[svn-r6252] Purpose:

    Lots of performance improvements & a couple new internal API interfaces.

Description:
    Performance Improvements:
        - Cached file offset & length sizes in shared file struct, to avoid
            constantly looking them up in the FCPL.
        - Generic property improvements:
            - Added "revision" number to generic property classes to speed
                up comparisons.
            - Changed method of storing properties from using a hash-table
                to the TBBT routines in the library.
            - Share the propery names between classes and the lists derived
                from them.
            - Removed redundant 'def_value' buffer from each property.
            - Switching code to use a "copy on write" strategy for
                properties in each list, where the properties in each list
                are shared with the properties in the class, until a
                property's value is changed in a list.
        - Fixed error in layout code which was allocating too many buffers.
        - Redefined public macros of the form (H5open()/H5check, <variable>)
            internally to only be (<variable>), avoiding innumerable useless
            calls to H5open() and H5check_version().
        - Reuse already zeroed buffers in H5F_contig_fill instead of
            constantly re-zeroing them.
        - Don't write fill values if writing entire dataset.
        - Use gettimeofday() system call instead of time() system when
            checking the modification time of a dataset.
        - Added reference counted string API and use it for tracking the
            names of objects opening in a file (for the ID->name code).
        - Removed redundant H5P_get() calls in B-tree routines.
        - Redefine H5T datatype macros internally to the library, to avoid
            calling H5check redundantly.
        - Keep dataspace information for dataset locally instead of reading
            from disk each time.  Added new module to track open objects
            in a file, to allow this (which will be useful eventually for
            some FPH5 metadata caching issues).
        - Remove H5AC_find macro which was inlining metadata cache lookups,
            and call function instead.
        - Remove redundant memset() calls from H5G_namei() routine.
        - Remove redundant checking of object type when locating objects
            in metadata cache and rely on the address only.
        - Create default dataset object to use when default dataset creation
            property list is used to create datasets, bypassing querying
            for all the property list values.
        - Use default I/O vector size when performing raw data with the
            default dataset transfer property list, instead of querying for
            I/O vector size.
        - Remove H5P_DEFAULT internally to the library, replacing it with
            more specific default property list based on the type of
            property list needed.
        - Remove redundant memset() calls in object header message (H5O*)
            routines.
        - Remove redunant memset() calls in data I/O routines.
        - Split free-list allocation routines into malloc() and calloc()-
            like routines, instead of one combined routine.
        - Remove lots of indirection in H5O*() routines.
        - Simplify metadata cache entry comparison routine (used when
            flushing entire cache out).
        - Only enable metadata cache statistics when H5AC_DEBUG is turned
            on, instead of always tracking them.
        - Simplify address comparison macro (H5F_addr_eq).
        - Remove redundant metadata cache entry protections during dataset
            creation by protecting the object header once and making all
            the modifications necessary for the dataset creation before
            unprotecting it.
        - Reduce # of "number of element in extent" computations performed
            by computing and storing the value during dataspace creation.
        - Simplify checking for group location's file information, when file
            has not been involving in file-mounting operations.
        - Use binary encoding for modification time, instead of ASCII.
        - Hoist H5HL_peek calls (to get information in a local heap)
            out of loops in many group routine.
        - Use static variable for iterators of selections, instead of
            dynamically allocation them each time.
        - Lookup & insert new entries in one step, avoiding traversing
            group's B-tree twice.
        - Fixed memory leak in H5Gget_objname_idx() routine (tangential to
            performance improvements, but fixed along the way).
        - Use free-list for reference counted strings.
        - Don't bother copying object names into cached group entries,
            since they are re-created when an object is opened.

        The benchmark I used to measure these results created several thousand
        small (2K) datasets in a file and wrote out the data for them.  This is
        Elena's "regular.c" benchmark.

        These changes resulted in approximately ~4.3x speedup of the
        development branch when compared to the previous code in the
        development branch and ~1.4x speedup compared to the release
        branch.

        Additionally, these changes reduce the total memory used (code and
        data) by the development branch by ~800KB, bringing the development
        branch back into the same ballpark as the release branch.

        I'll send out a more detailed description of the benchmark results
        as a followup note.

    New internal API routines:
        Added "reference counted strings" API for tracking strings that get
            used by multiple owners without duplicating the strings.
        Added "ternary search tree" API for text->object mappings.

Platforms tested:
    Tested h5committest {arabica (fortran), eirene (fortran, C++)
	modi4 (parallel, fortran)}
    Other platforms/configurations tested?
        FreeBSD 4.7 (sleipnir) serial & parallel
        Solaris 2.6 (baldric) serial

1 files changed, 277 insertions, 0 deletions

diff --git a/src/H5FO.c b/src/H5FO.c
new file mode 100644
index 0000000..29f3998
--- /dev/null
+++ b/src/H5FO.c
@@ -0,0 +1,277 @@
+/****************************************************************************
+ * NCSA HDF								    *
+ * Software Development Group						    *
+ * National Center for Supercomputing Applications			    *
+ * University of Illinois at Urbana-Champaign				    *
+ * 605 E. Springfield, Champaign IL 61820				    *
+ *									    *
+ * For conditions of distribution and use, see the accompanying		    *
+ * hdf/COPYING file.							    *
+ *									    *
+ ****************************************************************************/
+
+/*
+ * Open object info algorithms.
+ *
+ * These are used to track the objects currently open in a file, for various
+ * internal mechanisms which need to be aware of such things.
+ *
+ */
+
+#define H5F_PACKAGE		/*suppress error about including H5Fpkg	  */
+
+#include "H5Eprivate.h"		/* Error handling		  	*/
+#include "H5Fpkg.h"             /* File access                          */
+#include "H5FLprivate.h"	/* Free lists                           */
+#include "H5FOprivate.h"        /* File objects                         */
+
+#define PABLO_MASK	H5FO_mask
+
+/* Interface initialization */
+static int		interface_initialize_g = 0;
+#define INTERFACE_INIT	NULL
+
+/* Private typedefs */
+
+/* Information about object objects in a file */
+typedef struct H5FO_open_obj_t {
+    haddr_t addr;                       /* Address of object header for object */
+                                        /* THIS MUST BE FIRST FOR TBBT ROUTINES */
+    hid_t id;                           /* Current ID for object            */
+} H5FO_open_obj_t;
+
+/* Declare a free list to manage the H5FO_open_obj_t struct */
+H5FL_DEFINE_STATIC(H5FO_open_obj_t);
+
+
+/*--------------------------------------------------------------------------
+ NAME
+    H5FO_create
+ PURPOSE
+    Create an open object info set
+ USAGE
+    herr_t H5FO_create(f)
+        H5F_t *f;       IN/OUT: File to create opened object info set for
+
+ RETURNS
+    Returns non-negative on success, negative on failure
+ DESCRIPTION
+    Create a new open object info set.
+ GLOBAL VARIABLES
+ COMMENTS, BUGS, ASSUMPTIONS
+ EXAMPLES
+ REVISION LOG
+--------------------------------------------------------------------------*/
+herr_t
+H5FO_create(H5F_t *f)
+{
+    herr_t ret_value=SUCCEED;          /* Return value */
+
+    FUNC_ENTER_NOAPI(H5FO_create,FAIL);
+
+    /* Sanity check */
+    assert(f);
+    assert(f->shared);
+
+    /* Create TBBT used to store open object info */
+    if((f->shared->open_objs=H5TB_fast_dmake(H5TB_FAST_HADDR_COMPARE))==NULL)
+        HGOTO_ERROR(H5E_FILE, H5E_CANTINIT, FAIL, "unable to create open object TBBT");
+
+done:
+    FUNC_LEAVE(ret_value);
+} /* end H5FO_create() */
+
+
+/*--------------------------------------------------------------------------
+ NAME
+    H5FO_opened
+ PURPOSE
+    Checks if an object at an address is already open in the file.
+ USAGE
+    hid_t H5FO_opened(f,addr)
+        const H5F_t *f;         IN: File to check opened object info set
+        haddr_t addr;           IN: Address of object to check
+
+ RETURNS
+    Returns a non-negative ID for the object on success, negative on failure
+ DESCRIPTION
+    Check is an object at an address (the address of the object's object header)
+    is already open in the file and return the ID for that object if it is open.
+ GLOBAL VARIABLES
+ COMMENTS, BUGS, ASSUMPTIONS
+ EXAMPLES
+ REVISION LOG
+--------------------------------------------------------------------------*/
+hid_t
+H5FO_opened(const H5F_t *f, haddr_t addr)
+{
+    H5TB_NODE *obj_node;        /* TBBT node holding open object */
+    H5FO_open_obj_t *open_obj;  /* Information about open object */
+    hid_t ret_value;            /* Return value */
+
+    FUNC_ENTER_NOAPI(H5FO_opened,FAIL);
+
+    /* Sanity check */
+    assert(f);
+    assert(f->shared);
+    assert(f->shared->open_objs);
+    assert(H5F_addr_defined(addr));
+
+    /* Get the object node from the TBBT */
+    if((obj_node=H5TB_dfind(f->shared->open_objs,&addr,NULL))!=NULL) {
+        open_obj=H5TB_NODE_DATA(obj_node);
+        assert(open_obj);
+        ret_value=open_obj->id;
+        assert(ret_value>0);
+    } /* end if */
+    else
+        ret_value=FAIL;
+
+done:
+    FUNC_LEAVE(ret_value);
+} /* end H5FO_opened() */
+
+
+/*--------------------------------------------------------------------------
+ NAME
+    H5FO_insert
+ PURPOSE
+    Insert a newly opened object/ID pair into the opened object info set
+ USAGE
+    herr_t H5FO_insert(f,addr,id)
+        H5F_t *f;               IN/OUT: File's opened object info set
+        haddr_t addr;           IN: Address of object to insert
+        hid_t id;               IN: ID of object to insert
+
+ RETURNS
+    Returns a non-negative on success, negative on failure
+ DESCRIPTION
+    Insert an object/ID pair into the opened object info set.
+ GLOBAL VARIABLES
+ COMMENTS, BUGS, ASSUMPTIONS
+ EXAMPLES
+ REVISION LOG
+--------------------------------------------------------------------------*/
+herr_t
+H5FO_insert(H5F_t *f, haddr_t addr, hid_t id)
+{
+    H5FO_open_obj_t *open_obj;  /* Information about open object */
+    herr_t ret_value=SUCCEED;   /* Return value */
+
+    FUNC_ENTER_NOAPI(H5FO_insert,FAIL);
+
+    /* Sanity check */
+    assert(f);
+    assert(f->shared);
+    assert(f->shared->open_objs);
+    assert(H5F_addr_defined(addr));
+    assert(id>0);
+
+    /* Allocate new opened object information structure */
+    if((open_obj=H5FL_MALLOC(H5FO_open_obj_t))==NULL)
+        HGOTO_ERROR(H5E_CACHE,H5E_NOSPACE,NULL,"memory allocation failed");
+
+    /* Assign information */
+    open_obj->addr=addr;
+    open_obj->id=id;
+
+    /* Insert into TBBT */
+    if(H5TB_dins(f->shared->open_objs,open_obj,open_obj)==NULL)
+        HGOTO_ERROR(H5E_CACHE,H5E_CANTINSERT,FAIL,"can't insert object into TBBT");
+
+done:
+    FUNC_LEAVE(ret_value);
+} /* end H5FO_insert() */
+
+
+/*--------------------------------------------------------------------------
+ NAME
+    H5FO_delete
+ PURPOSE
+    Remove an opened object/ID pair from the opened object info set
+ USAGE
+    herr_t H5FO_delete(f,addr)
+        H5F_t *f;               IN/OUT: File's opened object info set
+        haddr_t addr;           IN: Address of object to remove
+
+ RETURNS
+    Returns a non-negative on success, negative on failure
+ DESCRIPTION
+    Remove an object/ID pair from the opened object info.
+ GLOBAL VARIABLES
+ COMMENTS, BUGS, ASSUMPTIONS
+ EXAMPLES
+ REVISION LOG
+--------------------------------------------------------------------------*/
+herr_t
+H5FO_delete(H5F_t *f, haddr_t addr)
+{
+    H5TB_NODE *obj_node;        /* TBBT node holding open object */
+    H5FO_open_obj_t *open_obj;  /* Information about open object */
+    herr_t ret_value=SUCCEED;   /* Return value */
+
+    FUNC_ENTER_NOAPI(H5FO_delete,FAIL);
+
+    /* Sanity check */
+    assert(f);
+    assert(f->shared);
+    assert(f->shared->open_objs);
+    assert(H5F_addr_defined(addr));
+
+    /* Get the object node from the TBBT */
+    if((obj_node=H5TB_dfind(f->shared->open_objs,&addr,NULL))==NULL)
+        HGOTO_ERROR(H5E_CACHE,H5E_NOTFOUND,FAIL,"can't locate object in TBBT");
+
+    /* Remove from TBBT */
+    if((open_obj=H5TB_rem(&f->shared->open_objs->root,obj_node,NULL))==NULL)
+        HGOTO_ERROR(H5E_CACHE,H5E_CANTRELEASE,FAIL,"can't remove object from TBBT");
+
+    /* Release the object information */
+    H5FL_FREE(H5FO_open_obj_t,open_obj);
+
+done:
+    FUNC_LEAVE(ret_value);
+} /* end H5FO_delete() */
+
+
+/*--------------------------------------------------------------------------
+ NAME
+    H5FO_dest
+ PURPOSE
+    Destroy an open object info set
+ USAGE
+    herr_t H5FO_create(f)
+        H5F_t *f;               IN/OUT: File's opened object info set
+
+ RETURNS
+    Returns a non-negative on success, negative on failure
+ DESCRIPTION
+    Destroy an existing open object info set.
+ GLOBAL VARIABLES
+ COMMENTS, BUGS, ASSUMPTIONS
+ EXAMPLES
+ REVISION LOG
+--------------------------------------------------------------------------*/
+herr_t
+H5FO_dest(H5F_t *f)
+{
+    herr_t ret_value=SUCCEED;   /* Return value */
+
+    FUNC_ENTER_NOAPI(H5FO_dest,FAIL);
+
+    /* Sanity check */
+    assert(f);
+    assert(f->shared);
+    assert(f->shared->open_objs);
+
+    /* Check if the object info set is empty */
+    if(H5TB_count(f->shared->open_objs)!=0)
+        HGOTO_ERROR(H5E_CACHE, H5E_CANTRELEASE, FAIL, "objects still in open object info set");
+
+    /* Release the open object info set TBBT */
+    f->shared->open_objs=H5TB_dfree(f->shared->open_objs,NULL,NULL);
+
+done:
+    FUNC_LEAVE(ret_value);
+} /* end H5FO_dest() */
+