[Sparc] Make sure that we really don't emit quad-precision unless the "hard-quad...
authorMark Kettenis <mark.kettenis@xs4all.nl>
Thu, 19 May 2022 00:05:56 +0000 (20:05 -0400)
committerBrad Smith <brad@comstyle.com>
Thu, 19 May 2022 00:11:58 +0000 (20:11 -0400)
commit3d869c88bb91d6e5a973a493068c1af689f13445
treef8afbbde52a2e44a7e2af24684b1ef36edc22ddc
parent565f5dfa1f3ea52637fcfec4babd4a406d2e4c03
[Sparc] Make sure that we really don't emit quad-precision unless the "hard-quad-float" feature is available

Make sure that we really don't emit quad-precision unless the "hard-quad-float"
feature is available. Add missing replacement instruction patterns that are
needed to emit alternative code for conditional moves of quad-precision floats.

Test from koakuma.

Reviewed By: koakuma

Differential Revision: https://reviews.llvm.org/D119104
llvm/lib/Target/Sparc/SparcISelLowering.cpp
llvm/lib/Target/Sparc/SparcInstr64Bit.td
llvm/lib/Target/Sparc/SparcInstrInfo.td
llvm/test/CodeGen/SPARC/hard-quad-float.ll [new file with mode: 0644]