130     if ((res = mp_init (&t1)) != MP_OKAY) \{
131       goto LBL_Q;
132     \}
133
134     if ((res = mp_init (&t2)) != MP_OKAY) \{
135       goto LBL_T1;
136     \}
137
138     if ((res = mp_init_copy (&x, a)) != MP_OKAY) \{
139       goto LBL_T2;
140     \}
141
142     if ((res = mp_init_copy (&y, b)) != MP_OKAY) \{
143       goto LBL_X;
144     \}
145
146     /* fix the sign */
147     neg = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
148     x.sign = y.sign = MP_ZPOS;
149
150     /* normalize both x and y, ensure that y >= b/2, [b == 2**DIGIT_BIT] */
151     norm = mp_count_bits(&y) % DIGIT_BIT;
152     if (norm < (int)(DIGIT_BIT-1)) \{
153        norm = (DIGIT_BIT-1) - norm;
154        if ((res = mp_mul_2d (&x, norm, &x)) != MP_OKAY) \{
155          goto LBL_Y;
156        \}
157        if ((res = mp_mul_2d (&y, norm, &y)) != MP_OKAY) \{
158          goto LBL_Y;
159        \}
160     \} else \{
161        norm = 0;
162     \}
163
164     /* note hac does 0 based, so if used==5 then its 0,1,2,3,4, e.g. use 4 */
165     n = x.used - 1;
166     t = y.used - 1;
167
168     /* while (x >= y*b**n-t) do \{ q[n-t] += 1; x -= y*b**\{n-t\} \} */
169     if ((res = mp_lshd (&y, n - t)) != MP_OKAY) \{ /* y = y*b**\{n-t\} */
170       goto LBL_Y;
171     \}
172
173     while (mp_cmp (&x, &y) != MP_LT) \{
174       ++(q.dp[n - t]);
175       if ((res = mp_sub (&x, &y, &x)) != MP_OKAY) \{
176         goto LBL_Y;
177       \}
178     \}
179
180     /* reset y by shifting it back down */
181     mp_rshd (&y, n - t);
182
183     /* step 3. for i from n down to (t + 1) */
184     for (i = n; i >= (t + 1); i--) \{
185       if (i > x.used) \{
186         continue;
187       \}
188
189       /* step 3.1 if xi == yt then set q\{i-t-1\} to b-1,
190        * otherwise set q\{i-t-1\} to (xi*b + x\{i-1\})/yt */
191       if (x.dp[i] == y.dp[t]) \{
192         q.dp[(i - t) - 1] = ((((mp_digit)1) << DIGIT_BIT) - 1);
193       \} else \{
194         mp_word tmp;
195         tmp = ((mp_word) x.dp[i]) << ((mp_word) DIGIT_BIT);
196         tmp |= ((mp_word) x.dp[i - 1]);
197         tmp /= ((mp_word) y.dp[t]);
198         if (tmp > (mp_word) MP_MASK) \{
199           tmp = MP_MASK;
200         \}
201         q.dp[(i - t) - 1] = (mp_digit) (tmp & (mp_word) (MP_MASK));
202       \}
203
204       /* while (q\{i-t-1\} * (yt * b + y\{t-1\})) >
205                xi * b**2 + xi-1 * b + xi-2
206
207          do q\{i-t-1\} -= 1;
208       */
209       q.dp[(i - t) - 1] = (q.dp[(i - t) - 1] + 1) & MP_MASK;
210       do \{
211         q.dp[(i - t) - 1] = (q.dp[(i - t) - 1] - 1) & MP_MASK;
212
213         /* find left hand */
214         mp_zero (&t1);
215         t1.dp[0] = ((t - 1) < 0) ? 0 : y.dp[t - 1];
216         t1.dp[1] = y.dp[t];
217         t1.used = 2;
218         if ((res = mp_mul_d (&t1, q.dp[(i - t) - 1], &t1)) != MP_OKAY) \{
219           goto LBL_Y;
220         \}
221
222         /* find right hand */
223         t2.dp[0] = ((i - 2) < 0) ? 0 : x.dp[i - 2];
224         t2.dp[1] = ((i - 1) < 0) ? 0 : x.dp[i - 1];
225         t2.dp[2] = x.dp[i];
226         t2.used = 3;
227       \} while (mp_cmp_mag(&t1, &t2) == MP_GT);
228
229       /* step 3.3 x = x - q\{i-t-1\} * y * b**\{i-t-1\} */
230       if ((res = mp_mul_d (&y, q.dp[(i - t) - 1], &t1)) != MP_OKAY) \{
231         goto LBL_Y;
232       \}
233
234       if ((res = mp_lshd (&t1, (i - t) - 1)) != MP_OKAY) \{
235         goto LBL_Y;
236       \}
237
238       if ((res = mp_sub (&x, &t1, &x)) != MP_OKAY) \{
239         goto LBL_Y;
240       \}
241
242       /* if x < 0 then \{ x = x + y*b**\{i-t-1\}; q\{i-t-1\} -= 1; \} */
243       if (x.sign == MP_NEG) \{
244         if ((res = mp_copy (&y, &t1)) != MP_OKAY) \{
245           goto LBL_Y;
246         \}
247         if ((res = mp_lshd (&t1, (i - t) - 1)) != MP_OKAY) \{
248           goto LBL_Y;
249         \}
250         if ((res = mp_add (&x, &t1, &x)) != MP_OKAY) \{
251           goto LBL_Y;
252         \}
253
254         q.dp[(i - t) - 1] = (q.dp[(i - t) - 1] - 1UL) & MP_MASK;
255       \}
256     \}
257
258     /* now q is the quotient and x is the remainder
259      * [which we have to normalize]
260      */
261
262     /* get sign before writing to c */
263     x.sign = (x.used == 0) ? MP_ZPOS : a->sign;
264
265     if (c != NULL) \{
266       mp_clamp (&q);
267       mp_exch (&q, c);
268       c->sign = neg;
269     \}
270
271     if (d != NULL) \{
272       if ((res = mp_div_2d (&x, norm, &x, NULL)) != MP_OKAY) \{
273         goto LBL_Y;
274       \}
275       mp_exch (&x, d);
276     \}
277
278     res = MP_OKAY;
279
280   LBL_Y:mp_clear (&y);
281   LBL_X:mp_clear (&x);
282   LBL_T2:mp_clear (&t2);
283   LBL_T1:mp_clear (&t1);
284   LBL_Q:mp_clear (&q);
285     return res;
286   \}
287
288   #endif
289
290   #endif
291
\end{alltt}
\end{small}

The implementation of this algorithm differs slightly from the pseudo code presented previously.  In this algorithm either of the quotient $c$ or
remainder $d$ may be passed as a \textbf{NULL} pointer which indicates their value is not desired.  For example, the C code to call the division
algorithm with only the quotient is

\begin{verbatim}
mp_div(&a, &b, &c, NULL);  /* c = [a/b] */
\end{verbatim}

Lines 108 and 113 handle the two trivial cases of inputs which are division by zero and dividend smaller than the divisor
respectively.  After the two trivial cases all of the temporary variables are initialized.  Line 147 determines the sign of
the quotient and line 148 ensures that both $x$ and $y$ are positive.

The number of bits in the leading digit is calculated on line 151.  Implictly an mp\_int with $r$ digits will require $lg(\beta)(r-1) + k$ bits
of precision which when reduced modulo $lg(\beta)$ produces the value of $k$.  In this case $k$ is the number of bits in the leading digit which is
exactly what is required.  For the algorithm to operate $k$ must equal $lg(\beta) - 1$ and when it does not the inputs must be normalized by shifting
them to the left by $lg(\beta) - 1 - k$ bits.

Throughout the variables $n$ and $t$ will represent the highest digit of $x$ and $y$ respectively.  These are first used to produce the
leading digit of the quotient.  The loop beginning on line 184 will produce the remainder of the quotient digits.

The conditional ``continue'' on line 186 is used to prevent the algorithm from reading past the leading edge of $x$ which can occur when the
algorithm eliminates multiple non-zero digits in a single iteration.  This ensures that $x_i$ is always non-zero since by definition the digits
above the $i$'th position $x$ must be zero in order for the quotient to be precise\footnote{Precise as far as integer division is concerned.}.

Lines 214, 216 and 223 through 225 manually construct the high accuracy estimations by setting the digits of the two mp\_int
variables directly.

\section{Single Digit Helpers}

This section briefly describes a series of single digit helper algorithms which come in handy when working with small constants.  All of
the helper functions assume the single digit input is positive and will treat them as such.

\subsection{Single Digit Addition and Subtraction}

Both addition and subtraction are performed by ``cheating'' and using mp\_set followed by the higher level addition or subtraction
algorithms.   As a result these algorithms are subtantially simpler with a slight cost in performance.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_add\_d}. \\
\textbf{Input}.   mp\_int $a$ and a mp\_digit $b$ \\
\textbf{Output}.  $c = a + b$ \\
\hline \\
1.  $t \leftarrow b$ (\textit{mp\_set}) \\
2.  $c \leftarrow a + t$ \\
3.  Return(\textit{MP\_OKAY}) \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_add\_d}
\end{figure}

\textbf{Algorithm mp\_add\_d.}
This algorithm initiates a temporary mp\_int with the value of the single digit and uses algorithm mp\_add to add the two values together.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_add\_d.c
\vspace{-3mm}
\begin{alltt}
016
017   /* single digit addition */
018   int
019   mp_add_d (mp_int * a, mp_digit b, mp_int * c)
020   \{
021     int     res, ix, oldused;
022     mp_digit *tmpa, *tmpc, mu;
023
024     /* grow c as required */
025     if (c->alloc < (a->used + 1)) \{
026        if ((res = mp_grow(c, a->used + 1)) != MP_OKAY) \{
027           return res;
028        \}
029     \}
030
031     /* if a is negative and |a| >= b, call c = |a| - b */
032     if ((a->sign == MP_NEG) && ((a->used > 1) || (a->dp[0] >= b))) \{
033        /* temporarily fix sign of a */
034        a->sign = MP_ZPOS;
035
036        /* c = |a| - b */
037        res = mp_sub_d(a, b, c);
038
039        /* fix sign  */
040        a->sign = c->sign = MP_NEG;
041
042        /* clamp */
043        mp_clamp(c);
044
045        return res;
046     \}
047
048     /* old number of used digits in c */
049     oldused = c->used;
050
051     /* sign always positive */
052     c->sign = MP_ZPOS;
053
054     /* source alias */
055     tmpa    = a->dp;
056
057     /* destination alias */
058     tmpc    = c->dp;
059
060     /* if a is positive */
061     if (a->sign == MP_ZPOS) \{
062        /* add digit, after this we're propagating
063         * the carry.
064         */
065        *tmpc   = *tmpa++ + b;
066        mu      = *tmpc >> DIGIT_BIT;
067        *tmpc++ &= MP_MASK;
068
069        /* now handle rest of the digits */
070        for (ix = 1; ix < a->used; ix++) \{
071           *tmpc   = *tmpa++ + mu;
072           mu      = *tmpc >> DIGIT_BIT;
073           *tmpc++ &= MP_MASK;
074        \}
075        /* set final carry */
076        ix++;
077        *tmpc++  = mu;
078
079        /* setup size */
080        c->used = a->used + 1;
081     \} else \{
082        /* a was negative and |a| < b */
083        c->used  = 1;
084
085        /* the result is a single digit */
086        if (a->used == 1) \{
087           *tmpc++  =  b - a->dp[0];
088        \} else \{
089           *tmpc++  =  b;
090        \}
091
092        /* setup count so the clearing of oldused
093         * can fall through correctly
094         */
095        ix       = 1;
096     \}
097
098     /* now zero to oldused */
099     while (ix++ < oldused) \{
100        *tmpc++ = 0;
101     \}
102     mp_clamp(c);
103
104     return MP_OKAY;
105   \}
106
107   #endif
108
\end{alltt}
\end{small}

Clever use of the letter 't'.

\subsubsection{Subtraction}
The single digit subtraction algorithm mp\_sub\_d is essentially the same except it uses mp\_sub to subtract the digit from the mp\_int.

\subsection{Single Digit Multiplication}
Single digit multiplication arises enough in division and radix conversion that it ought to be implement as a special case of the baseline
multiplication algorithm.  Essentially this algorithm is a modified version of algorithm s\_mp\_mul\_digs where one of the multiplicands
only has one digit.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_mul\_d}. \\
\textbf{Input}.   mp\_int $a$ and a mp\_digit $b$ \\
\textbf{Output}.  $c = ab$ \\
\hline \\
1.  $pa \leftarrow a.used$ \\
2.  Grow $c$ to at least $pa + 1$ digits. \\
3.  $oldused \leftarrow c.used$ \\
4.  $c.used \leftarrow pa + 1$ \\
5.  $c.sign \leftarrow a.sign$ \\
6.  $\mu \leftarrow 0$ \\
7.  for $ix$ from $0$ to $pa - 1$ do \\
\hspace{3mm}7.1  $\hat r \leftarrow \mu + a_{ix}b$ \\
\hspace{3mm}7.2  $c_{ix} \leftarrow \hat r \mbox{ (mod }\beta\mbox{)}$ \\
\hspace{3mm}7.3  $\mu \leftarrow \lfloor \hat r / \beta \rfloor$ \\
8.  $c_{pa} \leftarrow \mu$ \\
9.  for $ix$ from $pa + 1$ to $oldused$ do \\
\hspace{3mm}9.1  $c_{ix} \leftarrow 0$ \\
10.  Clamp excess digits of $c$. \\
11.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_mul\_d}
\end{figure}
\textbf{Algorithm mp\_mul\_d.}
This algorithm quickly multiplies an mp\_int by a small single digit value.  It is specially tailored to the job and has a minimal of overhead.
Unlike the full multiplication algorithms this algorithm does not require any significnat temporary storage or memory allocations.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_mul\_d.c
\vspace{-3mm}
\begin{alltt}
016
017   /* multiply by a digit */
018   int
019   mp_mul_d (mp_int * a, mp_digit b, mp_int * c)
020   \{
021     mp_digit u, *tmpa, *tmpc;
022     mp_word  r;
023     int      ix, res, olduse;
024
025     /* make sure c is big enough to hold a*b */
026     if (c->alloc < (a->used + 1)) \{
027       if ((res = mp_grow (c, a->used + 1)) != MP_OKAY) \{
028         return res;
029       \}
030     \}
031
032     /* get the original destinations used count */
033     olduse = c->used;
034
035     /* set the sign */
036     c->sign = a->sign;
037
038     /* alias for a->dp [source] */
039     tmpa = a->dp;
040
041     /* alias for c->dp [dest] */
042     tmpc = c->dp;
043
044     /* zero carry */
045     u = 0;
046
047     /* compute columns */
048     for (ix = 0; ix < a->used; ix++) \{
049       /* compute product and carry sum for this term */
050       r       = (mp_word)u + ((mp_word)*tmpa++ * (mp_word)b);
051
052       /* mask off higher bits to get a single digit */
053       *tmpc++ = (mp_digit) (r & ((mp_word) MP_MASK));
054
055       /* send carry into next iteration */
056       u       = (mp_digit) (r >> ((mp_word) DIGIT_BIT));
057     \}
058
059     /* store final carry [if any] and increment ix offset  */
060     *tmpc++ = u;
061     ++ix;
062
063     /* now zero digits above the top */
064     while (ix++ < olduse) \{
065        *tmpc++ = 0;
066     \}
067
068     /* set used count */
069     c->used = a->used + 1;
070     mp_clamp(c);
071
072     return MP_OKAY;
073   \}
074   #endif
075
\end{alltt}
\end{small}

In this implementation the destination $c$ may point to the same mp\_int as the source $a$ since the result is written after the digit is
read from the source.  This function uses pointer aliases $tmpa$ and $tmpc$ for the digits of $a$ and $c$ respectively.

\subsection{Single Digit Division}
Like the single digit multiplication algorithm, single digit division is also a fairly common algorithm used in radix conversion.  Since the
divisor is only a single digit a specialized variant of the division algorithm can be used to compute the quotient.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_div\_d}. \\
\textbf{Input}.   mp\_int $a$ and a mp\_digit $b$ \\
\textbf{Output}.  $c = \lfloor a / b \rfloor, d = a - cb$ \\
\hline \\
1.  If $b = 0$ then return(\textit{MP\_VAL}).\\
2.  If $b = 3$ then use algorithm mp\_div\_3 instead. \\
3.  Init $q$ to $a.used$ digits.  \\
4.  $q.used \leftarrow a.used$ \\
5.  $q.sign \leftarrow a.sign$ \\
6.  $\hat w \leftarrow 0$ \\
7.  for $ix$ from $a.used - 1$ down to $0$ do \\
\hspace{3mm}7.1  $\hat w \leftarrow \hat w \beta + a_{ix}$ \\
\hspace{3mm}7.2  If $\hat w \ge b$ then \\
\hspace{6mm}7.2.1  $t \leftarrow \lfloor \hat w / b \rfloor$ \\
\hspace{6mm}7.2.2  $\hat w \leftarrow \hat w \mbox{ (mod }b\mbox{)}$ \\
\hspace{3mm}7.3  else\\
\hspace{6mm}7.3.1  $t \leftarrow 0$ \\
\hspace{3mm}7.4  $q_{ix} \leftarrow t$ \\
8.  $d \leftarrow \hat w$ \\
9.  Clamp excess digits of $q$. \\
10.  $c \leftarrow q$ \\
11.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_div\_d}
\end{figure}
\textbf{Algorithm mp\_div\_d.}
This algorithm divides the mp\_int $a$ by the single mp\_digit $b$ using an optimized approach.  Essentially in every iteration of the
algorithm another digit of the dividend is reduced and another digit of quotient produced.  Provided $b < \beta$ the value of $\hat w$
after step 7.1 will be limited such that $0 \le \lfloor \hat w / b \rfloor < \beta$.

If the divisor $b$ is equal to three a variant of this algorithm is used which is called mp\_div\_3.  It replaces the division by three with
a multiplication by $\lfloor \beta / 3 \rfloor$ and the appropriate shift and residual fixup.  In essence it is much like the Barrett reduction
from chapter seven.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_div\_d.c
\vspace{-3mm}
\begin{alltt}
016
017   static int s_is_power_of_two(mp_digit b, int *p)
018   \{
019      int x;
020
021      /* fast return if no power of two */
022      if ((b == 0) || ((b & (b-1)) != 0)) \{
023         return 0;
024      \}
025
026      for (x = 0; x < DIGIT_BIT; x++) \{
027         if (b == (((mp_digit)1)<<x)) \{
028            *p = x;
029            return 1;
030         \}
031      \}
032      return 0;
033   \}
034
035   /* single digit division (based on routine from MPI) */
036   int mp_div_d (mp_int * a, mp_digit b, mp_int * c, mp_digit * d)
037   \{
038     mp_int  q;
039     mp_word w;
040     mp_digit t;
041     int     res, ix;
042
043     /* cannot divide by zero */
044     if (b == 0) \{
045        return MP_VAL;
046     \}
047
048     /* quick outs */
049     if ((b == 1) || (mp_iszero(a) == MP_YES)) \{
050        if (d != NULL) \{
051           *d = 0;
052        \}
053        if (c != NULL) \{
054           return mp_copy(a, c);
055        \}
056        return MP_OKAY;
057     \}
058
059     /* power of two ? */
060     if (s_is_power_of_two(b, &ix) == 1) \{
061        if (d != NULL) \{
062           *d = a->dp[0] & ((((mp_digit)1)<<ix) - 1);
063        \}
064        if (c != NULL) \{
065           return mp_div_2d(a, ix, c, NULL);
066        \}
067        return MP_OKAY;
068     \}
069
070   #ifdef BN_MP_DIV_3_C
071     /* three? */
072     if (b == 3) \{
073        return mp_div_3(a, c, d);
074     \}
075   #endif
076
077     /* no easy answer [c'est la vie].  Just division */
078     if ((res = mp_init_size(&q, a->used)) != MP_OKAY) \{
079        return res;
080     \}
081
082     q.used = a->used;
083     q.sign = a->sign;
084     w = 0;
085     for (ix = a->used - 1; ix >= 0; ix--) \{
086        w = (w << ((mp_word)DIGIT_BIT)) | ((mp_word)a->dp[ix]);
087
088        if (w >= b) \{
089           t = (mp_digit)(w / b);
090           w -= ((mp_word)t) * ((mp_word)b);
091         \} else \{
092           t = 0;
093         \}
094         q.dp[ix] = (mp_digit)t;
095     \}
096
097     if (d != NULL) \{
098        *d = (mp_digit)w;
099     \}
100
101     if (c != NULL) \{
102        mp_clamp(&q);
103        mp_exch(&q, c);
104     \}
105     mp_clear(&q);
106
107     return res;
108   \}
109
110   #endif
111
\end{alltt}
\end{small}

Like the implementation of algorithm mp\_div this algorithm allows either of the quotient or remainder to be passed as a \textbf{NULL} pointer to
indicate the respective value is not required.  This allows a trivial single digit modular reduction algorithm, mp\_mod\_d to be created.

The division and remainder on lines 89 and 90 can be replaced often by a single division on most processors.  For example, the 32-bit x86 based
processors can divide a 64-bit quantity by a 32-bit quantity and produce the quotient and remainder simultaneously.  Unfortunately the GCC
compiler does not recognize that optimization and will actually produce two function calls to find the quotient and remainder respectively.

\subsection{Single Digit Root Extraction}

Finding the $n$'th root of an integer is fairly easy as far as numerical analysis is concerned.  Algorithms such as the Newton-Raphson approximation
(\ref{eqn:newton}) series will converge very quickly to a root for any continuous function $f(x)$.

\begin{equation}
x_{i+1} = x_i - {f(x_i) \over f'(x_i)}
\label{eqn:newton}
\end{equation}

In this case the $n$'th root is desired and $f(x) = x^n - a$ where $a$ is the integer of which the root is desired.  The derivative of $f(x)$ is
simply $f'(x) = nx^{n - 1}$.  Of particular importance is that this algorithm will be used over the integers not over the a more continuous domain
such as the real numbers.  As a result the root found can be above the true root by few and must be manually adjusted.  Ideally at the end of the
algorithm the $n$'th root $b$ of an integer $a$ is desired such that $b^n \le a$.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_n\_root}. \\
\textbf{Input}.   mp\_int $a$ and a mp\_digit $b$ \\
\textbf{Output}.  $c^b \le a$ \\
\hline \\
1.  If $b$ is even and $a.sign = MP\_NEG$ return(\textit{MP\_VAL}). \\
2.  $sign \leftarrow a.sign$ \\
3.  $a.sign \leftarrow MP\_ZPOS$ \\
4.  t$2 \leftarrow 2$ \\
5.  Loop \\
\hspace{3mm}5.1  t$1 \leftarrow $ t$2$ \\
\hspace{3mm}5.2  t$3 \leftarrow $ t$1^{b - 1}$ \\
\hspace{3mm}5.3  t$2 \leftarrow $ t$3 $ $\cdot$ t$1$ \\
\hspace{3mm}5.4  t$2 \leftarrow $ t$2 - a$ \\
\hspace{3mm}5.5  t$3 \leftarrow $ t$3 \cdot b$ \\
\hspace{3mm}5.6  t$3 \leftarrow \lfloor $t$2 / $t$3 \rfloor$ \\
\hspace{3mm}5.7  t$2 \leftarrow $ t$1 - $ t$3$ \\
\hspace{3mm}5.8  If t$1 \ne $ t$2$ then goto step 5.  \\
6.  Loop \\
\hspace{3mm}6.1  t$2 \leftarrow $ t$1^b$ \\
\hspace{3mm}6.2  If t$2 > a$ then \\
\hspace{6mm}6.2.1  t$1 \leftarrow $ t$1 - 1$ \\
\hspace{6mm}6.2.2  Goto step 6. \\
7.  $a.sign \leftarrow sign$ \\
8.  $c \leftarrow $ t$1$ \\
9.  $c.sign \leftarrow sign$  \\
10.  Return(\textit{MP\_OKAY}).  \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_n\_root}
\end{figure}
\textbf{Algorithm mp\_n\_root.}
This algorithm finds the integer $n$'th root of an input using the Newton-Raphson approach.  It is partially optimized based on the observation
that the numerator of ${f(x) \over f'(x)}$ can be derived from a partial denominator.  That is at first the denominator is calculated by finding
$x^{b - 1}$.  This value can then be multiplied by $x$ and have $a$ subtracted from it to find the numerator.  This saves a total of $b - 1$
multiplications by t$1$ inside the loop.

The initial value of the approximation is t$2 = 2$ which allows the algorithm to start with very small values and quickly converge on the
root.  Ideally this algorithm is meant to find the $n$'th root of an input where $n$ is bounded by $2 \le n \le 5$.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_n\_root.c
\vspace{-3mm}
\begin{alltt}
016
017   /* wrapper function for mp_n_root_ex()
018    * computes c = (a)**(1/b) such that (c)**b <= a and (c+1)**b > a
019    */
020   int mp_n_root (mp_int * a, mp_digit b, mp_int * c)
021   \{
022     return mp_n_root_ex(a, b, c, 0);
023   \}
024
025   #endif
026
\end{alltt}
\end{small}

\section{Random Number Generation}

Random numbers come up in a variety of activities from public key cryptography to simple simulations and various randomized algorithms.  Pollard-Rho
factoring for example, can make use of random values as starting points to find factors of a composite integer.  In this case the algorithm presented
is solely for simulations and not intended for cryptographic use.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_rand}. \\
\textbf{Input}.   An integer $b$ \\
\textbf{Output}.  A pseudo-random number of $b$ digits \\
\hline \\
1.  $a \leftarrow 0$ \\
2.  If $b \le 0$ return(\textit{MP\_OKAY}) \\
3.  Pick a non-zero random digit $d$. \\
4.  $a \leftarrow a + d$ \\
5.  for $ix$ from 1 to $d - 1$ do \\
\hspace{3mm}5.1  $a \leftarrow a \cdot \beta$ \\
\hspace{3mm}5.2  Pick a random digit $d$. \\
\hspace{3mm}5.3  $a \leftarrow a + d$ \\
6.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_rand}
\end{figure}
\textbf{Algorithm mp\_rand.}
This algorithm produces a pseudo-random integer of $b$ digits.  By ensuring that the first digit is non-zero the algorithm also guarantees that the
final result has at least $b$ digits.  It relies heavily on a third-part random number generator which should ideally generate uniformly all of
the integers from $0$ to $\beta - 1$.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_rand.c
\vspace{-3mm}
\begin{alltt}
016
017   /* makes a pseudo-random int of a given size */
018   int
019   mp_rand (mp_int * a, int digits)
020   \{
021     int     res;
022     mp_digit d;
023
024     mp_zero (a);
025     if (digits <= 0) \{
026       return MP_OKAY;
027     \}
028
029     /* first place a random non-zero digit */
030     do \{
031       d = ((mp_digit) abs (MP_GEN_RANDOM())) & MP_MASK;
032     \} while (d == 0);
033
034     if ((res = mp_add_d (a, d, a)) != MP_OKAY) \{
035       return res;
036     \}
037
038     while (--digits > 0) \{
039       if ((res = mp_lshd (a, 1)) != MP_OKAY) \{
040         return res;
041       \}
042
043       if ((res = mp_add_d (a, ((mp_digit) abs (MP_GEN_RANDOM())), a)) != MP_OK
      AY) \{
044         return res;
045       \}
046     \}
047
048     return MP_OKAY;
049   \}
050   #endif
051
\end{alltt}
\end{small}

\section{Formatted Representations}
The ability to emit a radix-$n$ textual representation of an integer is useful for interacting with human parties.  For example, the ability to
be given a string of characters such as ``114585'' and turn it into the radix-$\beta$ equivalent would make it easier to enter numbers
into a program.

\subsection{Reading Radix-n Input}
For the purposes of this text we will assume that a simple lower ASCII map (\ref{fig:ASC}) is used for the values of from $0$ to $63$ to
printable characters.  For example, when the character ``N'' is read it represents the integer $23$.  The first $16$ characters of the
map are for the common representations up to hexadecimal.  After that they match the ``base64'' encoding scheme which are suitable chosen
such that they are printable.  While outputting as base64 may not be too helpful for human operators it does allow communication via non binary
mediums.

\newpage\begin{figure}[here]
\begin{center}
\begin{tabular}{cc|cc|cc|cc}
\hline \textbf{Value} & \textbf{Char} & \textbf{Value} & \textbf{Char} & \textbf{Value} & \textbf{Char} &  \textbf{Value} & \textbf{Char} \\
\hline
0 & 0 & 1 & 1 & 2 & 2 & 3 & 3 \\
4 & 4 & 5 & 5 & 6 & 6 & 7 & 7 \\
8 & 8 & 9 & 9 & 10 & A & 11 & B \\
12 & C & 13 & D & 14 & E & 15 & F \\
16 & G & 17 & H & 18 & I & 19 & J \\
20 & K & 21 & L & 22 & M & 23 & N \\
24 & O & 25 & P & 26 & Q & 27 & R \\
28 & S & 29 & T & 30 & U & 31 & V \\
32 & W & 33 & X & 34 & Y & 35 & Z \\
36 & a & 37 & b & 38 & c & 39 & d \\
40 & e & 41 & f & 42 & g & 43 & h \\
44 & i & 45 & j & 46 & k & 47 & l \\
48 & m & 49 & n & 50 & o & 51 & p \\
52 & q & 53 & r & 54 & s & 55 & t \\
56 & u & 57 & v & 58 & w & 59 & x \\
60 & y & 61 & z & 62 & $+$ & 63 & $/$ \\
\hline
\end{tabular}
\end{center}
\caption{Lower ASCII Map}
\label{fig:ASC}
\end{figure}

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_read\_radix}. \\
\textbf{Input}.   A string $str$ of length $sn$ and radix $r$. \\
\textbf{Output}.  The radix-$\beta$ equivalent mp\_int. \\
\hline \\
1.  If $r < 2$ or $r > 64$ return(\textit{MP\_VAL}). \\
2.  $ix \leftarrow 0$ \\
3.  If $str_0 =$ ``-'' then do \\
\hspace{3mm}3.1  $ix \leftarrow ix + 1$ \\
\hspace{3mm}3.2  $sign \leftarrow MP\_NEG$ \\
4.  else \\
\hspace{3mm}4.1  $sign \leftarrow MP\_ZPOS$ \\
5.  $a \leftarrow 0$ \\
6.  for $iy$ from $ix$ to $sn - 1$ do \\
\hspace{3mm}6.1  Let $y$ denote the position in the map of $str_{iy}$. \\
\hspace{3mm}6.2  If $str_{iy}$ is not in the map or $y \ge r$ then goto step 7. \\
\hspace{3mm}6.3  $a \leftarrow a \cdot r$ \\
\hspace{3mm}6.4  $a \leftarrow a + y$ \\
7.  If $a \ne 0$ then $a.sign \leftarrow sign$ \\
8.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_read\_radix}
\end{figure}
\textbf{Algorithm mp\_read\_radix.}
This algorithm will read an ASCII string and produce the radix-$\beta$ mp\_int representation of the same integer.  A minus symbol ``-'' may precede the
string  to indicate the value is negative, otherwise it is assumed to be positive.  The algorithm will read up to $sn$ characters from the input
and will stop when it reads a character it cannot map the algorithm stops reading characters from the string.  This allows numbers to be embedded
as part of larger input without any significant problem.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_read\_radix.c
\vspace{-3mm}
\begin{alltt}
016
017   /* read a string [ASCII] in a given radix */
018   int mp_read_radix (mp_int * a, const char *str, int radix)
019   \{
020     int     y, res, neg;
021     char    ch;
022
023     /* zero the digit bignum */
024     mp_zero(a);
025
026     /* make sure the radix is ok */
027     if ((radix < 2) || (radix > 64)) \{
028       return MP_VAL;
029     \}
030
031     /* if the leading digit is a
032      * minus set the sign to negative.
033      */
034     if (*str == '-') \{
035       ++str;
036       neg = MP_NEG;
037     \} else \{
038       neg = MP_ZPOS;
039     \}
040
041     /* set the integer to the default of zero */
042     mp_zero (a);
043
044     /* process each digit of the string */
045     while (*str != '\symbol{92}0') \{
046       /* if the radix <= 36 the conversion is case insensitive
047        * this allows numbers like 1AB and 1ab to represent the same  value
048        * [e.g. in hex]
049        */
050       ch = (radix <= 36) ? (char)toupper((int)*str) : *str;
051       for (y = 0; y < 64; y++) \{
052         if (ch == mp_s_rmap[y]) \{
053            break;
054         \}
055       \}
056
057       /* if the char was found in the map
058        * and is less than the given radix add it
059        * to the number, otherwise exit the loop.
060        */
061       if (y < radix) \{
062         if ((res = mp_mul_d (a, (mp_digit) radix, a)) != MP_OKAY) \{
063            return res;
064         \}
065         if ((res = mp_add_d (a, (mp_digit) y, a)) != MP_OKAY) \{
066            return res;
067         \}
068       \} else \{
069         break;
070       \}
071       ++str;
072     \}
073
074     /* set the sign only if a != 0 */
075     if (mp_iszero(a) != MP_YES) \{
076        a->sign = neg;
077     \}
078     return MP_OKAY;
079   \}
080   #endif
081
\end{alltt}
\end{small}

\subsection{Generating Radix-$n$ Output}
Generating radix-$n$ output is fairly trivial with a division and remainder algorithm.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_toradix}. \\
\textbf{Input}.   A mp\_int $a$ and an integer $r$\\
\textbf{Output}.  The radix-$r$ representation of $a$ \\
\hline \\
1.  If $r < 2$ or $r > 64$ return(\textit{MP\_VAL}). \\
2.  If $a = 0$ then $str = $ ``$0$'' and return(\textit{MP\_OKAY}).  \\
3.  $t \leftarrow a$ \\
4.  $str \leftarrow$ ``'' \\
5.  if $t.sign = MP\_NEG$ then \\
\hspace{3mm}5.1  $str \leftarrow str + $ ``-'' \\
\hspace{3mm}5.2  $t.sign = MP\_ZPOS$ \\
6.  While ($t \ne 0$) do \\
\hspace{3mm}6.1  $d \leftarrow t \mbox{ (mod }r\mbox{)}$ \\
\hspace{3mm}6.2  $t \leftarrow \lfloor t / r \rfloor$ \\
\hspace{3mm}6.3  Look up $d$ in the map and store the equivalent character in $y$. \\
\hspace{3mm}6.4  $str \leftarrow str + y$ \\
7.  If $str_0 = $``$-$'' then \\
\hspace{3mm}7.1  Reverse the digits $str_1, str_2, \ldots str_n$. \\
8.  Otherwise \\
\hspace{3mm}8.1  Reverse the digits $str_0, str_1, \ldots str_n$. \\
9.  Return(\textit{MP\_OKAY}).\\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_toradix}
\end{figure}
\textbf{Algorithm mp\_toradix.}
This algorithm computes the radix-$r$ representation of an mp\_int $a$.  The ``digits'' of the representation are extracted by reducing
successive powers of $\lfloor a / r^k \rfloor$ the input modulo $r$ until $r^k > a$.  Note that instead of actually dividing by $r^k$ in
each iteration the quotient $\lfloor a / r \rfloor$ is saved for the next iteration.  As a result a series of trivial $n \times 1$ divisions
are required instead of a series of $n \times k$ divisions.  One design flaw of this approach is that the digits are produced in the reverse order
(see~\ref{fig:mpradix}).  To remedy this flaw the digits must be swapped or simply ``reversed''.

\begin{figure}
\begin{center}
\begin{tabular}{|c|c|c|}
\hline \textbf{Value of $a$} & \textbf{Value of $d$} & \textbf{Value of $str$} \\
\hline $1234$ & -- & -- \\
\hline $123$  & $4$ & ``4'' \\
\hline $12$   & $3$ & ``43'' \\
\hline $1$    & $2$ & ``432'' \\
\hline $0$    & $1$ & ``4321'' \\
\hline
\end{tabular}
\end{center}
\caption{Example of Algorithm mp\_toradix.}
\label{fig:mpradix}
\end{figure}

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_toradix.c
\vspace{-3mm}
\begin{alltt}
016
017   /* stores a bignum as a ASCII string in a given radix (2..64) */
018   int mp_toradix (mp_int * a, char *str, int radix)
019   \{
020     int     res, digs;
021     mp_int  t;
022     mp_digit d;
023     char   *_s = str;
024
025     /* check range of the radix */
026     if ((radix < 2) || (radix > 64)) \{
027       return MP_VAL;
028     \}
029
030     /* quick out if its zero */
031     if (mp_iszero(a) == MP_YES) \{
032        *str++ = '0';
033        *str = '\symbol{92}0';
034        return MP_OKAY;
035     \}
036
037     if ((res = mp_init_copy (&t, a)) != MP_OKAY) \{
038       return res;
039     \}
040
041     /* if it is negative output a - */
042     if (t.sign == MP_NEG) \{
043       ++_s;
044       *str++ = '-';
045       t.sign = MP_ZPOS;
046     \}
047
048     digs = 0;
049     while (mp_iszero (&t) == MP_NO) \{
050       if ((res = mp_div_d (&t, (mp_digit) radix, &t, &d)) != MP_OKAY) \{
051         mp_clear (&t);
052         return res;
053       \}
054       *str++ = mp_s_rmap[d];
055       ++digs;
056     \}
057
058     /* reverse the digits of the string.  In this case _s points
059      * to the first digit [exluding the sign] of the number]
060      */
061     bn_reverse ((unsigned char *)_s, digs);
062
063     /* append a NULL so the string is properly terminated */
064     *str = '\symbol{92}0';
065
066     mp_clear (&t);
067     return MP_OKAY;
068   \}
069
070   #endif
071
\end{alltt}
\end{small}

\chapter{Number Theoretic Algorithms}
This chapter discusses several fundamental number theoretic algorithms such as the greatest common divisor, least common multiple and Jacobi
symbol computation.  These algorithms arise as essential components in several key cryptographic algorithms such as the RSA public key algorithm and
various Sieve based factoring algorithms.

\section{Greatest Common Divisor}
The greatest common divisor of two integers $a$ and $b$, often denoted as $(a, b)$ is the largest integer $k$ that is a proper divisor of
both $a$ and $b$.  That is, $k$ is the largest integer such that $0 \equiv a \mbox{ (mod }k\mbox{)}$ and $0 \equiv b \mbox{ (mod }k\mbox{)}$ occur
simultaneously.

The most common approach (cite) is to reduce one input modulo another.  That is if $a$ and $b$ are divisible by some integer $k$ and if $qa + r = b$ then
$r$ is also divisible by $k$.  The reduction pattern follows $\left < a , b \right > \rightarrow \left < b, a \mbox{ mod } b \right >$.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Greatest Common Divisor (I)}. \\
\textbf{Input}.   Two positive integers $a$ and $b$ greater than zero. \\
\textbf{Output}.  The greatest common divisor $(a, b)$.  \\
\hline \\
1.  While ($b > 0$) do \\
\hspace{3mm}1.1  $r \leftarrow a \mbox{ (mod }b\mbox{)}$ \\
\hspace{3mm}1.2  $a \leftarrow b$ \\
\hspace{3mm}1.3  $b \leftarrow r$ \\
2.  Return($a$). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Greatest Common Divisor (I)}
\label{fig:gcd1}
\end{figure}

This algorithm will quickly converge on the greatest common divisor since the residue $r$ tends diminish rapidly.  However, divisions are
relatively expensive operations to perform and should ideally be avoided.  There is another approach based on a similar relationship of
greatest common divisors.  The faster approach is based on the observation that if $k$ divides both $a$ and $b$ it will also divide $a - b$.
In particular, we would like $a - b$ to decrease in magnitude which implies that $b \ge a$.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Greatest Common Divisor (II)}. \\
\textbf{Input}.   Two positive integers $a$ and $b$ greater than zero. \\
\textbf{Output}.  The greatest common divisor $(a, b)$.  \\
\hline \\
1.  While ($b > 0$) do \\
\hspace{3mm}1.1  Swap $a$ and $b$ such that $a$ is the smallest of the two. \\
\hspace{3mm}1.2  $b \leftarrow b - a$ \\
2.  Return($a$). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Greatest Common Divisor (II)}
\label{fig:gcd2}
\end{figure}

\textbf{Proof} \textit{Algorithm~\ref{fig:gcd2} will return the greatest common divisor of $a$ and $b$.}
The algorithm in figure~\ref{fig:gcd2} will eventually terminate since $b \ge a$ the subtraction in step 1.2 will be a value less than $b$.  In other
words in every iteration that tuple $\left < a, b \right >$ decrease in magnitude until eventually $a = b$.  Since both $a$ and $b$ are always
divisible by the greatest common divisor (\textit{until the last iteration}) and in the last iteration of the algorithm $b = 0$, therefore, in the
second to last iteration of the algorithm $b = a$ and clearly $(a, a) = a$ which concludes the proof.  \textbf{QED}.

As a matter of practicality algorithm \ref{fig:gcd1} decreases far too slowly to be useful.  Specially if $b$ is much larger than $a$ such that
$b - a$ is still very much larger than $a$.  A simple addition to the algorithm is to divide $b - a$ by a power of some integer $p$ which does
not divide the greatest common divisor but will divide $b - a$.  In this case ${b - a} \over p$ is also an integer and still divisible by
the greatest common divisor.

However, instead of factoring $b - a$ to find a suitable value of $p$ the powers of $p$ can be removed from $a$ and $b$ that are in common first.
Then inside the loop whenever $b - a$ is divisible by some power of $p$ it can be safely removed.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{Greatest Common Divisor (III)}. \\
\textbf{Input}.   Two positive integers $a$ and $b$ greater than zero. \\
\textbf{Output}.  The greatest common divisor $(a, b)$.  \\
\hline \\
1.  $k \leftarrow 0$ \\
2.  While $a$ and $b$ are both divisible by $p$ do \\
\hspace{3mm}2.1  $a \leftarrow \lfloor a / p \rfloor$ \\
\hspace{3mm}2.2  $b \leftarrow \lfloor b / p \rfloor$ \\
\hspace{3mm}2.3  $k \leftarrow k + 1$ \\
3.  While $a$ is divisible by $p$ do \\
\hspace{3mm}3.1  $a \leftarrow \lfloor a / p \rfloor$ \\
4.  While $b$ is divisible by $p$ do \\
\hspace{3mm}4.1  $b \leftarrow \lfloor b / p \rfloor$ \\
5.  While ($b > 0$) do \\
\hspace{3mm}5.1  Swap $a$ and $b$ such that $a$ is the smallest of the two. \\
\hspace{3mm}5.2  $b \leftarrow b - a$ \\
\hspace{3mm}5.3  While $b$ is divisible by $p$ do \\
\hspace{6mm}5.3.1  $b \leftarrow \lfloor b / p \rfloor$ \\
6.  Return($a \cdot p^k$). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm Greatest Common Divisor (III)}
\label{fig:gcd3}
\end{figure}

This algorithm is based on the first except it removes powers of $p$ first and inside the main loop to ensure the tuple $\left < a, b \right >$
decreases more rapidly.  The first loop on step two removes powers of $p$ that are in common.  A count, $k$, is kept which will present a common
divisor of $p^k$.  After step two the remaining common divisor of $a$ and $b$ cannot be divisible by $p$.  This means that $p$ can be safely
divided out of the difference $b - a$ so long as the division leaves no remainder.

In particular the value of $p$ should be chosen such that the division on step 5.3.1 occur often.  It also helps that division by $p$ be easy
to compute.  The ideal choice of $p$ is two since division by two amounts to a right logical shift.  Another important observation is that by
step five both $a$ and $b$ are odd.  Therefore, the diffrence $b - a$ must be even which means that each iteration removes one bit from the
largest of the pair.

\subsection{Complete Greatest Common Divisor}
The algorithms presented so far cannot handle inputs which are zero or negative.  The following algorithm can handle all input cases properly
and will produce the greatest common divisor.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_gcd}. \\
\textbf{Input}.   mp\_int $a$ and $b$ \\
\textbf{Output}.  The greatest common divisor $c = (a, b)$.  \\
\hline \\
1.  If $a = 0$ then \\
\hspace{3mm}1.1  $c \leftarrow \vert b \vert $ \\
\hspace{3mm}1.2  Return(\textit{MP\_OKAY}). \\
2.  If $b = 0$ then \\
\hspace{3mm}2.1  $c \leftarrow \vert a \vert $ \\
\hspace{3mm}2.2  Return(\textit{MP\_OKAY}). \\
3.  $u \leftarrow \vert a \vert, v \leftarrow \vert b \vert$ \\
4.  $k \leftarrow 0$ \\
5.  While $u.used > 0$ and $v.used > 0$ and $u_0 \equiv v_0 \equiv 0 \mbox{ (mod }2\mbox{)}$ \\
\hspace{3mm}5.1  $k \leftarrow k + 1$ \\
\hspace{3mm}5.2  $u \leftarrow \lfloor u / 2 \rfloor$ \\
\hspace{3mm}5.3  $v \leftarrow \lfloor v / 2 \rfloor$ \\
6.  While $u.used > 0$ and $u_0 \equiv 0 \mbox{ (mod }2\mbox{)}$ \\
\hspace{3mm}6.1  $u \leftarrow \lfloor u / 2 \rfloor$ \\
7.  While $v.used > 0$ and $v_0 \equiv 0 \mbox{ (mod }2\mbox{)}$ \\
\hspace{3mm}7.1  $v \leftarrow \lfloor v / 2 \rfloor$ \\
8.  While $v.used > 0$ \\
\hspace{3mm}8.1  If $\vert u \vert > \vert v \vert$ then \\
\hspace{6mm}8.1.1  Swap $u$ and $v$. \\
\hspace{3mm}8.2  $v \leftarrow \vert v \vert - \vert u \vert$ \\
\hspace{3mm}8.3  While $v.used > 0$ and $v_0 \equiv 0 \mbox{ (mod }2\mbox{)}$ \\
\hspace{6mm}8.3.1  $v \leftarrow \lfloor v / 2 \rfloor$ \\
9.  $c \leftarrow u \cdot 2^k$ \\
10.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_gcd}
\end{figure}
\textbf{Algorithm mp\_gcd.}
This algorithm will produce the greatest common divisor of two mp\_ints $a$ and $b$.  The algorithm was originally based on Algorithm B of
Knuth \cite[pp. 338]{TAOCPV2} but has been modified to be simpler to explain.  In theory it achieves the same asymptotic working time as
Algorithm B and in practice this appears to be true.

The first two steps handle the cases where either one of or both inputs are zero.  If either input is zero the greatest common divisor is the
largest input or zero if they are both zero.  If the inputs are not trivial than $u$ and $v$ are assigned the absolute values of
$a$ and $b$ respectively and the algorithm will proceed to reduce the pair.

Step five will divide out any common factors of two and keep track of the count in the variable $k$.  After this step, two is no longer a
factor of the remaining greatest common divisor between $u$ and $v$ and can be safely evenly divided out of either whenever they are even.  Step
six and seven ensure that the $u$ and $v$ respectively have no more factors of two.  At most only one of the while--loops will iterate since
they cannot both be even.

By step eight both of $u$ and $v$ are odd which is required for the inner logic.  First the pair are swapped such that $v$ is equal to
or greater than $u$.  This ensures that the subtraction on step 8.2 will always produce a positive and even result.  Step 8.3 removes any
factors of two from the difference $u$ to ensure that in the next iteration of the loop both are once again odd.

After $v = 0$ occurs the variable $u$ has the greatest common divisor of the pair $\left < u, v \right >$ just after step six.  The result
must be adjusted by multiplying by the common factors of two ($2^k$) removed earlier.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_gcd.c
\vspace{-3mm}
\begin{alltt}
016
017   /* Greatest Common Divisor using the binary method */
018   int mp_gcd (mp_int * a, mp_int * b, mp_int * c)
019   \{
020     mp_int  u, v;
021     int     k, u_lsb, v_lsb, res;
022
023     /* either zero than gcd is the largest */
024     if (mp_iszero (a) == MP_YES) \{
025       return mp_abs (b, c);
026     \}
027     if (mp_iszero (b) == MP_YES) \{
028       return mp_abs (a, c);
029     \}
030
031     /* get copies of a and b we can modify */
032     if ((res = mp_init_copy (&u, a)) != MP_OKAY) \{
033       return res;
034     \}
035
036     if ((res = mp_init_copy (&v, b)) != MP_OKAY) \{
037       goto LBL_U;
038     \}
039
040     /* must be positive for the remainder of the algorithm */
041     u.sign = v.sign = MP_ZPOS;
042
043     /* B1.  Find the common power of two for u and v */
044     u_lsb = mp_cnt_lsb(&u);
045     v_lsb = mp_cnt_lsb(&v);
046     k     = MIN(u_lsb, v_lsb);
047
048     if (k > 0) \{
049        /* divide the power of two out */
050        if ((res = mp_div_2d(&u, k, &u, NULL)) != MP_OKAY) \{
051           goto LBL_V;
052        \}
053
054        if ((res = mp_div_2d(&v, k, &v, NULL)) != MP_OKAY) \{
055           goto LBL_V;
056        \}
057     \}
058
059     /* divide any remaining factors of two out */
060     if (u_lsb != k) \{
061        if ((res = mp_div_2d(&u, u_lsb - k, &u, NULL)) != MP_OKAY) \{
062           goto LBL_V;
063        \}
064     \}
065
066     if (v_lsb != k) \{
067        if ((res = mp_div_2d(&v, v_lsb - k, &v, NULL)) != MP_OKAY) \{
068           goto LBL_V;
069        \}
070     \}
071
072     while (mp_iszero(&v) == MP_NO) \{
073        /* make sure v is the largest */
074        if (mp_cmp_mag(&u, &v) == MP_GT) \{
075           /* swap u and v to make sure v is >= u */
076           mp_exch(&u, &v);
077        \}
078
079        /* subtract smallest from largest */
080        if ((res = s_mp_sub(&v, &u, &v)) != MP_OKAY) \{
081           goto LBL_V;
082        \}
083
084        /* Divide out all factors of two */
085        if ((res = mp_div_2d(&v, mp_cnt_lsb(&v), &v, NULL)) != MP_OKAY) \{
086           goto LBL_V;
087        \}
088     \}
089
090     /* multiply by 2**k which we divided out at the beginning */
091     if ((res = mp_mul_2d (&u, k, c)) != MP_OKAY) \{
092        goto LBL_V;
093     \}
094     c->sign = MP_ZPOS;
095     res = MP_OKAY;
096   LBL_V:mp_clear (&u);
097   LBL_U:mp_clear (&v);
098     return res;
099   \}
100   #endif
101
\end{alltt}
\end{small}

This function makes use of the macros mp\_iszero and mp\_iseven.  The former evaluates to $1$ if the input mp\_int is equivalent to the
integer zero otherwise it evaluates to $0$.  The latter evaluates to $1$ if the input mp\_int represents a non-zero even integer otherwise
it evaluates to $0$.  Note that just because mp\_iseven may evaluate to $0$ does not mean the input is odd, it could also be zero.  The three
trivial cases of inputs are handled on lines 23 through 29.  After those lines the inputs are assumed to be non-zero.

Lines 32 and 36 make local copies $u$ and $v$ of the inputs $a$ and $b$ respectively.  At this point the common factors of two
must be divided out of the two inputs.  The block starting at line 43 removes common factors of two by first counting the number of trailing
zero bits in both.  The local integer $k$ is used to keep track of how many factors of $2$ are pulled out of both values.  It is assumed that
the number of factors will not exceed the maximum value of a C ``int'' data type\footnote{Strictly speaking no array in C may have more than
entries than are accessible by an ``int'' so this is not a limitation.}.

At this point there are no more common factors of two in the two values.  The divisions by a power of two on lines 61 and 67 remove
any independent factors of two such that both $u$ and $v$ are guaranteed to be an odd integer before hitting the main body of the algorithm.  The while loop
on line 72 performs the reduction of the pair until $v$ is equal to zero.  The unsigned comparison and subtraction algorithms are used in
place of the full signed routines since both values are guaranteed to be positive and the result of the subtraction is guaranteed to be non-negative.

\section{Least Common Multiple}
The least common multiple of a pair of integers is their product divided by their greatest common divisor.  For two integers $a$ and $b$ the
least common multiple is normally denoted as $[ a, b ]$ and numerically equivalent to ${ab} \over {(a, b)}$.  For example, if $a = 2 \cdot 2 \cdot 3 = 12$
and $b = 2 \cdot 3 \cdot 3 \cdot 7 = 126$ the least common multiple is ${126 \over {(12, 126)}} = {126 \over 6} = 21$.

The least common multiple arises often in coding theory as well as number theory.  If two functions have periods of $a$ and $b$ respectively they will
collide, that is be in synchronous states, after only $[ a, b ]$ iterations.  This is why, for example, random number generators based on
Linear Feedback Shift Registers (LFSR) tend to use registers with periods which are co-prime (\textit{e.g. the greatest common divisor is one.}).
Similarly in number theory if a composite $n$ has two prime factors $p$ and $q$ then maximal order of any unit of $\Z/n\Z$ will be $[ p - 1, q - 1] $.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_lcm}. \\
\textbf{Input}.   mp\_int $a$ and $b$ \\
\textbf{Output}.  The least common multiple $c = [a, b]$.  \\
\hline \\
1.  $c \leftarrow (a, b)$ \\
2.  $t \leftarrow a \cdot b$ \\
3.  $c \leftarrow \lfloor t / c \rfloor$ \\
4.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_lcm}
\end{figure}
\textbf{Algorithm mp\_lcm.}
This algorithm computes the least common multiple of two mp\_int inputs $a$ and $b$.  It computes the least common multiple directly by
dividing the product of the two inputs by their greatest common divisor.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_lcm.c
\vspace{-3mm}
\begin{alltt}
016
017   /* computes least common multiple as |a*b|/(a, b) */
018   int mp_lcm (mp_int * a, mp_int * b, mp_int * c)
019   \{
020     int     res;
021     mp_int  t1, t2;
022
023
024     if ((res = mp_init_multi (&t1, &t2, NULL)) != MP_OKAY) \{
025       return res;
026     \}
027
028     /* t1 = get the GCD of the two inputs */
029     if ((res = mp_gcd (a, b, &t1)) != MP_OKAY) \{
030       goto LBL_T;
031     \}
032
033     /* divide the smallest by the GCD */
034     if (mp_cmp_mag(a, b) == MP_LT) \{
035        /* store quotient in t2 such that t2 * b is the LCM */
036        if ((res = mp_div(a, &t1, &t2, NULL)) != MP_OKAY) \{
037           goto LBL_T;
038        \}
039        res = mp_mul(b, &t2, c);
040     \} else \{
041        /* store quotient in t2 such that t2 * a is the LCM */
042        if ((res = mp_div(b, &t1, &t2, NULL)) != MP_OKAY) \{
043           goto LBL_T;
044        \}
045        res = mp_mul(a, &t2, c);
046     \}
047
048     /* fix the sign to positive */
049     c->sign = MP_ZPOS;
050
051   LBL_T:
052     mp_clear_multi (&t1, &t2, NULL);
053     return res;
054   \}
055   #endif
056
\end{alltt}
\end{small}

\section{Jacobi Symbol Computation}
To explain the Jacobi Symbol we shall first discuss the Legendre function\footnote{Arrg.  What is the name of this?} off which the Jacobi symbol is
defined.  The Legendre function computes whether or not an integer $a$ is a quadratic residue modulo an odd prime $p$.  Numerically it is
equivalent to equation \ref{eqn:legendre}.

\textit{-- Tom, don't be an ass, cite your source here...!}

\begin{equation}
a^{(p-1)/2} \equiv \begin{array}{rl}
                              -1 &  \mbox{if }a\mbox{ is a quadratic non-residue.} \\
                              0  &  \mbox{if }a\mbox{ divides }p\mbox{.} \\
                              1  &  \mbox{if }a\mbox{ is a quadratic residue}.
                              \end{array} \mbox{ (mod }p\mbox{)}
\label{eqn:legendre}
\end{equation}

\textbf{Proof.} \textit{Equation \ref{eqn:legendre} correctly identifies the residue status of an integer $a$ modulo a prime $p$.}
An integer $a$ is a quadratic residue if the following equation has a solution.

\begin{equation}
x^2 \equiv a \mbox{ (mod }p\mbox{)}
\label{eqn:root}
\end{equation}

Consider the following equation.

\begin{equation}
0 \equiv x^{p-1} - 1 \equiv \left \lbrace \left (x^2 \right )^{(p-1)/2} - a^{(p-1)/2} \right \rbrace + \left ( a^{(p-1)/2} - 1 \right ) \mbox{ (mod }p\mbox{)}
\label{eqn:rooti}
\end{equation}

Whether equation \ref{eqn:root} has a solution or not equation \ref{eqn:rooti} is always true.  If $a^{(p-1)/2} - 1 \equiv 0 \mbox{ (mod }p\mbox{)}$
then the quantity in the braces must be zero.  By reduction,

\begin{eqnarray}
\left (x^2 \right )^{(p-1)/2} - a^{(p-1)/2} \equiv 0  \nonumber \\
\left (x^2 \right )^{(p-1)/2} \equiv a^{(p-1)/2} \nonumber \\
x^2 \equiv a \mbox{ (mod }p\mbox{)}
\end{eqnarray}

As a result there must be a solution to the quadratic equation and in turn $a$ must be a quadratic residue.  If $a$ does not divide $p$ and $a$
is not a quadratic residue then the only other value $a^{(p-1)/2}$ may be congruent to is $-1$ since
\begin{equation}
0 \equiv a^{p - 1} - 1 \equiv (a^{(p-1)/2} + 1)(a^{(p-1)/2} - 1) \mbox{ (mod }p\mbox{)}
\end{equation}
One of the terms on the right hand side must be zero.  \textbf{QED}

\subsection{Jacobi Symbol}
The Jacobi symbol is a generalization of the Legendre function for any odd non prime moduli $p$ greater than 2.  If $p = \prod_{i=0}^n p_i$ then
the Jacobi symbol $\left ( { a \over p } \right )$ is equal to the following equation.

\begin{equation}
\left ( { a \over p } \right ) = \left ( { a \over p_0} \right ) \left ( { a \over p_1} \right ) \ldots \left ( { a \over p_n} \right )
\end{equation}

By inspection if $p$ is prime the Jacobi symbol is equivalent to the Legendre function.  The following facts\footnote{See HAC \cite[pp. 72-74]{HAC} for
further details.} will be used to derive an efficient Jacobi symbol algorithm.  Where $p$ is an odd integer greater than two and $a, b \in \Z$ the
following are true.

\begin{enumerate}
\item $\left ( { a \over p} \right )$ equals $-1$, $0$ or $1$.
\item $\left ( { ab \over p} \right ) = \left ( { a \over p} \right )\left ( { b \over p} \right )$.
\item If $a \equiv b$ then $\left ( { a \over p} \right ) = \left ( { b \over p} \right )$.
\item $\left ( { 2 \over p} \right )$ equals $1$ if $p \equiv 1$ or $7 \mbox{ (mod }8\mbox{)}$.  Otherwise, it equals $-1$.
\item $\left ( { a \over p} \right ) \equiv \left ( { p \over a} \right ) \cdot (-1)^{(p-1)(a-1)/4}$.  More specifically
$\left ( { a \over p} \right ) = \left ( { p \over a} \right )$ if $p \equiv a \equiv 1 \mbox{ (mod }4\mbox{)}$.
\end{enumerate}

Using these facts if $a = 2^k \cdot a'$ then

\begin{eqnarray}
\left ( { a \over p } \right ) = \left ( {{2^k} \over p } \right ) \left ( {a' \over p} \right ) \nonumber \\
                               = \left ( {2 \over p } \right )^k \left ( {a' \over p} \right )
\label{eqn:jacobi}
\end{eqnarray}

By fact five,

\begin{equation}
\left ( { a \over p } \right ) = \left ( { p \over a } \right ) \cdot (-1)^{(p-1)(a-1)/4}
\end{equation}

Subsequently by fact three since $p \equiv (p \mbox{ mod }a) \mbox{ (mod }a\mbox{)}$ then

\begin{equation}
\left ( { a \over p } \right ) = \left ( { {p \mbox{ mod } a} \over a } \right ) \cdot (-1)^{(p-1)(a-1)/4}
\end{equation}

By putting both observations into equation \ref{eqn:jacobi} the following simplified equation is formed.

\begin{equation}
\left ( { a \over p } \right ) = \left ( {2 \over p } \right )^k \left ( {{p\mbox{ mod }a'} \over a'} \right )  \cdot (-1)^{(p-1)(a'-1)/4}
\end{equation}

The value of $\left ( {{p \mbox{ mod }a'} \over a'} \right )$ can be found by using the same equation recursively.  The value of
$\left ( {2 \over p } \right )^k$ equals $1$ if $k$ is even otherwise it equals $\left ( {2 \over p } \right )$.  Using this approach the
factors of $p$ do not have to be known.  Furthermore, if $(a, p) = 1$ then the algorithm will terminate when the recursion requests the
Jacobi symbol computation of $\left ( {1 \over a'} \right )$ which is simply $1$.

\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_jacobi}. \\
\textbf{Input}.   mp\_int $a$ and $p$, $a \ge 0$, $p \ge 3$, $p \equiv 1 \mbox{ (mod }2\mbox{)}$ \\
\textbf{Output}.  The Jacobi symbol $c = \left ( {a \over p } \right )$. \\
\hline \\
1.  If $a = 0$ then \\
\hspace{3mm}1.1  $c \leftarrow 0$ \\
\hspace{3mm}1.2  Return(\textit{MP\_OKAY}). \\
2.  If $a = 1$ then \\
\hspace{3mm}2.1  $c \leftarrow 1$ \\
\hspace{3mm}2.2  Return(\textit{MP\_OKAY}). \\
3.  $a' \leftarrow a$ \\
4.  $k \leftarrow 0$ \\
5.  While $a'.used > 0$ and $a'_0 \equiv 0 \mbox{ (mod }2\mbox{)}$ \\
\hspace{3mm}5.1  $k \leftarrow k + 1$ \\
\hspace{3mm}5.2  $a' \leftarrow \lfloor a' / 2 \rfloor$ \\
6.  If $k \equiv 0 \mbox{ (mod }2\mbox{)}$ then \\
\hspace{3mm}6.1  $s \leftarrow 1$ \\
7.  else \\
\hspace{3mm}7.1  $r \leftarrow p_0 \mbox{ (mod }8\mbox{)}$ \\
\hspace{3mm}7.2  If $r = 1$ or $r = 7$ then \\
\hspace{6mm}7.2.1  $s \leftarrow 1$ \\
\hspace{3mm}7.3  else \\
\hspace{6mm}7.3.1  $s \leftarrow -1$ \\
8.  If $p_0 \equiv a'_0 \equiv 3 \mbox{ (mod }4\mbox{)}$ then \\
\hspace{3mm}8.1  $s \leftarrow -s$ \\
9.  If $a' \ne 1$ then \\
\hspace{3mm}9.1  $p' \leftarrow p \mbox{ (mod }a'\mbox{)}$ \\
\hspace{3mm}9.2  $s \leftarrow s \cdot \mbox{mp\_jacobi}(p', a')$ \\
10.  $c \leftarrow s$ \\
11.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_jacobi}
\end{figure}
\textbf{Algorithm mp\_jacobi.}
This algorithm computes the Jacobi symbol for an arbitrary positive integer $a$ with respect to an odd integer $p$ greater than three.  The algorithm
is based on algorithm 2.149 of HAC \cite[pp. 73]{HAC}.

Step numbers one and two handle the trivial cases of $a = 0$ and $a = 1$ respectively.  Step five determines the number of two factors in the
input $a$.  If $k$ is even than the term $\left ( { 2 \over p } \right )^k$ must always evaluate to one.  If $k$ is odd than the term evaluates to one
if $p_0$ is congruent to one or seven modulo eight, otherwise it evaluates to $-1$. After the the $\left ( { 2 \over p } \right )^k$ term is handled
the $(-1)^{(p-1)(a'-1)/4}$ is computed and multiplied against the current product $s$.  The latter term evaluates to one if both $p$ and $a'$
are congruent to one modulo four, otherwise it evaluates to negative one.

By step nine if $a'$ does not equal one a recursion is required.  Step 9.1 computes $p' \equiv p \mbox{ (mod }a'\mbox{)}$ and will recurse to compute
$\left ( {p' \over a'} \right )$ which is multiplied against the current Jacobi product.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_jacobi.c
\vspace{-3mm}
\begin{alltt}
016
017   /* computes the jacobi c = (a | n) (or Legendre if n is prime)
018    * HAC pp. 73 Algorithm 2.149
019    * HAC is wrong here, as the special case of (0 | 1) is not
020    * handled correctly.
021    */
022   int mp_jacobi (mp_int * a, mp_int * n, int *c)
023   \{
024     mp_int  a1, p1;
025     int     k, s, r, res;
026     mp_digit residue;
027
028     /* if a < 0 return MP_VAL */
029     if (mp_isneg(a) == MP_YES) \{
030        return MP_VAL;
031     \}
032
033     /* if n <= 0 return MP_VAL */
034     if (mp_cmp_d(n, 0) != MP_GT) \{
035        return MP_VAL;
036     \}
037
038     /* step 1. handle case of a == 0 */
039     if (mp_iszero (a) == MP_YES) \{
040        /* special case of a == 0 and n == 1 */
041        if (mp_cmp_d (n, 1) == MP_EQ) \{
042          *c = 1;
043        \} else \{
044          *c = 0;
045        \}
046        return MP_OKAY;
047     \}
048
049     /* step 2.  if a == 1, return 1 */
050     if (mp_cmp_d (a, 1) == MP_EQ) \{
051       *c = 1;
052       return MP_OKAY;
053     \}
054
055     /* default */
056     s = 0;
057
058     /* step 3.  write a = a1 * 2**k  */
059     if ((res = mp_init_copy (&a1, a)) != MP_OKAY) \{
060       return res;
061     \}
062
063     if ((res = mp_init (&p1)) != MP_OKAY) \{
064       goto LBL_A1;
065     \}
066
067     /* divide out larger power of two */
068     k = mp_cnt_lsb(&a1);
069     if ((res = mp_div_2d(&a1, k, &a1, NULL)) != MP_OKAY) \{
070        goto LBL_P1;
071     \}
072
073     /* step 4.  if e is even set s=1 */
074     if ((k & 1) == 0) \{
075       s = 1;
076     \} else \{
077       /* else set s=1 if p = 1/7 (mod 8) or s=-1 if p = 3/5 (mod 8) */
078       residue = n->dp[0] & 7;
079
080       if ((residue == 1) || (residue == 7)) \{
081         s = 1;
082       \} else if ((residue == 3) || (residue == 5)) \{
083         s = -1;
084       \}
085     \}
086
087     /* step 5.  if p == 3 (mod 4) *and* a1 == 3 (mod 4) then s = -s */
088     if ( ((n->dp[0] & 3) == 3) && ((a1.dp[0] & 3) == 3)) \{
089       s = -s;
090     \}
091
092     /* if a1 == 1 we're done */
093     if (mp_cmp_d (&a1, 1) == MP_EQ) \{
094       *c = s;
095     \} else \{
096       /* n1 = n mod a1 */
097       if ((res = mp_mod (n, &a1, &p1)) != MP_OKAY) \{
098         goto LBL_P1;
099       \}
100       if ((res = mp_jacobi (&p1, &a1, &r)) != MP_OKAY) \{
101         goto LBL_P1;
102       \}
103       *c = s * r;
104     \}
105
106     /* done */
107     res = MP_OKAY;
108   LBL_P1:mp_clear (&p1);
109   LBL_A1:mp_clear (&a1);
110     return res;
111   \}
112   #endif
113
\end{alltt}
\end{small}

As a matter of practicality the variable $a'$ as per the pseudo-code is reprensented by the variable $a1$ since the $'$ symbol is not valid for a C
variable name character.

The two simple cases of $a = 0$ and $a = 1$ are handled at the very beginning to simplify the algorithm.  If the input is non-trivial the algorithm
has to proceed compute the Jacobi.  The variable $s$ is used to hold the current Jacobi product.  Note that $s$ is merely a C ``int'' data type since
the values it may obtain are merely $-1$, $0$ and $1$.

After a local copy of $a$ is made all of the factors of two are divided out and the total stored in $k$.  Technically only the least significant
bit of $k$ is required, however, it makes the algorithm simpler to follow to perform an addition. In practice an exclusive-or and addition have the same
processor requirements and neither is faster than the other.

Line 59 through 71 determines the value of $\left ( { 2 \over p } \right )^k$.  If the least significant bit of $k$ is zero than
$k$ is even and the value is one.  Otherwise, the value of $s$ depends on which residue class $p$ belongs to modulo eight.  The value of
$(-1)^{(p-1)(a'-1)/4}$ is compute and multiplied against $s$ on lines 73 through 76.

Finally, if $a1$ does not equal one the algorithm must recurse and compute $\left ( {p' \over a'} \right )$.

\textit{-- Comment about default $s$ and such...}

\section{Modular Inverse}
\label{sec:modinv}
The modular inverse of a number actually refers to the modular multiplicative inverse.  Essentially for any integer $a$ such that $(a, p) = 1$ there
exist another integer $b$ such that $ab \equiv 1 \mbox{ (mod }p\mbox{)}$.  The integer $b$ is called the multiplicative inverse of $a$ which is
denoted as $b = a^{-1}$.  Technically speaking modular inversion is a well defined operation for any finite ring or field not just for rings and
fields of integers.  However, the former will be the matter of discussion.

The simplest approach is to compute the algebraic inverse of the input.  That is to compute $b \equiv a^{\Phi(p) - 1}$.  If $\Phi(p)$ is the
order of the multiplicative subgroup modulo $p$ then $b$ must be the multiplicative inverse of $a$.  The proof of which is trivial.

\begin{equation}
ab \equiv a \left (a^{\Phi(p) - 1} \right ) \equiv a^{\Phi(p)} \equiv a^0 \equiv 1 \mbox{ (mod }p\mbox{)}
\end{equation}

However, as simple as this approach may be it has two serious flaws.  It requires that the value of $\Phi(p)$ be known which if $p$ is composite
requires all of the prime factors.  This approach also is very slow as the size of $p$ grows.

A simpler approach is based on the observation that solving for the multiplicative inverse is equivalent to solving the linear
Diophantine\footnote{See LeVeque \cite[pp. 40-43]{LeVeque} for more information.} equation.

\begin{equation}
ab + pq = 1
\end{equation}

Where $a$, $b$, $p$ and $q$ are all integers.  If such a pair of integers $ \left < b, q \right >$ exist than $b$ is the multiplicative inverse of
$a$ modulo $p$.  The extended Euclidean algorithm (Knuth \cite[pp. 342]{TAOCPV2}) can be used to solve such equations provided $(a, p) = 1$.
However, instead of using that algorithm directly a variant known as the binary Extended Euclidean algorithm will be used in its place.  The
binary approach is very similar to the binary greatest common divisor algorithm except it will produce a full solution to the Diophantine
equation.

\subsection{General Case}
\newpage\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_invmod}. \\
\textbf{Input}.   mp\_int $a$ and $b$, $(a, b) = 1$, $p \ge 2$, $0 < a < p$.  \\
\textbf{Output}.  The modular inverse $c \equiv a^{-1} \mbox{ (mod }b\mbox{)}$. \\
\hline \\
1.  If $b \le 0$ then return(\textit{MP\_VAL}). \\
2.  If $b_0 \equiv 1 \mbox{ (mod }2\mbox{)}$ then use algorithm fast\_mp\_invmod. \\
3.  $x \leftarrow \vert a \vert, y \leftarrow b$ \\
4.  If $x_0 \equiv y_0  \equiv 0 \mbox{ (mod }2\mbox{)}$ then return(\textit{MP\_VAL}). \\
5.  $B \leftarrow 0, C \leftarrow 0, A \leftarrow 1, D \leftarrow 1$ \\
6.  While $u.used > 0$ and $u_0 \equiv 0 \mbox{ (mod }2\mbox{)}$ \\
\hspace{3mm}6.1  $u \leftarrow \lfloor u / 2 \rfloor$ \\
\hspace{3mm}6.2  If ($A.used > 0$ and $A_0 \equiv 1 \mbox{ (mod }2\mbox{)}$) or ($B.used > 0$ and $B_0 \equiv 1 \mbox{ (mod }2\mbox{)}$) then \\
\hspace{6mm}6.2.1  $A \leftarrow A + y$ \\
\hspace{6mm}6.2.2  $B \leftarrow B - x$ \\
\hspace{3mm}6.3  $A \leftarrow \lfloor A / 2 \rfloor$ \\
\hspace{3mm}6.4  $B \leftarrow \lfloor B / 2 \rfloor$ \\
7.  While $v.used > 0$ and $v_0 \equiv 0 \mbox{ (mod }2\mbox{)}$ \\
\hspace{3mm}7.1  $v \leftarrow \lfloor v / 2 \rfloor$ \\
\hspace{3mm}7.2  If ($C.used > 0$ and $C_0 \equiv 1 \mbox{ (mod }2\mbox{)}$) or ($D.used > 0$ and $D_0 \equiv 1 \mbox{ (mod }2\mbox{)}$) then \\
\hspace{6mm}7.2.1  $C \leftarrow C + y$ \\
\hspace{6mm}7.2.2  $D \leftarrow D - x$ \\
\hspace{3mm}7.3  $C \leftarrow \lfloor C / 2 \rfloor$ \\
\hspace{3mm}7.4  $D \leftarrow \lfloor D / 2 \rfloor$ \\
8.  If $u \ge v$ then \\
\hspace{3mm}8.1  $u \leftarrow u - v$ \\
\hspace{3mm}8.2  $A \leftarrow A - C$ \\
\hspace{3mm}8.3  $B \leftarrow B - D$ \\
9.  else \\
\hspace{3mm}9.1  $v \leftarrow v - u$ \\
\hspace{3mm}9.2  $C \leftarrow C - A$ \\
\hspace{3mm}9.3  $D \leftarrow D - B$ \\
10.  If $u \ne 0$ goto step 6. \\
11.  If $v \ne 1$ return(\textit{MP\_VAL}). \\
12.  While $C \le 0$ do \\
\hspace{3mm}12.1  $C \leftarrow C + b$ \\
13.  While $C \ge b$ do \\
\hspace{3mm}13.1  $C \leftarrow C - b$ \\
14.  $c \leftarrow C$ \\
15.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\end{figure}
\textbf{Algorithm mp\_invmod.}
This algorithm computes the modular multiplicative inverse of an integer $a$ modulo an integer $b$.  This algorithm is a variation of the
extended binary Euclidean algorithm from HAC \cite[pp. 608]{HAC}.  It has been modified to only compute the modular inverse and not a complete
Diophantine solution.

If $b \le 0$ than the modulus is invalid and MP\_VAL is returned.  Similarly if both $a$ and $b$ are even then there cannot be a multiplicative
inverse for $a$ and the error is reported.

The astute reader will observe that steps seven through nine are very similar to the binary greatest common divisor algorithm mp\_gcd.  In this case
the other variables to the Diophantine equation are solved.  The algorithm terminates when $u = 0$ in which case the solution is

\begin{equation}
Ca + Db = v
\end{equation}

If $v$, the greatest common divisor of $a$ and $b$ is not equal to one then the algorithm will report an error as no inverse exists.  Otherwise, $C$
is the modular inverse of $a$.  The actual value of $C$ is congruent to, but not necessarily equal to, the ideal modular inverse which should lie
within $1 \le a^{-1} < b$.  Step numbers twelve and thirteen adjust the inverse until it is in range.  If the original input $a$ is within $0 < a < p$
then only a couple of additions or subtractions will be required to adjust the inverse.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_invmod.c
\vspace{-3mm}
\begin{alltt}
016
017   /* hac 14.61, pp608 */
018   int mp_invmod (mp_int * a, mp_int * b, mp_int * c)
019   \{
020     /* b cannot be negative */
021     if ((b->sign == MP_NEG) || (mp_iszero(b) == MP_YES)) \{
022       return MP_VAL;
023     \}
024
025   #ifdef BN_FAST_MP_INVMOD_C
026     /* if the modulus is odd we can use a faster routine instead */
027     if (mp_isodd (b) == MP_YES) \{
028       return fast_mp_invmod (a, b, c);
029     \}
030   #endif
031
032   #ifdef BN_MP_INVMOD_SLOW_C
033     return mp_invmod_slow(a, b, c);
034   #else
035     return MP_VAL;
036   #endif
037   \}
038   #endif
039
\end{alltt}
\end{small}

\subsubsection{Odd Moduli}

When the modulus $b$ is odd the variables $A$ and $C$ are fixed and are not required to compute the inverse.  In particular by attempting to solve
the Diophantine $Cb + Da = 1$ only $B$ and $D$ are required to find the inverse of $a$.

The algorithm fast\_mp\_invmod is a direct adaptation of algorithm mp\_invmod with all all steps involving either $A$ or $C$ removed.  This
optimization will halve the time required to compute the modular inverse.

\section{Primality Tests}

A non-zero integer $a$ is said to be prime if it is not divisible by any other integer excluding one and itself.  For example, $a = 7$ is prime
since the integers $2 \ldots 6$ do not evenly divide $a$.  By contrast, $a = 6$ is not prime since $a = 6 = 2 \cdot 3$.

Prime numbers arise in cryptography considerably as they allow finite fields to be formed.  The ability to determine whether an integer is prime or
not quickly has been a viable subject in cryptography and number theory for considerable time.  The algorithms that will be presented are all
probablistic algorithms in that when they report an integer is composite it must be composite.  However, when the algorithms report an integer is
prime the algorithm may be incorrect.

As will be discussed it is possible to limit the probability of error so well that for practical purposes the probablity of error might as
well be zero.  For the purposes of these discussions let $n$ represent the candidate integer of which the primality is in question.

\subsection{Trial Division}

Trial division means to attempt to evenly divide a candidate integer by small prime integers.  If the candidate can be evenly divided it obviously
cannot be prime.  By dividing by all primes $1 < p \le \sqrt{n}$ this test can actually prove whether an integer is prime.  However, such a test
would require a prohibitive amount of time as $n$ grows.

Instead of dividing by every prime, a smaller, more mangeable set of primes may be used instead.  By performing trial division with only a subset
of the primes less than $\sqrt{n} + 1$ the algorithm cannot prove if a candidate is prime.  However, often it can prove a candidate is not prime.

The benefit of this test is that trial division by small values is fairly efficient.  Specially compared to the other algorithms that will be
discussed shortly.  The probability that this approach correctly identifies a composite candidate when tested with all primes upto $q$ is given by
$1 - {1.12 \over ln(q)}$.  The graph (\ref{pic:primality}, will be added later) demonstrates the probability of success for the range
$3 \le q \le 100$.

At approximately $q = 30$ the gain of performing further tests diminishes fairly quickly.  At $q = 90$ further testing is generally not going to
be of any practical use.  In the case of LibTomMath the default limit $q = 256$ was chosen since it is not too high and will eliminate
approximately $80\%$ of all candidate integers.  The constant \textbf{PRIME\_SIZE} is equal to the number of primes in the test base.  The
array \_\_prime\_tab is an array of the first \textbf{PRIME\_SIZE} prime numbers.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_prime\_is\_divisible}. \\
\textbf{Input}.   mp\_int $a$ \\
\textbf{Output}.  $c = 1$ if $n$ is divisible by a small prime, otherwise $c = 0$.  \\
\hline \\
1.  for $ix$ from $0$ to $PRIME\_SIZE$ do \\
\hspace{3mm}1.1  $d \leftarrow n \mbox{ (mod }\_\_prime\_tab_{ix}\mbox{)}$ \\
\hspace{3mm}1.2  If $d = 0$ then \\
\hspace{6mm}1.2.1  $c \leftarrow 1$ \\
\hspace{6mm}1.2.2  Return(\textit{MP\_OKAY}). \\
2.  $c \leftarrow 0$ \\
3.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_prime\_is\_divisible}
\end{figure}
\textbf{Algorithm mp\_prime\_is\_divisible.}
This algorithm attempts to determine if a candidate integer $n$ is composite by performing trial divisions.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_prime\_is\_divisible.c
\vspace{-3mm}
\begin{alltt}
016
017   /* determines if an integers is divisible by one
018    * of the first PRIME_SIZE primes or not
019    *
020    * sets result to 0 if not, 1 if yes
021    */
022   int mp_prime_is_divisible (mp_int * a, int *result)
023   \{
024     int     err, ix;
025     mp_digit res;
026
027     /* default to not */
028     *result = MP_NO;
029
030     for (ix = 0; ix < PRIME_SIZE; ix++) \{
031       /* what is a mod LBL_prime_tab[ix] */
032       if ((err = mp_mod_d (a, ltm_prime_tab[ix], &res)) != MP_OKAY) \{
033         return err;
034       \}
035
036       /* is the residue zero? */
037       if (res == 0) \{
038         *result = MP_YES;
039         return MP_OKAY;
040       \}
041     \}
042
043     return MP_OKAY;
044   \}
045   #endif
046
\end{alltt}
\end{small}

The algorithm defaults to a return of $0$ in case an error occurs.  The values in the prime table are all specified to be in the range of a
mp\_digit.  The table \_\_prime\_tab is defined in the following file.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_prime\_tab.c
\vspace{-3mm}
\begin{alltt}
016   const mp_digit ltm_prime_tab[] = \{
017     0x0002, 0x0003, 0x0005, 0x0007, 0x000B, 0x000D, 0x0011, 0x0013,
018     0x0017, 0x001D, 0x001F, 0x0025, 0x0029, 0x002B, 0x002F, 0x0035,
019     0x003B, 0x003D, 0x0043, 0x0047, 0x0049, 0x004F, 0x0053, 0x0059,
020     0x0061, 0x0065, 0x0067, 0x006B, 0x006D, 0x0071, 0x007F,
021   #ifndef MP_8BIT
022     0x0083,
023     0x0089, 0x008B, 0x0095, 0x0097, 0x009D, 0x00A3, 0x00A7, 0x00AD,
024     0x00B3, 0x00B5, 0x00BF, 0x00C1, 0x00C5, 0x00C7, 0x00D3, 0x00DF,
025     0x00E3, 0x00E5, 0x00E9, 0x00EF, 0x00F1, 0x00FB, 0x0101, 0x0107,
026     0x010D, 0x010F, 0x0115, 0x0119, 0x011B, 0x0125, 0x0133, 0x0137,
027
028     0x0139, 0x013D, 0x014B, 0x0151, 0x015B, 0x015D, 0x0161, 0x0167,
029     0x016F, 0x0175, 0x017B, 0x017F, 0x0185, 0x018D, 0x0191, 0x0199,
030     0x01A3, 0x01A5, 0x01AF, 0x01B1, 0x01B7, 0x01BB, 0x01C1, 0x01C9,
031     0x01CD, 0x01CF, 0x01D3, 0x01DF, 0x01E7, 0x01EB, 0x01F3, 0x01F7,
032     0x01FD, 0x0209, 0x020B, 0x021D, 0x0223, 0x022D, 0x0233, 0x0239,
033     0x023B, 0x0241, 0x024B, 0x0251, 0x0257, 0x0259, 0x025F, 0x0265,
034     0x0269, 0x026B, 0x0277, 0x0281, 0x0283, 0x0287, 0x028D, 0x0293,
035     0x0295, 0x02A1, 0x02A5, 0x02AB, 0x02B3, 0x02BD, 0x02C5, 0x02CF,
036
037     0x02D7, 0x02DD, 0x02E3, 0x02E7, 0x02EF, 0x02F5, 0x02F9, 0x0301,
038     0x0305, 0x0313, 0x031D, 0x0329, 0x032B, 0x0335, 0x0337, 0x033B,
039     0x033D, 0x0347, 0x0355, 0x0359, 0x035B, 0x035F, 0x036D, 0x0371,
040     0x0373, 0x0377, 0x038B, 0x038F, 0x0397, 0x03A1, 0x03A9, 0x03AD,
041     0x03B3, 0x03B9, 0x03C7, 0x03CB, 0x03D1, 0x03D7, 0x03DF, 0x03E5,
042     0x03F1, 0x03F5, 0x03FB, 0x03FD, 0x0407, 0x0409, 0x040F, 0x0419,
043     0x041B, 0x0425, 0x0427, 0x042D, 0x043F, 0x0443, 0x0445, 0x0449,
044     0x044F, 0x0455, 0x045D, 0x0463, 0x0469, 0x047F, 0x0481, 0x048B,
045
046     0x0493, 0x049D, 0x04A3, 0x04A9, 0x04B1, 0x04BD, 0x04C1, 0x04C7,
047     0x04CD, 0x04CF, 0x04D5, 0x04E1, 0x04EB, 0x04FD, 0x04FF, 0x0503,
048     0x0509, 0x050B, 0x0511, 0x0515, 0x0517, 0x051B, 0x0527, 0x0529,
049     0x052F, 0x0551, 0x0557, 0x055D, 0x0565, 0x0577, 0x0581, 0x058F,
050     0x0593, 0x0595, 0x0599, 0x059F, 0x05A7, 0x05AB, 0x05AD, 0x05B3,
051     0x05BF, 0x05C9, 0x05CB, 0x05CF, 0x05D1, 0x05D5, 0x05DB, 0x05E7,
052     0x05F3, 0x05FB, 0x0607, 0x060D, 0x0611, 0x0617, 0x061F, 0x0623,
053     0x062B, 0x062F, 0x063D, 0x0641, 0x0647, 0x0649, 0x064D, 0x0653
054   #endif
055   \};
056   #endif
057
\end{alltt}
\end{small}

Note that there are two possible tables.  When an mp\_digit is 7-bits long only the primes upto $127$ may be included, otherwise the primes
upto $1619$ are used.  Note that the value of \textbf{PRIME\_SIZE} is a constant dependent on the size of a mp\_digit.

\subsection{The Fermat Test}
The Fermat test is probably one the oldest tests to have a non-trivial probability of success.  It is based on the fact that if $n$ is in
fact prime then $a^{n} \equiv a \mbox{ (mod }n\mbox{)}$ for all $0 < a < n$.  The reason being that if $n$ is prime than the order of
the multiplicative sub group is $n - 1$.  Any base $a$ must have an order which divides $n - 1$ and as such $a^n$ is equivalent to
$a^1 = a$.

If $n$ is composite then any given base $a$ does not have to have a period which divides $n - 1$.  In which case
it is possible that $a^n \nequiv a \mbox{ (mod }n\mbox{)}$.  However, this test is not absolute as it is possible that the order
of a base will divide $n - 1$ which would then be reported as prime.  Such a base yields what is known as a Fermat pseudo-prime.  Several
integers known as Carmichael numbers will be a pseudo-prime to all valid bases.  Fortunately such numbers are extremely rare as $n$ grows
in size.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_prime\_fermat}. \\
\textbf{Input}.   mp\_int $a$ and $b$, $a \ge 2$, $0 < b < a$.  \\
\textbf{Output}.  $c = 1$ if $b^a \equiv b \mbox{ (mod }a\mbox{)}$, otherwise $c = 0$.  \\
\hline \\
1.  $t \leftarrow b^a \mbox{ (mod }a\mbox{)}$ \\
2.  If $t = b$ then \\
\hspace{3mm}2.1  $c = 1$ \\
3.  else \\
\hspace{3mm}3.1  $c = 0$ \\
4.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_prime\_fermat}
\end{figure}
\textbf{Algorithm mp\_prime\_fermat.}
This algorithm determines whether an mp\_int $a$ is a Fermat prime to the base $b$ or not.  It uses a single modular exponentiation to
determine the result.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_prime\_fermat.c
\vspace{-3mm}
\begin{alltt}
016
017   /* performs one Fermat test.
018    *
019    * If "a" were prime then b**a == b (mod a) since the order of
020    * the multiplicative sub-group would be phi(a) = a-1.  That means
021    * it would be the same as b**(a mod (a-1)) == b**1 == b (mod a).
022    *
023    * Sets result to 1 if the congruence holds, or zero otherwise.
024    */
025   int mp_prime_fermat (mp_int * a, mp_int * b, int *result)
026   \{
027     mp_int  t;
028     int     err;
029
030     /* default to composite  */
031     *result = MP_NO;
032
033     /* ensure b > 1 */
034     if (mp_cmp_d(b, 1) != MP_GT) \{
035        return MP_VAL;
036     \}
037
038     /* init t */
039     if ((err = mp_init (&t)) != MP_OKAY) \{
040       return err;
041     \}
042
043     /* compute t = b**a mod a */
044     if ((err = mp_exptmod (b, a, a, &t)) != MP_OKAY) \{
045       goto LBL_T;
046     \}
047
048     /* is it equal to b? */
049     if (mp_cmp (&t, b) == MP_EQ) \{
050       *result = MP_YES;
051     \}
052
053     err = MP_OKAY;
054   LBL_T:mp_clear (&t);
055     return err;
056   \}
057   #endif
058
\end{alltt}
\end{small}

\subsection{The Miller-Rabin Test}
The Miller-Rabin (citation) test is another primality test which has tighter error bounds than the Fermat test specifically with sequentially chosen
candidate  integers.  The algorithm is based on the observation that if $n - 1 = 2^kr$ and if $b^r \nequiv \pm 1$ then after upto $k - 1$ squarings the
value must be equal to $-1$.  The squarings are stopped as soon as $-1$ is observed.  If the value of $1$ is observed first it means that
some value not congruent to $\pm 1$ when squared equals one which cannot occur if $n$ is prime.

\begin{figure}[!here]
\begin{small}
\begin{center}
\begin{tabular}{l}
\hline Algorithm \textbf{mp\_prime\_miller\_rabin}. \\
\textbf{Input}.   mp\_int $a$ and $b$, $a \ge 2$, $0 < b < a$.  \\
\textbf{Output}.  $c = 1$ if $a$ is a Miller-Rabin prime to the base $a$, otherwise $c = 0$.  \\
\hline
1.  $a' \leftarrow a - 1$ \\
2.  $r  \leftarrow n1$    \\
3.  $c \leftarrow 0, s  \leftarrow 0$ \\
4.  While $r.used > 0$ and $r_0 \equiv 0 \mbox{ (mod }2\mbox{)}$ \\
\hspace{3mm}4.1  $s \leftarrow s + 1$ \\
\hspace{3mm}4.2  $r \leftarrow \lfloor r / 2 \rfloor$ \\
5.  $y \leftarrow b^r \mbox{ (mod }a\mbox{)}$ \\
6.  If $y \nequiv \pm 1$ then \\
\hspace{3mm}6.1  $j \leftarrow 1$ \\
\hspace{3mm}6.2  While $j \le (s - 1)$ and $y \nequiv a'$ \\
\hspace{6mm}6.2.1  $y \leftarrow y^2 \mbox{ (mod }a\mbox{)}$ \\
\hspace{6mm}6.2.2  If $y = 1$ then goto step 8. \\
\hspace{6mm}6.2.3  $j \leftarrow j + 1$ \\
\hspace{3mm}6.3  If $y \nequiv a'$ goto step 8. \\
7.  $c \leftarrow 1$\\
8.  Return(\textit{MP\_OKAY}). \\
\hline
\end{tabular}
\end{center}
\end{small}
\caption{Algorithm mp\_prime\_miller\_rabin}
\end{figure}
\textbf{Algorithm mp\_prime\_miller\_rabin.}
This algorithm performs one trial round of the Miller-Rabin algorithm to the base $b$.  It will set $c = 1$ if the algorithm cannot determine
if $b$ is composite or $c = 0$ if $b$ is provably composite.  The values of $s$ and $r$ are computed such that $a' = a - 1 = 2^sr$.

If the value $y \equiv b^r$ is congruent to $\pm 1$ then the algorithm cannot prove if $a$ is composite or not.  Otherwise, the algorithm will
square $y$ upto $s - 1$ times stopping only when $y \equiv -1$.  If $y^2 \equiv 1$ and $y \nequiv \pm 1$ then the algorithm can report that $a$
is provably composite.  If the algorithm performs $s - 1$ squarings and $y \nequiv -1$ then $a$ is provably composite.  If $a$ is not provably
composite then it is \textit{probably} prime.

\vspace{+3mm}\begin{small}
\hspace{-5.1mm}{\bf File}: bn\_mp\_prime\_miller\_rabin.c
\vspace{-3mm}
\begin{alltt}
016
017   /* Miller-Rabin test of "a" to the base of "b" as described in
018    * HAC pp. 139 Algorithm 4.24
019    *
020    * Sets result to 0 if definitely composite or 1 if probably prime.
021    * Randomly the chance of error is no more than 1/4 and often
022    * very much lower.
023    */
024   int mp_prime_miller_rabin (mp_int * a, mp_int * b, int *result)
025   \{
026     mp_int  n1, y, r;
027     int     s, j, err;
028
029     /* default */
030     *result = MP_NO;
031
032     /* ensure b > 1 */
033     if (mp_cmp_d(b, 1) != MP_GT) \{
034        return MP_VAL;
035     \}
036
037     /* get n1 = a - 1 */
038     if ((err = mp_init_copy (&n1, a)) != MP_OKAY) \{
039       return err;
040     \}
041     if ((err = mp_sub_d (&n1, 1, &n1)) != MP_OKAY) \{
042       goto LBL_N1;
043     \}
044
045     /* set 2**s * r = n1 */
046     if ((err = mp_init_copy (&r, &n1)) != MP_OKAY) \{
047       goto LBL_N1;
048     \}
049
050     /* count the number of least significant bits
051      * which are zero
052      */
053     s = mp_cnt_lsb(&r);
054
055     /* now divide n - 1 by 2**s */
056     if ((err = mp_div_2d (&r, s, &r, NULL)) != MP_OKAY) \{
057       goto LBL_R;
058     \}
059
060     /* compute y = b**r mod a */
061     if ((err = mp_init (&y)) != MP_OKAY) \{
062       goto LBL_R;
063     \}
064     if ((err = mp_exptmod (b, &r, a, &y)) != MP_OKAY) \{
065       goto LBL_Y;
066     \}
067
068     /* if y != 1 and y != n1 do */
069     if ((mp_cmp_d (&y, 1) != MP_EQ) && (mp_cmp (&y, &n1) != MP_EQ)) \{
070       j = 1;
071       /* while j <= s-1 and y != n1 */
072       while ((j <= (s - 1)) && (mp_cmp (&y, &n1) != MP_EQ)) \{
073         if ((err = mp_sqrmod (&y, a, &y)) != MP_OKAY) \{
074            goto LBL_Y;
075         \}
076
077         /* if y == 1 then composite */
078         if (mp_cmp_d (&y, 1) == MP_EQ) \{
079            goto LBL_Y;
080         \}
081
082         ++j;
083       \}
084
085       /* if y != n1 then composite */
086       if (mp_cmp (&y, &n1) != MP_EQ) \{
087         goto LBL_Y;
088       \}
089     \}
090
091     /* probably prime now */
092     *result = MP_YES;
093   LBL_Y:mp_clear (&y);
094   LBL_R:mp_clear (&r);
095   LBL_N1:mp_clear (&n1);
096     return err;
097   \}
098   #endif
099
\end{alltt}
\end{small}




\backmatter
\appendix
\begin{thebibliography}{ABCDEF}
\bibitem[1]{TAOCPV2}
Donald Knuth, \textit{The Art of Computer Programming}, Third Edition, Volume Two, Seminumerical Algorithms, Addison-Wesley, 1998

\bibitem[2]{HAC}
A. Menezes, P. van Oorschot, S. Vanstone, \textit{Handbook of Applied Cryptography}, CRC Press, 1996

\bibitem[3]{ROSE}
Michael Rosing, \textit{Implementing Elliptic Curve Cryptography}, Manning Publications, 1999

\bibitem[4]{COMBA}
Paul G. Comba, \textit{Exponentiation Cryptosystems on the IBM PC}. IBM Systems Journal 29(4): 526-538 (1990)

\bibitem[5]{KARA}
A. Karatsuba, Doklay Akad. Nauk SSSR 145 (1962), pp.293-294

\bibitem[6]{KARAP}
Andre Weimerskirch and Christof Paar, \textit{Generalizations of the Karatsuba Algorithm for Polynomial Multiplication}, Submitted to Design, Codes and Cryptography, March 2002

\bibitem[7]{BARRETT}
Paul Barrett, \textit{Implementing the Rivest Shamir and Adleman Public Key Encryption Algorithm on a Standard Digital Signal Processor}, Advances in Cryptology, Crypto '86, Springer-Verlag.

\bibitem[8]{MONT}
P.L.Montgomery. \textit{Modular multiplication without trial division}. Mathematics of Computation, 44(170):519-521, April 1985.

\bibitem[9]{DRMET}
Chae Hoon Lim and Pil Joong Lee, \textit{Generating Efficient Primes for Discrete Log Cryptosystems}, POSTECH Information Research Laboratories

\bibitem[10]{MMB}
J. Daemen and R. Govaerts and J. Vandewalle, \textit{Block ciphers based on Modular Arithmetic}, State and {P}rogress in the {R}esearch of {C}ryptography, 1993, pp. 80-89

\bibitem[11]{RSAREF}
R.L. Rivest, A. Shamir, L. Adleman, \textit{A Method for Obtaining Digital Signatures and Public-Key Cryptosystems}

\bibitem[12]{DHREF}
Whitfield Diffie, Martin E. Hellman, \textit{New Directions in Cryptography}, IEEE Transactions on Information Theory, 1976

\bibitem[13]{IEEE}
IEEE Standard for Binary Floating-Point Arithmetic (ANSI/IEEE Std 754-1985)

\bibitem[14]{GMP}
GNU Multiple Precision (GMP), \url{http://www.swox.com/gmp/}

\bibitem[15]{MPI}
Multiple Precision Integer Library (MPI), Michael Fromberger, \url{http://thayer.dartmouth.edu/~sting/mpi/}

\bibitem[16]{OPENSSL}
OpenSSL Cryptographic Toolkit, \url{http://openssl.org}

\bibitem[17]{LIP}
Large Integer Package, \url{http://home.hetnet.nl/~ecstr/LIP.zip}

\bibitem[18]{ISOC}
JTC1/SC22/WG14, ISO/IEC 9899:1999, ``A draft rationale for the C99 standard.''

\bibitem[19]{JAVA}
The Sun Java Website, \url{http://java.sun.com/}

\end{thebibliography}

\input{tommath.ind}

\end{document}
generated by cgit v0.12 at 2024-10-01 10:38:15 (GMT)