FW: D11316: [X86] -- Fix fptoui i64 conversions for IA32 (performance and correctness) -- Request for code review

[Bodart, Mitch L]

Hello,

This is a request for volunteers to review the D11316 changes.

It fixes several issues I uncovered when looking at fptoui lowering for X86.

There are still some issues with half float, and with ui/sitofp (f80 with avx512),
but those will be addressed separately.

best regards,
Mitch Bodart

-----Original Message-----
From: Mitch Bodart [mailto:mitch.l.bodart at intel.com] 
Sent: Friday, July 17, 2015 4:31 PM
To: Bodart, Mitch L
Cc: llvm-commits at cs.uiuc.edu
Subject: [PATCH] D11316: [X86] -- Fix fptoui i64 conversions for IA32 (performance and correctness)

mbodart created this revision.
mbodart added a subscriber: llvm-commits.

Fixes assertion failures when using fptosi or fptoui with f80 and AVX512.  The operation actions need to be left as Custom, not Legal.  The legality for f32/f64 is handled by having FP_TO_INTHelper do nothing for those cases.

Suppresses use of the MSVC ftol2 library function for fptoui i64.  That function performs a conversion to *signed* i64, so the results were incorrect for source values >= 2^63.  I didn't rip out the now dead references to ftol2, but that can be done with a small followup change set if desired.

Implements an inline sequence for fptoui i64 for 32-bit X86.  This is mostly in FP_TO_INTHelper, replacing the ftol2 usage on Windows, and replacing the calls to fixuns{sf,df,xf}di for non-windows.  Improves performance by 6X under SSE3, 3X otherwise.

http://reviews.llvm.org/D11316

Files:
  lib/Target/X86/X86ISelLowering.cpp
  test/CodeGen/X86/pr17631.ll
  test/CodeGen/X86/scalar-fp-to-i64.ll
  test/CodeGen/X86/win_ftol2.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D11316.30048.patch
Type: text/x-patch
Size: 17980 bytes
Desc: D11316.30048.patch
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150729/b0253ddd/attachment.bin>