Commit Graph

3 Commits

Author SHA1 Message Date
Sjoerd Meijer d5f1131c81 [AArch64] Default to zero-cycle-zeroing FP registers
It is generally beneficial to prefer "movi d0, #0" over "fmov s0, wzr" as this
is most efficient across all cores; it is recognised as a zeroing idiom. For
newer cores, fmov instructions can also be eliminated early and there is no
difference with movi, but some implementations lack this so is not true for
other/older cores. Thus this standardises on using movi as this should always
gives the same or better performance than the fmov with wzr.

Differential Revision: https://reviews.llvm.org/D99586
2021-04-06 09:47:50 +01:00
Chad Rosier 384ade9b11 [AArch64] Add FileCheck that was missing from test in r232967.
llvm-svn: 233013
2015-03-23 20:25:15 +00:00
Chad Rosier affe181b39 [AArch64] Enable rematerialization of float 0 values.
Patch by Geoff Berry<gberry@codeaurora.org>.

llvm-svn: 232967
2015-03-23 17:19:34 +00:00