Hi,
We've met several performance issues (up to 15%) on Silvermont caused
by the PARTIAL_REG_DEPENDENCY tuning.
Previously discussed here: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57954
Propose removing Silvermont related tune from PARTIAL_REG_DEPENDENCY.
The patch passed bootstrap, make check.
Is it ok for trunk?
Thanks,
Evgeny
2014-10-10 Evgeny Stupachenko <evstu...@gmail.com>
* config/i386/x86-tune.def (X86_TUNE_PARTIAL_REG_DEPENDENCY):
Remove m_SILVERMONT and m_INTEL from the tune.
diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def
index 215c63c..b6b210e 100644
--- a/gcc/config/i386/x86-tune.def
+++ b/gcc/config/i386/x86-tune.def
@@ -58,8 +58,8 @@ DEF_TUNE (X86_TUNE_PARTIAL_REG_DEPENDENCY,
"partial_reg_dependency",
SPECfp regression, while enabling it on K8 brings roughly 2.4% regression
that can be partly masked by careful scheduling of moves. */
DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY, "sse_partial_reg_dependency",
- m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_SILVERMONT
- | m_INTEL | m_AMDFAM10 | m_BDVER | m_GENERIC)
+ m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_AMDFAM10
+ | m_BDVER | m_GENERIC)
/* X86_TUNE_SSE_SPLIT_REGS: Set for machines where the type and dependencies
are resolved on SSE register parts instead of whole registers, so we may
