git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH v3 01/11] Add more large blob test cases

From
Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Date
Mar 5, 2012, 03:43 UTC
Message-ID
<1330919028-6611-2-git-send-email-pclouds@gmail.com>
In-Reply-To
<1330865996-2069-1-git-send-email-pclouds@gmail.com>

New test cases list commands that should work when memory is limited. All memory allocation functions (*) learn to reject any allocation larger than $GIT_ALLOC_LIMIT if set.

(*) Not exactly all. Some places do not use x* functions, but malloc/calloc directly, notably diff-delta. These code path should never be run on large blobs.

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
---
 t/t1050-large.sh |   63 ++++++++++++++++++++++++++++++++++++++++++++++++++++-
 wrapper.c        |   27 ++++++++++++++++++++--
 2 files changed, 85 insertions(+), 5 deletions(-)
diff --git a/t/t1050-large.sh b/t/t1050-large.sh
index 29d6024..80f157a 100755
--- a/t/t1050-large.sh
+++ b/t/t1050-large.sh
@@ -6,11 +6,15 @@ test_description='adding and checking out large blobs'
 . ./test-lib.sh
 
 test_expect_success setup '
-	git config core.bigfilethreshold 200k &&
+	# clone does not allow us to pass core.bigfilethreshold to
+	# new repos, so set core.bigfilethreshold globally
+	git config --global core.bigfilethreshold 200k &&
 	echo X | dd of=large1 bs=1k seek=2000 &&
 	echo X | dd of=large2 bs=1k seek=2000 &&
 	echo X | dd of=large3 bs=1k seek=2000 &&
-	echo Y | dd of=huge bs=1k seek=2500
+	echo Y | dd of=huge bs=1k seek=2500 &&
+	GIT_ALLOC_LIMIT=1500 &&
+	export GIT_ALLOC_LIMIT
 '
 
 test_expect_success 'add a large file or two' '
@@ -100,4 +104,59 @@ test_expect_success 'packsize limit' '
 	)
 '
 
+test_expect_success 'diff --raw' '
+	git commit -q -m initial &&
+	echo modified >>large1 &&
+	git add large1 &&
+	git commit -q -m modified &&
+	git diff --raw HEAD^
+'
+
+test_expect_success 'hash-object' '
+	git hash-object large1
+'
+
+test_expect_failure 'cat-file a large file' '
+	git cat-file blob :large1 >/dev/null
+'
+
+test_expect_failure 'cat-file a large file from a tag' '
+	git tag -m largefile largefiletag :large1 &&
+	git cat-file blob largefiletag >/dev/null
+'
+
+test_expect_failure 'git-show a large file' '
+	git show :large1 >/dev/null
+
+'
+
+test_expect_failure 'clone' '
+	git clone file://"$PWD"/.git new
+'
+
+test_expect_failure 'fetch updates' '
+	echo modified >> large1 &&
+	git commit -q -a -m updated &&
+	(
+	cd new &&
+	git fetch --keep # FIXME should not need --keep
+	)
+'
+
+test_expect_failure 'fsck' '
+	git fsck --full
+'
+
+test_expect_failure 'repack' '
+	git repack -ad
+'
+
+test_expect_failure 'tar achiving' '
+	git archive --format=tar HEAD >/dev/null
+'
+
+test_expect_failure 'zip achiving' '
+	git archive --format=zip HEAD >/dev/null
+'
+
 test_done
diff --git a/wrapper.c b/wrapper.c
index 85f09df..d4c0972 100644
--- a/wrapper.c
+++ b/wrapper.c
@@ -9,6 +9,18 @@ static void do_nothing(size_t size)
 
 static void (*try_to_free_routine)(size_t size) = do_nothing;
 
+static void memory_limit_check(size_t size)
+{
+	static int limit = -1;
+	if (limit == -1) {
+		const char *env = getenv("GIT_ALLOC_LIMIT");
+		limit = env ? atoi(env) * 1024 : 0;
+	}
+	if (limit && size > limit)
+		die("attempting to allocate %d over limit %d",
+		    size, limit);
+}
+
 try_to_free_t set_try_to_free_routine(try_to_free_t routine)
 {
 	try_to_free_t old = try_to_free_routine;
@@ -32,7 +44,10 @@ char *xstrdup(const char *str)
 
 void *xmalloc(size_t size)
 {
-	void *ret = malloc(size);
+	void *ret;
+
+	memory_limit_check(size);
+	ret = malloc(size);
 	if (!ret && !size)
 		ret = malloc(1);
 	if (!ret) {
@@ -79,7 +94,10 @@ char *xstrndup(const char *str, size_t len)
 
 void *xrealloc(void *ptr, size_t size)
 {
-	void *ret = realloc(ptr, size);
+	void *ret;
+
+	memory_limit_check(size);
+	ret = realloc(ptr, size);
 	if (!ret && !size)
 		ret = realloc(ptr, 1);
 	if (!ret) {
@@ -95,7 +113,10 @@ void *xrealloc(void *ptr, size_t size)
 
 void *xcalloc(size_t nmemb, size_t size)
 {
-	void *ret = calloc(nmemb, size);
+	void *ret;
+
+	memory_limit_check(size * nmemb);
+	ret = calloc(nmemb, size);
 	if (!ret && (!nmemb || !size))
 		ret = calloc(1, 1);
 	if (!ret) {
-- 
1.7.3.1.256.g2539c.dirty
Previous: Nguyễn Thái Ngọc DuyNext: Nguyễn Thái Ngọc Duy
Message 36 of 48 in “Large blob fixes”
  1. 00/11 Large blob fixesNguyễn Thái Ngọc Duy, Feb 27, 2012
  2. 01/11 Add more large blob test casesNguyễn Thái Ngọc Duy, Feb 27, 2012
  3. Peter BaumannFeb 27, 2012
  4. 02/11 Factor out and export large blob writing code to arbitrary file handleNguyễn Thái Ngọc Duy, Feb 27, 2012
  5. Junio C HamanoFeb 27, 2012
  6. Junio C HamanoFeb 27, 2012
  7. 03/11 cat-file: use streaming interface to print blobsNguyễn Thái Ngọc Duy, Feb 27, 2012
  8. Junio C HamanoFeb 27, 2012
  9. Nguyen Thai Ngoc DuyFeb 28, 2012
  10. 04/11 parse_object: special code path for blobs to avoid putting whole object in memoryNguyễn Thái Ngọc Duy, Feb 27, 2012
  11. 05/11 show: use streaming interface for showing blobsNguyễn Thái Ngọc Duy, Feb 27, 2012
  12. Junio C HamanoFeb 27, 2012
  13. 06/11 index-pack --verify: skip sha-1 collision testNguyễn Thái Ngọc Duy, Feb 27, 2012
  14. 07/11 index-pack: split second pass obj handling into own functionNguyễn Thái Ngọc Duy, Feb 27, 2012
  15. 08/11 index-pack: reduce memory usage when the pack has large blobsNguyễn Thái Ngọc Duy, Feb 27, 2012
  16. 09/11 pack-check: do not unpack blobsNguyễn Thái Ngọc Duy, Feb 27, 2012
  17. 10/11 archive: support streaming large files to a tar archiveNguyễn Thái Ngọc Duy, Feb 27, 2012
  18. 11/11 fsck: use streaming interface for writing lost-found blobsNguyễn Thái Ngọc Duy, Feb 27, 2012
  19. Junio C HamanoFeb 27, 2012
  20. Nguyen Thai Ngoc DuyFeb 28, 2012
  21. 00/10 Large blob fixesNguyễn Thái Ngọc Duy, Mar 4, 2012
  22. 01/10 Add more large blob test casesNguyễn Thái Ngọc Duy, Mar 4, 2012
  23. Junio C HamanoMar 6, 2012
  24. 02/10 streaming: make streaming-write-entry to be more reusableNguyễn Thái Ngọc Duy, Mar 4, 2012
  25. 03/10 cat-file: use streaming interface to print blobsNguyễn Thái Ngọc Duy, Mar 4, 2012
  26. Junio C HamanoMar 4, 2012
  27. Nguyen Thai Ngoc DuyMar 5, 2012
  28. 04/10 parse_object: special code path for blobs to avoid putting whole object in memoryNguyễn Thái Ngọc Duy, Mar 4, 2012
  29. 05/10 show: use streaming interface for showing blobsNguyễn Thái Ngọc Duy, Mar 4, 2012
  30. 06/10 index-pack: split second pass obj handling into own functionNguyễn Thái Ngọc Duy, Mar 4, 2012
  31. 07/10 index-pack: reduce memory usage when the pack has large blobsNguyễn Thái Ngọc Duy, Mar 4, 2012
  32. 08/10 pack-check: do not unpack blobsNguyễn Thái Ngọc Duy, Mar 4, 2012
  33. 09/10 archive: support streaming large files to a tar archiveNguyễn Thái Ngọc Duy, Mar 4, 2012
  34. 10/10 fsck: use streaming interface for writing lost-found blobsNguyễn Thái Ngọc Duy, Mar 4, 2012
  35. 00/11 Large blob fixesNguyễn Thái Ngọc Duy, Mar 5, 2012
  36. 01/11 Add more large blob test casesNguyễn Thái Ngọc Duy, Mar 5, 2012
  37. 02/11 streaming: make streaming-write-entry to be more reusableNguyễn Thái Ngọc Duy, Mar 5, 2012
  38. 03/11 cat-file: use streaming interface to print blobsNguyễn Thái Ngọc Duy, Mar 5, 2012
  39. 04/11 parse_object: special code path for blobs to avoid putting whole object in memoryNguyễn Thái Ngọc Duy, Mar 5, 2012
  40. Junio C HamanoMar 6, 2012
  41. 05/11 show: use streaming interface for showing blobsNguyễn Thái Ngọc Duy, Mar 5, 2012
  42. 06/11 index-pack: split second pass obj handling into own functionNguyễn Thái Ngọc Duy, Mar 5, 2012
  43. 07/11 index-pack: reduce memory usage when the pack has large blobsNguyễn Thái Ngọc Duy, Mar 5, 2012
  44. 08/11 pack-check: do not unpack blobsNguyễn Thái Ngọc Duy, Mar 5, 2012
  45. 09/11 archive: support streaming large files to a tar archiveNguyễn Thái Ngọc Duy, Mar 5, 2012
  46. Junio C HamanoMar 6, 2012
  47. 10/11 fsck: use streaming interface for writing lost-found blobsNguyễn Thái Ngọc Duy, Mar 5, 2012
  48. 11/11 update-server-info: respect core.bigfilethresholdNguyễn Thái Ngọc Duy, Mar 5, 2012

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.