index
:
llvm
master
release_1
release_16
release_20
release_21
release_22
release_23
release_24
release_25
release_26
release_27
release_28
release_29
release_30
release_31
release_32
release_33
stable
svn-tags/RELEASE_1
svn-tags/RELEASE_20
svn-tags/RELEASE_21
svn-tags/RELEASE_22
svn-tags/RELEASE_23
svn-tags/RELEASE_24
svn-tags/RELEASE_25
svn-tags/RELEASE_26
svn-tags/RELEASE_27
svn-tags/RELEASE_28
svn-tags/RELEASE_29
svn-tags/RELEASE_30
svn-tags/RELEASE_31
svn-tags/RELEASE_32
testing
http://llvm.org
git repository hosting
about
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
lib
/
Target
/
X86
/
X86InstrSSE.td
Age
Commit message (
Expand
)
Author
2012-12-06
Remove intrinsic specific instructions for (V)MOVQUmr with patterns pointing ...
Craig Topper
2012-12-06
Mark MOVDQ(A/U)rm as ReMaterializable. Mark all MOVDQ(A/U) instructions as ne...
Craig Topper
2012-12-05
Simplified BLEND pattern matching for shuffles.
Elena Demikhovsky
2012-11-16
Use roundps/pd for llvm.ceil, llvm.trunc, llvm.rint, and llvm.nearbyint of ve...
Craig Topper
2012-11-10
Cleanup pcmp(e/i)str(m/i) instruction definitions and load folding support.
Craig Topper
2012-11-06
Remove tailing whitespaces
Michael Liao
2012-10-30
X86 SSE: update rsqrtss and rcpss to use two source operands and
Manman Ren
2012-10-29
Fix PR14204
Michael Liao
2012-10-24
Add custom conversion from v2u32 to v2f32 in 32-bit mode
Michael Liao
2012-10-23
Enable lowering ZERO_EXTEND/ANY_EXTEND to PMOVZX from SSE4.1
Michael Liao
2012-10-10
Add support for FP_ROUND from v2f64 to v2f32
Michael Liao
2012-09-19
Remove code for setting the VEX L-bit as a function of operand size from the ...
Craig Topper
2012-09-19
Add explicit VEX_L tags to all 256-bit instructions. This will allow us to re...
Craig Topper
2012-09-16
The PMOVZXWD family of functions had patterns extends narrow vector types to ...
Nadav Rotem
2012-09-10
Enhance PR11334 fix to support extload from v2f32/v4f32
Michael Liao
2012-09-08
Add instruction selection for ffloor of vectors when SSE4.1 or AVX is enabled.
Craig Topper
2012-09-06
Use iPTR instead of i32 for extract_subvector/insert_subvector index in lower...
Craig Topper
2012-09-06
Add patterns for converting stores of subvector_extracts of lower 128-bits of...
Craig Topper
2012-09-05
Remove some of the patterns added in r163196. Increasing the complexity on in...
Craig Topper
2012-09-05
Add patterns for integer forms of VINSERTF128/VINSERTI128 folded with loads. ...
Craig Topper
2012-09-05
Convert vextracti128/vextractf128 intrinsics to extract_subvector at DAG buil...
Craig Topper
2012-09-01
Typos
Craig Topper
2012-08-31
Clean up AddedComplexity further after adding UseSSEx
Michael Liao
2012-08-31
X86: Fix encoding of 'movd %xmm0, %rax'
Jim Grosbach
2012-08-30
Introduce 'UseSSEx' to force SSE legacy encoding
Michael Liao
2012-08-28
The commutative flag is already correctly set within the multiclass. If we set
Bill Wendling
2012-08-28
Convert V_SETALLONES/AVX_SETALLONES/AVX2_SETALLONES to Post-RA pseudos.
Craig Topper
2012-08-28
Merge AVX_SET0PSY/AVX_SET0PDY/AVX2_SET0 into a single post-RA pseudo.
Craig Topper
2012-08-28
More missing mayLoad flags on AVX multiclasses.
Jakob Stoklund Olesen
2012-08-27
Don't allow vextractf128 to be folded with unaligned stores. We don't fold un...
Craig Topper
2012-08-27
Fold some patterns into instruction definitons so tablegen can infer flags re...
Craig Topper
2012-08-27
Add HasAVX1Only predicate and use it for patterns that have an AVX1 instructi...
Craig Topper
2012-08-24
Add missing mayLoad flags to a large class of AVX *_Int instructions.
Jakob Stoklund Olesen
2012-08-24
Remove some spurious mayLoad = 0 flags.
Jakob Stoklund Olesen
2012-08-19
When unsafe math is used, we can use commutative FMAX and FMIN. In some cases
Nadav Rotem
2012-08-14
fix PR11334
Michael Liao
2012-08-06
Implement proper handling for pcmpistri/pcmpestri intrinsics. Requires custom...
Craig Topper
2012-08-05
Remove custom inserter for MWAIT. It doesn't do anything that couldn't be rep...
Craig Topper
2012-08-01
X86: mark GATHER instructios as mayLoad
Manman Ren
2012-07-30
Give VCVTTPD2DQ priority over CVTTPD2DQ.
Craig Topper
2012-07-30
Fix patterns for CVTTPS2DQ to specify SSE2 instead of SSE1.
Craig Topper
2012-07-30
Fix up patterns for VCVTSS2SD. Specifically give it priority over SSE form. A...
Craig Topper
2012-07-29
Fix load types on intrinsic forms of SS2SD and SD2SS AVX/SSE convert instruct...
Craig Topper
2012-07-29
Move more SSE/AVX convert instruction patterns into their definitions.
Craig Topper
2012-07-28
Fold patterns for some of the SSE/AVX convert instructions into their instruc...
Craig Topper
2012-07-28
Mark some of the SSE/AVX convert instructions as mayLoad/neverHasSideEffects.
Craig Topper
2012-07-28
Make CVTSS2SI instruction definition consistent with CVTSD2SI.
Craig Topper
2012-07-28
Fix up memory load types for SSE scalar convert intrinsic patterns.
Craig Topper
2012-07-26
Remove the last mentions of sub_ss and sub_sd from patterns.
Jakob Stoklund Olesen
2012-07-26
Eliminate sub_ss, sub_sd from broadcast patterns.
Jakob Stoklund Olesen
[next]