Signed-off-by: Sun He <>
 bulk-checkin.c |   10 +++++++++-
 1 files changed, 9 insertions(+), 1 deletions(-)

diff --git a/bulk-checkin.c b/bulk-checkin.c
index 118c625..e3c7fb2 100644
--- a/bulk-checkin.c
+++ b/bulk-checkin.c
@@ -23,7 +23,8 @@ static struct bulk_checkin_state {
 static void finish_bulk_checkin(struct bulk_checkin_state *state)
        unsigned char sha1[20];
-       char packname[PATH_MAX];
+    char *packname;
+    struct strbuf sb;
        int i;
        if (!state->f)
@@ -43,6 +44,10 @@ static void finish_bulk_checkin(struct bulk_checkin_state 
+     /* 64-1 is more than the sum of len(sha1_to_hex(sha1)) and len(".pack") */
+    strbuf_init(&sb,strlen(get_object_directory())+64);
+    packname = sb.buf;
        sprintf(packname, "%s/pack/pack-", get_object_directory());
        finish_tmp_packfile(packname, state->pack_tmp_name,
                            state->written, state->nr_written,
@@ -54,6 +59,9 @@ clear_exit:
        memset(state, 0, sizeof(*state));
+    /* release sb space */
+    strbuf_release(&sb);
        /* Make objects we just wrote available to ourselves */

> Rewrite bulk-checkin.c:finish_bulk_checkin() to use a strbuf for handling 
> packname, and explain why this is useful.

char packname[PATH_MAX] (PATH_MAX=4096) in a local function costs too much 
space, it may cause stack overflow.
Using strbuf to deal with packname is more space-friendly.
I only use the API strbuf_init to alloc enough space for packname. I didn't use 
other APIs because I want to make the change as little as possible that I don't 
have to fix other functions which may import new bugs.
Because the space spared for sb is on heap, we must release it after it is not 

> Also check if the first argument of pack-write.c:finish_tmp_packfile() can be 
> made const.

Tht first argument of pack-write.c:finish_tmp_packfile() can be made const 
because we didn't use *name_buffer to change anything.

He Sun

Why I cannot sent email to via my Firefox?
Can you receive my former emails?
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to
More majordomo info at

Reply via email to