projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d4c1f22) | patch
Fix test-rsqrt.mlir to accept AMD's approximation of rsqrt as well
authorMehdi Amini <joker.eph@gmail.com>
Fri, 8 Oct 2021 01:21:24 +0000 (01:21 +0000)
committerMehdi Amini <joker.eph@gmail.com>
Fri, 8 Oct 2021 04:24:24 +0000 (04:24 +0000)
commit82cd8b81aad5f33b054e6576b5ee85ceb9063107
treeb0e968ad9bc15ee268a240f371d05a464a61330ftree | snapshot
parentd4c1f222f2afe40567fb2602955cf5930ce2a5f2commit | diff
Fix test-rsqrt.mlir to accept AMD's approximation of rsqrt as well

These kind of function can behave differently on these X86 chips, there
isn't really "one true answer" so we'll accept both.

Also remove spurious passes and use mattr="avx" to match the instruction
used here.

Differential Revision: https://reviews.llvm.org/D111373
mlir/test/Integration/Dialect/Vector/CPU/X86Vector/test-rsqrt.mlir diff | blob | history
Domain: System / Toolchain;