https://issues.dlang.org/show_bug.cgi?id=17484

--- Comment #1 from github-bugzilla@puremagic.com ---
Commit pushed to master at https://github.com/dlang/dmd

https://github.com/dlang/dmd/commit/1f11aa0eb8f6087b7dbadeb770e4526ec9808ccc fix Issue 17484 - high penalty for AVX-256 instructions with AVX-128 regs

- as the upper 128-bits are no longer zero, the CPU will save/restore
  them when that register is used with legacy SSE instructions
- avoid using vbroadcastsd which is a AVX-256 only instruction to
  initialize 128-bit XMM vectors

--

https://issues.dlang.org/show_bug.cgi?id=17484

github-bugzilla@puremagic.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--

https://issues.dlang.org/show_bug.cgi?id=17484

--- Comment #2 from github-bugzilla@puremagic.com ---
Commit pushed to newCTFE at https://github.com/dlang/dmd

https://github.com/dlang/dmd/commit/1f11aa0eb8f6087b7dbadeb770e4526ec9808ccc fix Issue 17484 - high penalty for AVX-256 instructions with AVX-128 regs

--

https://issues.dlang.org/show_bug.cgi?id=17484

--- Comment #3 from github-bugzilla@puremagic.com ---
Commit pushed to stable at https://github.com/dlang/dmd

https://github.com/dlang/dmd/commit/1f11aa0eb8f6087b7dbadeb770e4526ec9808ccc fix Issue 17484 - high penalty for AVX-256 instructions with AVX-128 regs

--