summaryrefslogtreecommitdiffstats
path: root/meta/recipes-core/busybox
diff options
context:
space:
mode:
authorAndrej Valek <andrej.valek@siemens.com>2018-03-02 09:13:51 +0100
committerRichard Purdie <richard.purdie@linuxfoundation.org>2018-03-04 11:35:42 +0000
commitd7d3d341d24687c10b7bec234cea505865ded9cf (patch)
tree7b5311546aaa337f8030803e082cc8b695afa312 /meta/recipes-core/busybox
parent615e700f7ace78b04cdfb9f5a7f70cf6335449fa (diff)
downloadpoky-d7d3d341d24687c10b7bec234cea505865ded9cf.tar.gz
busybox: fix CVE-2017-15873
(From OE-Core rev: 91829d07cdeb42e213500e053e20a9db68848d2e) Signed-off-by: Radovan Scasny <radovan.scasny@siemens.com> Signed-off-by: Andrej Valek <andrej.valek@siemens.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Diffstat (limited to 'meta/recipes-core/busybox')
-rw-r--r--meta/recipes-core/busybox/busybox/CVE-2017-15873.patch95
-rw-r--r--meta/recipes-core/busybox/busybox_1.27.2.bb1
2 files changed, 96 insertions, 0 deletions
diff --git a/meta/recipes-core/busybox/busybox/CVE-2017-15873.patch b/meta/recipes-core/busybox/busybox/CVE-2017-15873.patch
new file mode 100644
index 0000000000..5a027c9bcc
--- /dev/null
+++ b/meta/recipes-core/busybox/busybox/CVE-2017-15873.patch
@@ -0,0 +1,95 @@
1busybox-1.27.2: Fix CVE-2017-15873
2
3[No upstream tracking] -- https://bugs.busybox.net/show_bug.cgi?id=10431
4
5bunzip2: fix runCnt overflow
6
7The get_next_block function in archival/libarchive/decompress_bunzip2.c
8in BusyBox 1.27.2 has an Integer Overflow that may lead to a write
9access violation.
10
11Upstream-Status: Backport [https://git.busybox.net/busybox/commit/?id=0402cb32df015d9372578e3db27db47b33d5c7b0]
12CVE: CVE-2017-15873
13bug: 10431
14Signed-off-by: Radovan Scasny <radovan.scasny@siemens.com>
15
16diff --git a/archival/libarchive/decompress_bunzip2.c b/archival/libarchive/decompress_bunzip2.c
17index 7cd18f5..bec89ed 100644
18--- a/archival/libarchive/decompress_bunzip2.c
19+++ b/archival/libarchive/decompress_bunzip2.c
20@@ -156,15 +156,15 @@ static unsigned get_bits(bunzip_data *bd, int bits_wanted)
21 static int get_next_block(bunzip_data *bd)
22 {
23 struct group_data *hufGroup;
24- int dbufCount, dbufSize, groupCount, *base, *limit, selector,
25- i, j, runPos, symCount, symTotal, nSelectors, byteCount[256];
26- int runCnt = runCnt; /* for compiler */
27+ int groupCount, *base, *limit, selector,
28+ i, j, symCount, symTotal, nSelectors, byteCount[256];
29 uint8_t uc, symToByte[256], mtfSymbol[256], *selectors;
30 uint32_t *dbuf;
31 unsigned origPtr, t;
32+ unsigned dbufCount, runPos;
33+ unsigned runCnt = runCnt; /* for compiler */
34
35 dbuf = bd->dbuf;
36- dbufSize = bd->dbufSize;
37 selectors = bd->selectors;
38
39 /* In bbox, we are ok with aborting through setjmp which is set up in start_bunzip */
40@@ -187,7 +187,7 @@ static int get_next_block(bunzip_data *bd)
41 it didn't actually work. */
42 if (get_bits(bd, 1)) return RETVAL_OBSOLETE_INPUT;
43 origPtr = get_bits(bd, 24);
44- if ((int)origPtr > dbufSize) return RETVAL_DATA_ERROR;
45+ if (origPtr > bd->dbufSize) return RETVAL_DATA_ERROR;
46
47 /* mapping table: if some byte values are never used (encoding things
48 like ascii text), the compression code removes the gaps to have fewer
49@@ -435,7 +435,14 @@ static int get_next_block(bunzip_data *bd)
50 symbols, but a run of length 0 doesn't mean anything in this
51 context). Thus space is saved. */
52 runCnt += (runPos << nextSym); /* +runPos if RUNA; +2*runPos if RUNB */
53- if (runPos < dbufSize) runPos <<= 1;
54+//The 32-bit overflow of runCnt wasn't yet seen, but probably can happen.
55+//This would be the fix (catches too large count way before it can overflow):
56+// if (runCnt > bd->dbufSize) {
57+// dbg("runCnt:%u > dbufSize:%u RETVAL_DATA_ERROR",
58+// runCnt, bd->dbufSize);
59+// return RETVAL_DATA_ERROR;
60+// }
61+ if (runPos < bd->dbufSize) runPos <<= 1;
62 goto end_of_huffman_loop;
63 }
64
65@@ -445,14 +452,15 @@ static int get_next_block(bunzip_data *bd)
66 literal used is the one at the head of the mtfSymbol array.) */
67 if (runPos != 0) {
68 uint8_t tmp_byte;
69- if (dbufCount + runCnt > dbufSize) {
70- dbg("dbufCount:%d+runCnt:%d %d > dbufSize:%d RETVAL_DATA_ERROR",
71- dbufCount, runCnt, dbufCount + runCnt, dbufSize);
72+ if (dbufCount + runCnt > bd->dbufSize) {
73+ dbg("dbufCount:%u+runCnt:%u %u > dbufSize:%u RETVAL_DATA_ERROR",
74+ dbufCount, runCnt, dbufCount + runCnt, bd->dbufSize);
75 return RETVAL_DATA_ERROR;
76 }
77 tmp_byte = symToByte[mtfSymbol[0]];
78 byteCount[tmp_byte] += runCnt;
79- while (--runCnt >= 0) dbuf[dbufCount++] = (uint32_t)tmp_byte;
80+ while ((int)--runCnt >= 0)
81+ dbuf[dbufCount++] = (uint32_t)tmp_byte;
82 runPos = 0;
83 }
84
85@@ -466,7 +474,7 @@ static int get_next_block(bunzip_data *bd)
86 first symbol in the mtf array, position 0, would have been handled
87 as part of a run above. Therefore 1 unused mtf position minus
88 2 non-literal nextSym values equals -1.) */
89- if (dbufCount >= dbufSize) return RETVAL_DATA_ERROR;
90+ if (dbufCount >= bd->dbufSize) return RETVAL_DATA_ERROR;
91 i = nextSym - 1;
92 uc = mtfSymbol[i];
93
94--
95cgit v0.12
diff --git a/meta/recipes-core/busybox/busybox_1.27.2.bb b/meta/recipes-core/busybox/busybox_1.27.2.bb
index 6c1f4888cf..36a6342aaf 100644
--- a/meta/recipes-core/busybox/busybox_1.27.2.bb
+++ b/meta/recipes-core/busybox/busybox_1.27.2.bb
@@ -43,6 +43,7 @@ SRC_URI = "http://www.busybox.net/downloads/busybox-${PV}.tar.bz2;name=tarball \
43 file://runlevel \ 43 file://runlevel \
44 file://makefile-libbb-race.patch \ 44 file://makefile-libbb-race.patch \
45 file://CVE-2011-5325.patch \ 45 file://CVE-2011-5325.patch \
46 file://CVE-2017-15873.patch \
46 file://busybox-CVE-2017-16544.patch \ 47 file://busybox-CVE-2017-16544.patch \
47" 48"
48SRC_URI_append_libc-musl = " file://musl.cfg " 49SRC_URI_append_libc-musl = " file://musl.cfg "