diff options
Diffstat (limited to 'src/H5AC.c')
| -rw-r--r-- | src/H5AC.c | 42 |
1 files changed, 4 insertions, 38 deletions
@@ -413,14 +413,6 @@ done: * Programmer: Robb Matzke * Jul 9 1997 * - * Changes: - * - * In the parallel case, added code to setup the MDC slist - * before the call to H5AC__flush_entries() and take it down - * afterwards. - * - * JRM -- 7/29/20 - * *------------------------------------------------------------------------- */ herr_t @@ -1191,13 +1183,12 @@ done: * * Function: H5AC_prep_for_file_flush * - * Purpose: This function should be called just prior to the first - * call to H5AC_flush() during a file flush. + * Purpose: Handle any setup required prior to metadata cache flush. * - * Its purpose is to handly any setup required prior to - * metadata cache flush. + * This function should be called just prior to the first + * call to H5AC_flush() during a file flush. * - * Initially, this means setting up the slist prior to the + * Initially, this means setting up the skip list prior to the * flush. We do this in a separate call because * H5F__flush_phase2() make repeated calls to H5AC_flush(). * Handling this detail in separate calls allows us to avoid @@ -1209,8 +1200,6 @@ done: * Programmer: John Mainzer * 5/5/20 * - * Changes: None. - * *------------------------------------------------------------------------- */ herr_t @@ -1242,9 +1231,6 @@ done: * Purpose: This function should be called just after the last * call to H5AC_flush() during a file flush. * - * Its purpose is to perform any necessary cleanup after the - * metadata cache flush. - * * The objective of the call is to allow the metadata cache * to do any necessary necessary cleanup work after a cache * flush. @@ -1261,8 +1247,6 @@ done: * Programmer: John Mainzer * 5/5/20 * - * Changes: None. - * *------------------------------------------------------------------------- */ herr_t @@ -1474,24 +1458,6 @@ H5AC_resize_entry(void *thing, size_t new_size) * amounts of dirty metadata creation in other areas -- which will * cause aux_ptr->dirty_bytes to be incremented. * - * The bottom line is that this code is probably OK, but the above - * points should be kept in mind. - * - * One final observation: This comment is occasioned by a bug caused - * by moving the call to H5AC__log_dirtied_entry() after the call to - * H5C_resize_entry(), and then only calling H5AC__log_dirtied_entry() - * if entry_ptr->is_dirty was false. - * - * Since H5C_resize_entry() marks the target entry dirty unless there - * is not change in size, this had the effect of not calling - * H5AC__log_dirtied_entry() when it should be, and corrupting - * the cleaned and dirtied lists used by rank 0 in the parallel - * version of the metadata cache. - * - * The point here is that you should be very careful when working with - * this code, and not modify it unless you fully understand it. - * - * JRM -- 2/28/22 */ if ((!entry_ptr->is_dirty) && (entry_ptr->size != new_size)) { |