glibc/SOURCES/glibc-RHEL-49489-1.patch

52 lines
2.0 KiB
Diff

commit 9b7651410375ec8848a1944992d663d514db4ba7
Author: Stefan Liebler <stli@linux.ibm.com>
Date: Thu Jul 11 11:28:53 2024 +0200
s390x: Fix segfault in wcsncmp [BZ #31934]
The z13/vector-optimized wcsncmp implementation segfaults if n=1
and there is only one character (equal on both strings) before
the page end. Then it loads and compares one character and misses
to check n again. The following load fails.
This patch removes the extra load and compare of the first character
and just start with the loop which uses vector-load-to-block-boundary.
This code-path also checks n.
With this patch both tests are passing:
- the simplified one mentioned in the bugzilla 31934
- the full one in Florian Weimer's patch:
"manual: Document a GNU extension for strncmp/wcsncmp"
(https://patchwork.sourceware.org/project/glibc/patch/874j9eml6y.fsf@oldenburg.str.redhat.com/):
On s390x-linux-gnu (z16), the new wcsncmp test fails due to bug 31934.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
diff --git a/sysdeps/s390/wcsncmp-vx.S b/sysdeps/s390/wcsncmp-vx.S
index bf6dfa6bc2..8b081567a2 100644
--- a/sysdeps/s390/wcsncmp-vx.S
+++ b/sysdeps/s390/wcsncmp-vx.S
@@ -59,14 +59,7 @@ ENTRY(WCSNCMP_Z13)
sllg %r4,%r4,2 /* Convert character-count to byte-count. */
locgrne %r4,%r1 /* Use max byte-count, if bit 0/1 was one. */
- /* Check first character without vector load. */
- lghi %r5,4 /* current_len = 4 bytes. */
- /* Check s1/2[0]. */
- lt %r0,0(%r2)
- l %r1,0(%r3)
- je .Lend_cmp_one_char
- crjne %r0,%r1,.Lend_cmp_one_char
-
+ lghi %r5,0 /* current_len = 0 bytes. */
.Lloop:
vlbb %v17,0(%r5,%r3),6 /* Load s2 to block boundary. */
vlbb %v16,0(%r5,%r2),6 /* Load s1 to block boundary. */
@@ -167,7 +160,6 @@ ENTRY(WCSNCMP_Z13)
srl %r4,2 /* And convert it to character-index. */
vlgvf %r0,%v16,0(%r4) /* Load character-values. */
vlgvf %r1,%v17,0(%r4)
-.Lend_cmp_one_char:
cr %r0,%r1
je .Lend_equal
lghi %r2,1