aboutsummaryrefslogtreecommitdiff
path: root/ChangeLog
diff options
context:
space:
mode:
authorHarsha Jagasia <harsha.jagasia@amd.com>2011-03-04 23:30:08 -0500
committerUlrich Drepper <drepper@gmail.com>2011-03-04 23:30:08 -0500
commit7e4ba49cd365555ddaff2ae8bba7b912464ad6e5 (patch)
treeec4eaf0ea436e74b584daefdceeb4ab66c52728d /ChangeLog
parent13a804de8f3091e8ccd9b650f61becd6e1304227 (diff)
downloadglibc-7e4ba49cd365555ddaff2ae8bba7b912464ad6e5.tar
glibc-7e4ba49cd365555ddaff2ae8bba7b912464ad6e5.tar.gz
glibc-7e4ba49cd365555ddaff2ae8bba7b912464ad6e5.tar.bz2
glibc-7e4ba49cd365555ddaff2ae8bba7b912464ad6e5.zip
Enable SSE2 memset for AMD'supcoming Orochi processor.
This patch enables SSE2 memset for AMD's upcoming Orochi processor. This patch also fixes the following bug: For misaligned blocks larger than > 144 Bytes, memset branches into the integer code path depending on the value of misalignment even if the startup code chooses the SSE2 code path upfront, when multiarch is enabled.
Diffstat (limited to 'ChangeLog')
-rw-r--r--ChangeLog16
1 files changed, 15 insertions, 1 deletions
diff --git a/ChangeLog b/ChangeLog
index c6d652d559..821a58194f 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,17 @@
+2011-03-02 Harsha Jagasia <harsha.jagasia@amd.com>
+ Ulrich Drepper <drepper@gmail.com>
+
+ * sysdeps/x86_64/memset.S: After aligning destination, code
+ branches to different locations depending on the value of
+ misalignment, when multiarch is enabled. Fix this.
+
+2011-03-02 Harsha Jagasia <harsha.jagasia@amd.com>
+
+ * sysdeps/x86_64/cacheinfo.c (init_cacheinfo):
+ Set _x86_64_preferred_memory_instruction for AMD processsors.
+ * sysdeps/x86_64/multiarch/init-arch.c (__init_cpu_features):
+ Set bit_Prefer_SSE_for_memop for AMD processors.
+
2011-03-04 Ulrich Drepper <drepper@gmail.com>
* libio/fmemopen.c (fmemopen): Optimize a bit.
@@ -12,7 +26,7 @@
2011-02-28 Aurelien Jarno <aurelien@aurel32.net>
- * sysdeps/sparc/sparc64/multiarch/memset.S(__bzero): call
+ * sysdeps/sparc/sparc64/multiarch/memset.S(__bzero): Call
__bzero_ultra1 instead of __memset_ultra1.
2011-02-23 Andreas Schwab <schwab@redhat.com>