4 Commits

Author SHA1 Message Date
Mirko Brkusanin
a657deb42e [AMDGPU] Update RUN line in test (NFC) 2023-09-22 12:41:54 +02:00
Mirko Brkušanin
ecfdc23dd2
[AMDGPU] Select gfx1150 SALU Float instructions (#66885) 2023-09-21 12:22:55 +02:00
Mirko Brkusanin
1e5359c6ba [AMDGPU] Treat KIMM32 and KIMM16 operand types as noninlinable
While they are represent 32/16 bit immediate values they are already
included in encoding of the instructions that use them and are not true
literals. FMAMK and FMAAK instructions that use them are marked with fixed
size so getInstSizeInBytes will not increase the size for these operands.

We also add tests whose logic relies on KIMM16 and KIMM32 being considered
not inlinable.

Differential Revision: https://reviews.llvm.org/D157624
2023-08-11 18:46:39 +02:00
Matt Arsenault
4b1702e87a AMDGPU: Fix counting source modifiers as literal constants
This fixes over estimating code size. This was broken by
79f52af4cd9a76485dd50bcdbb5d393eb7a70103.

https://reviews.llvm.org/D157103
2023-08-07 18:40:16 -04:00