summaryrefslogtreecommitdiffstats
path: root/meta/recipes-devtools/git/files/CVE-2018-11233.patch
diff options
context:
space:
mode:
authorSinan Kaya <okaya@kernel.org>2018-10-05 03:55:15 +0000
committerRichard Purdie <richard.purdie@linuxfoundation.org>2018-10-18 11:08:53 +0100
commitfbc735796f473d85055feb9c3927cd5293061bf6 (patch)
tree14e95a758a49ca93b19aeea965e4e4e4383fdbd8 /meta/recipes-devtools/git/files/CVE-2018-11233.patch
parent97ee1f80870745bf6c542ee2f184c9e468672714 (diff)
downloadpoky-fbc735796f473d85055feb9c3927cd5293061bf6.tar.gz
git: CVE-2018-11233
* CVE-2018-11233 Code to sanity-check pathnames on NTFS can result in reading out-of-bounds memory. Affects < 2.17.1 CVE: CVE-2018-11233 Ref: https://bugzilla.redhat.com/show_bug.cgi?id=1583888 (From OE-Core rev: d145f605c274386baf0dde023f15cddf37523f3b) Signed-off-by: Sinan Kaya <okaya@kernel.org> Signed-off-by: Armin Kuster <akuster808@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Diffstat (limited to 'meta/recipes-devtools/git/files/CVE-2018-11233.patch')
-rw-r--r--meta/recipes-devtools/git/files/CVE-2018-11233.patch44
1 files changed, 44 insertions, 0 deletions
diff --git a/meta/recipes-devtools/git/files/CVE-2018-11233.patch b/meta/recipes-devtools/git/files/CVE-2018-11233.patch
new file mode 100644
index 0000000000..f4468cf2fd
--- /dev/null
+++ b/meta/recipes-devtools/git/files/CVE-2018-11233.patch
@@ -0,0 +1,44 @@
1From 014281e62b7920a6d710a85089e00ca012b0744c Mon Sep 17 00:00:00 2001
2From: Jeff King <peff@peff.net>
3Date: Sun, 13 May 2018 12:09:42 -0400
4Subject: [PATCH] is_ntfs_dotgit: use a size_t for traversing string
5
6We walk through the "name" string using an int, which can
7wrap to a negative value and cause us to read random memory
8before our array (e.g., by creating a tree with a name >2GB,
9since "int" is still 32 bits even on most 64-bit platforms).
10Worse, this is easy to trigger during the fsck_tree() check,
11which is supposed to be protecting us from malicious
12garbage.
13
14Note one bit of trickiness in the existing code: we
15sometimes assign -1 to "len" at the end of the loop, and
16then rely on the "len++" in the for-loop's increment to take
17it back to 0. This is still legal with a size_t, since
18assigning -1 will turn into SIZE_MAX, which then wraps
19around to 0 on increment.
20
21Signed-off-by: Jeff King <peff@peff.net>
22CVE: CVE-2018-11233
23Upstream-Status: Backport[https://github.com/git/git/commit/11a9f4d807a0d71dc6eff51bb87baf4ca2cccf1d]
24Signed-off-by: Sinan Kaya <okaya@kernel.org>
25---
26 path.c | 2 +-
27 1 file changed, 1 insertion(+), 1 deletion(-)
28
29diff --git a/path.c b/path.c
30index da8b65573..d31c795ff 100644
31--- a/path.c
32+++ b/path.c
33@@ -1305,7 +1305,7 @@ static int only_spaces_and_periods(const char *path, size_t len, size_t skip)
34
35 int is_ntfs_dotgit(const char *name)
36 {
37- int len;
38+ size_t len;
39
40 for (len = 0; ; len++)
41 if (!name[len] || name[len] == '\\' || is_dir_sep(name[len])) {
42--
432.19.0
44