We need to ensure that these end up on FPR to allow imported patterns to select them.
This will also ensure that we get good regbank selection when dealing with instructions like G_PHI/G_LOAD/G_STORE which deduce their banks from their uses/users.
Paths
| Differential D108260
[AArch64][GlobalISel] Mark G_FMINNUM/G_FMAXNUM as floating point opcodes ClosedPublic Authored by paquette on Aug 17 2021, 5:43 PM.
Details Summary We need to ensure that these end up on FPR to allow imported patterns to select them. This will also ensure that we get good regbank selection when dealing with instructions like G_PHI/G_LOAD/G_STORE which deduce their banks from their uses/users.
Diff Detail
Event TimelineHerald added subscribers: hiraditya, kristof.beyls, rovka. · View Herald TranscriptAug 17 2021, 5:43 PM paquette added a parent revision: D108259: [AArch64][GlobalISel] Legalize scalar G_FMINNUM + G_FMAXNUM.Aug 17 2021, 5:43 PM Comment Actions LGTM
This revision is now accepted and ready to land.Aug 17 2021, 5:53 PM
Closed by commit rG3d91d5b757ff: [AArch64][GlobalISel] Mark G_FMINNUM/G_FMAXNUM as floating point opcodes (authored by paquette). · Explain WhyAug 18 2021, 1:32 PM This revision was automatically updated to reflect the committed changes.
Revision Contents
Diff 367316 llvm/lib/Target/AArch64/GISel/AArch64RegisterBankInfo.cpp
llvm/test/CodeGen/AArch64/GlobalISel/regbank-maxnum.mir
llvm/test/CodeGen/AArch64/GlobalISel/regbank-minnum.mir
|
how about these?