* x86-tune.def (partial_reg_stall): Disable for CoreI7 and newer.

	(sse_typeless_stores): Enable for core
	(sse_load0_by_pxor): Likewise.
	(four_jump_limit): Disable for core.
	(pad_returns): Likewise.
	(avoid_vector_decode): Likewise.
	(fuse_cmp_and_branch): Enable for cores.
	* i386.c (x86_accumulate_outgoing_args): Disable for cores.

git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@202813 138bc75d-0d04-0410-961f-82ee72b054a4

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 6f7454e..a4dbd39 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,14 @@
+2013-09-20  Jan Hubicka  <jh@suse.cz>
+
+       * x86-tune.def (partial_reg_stall): Disable for CoreI7 and newer.
+       (sse_typeless_stores): Enable for core
+       (sse_load0_by_pxor): Likewise.
+       (four_jump_limit): Disable for core.
+       (pad_returns): Likewise.
+       (avoid_vector_decode): Likewise.
+       (fuse_cmp_and_branch): Enable for cores.
+       * i386.c (x86_accumulate_outgoing_args): Disable for cores.
+
 2013-09-20  John David Anglin  <danglin@gcc.gnu.org>
 
        PR middle-end/56791

diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index 46c37d8..0c10e73 100644 (file)
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -1899,7 +1899,7 @@ static unsigned int initial_ix86_arch_features[X86_ARCH_LAST] = {
 };
 
 static const unsigned int x86_accumulate_outgoing_args
-  = m_PPRO | m_P4_NOCONA | m_ATOM | m_SLM | m_CORE_ALL | m_AMD_MULTIPLE | m_GENERIC;
+  = m_PPRO | m_P4_NOCONA | m_ATOM | m_SLM | m_AMD_MULTIPLE | m_GENERIC;
 
 static const unsigned int x86_arch_always_fancy_math_387
   = m_PENT | m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_ATOM | m_SLM | m_AMD_MULTIPLE | m_GENERIC;

diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def
index fc19df1..5f2e897 100644 (file)
--- a/gcc/config/i386/x86-tune.def
+++ b/gcc/config/i386/x86-tune.def
@@ -52,7 +52,7 @@ DEF_TUNE (X86_TUNE_MOVX, "movx",
    and can happen in caller/callee saving sequences.  */
 DEF_TUNE (X86_TUNE_PARTIAL_REG_STALL, "partial_reg_stall", m_PPRO)
 DEF_TUNE (X86_TUNE_PARTIAL_FLAG_REG_STALL, "partial_flag_reg_stall",
-          m_CORE_ALL | m_GENERIC)
+          m_CORE2 | m_GENERIC)
 /* X86_TUNE_LCP_STALL: Avoid an expensive length-changing prefix stall
  * on 16-bit immediate moves into memory on Core2 and Corei7.  */
 DEF_TUNE (X86_TUNE_LCP_STALL, "lcp_stall", m_CORE_ALL | m_GENERIC)
@@ -125,8 +125,10 @@ DEF_TUNE (X86_TUNE_SSE_PACKED_SINGLE_INSN_OPTIMAL, "sse_packed_single_insn_optim
    maintain just lower part of scalar values in proper format leaving the
    upper part undefined.  */
 DEF_TUNE (X86_TUNE_SSE_SPLIT_REGS, "sse_split_regs", m_ATHLON_K8)
-DEF_TUNE (X86_TUNE_SSE_TYPELESS_STORES, "sse_typeless_stores", m_AMD_MULTIPLE)
-DEF_TUNE (X86_TUNE_SSE_LOAD0_BY_PXOR, "sse_load0_by_pxor", m_PPRO | m_P4_NOCONA)
+DEF_TUNE (X86_TUNE_SSE_TYPELESS_STORES, "sse_typeless_stores",
+         m_AMD_MULTIPLE | m_CORE_ALL)
+DEF_TUNE (X86_TUNE_SSE_LOAD0_BY_PXOR, "sse_load0_by_pxor",
+         m_PPRO | m_P4_NOCONA | m_CORE_ALL)
 DEF_TUNE (X86_TUNE_MEMORY_MISMATCH_STALL, "memory_mismatch_stall",
           m_P4_NOCONA | m_CORE_ALL | m_ATOM | m_SLM | m_AMD_MULTIPLE | m_GENERIC)
 DEF_TUNE (X86_TUNE_PROLOGUE_USING_MOVE, "prologue_using_move", 
@@ -144,7 +146,7 @@ DEF_TUNE (X86_TUNE_INTER_UNIT_CONVERSIONS, "inter_unit_conversions",
 /* X86_TUNE_FOUR_JUMP_LIMIT: Some CPU cores are not able to predict more
    than 4 branch instructions in the 16 byte window.  */
 DEF_TUNE (X86_TUNE_FOUR_JUMP_LIMIT, "four_jump_limit",
-          m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_ATOM | m_SLM| m_AMD_MULTIPLE 
+          m_PPRO | m_P4_NOCONA | m_ATOM | m_SLM | m_AMD_MULTIPLE 
           | m_GENERIC)
 DEF_TUNE (X86_TUNE_SCHEDULE, "schedule",
           m_PENT | m_PPRO | m_CORE_ALL | m_ATOM | m_SLM | m_K6_GEODE 
@@ -154,13 +156,13 @@ DEF_TUNE (X86_TUNE_USE_BT, "use_bt",
 DEF_TUNE (X86_TUNE_USE_INCDEC, "use_incdec",
           ~(m_P4_NOCONA | m_CORE_ALL | m_ATOM | m_SLM | m_GENERIC))
 DEF_TUNE (X86_TUNE_PAD_RETURNS, "pad_returns",
-          m_CORE_ALL | m_AMD_MULTIPLE | m_GENERIC)
+          m_AMD_MULTIPLE | m_GENERIC)
 DEF_TUNE (X86_TUNE_PAD_SHORT_FUNCTION, "pad_short_function", m_ATOM)
 DEF_TUNE (X86_TUNE_EXT_80387_CONSTANTS, "ext_80387_constants",
           m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_ATOM | m_SLM | m_K6_GEODE
           | m_ATHLON_K8 | m_GENERIC)
 DEF_TUNE (X86_TUNE_AVOID_VECTOR_DECODE, "avoid_vector_decode",
-          m_CORE_ALL | m_K8 | m_GENERIC)
+          m_K8 | m_GENERIC)
 /* X86_TUNE_PROMOTE_HIMODE_IMUL: Modern CPUs have same latency for HImode
    and SImode multiply, but 386 and 486 do HImode multiply faster.  */
 DEF_TUNE (X86_TUNE_PROMOTE_HIMODE_IMUL, "promote_himode_imul",
@@ -193,7 +195,7 @@ DEF_TUNE (X86_TUNE_USE_VECTOR_CONVERTS, "use_vector_converts", m_AMDFAM10)
 /* X86_TUNE_FUSE_CMP_AND_BRANCH: Fuse a compare or test instruction
    with a subsequent conditional jump instruction into a single
    compare-and-branch uop.  */
-DEF_TUNE (X86_TUNE_FUSE_CMP_AND_BRANCH, "fuse_cmp_and_branch", m_BDVER)
+DEF_TUNE (X86_TUNE_FUSE_CMP_AND_BRANCH, "fuse_cmp_and_branch", m_BDVER | m_CORE_ALL)
 /* X86_TUNE_OPT_AGU: Optimize for Address Generation Unit. This flag
    will impact LEA instruction selection. */
 DEF_TUNE (X86_TUNE_OPT_AGU, "opt_agu", m_ATOM | m_SLM)