projects / gcc.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5e176a6) | patch
[AArch64] Remove unneeded FSUB alternatives and add a new one
authorRichard Sandiford <richard.sandiford@arm.com>
Thu, 15 Aug 2019 08:34:40 +0000 (08:34 +0000)
committerRichard Sandiford <rsandifo@gcc.gnu.org>
Thu, 15 Aug 2019 08:34:40 +0000 (08:34 +0000)
commit2ae21bd133c357fcd7b6e06dc7d7d9e0660abe2c
tree8a597f87da62736b9b03cf048fdaade3f756d26ftree
parent5e176a613ef2eda92aa65736763a562dc42a50fecommit | diff
[AArch64] Remove unneeded FSUB alternatives and add a new one

The floating-point subtraction patterns don't need to handle
subtraction of constants, since those go through the addition
patterns instead.  There was a missing MOVPRFX alternative for
FSUBR though.

2019-08-15  Richard Sandiford  <richard.sandiford@arm.com>

gcc/
* config/aarch64/aarch64-sve.md (*sub<SVE_F:mode>3): Remove immediate
FADD and FSUB alternatives.  Add a MOVPRFX alternative for FSUBR.

From-SVN: r274514
gcc/ChangeLog diff | blob | history
gcc/config/aarch64/aarch64-sve.md diff | blob | history