Matt Arsenault c7c561ef98 AMDGPU: Enable ExpandLargeFpConvert for > 64-bit types ...

Fixes casts between double/float/half and i128. The pass seems to be
broken for bfloat though. I also believe we could have a better implementation
which attempts to make use the native 32-bit conversion instructions like
the 64-bit expansion does.

2024-03-15 16:08:39 +05:30

..

AArch64

[MachineLICM] Give opportunity to analyze physregs for invariance. (#84779)

2024-03-15 09:30:46 +00:00

AMDGPU

AMDGPU: Enable ExpandLargeFpConvert for > 64-bit types

2024-03-15 16:08:39 +05:30

ARC

[ARC] Convert tests to opaque pointers (NFC)

2024-02-06 12:55:16 -08:00

ARM

[llvm][arm] add T1 and T2 assembly options for vlldm and vlstm

2024-03-11 14:27:28 +00:00

AVR

[AVR] Convert tests to opaque pointers (NFC)

2024-02-05 13:55:50 +01:00

BPF

[BPF] rename 'arena' to 'address_space' (#85161)

2024-03-14 19:20:06 -07:00

CSKY

[CSKY] Convert tests to opaque pointers (NFC)

2024-02-06 12:54:21 -08:00

DirectX

[DXIL] exp, any, lerp, & rcp Intrinsic Lowering (#84526)

2024-03-14 20:25:57 -04:00

Generic

[CodeGen][GC] Skip function without GC in GCLoweringPass (#84421)

2024-03-14 13:07:41 +08:00

Hexagon

[Tests] Drop inrange attribute from some tests (NFC)

2024-03-13 11:49:16 +01:00

Inputs

…

Lanai

[CodeGen] Convert tests to opaque pointers (NFC)

2024-02-05 14:07:09 +01:00

LoongArch

Revert "[llvm][LoongArch] Improve loongarch_lasx_xvpermi_q instrinsic" (#84708)

2024-03-13 11:51:47 +08:00

M68k

[M68k] Fix compilation pipeline check

2024-03-01 14:49:27 -05:00

Mips

[Mips] Fix missing sign extension in expansion of sub-word atomic max (#77072)

2024-03-08 15:41:31 -05:00

MIR

[AMDGPU] Don't fix the scavenge slot at offset 0 (#79136)

2024-02-09 09:20:25 +01:00

MLRegAlloc

…

MSP430

[CodeGen] Convert tests to opaque pointers (NFC)

2024-02-05 14:07:09 +01:00

NVPTX

Reland "[NVPTX] Add support for atomic add for f16 type" (#85197)

2024-03-15 08:13:59 +01:00

PowerPC

[PowerPC][NFC] Rename ADDItocL to match the 64-bit naming convention (#85099)

2024-03-13 11:57:07 -04:00

RISCV

[RISCV] Optimize lowering of VECREDUCE_FMINIMUM/VECREDUCE_FMAXIMUM. (#85165)

2024-03-14 12:51:29 -07:00

SPARC

[Sparc] limit MaxAtomicSizeInBitsSupported to 32 for 32-bit Sparc. (#81655)

2024-02-13 15:40:51 -05:00

SPIRV

[SPIRV] Add type inference of function parameters by call instances (#85077)

2024-03-14 10:50:11 +01:00

SystemZ

[FPEnv][SystemZ] Correct strictfp test.

2024-02-23 13:00:38 -05:00

Thumb

Revert "Convert many LivePhysRegs uses to LiveRegUnits (#83905)"

2024-03-07 08:20:26 +00:00

Thumb2

Revert "[ARM] R11 not pushed adjacent to link register with PAC-M and… (#84019)

2024-03-05 14:34:43 +00:00

VE

[DAGCombiner][RISCV] Optimize (zext nneg (truncate X)) if X has known sign bits. (#82227)

2024-02-19 10:45:11 -08:00

WebAssembly

[WebAssembly] Use RefTypeMem2Local instead of Mem2Reg (#83196)

2024-03-05 19:54:41 -08:00

WinCFGuard

[Tests] Drop inrange attribute from some tests (NFC)

2024-03-13 11:49:16 +01:00

WinEH

…

X86

[MachineCombiner] Don't ignore PHI depths (#82025)

2024-03-14 11:47:29 -04:00

XCore

[CodeGen] Convert tests to opaque pointers (NFC)

2024-02-05 14:07:09 +01:00

Xtensa

[Xtensa] Initial codegen support from IR (#78548)

2024-02-12 17:41:59 +01:00