diff options
| author | Liubov Dmitrieva <liubov.dmitrieva@intel.com> | 2013-06-14 20:46:15 +0200 |
|---|---|---|
| committer | Andreas Jaeger <aj@suse.de> | 2013-06-14 20:46:15 +0200 |
| commit | d086fc7ba0d4b58c9db901780cb8adf415e1f2b5 (patch) | |
| tree | 1008d0651263454a304dd23acf1d7df7d8ac0c2f | |
| parent | 747ef469ffc9c9179ca9d76854167925b4e44346 (diff) | |
| download | glibc-d086fc7ba0d4b58c9db901780cb8adf415e1f2b5.tar glibc-d086fc7ba0d4b58c9db901780cb8adf415e1f2b5.tar.gz glibc-d086fc7ba0d4b58c9db901780cb8adf415e1f2b5.tar.bz2 glibc-d086fc7ba0d4b58c9db901780cb8adf415e1f2b5.zip | |
Set fast unaligned load flag for new Intel microarchitecture
I have small patch for new Intel Silvermont machines.
http://newsroom.intel.com/community/intel_newsroom/blog/2013/05/06/intel-launches-low-power-high-performance-silvermont-microarchitecture
I checked this on my machine and see that strcpy, ... unaligned
versions are faster than ssse3 versions.
| -rw-r--r-- | ChangeLog | 5 | ||||
| -rw-r--r-- | sysdeps/x86_64/multiarch/init-arch.c | 7 |
2 files changed, 12 insertions, 0 deletions
@@ -1,3 +1,8 @@ +2013-06-14 Liubov Dmitrieva <liubov.dmitrieva@intel.com> + + * sysdeps/x86_64/multiarch/init-arch.c (__init_cpu_features): + Set bit_Fast_Unaligned_Load for Intel Silvermont architecture. + 2013-06-14 Siddhesh Poyarekar <siddhesh@redhat.com> H.J. Lu <hjl.tools@gmail.com> diff --git a/sysdeps/x86_64/multiarch/init-arch.c b/sysdeps/x86_64/multiarch/init-arch.c index 7daaf46099..9524aeea18 100644 --- a/sysdeps/x86_64/multiarch/init-arch.c +++ b/sysdeps/x86_64/multiarch/init-arch.c @@ -78,6 +78,13 @@ __init_cpu_features (void) __cpu_features.feature[index_Slow_BSF] |= bit_Slow_BSF; break; + case 0x37: + /* Unaligned load versions are faster than SSSE3 + on Silvermont. */ + __cpu_features.feature[index_Fast_Unaligned_Load] + |= bit_Fast_Unaligned_Load; + break; + default: /* Unknown family 0x06 processors. Assuming this is one of Core i3/i5/i7 processors if AVX is available. */ |