averello/swift-mirror
1
0
Fork 0
mirror of https://github.com/apple/swift.git synced 2025-12-21 12:14:44 +01:00
Code Issues Packages Projects Releases Wiki Activity
33,159 Commits 623 Branches 2,949 Tags
2b393bbcb15c268d9bb33a3bad2b6be21a87512e
Commit Graph

8 Commits

Author SHA1 Message Date
Jordan Rose
a4a4a80ac4 [CMake] Don't pass -sil-serialize-all under SWIFT_STDLIB_ENABLE_RESILIENCE. 
Thanks, Slava!
 2016-02-10 10:29:01 -08:00
Joe Groff
0b54e21629 simd overlay: Use LLVM vector types. 
This brings the David Owens benchmark from http://owensd.io/2015/06/27/performance-xcode7-beta-2.html from parity with simd.h-based C to 3x faster.

Before:

RenderGradient ([UInt32].withUnsafeMutablePointer (SIMD))                 │ 7.035851 │ 6.304739 │ 9.815832 │  1.212 │

After:

RenderGradient ([UInt32].withUnsafeMutablePointer (SIMD))                 │ 2.318357 │ 2.223325 │ 2.697981 │ 0.1490 │

This also addresses rdar://problem/21574425, since Builtin.add_VecNxIntM isn't overflow-checked, and overflow checks really aren't wanted when working with vector types directly.

Reapplying now that Nadav's fixed the ARM64 SelectionDAG issue this exposed before, and Arnold's fixed
yet another SelectionDAG issue exposed after that.

Swift SVN r30006
 2015-07-09 01:20:01 +00:00
Arnold Schwaighofer
8517b7ba08 Revert "Un-XFAIL test on Linux." 
Revert "simd overlay: Use LLVM vector types."

This reverts commit r29922 and r29924.

More arm64 instruction selection errors.

rdar://21703486

Swift SVN r29941
 2015-07-07 16:45:08 +00:00
Joe Groff
9a23fa9040 simd overlay: Use LLVM vector types. 
This brings the David Owens benchmark from http://owensd.io/2015/06/27/performance-xcode7-beta-2.html from parity with simd.h-based C to 3x faster.

Before:

RenderGradient ([UInt32].withUnsafeMutablePointer (SIMD))                 │ 7.035851 │ 6.304739 │ 9.815832 │  1.212 │

After:

RenderGradient ([UInt32].withUnsafeMutablePointer (SIMD))                 │ 2.318357 │ 2.223325 │ 2.697981 │ 0.1490 │

This also addresses rdar://problem/21574425, since Builtin.add_VecNxIntM isn't overflow-checked, and overflow checks really aren't wanted when working with vector types directly.

Reapplying now that Nadav's fixed the ARM64 SelectionDAG issue this exposed before.

Swift SVN r29922
 2015-07-06 22:56:42 +00:00
Joe Groff
e692c87bbf Revert "simd overlay: Use LLVM vector types." 
This reverts commit r29891. It causes an LLVM assertion failure in SelectionDAG on appletvos-arm64.

Swift SVN r29893
 2015-07-02 17:31:34 +00:00
Joe Groff
8103dfb825 simd overlay: Use LLVM vector types. 
This brings the David Owens benchmark from http://owensd.io/2015/06/27/performance-xcode7-beta-2.html from parity with simd.h-based C to 3x faster.

Before:

RenderGradient ([UInt32].withUnsafeMutablePointer (SIMD))                 │ 7.035851 │ 6.304739 │ 9.815832 │  1.212 │

After:

RenderGradient ([UInt32].withUnsafeMutablePointer (SIMD))                 │ 2.318357 │ 2.223325 │ 2.697981 │ 0.1490 │

This also addresses rdar://problem/21574425, since Builtin.add_VecNxIntM isn't overflow-checked, and overflow checks really aren't wanted when working with vector types directly.

Swift SVN r29891
 2015-07-02 17:05:29 +00:00
Jordan Rose
d3291512db Make the 'simd' module build like the rest of the overlays. 
At one point we were considering it to be a replacement rather than an
overlay, but that's not where we are today. We can revisit that later.

Necessary for next commit.

Swift SVN r29438
 2015-06-17 04:47:55 +00:00
Joe Groff
11fcbda0e9 Add Steve Canon's simd module to the stdlib. 
Swift SVN r28088
 2015-05-02 15:46:38 +00:00