core: log offset pack data accesses happened

In a workload other than "git log" (without pathspec nor any option that
causes us to inspect trees and blobs), the recency pack order is said to
cause the access jump around quite a bit. Add a hook to allow us observe
how bad it is.

"git config core.logpackaccess /var/tmp/pal.txt" will give you the log
in the specified file.

Signed-off-by: Junio C Hamano <gitster@pobox.com>

diff --git a/cache.h b/cache.h
index f4bb43ec6aa3e910e8a7d2ba2d9f06440e400fc2..16a8c7c23c9c12a1b13c219a64a47524c1443886 100644 (file)
--- a/cache.h
+++ b/cache.h
@@ -784,6 +784,9 @@ extern int force_object_loose(const unsigned char *sha1, time_t mtime);
 /* global flag to enable extra checks when accessing packed objects */
 extern int do_check_packed_object_crc;
 
+/* for development: log offset of pack access */
+extern const char *log_pack_access;
+
 extern int check_sha1_signature(const unsigned char *sha1, void *buf, unsigned long size, const char *type);
 
 extern int move_temp_to_file(const char *tmpfile, const char *filename);

diff --git a/config.c b/config.c
index e0b3b80d92ece9f37e0e76292bd61ce97271fcce..5ef3f397d51456f37d1efe34e5eb4547c756acdd 100644 (file)
--- a/config.c
+++ b/config.c
@@ -569,6 +569,9 @@ static int git_default_core_config(const char *var, const char *value)
                return 0;
        }
 
+       if (!strcmp(var, "core.logpackaccess"))
+               return git_config_string(&log_pack_access, var, value);
+
        if (!strcmp(var, "core.autocrlf")) {
                if (value && !strcasecmp(value, "input")) {
                        if (core_eol == EOL_CRLF)

diff --git a/environment.c b/environment.c
index 94d58fd24413ec1d1a5769029bd5149b609f0d4e..19351024f5ef61ba0e5237aa039fc9e657ad81a0 100644 (file)
--- a/environment.c
+++ b/environment.c
@@ -36,6 +36,7 @@ size_t packed_git_window_size = DEFAULT_PACKED_GIT_WINDOW_SIZE;
 size_t packed_git_limit = DEFAULT_PACKED_GIT_LIMIT;
 size_t delta_base_cache_limit = 16 * 1024 * 1024;
 unsigned long big_file_threshold = 512 * 1024 * 1024;
+const char *log_pack_access;
 const char *pager_program;
 int pager_use_color = 1;
 const char *editor_program;

diff --git a/sha1_file.c b/sha1_file.c
index 064a33040812ba8782bf602c693abf08613d6ec7..baf5da182301eb245187c72d0bd26c48c4aca3af 100644 (file)
--- a/sha1_file.c
+++ b/sha1_file.c
@@ -1839,6 +1839,24 @@ static void *unpack_delta_entry(struct packed_git *p,
        return result;
 }
 
+static void write_pack_access_log(struct packed_git *p, off_t obj_offset)
+{
+       static FILE *log_file;
+
+       if (!log_file) {
+               log_file = fopen(log_pack_access, "w");
+               if (!log_file) {
+                       error("cannot open pack access log '%s' for writing: %s",
+                             log_pack_access, strerror(errno));
+                       log_pack_access = NULL;
+                       return;
+               }
+       }
+       fprintf(log_file, "%s %"PRIuMAX"\n",
+               p->pack_name, (uintmax_t)obj_offset);
+       fflush(log_file);
+}
+
 int do_check_packed_object_crc;
 
 void *unpack_entry(struct packed_git *p, off_t obj_offset,
@@ -1848,6 +1866,9 @@ void *unpack_entry(struct packed_git *p, off_t obj_offset,
        off_t curpos = obj_offset;
        void *data;
 
+       if (log_pack_access)
+               write_pack_access_log(p, obj_offset);
+
        if (do_check_packed_object_crc && p->index_version > 1) {
                struct revindex_entry *revidx = find_pack_revindex(p, obj_offset);
                unsigned long len = revidx[1].offset - obj_offset;