git
/
gcc
mirror of git://gcc.gnu.org/git/gcc.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

x86-tune.def (X86_TUNE_SLOW_IMUL_IMM32_MEM, [...]): Keep enabled only for K8 and AMDFAM10. 

* config/i386/x86-tune.def (X86_TUNE_SLOW_IMUL_IMM32_MEM,
	X86_TUNE_SLOW_IMUL_IMM8): Keep enabled only for K8 and AMDFAM10.
	(X86_TUNE_USE_VECTOR_FP_CONVERTS): Disable for generic.

From-SVN: r203876
This commit is contained in:
Jan Hubicka 2013-10-20 13:18:12 +02:00 committed by Jan Hubicka
parent 603b781be4
commit 915e8e6eec
2 changed files with 9 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
gcc/ChangeLog
View File

@ -1,3 +1,9 @@
2013-10-20 Jan Hubicka <jh@suse.cz>
* config/i386/x86-tune.def (X86_TUNE_SLOW_IMUL_IMM32_MEM,
X86_TUNE_SLOW_IMUL_IMM8): Keep enabled only for K8 and AMDFAM10.
(X86_TUNE_USE_VECTOR_FP_CONVERTS): Disable for generic.
2013-10-20 Richard Sandiford <rdsandiford@googlemail.com>
* config/mips/mips.h (ISA_HAS_WSBH): Define.

6
gcc/config/i386/x86-tune.def
View File

@ -327,13 +327,13 @@ DEF_TUNE (X86_TUNE_PROMOTE_HIMODE_IMUL, "promote_himode_imul",
vector path on AMD machines.
FIXME: Do we need to enable this for core? */
DEF_TUNE (X86_TUNE_SLOW_IMUL_IMM32_MEM, "slow_imul_imm32_mem",
m_CORE_ALL | m_K8 | m_AMDFAM10 | m_BDVER | m_BTVER | m_GENERIC)
m_K8 | m_AMDFAM10)
/* X86_TUNE_SLOW_IMUL_IMM8: Imul of 8-bit constant is vector path on AMD
machines.
FIXME: Do we need to enable this for core? */
DEF_TUNE (X86_TUNE_SLOW_IMUL_IMM8, "slow_imul_imm8",
m_CORE_ALL | m_K8 | m_AMDFAM10 | m_BDVER | m_BTVER | m_GENERIC)
m_K8 | m_AMDFAM10)
/* X86_TUNE_MOVE_M1_VIA_OR: On pentiums, it is faster to load -1 via OR
than a MOV. */
@ -352,7 +352,7 @@ DEF_TUNE (X86_TUNE_NOT_VECTORMODE, "not_vectormode", m_K6)
from FP to FP. This form of instructions avoids partial write to the
destination. */
DEF_TUNE (X86_TUNE_USE_VECTOR_FP_CONVERTS, "use_vector_fp_converts",
m_AMDFAM10 | m_GENERIC)
m_AMDFAM10)
/* X86_TUNE_USE_VECTOR_CONVERTS: Prefer vector packed SSE conversion
from integer to FP. */