[PATCH] D89697: [X86] Implement smarter instruction lowering for FP_TO_UINT from vXf32/vXf64 to vXi32 for SSE2 and AVX2 by using the exact semantic of the CVTTPS2SI instruction.

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added a comment.

In D89697#2893519 <https://reviews.llvm.org/D89697#2893519>, @aeubanks wrote:

> Not yet, it's hard to figure out exactly which file is causing the problem.
> Does this patch exploit UB? Or should it just work for all inputs?

If the floating point value is negative (apart from -0.0) then there can be diffs - but given its fp_to_uint then yes that ub.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D89697/new/

https://reviews.llvm.org/D89697