Jay Foad 2012b25420

[AMDGPU][GlobalISel] Disable fixed-point iteration in all Combiners (#105517) ...

Disable fixed-point iteration in all AMDGPU Combiners after #102163.

This saves around 2% compile time in ad hoc testing on some large
graphics shaders. I did not notice any regressions in the generated
code, just a bunch of harmless differences in instruction selection and
register allocation.

2024-08-22 17:14:53 +01:00

..

AArch64

[AArch64] Lower aarch64_neon_saddlv via SADDLV nodes. (#103307)

2024-08-22 16:43:24 +01:00

AMDGPU

[AMDGPU][GlobalISel] Disable fixed-point iteration in all Combiners (#105517)

2024-08-22 17:14:53 +01:00

ARC

…

ARM

[GlobalISel] Bail out early for big-endian (#103310)

2024-08-19 18:50:47 +01:00

AVR

…

BPF

[BPF] introduce __attribute__((bpf_fastcall)) (#105417)

2024-08-22 03:40:56 +03:00

CSKY

…

DirectX

[DirectX] Register a few DXIL passes with the new PM

2024-08-20 16:33:25 -07:00

Generic

[VP] Merge ExpandVP pass into PreISelIntrinsicLowering (#101652)

2024-08-06 09:27:59 +02:00

Hexagon

[SimplifyCFG] Don't block sinking for allocas if no phi created (#104579)

2024-08-19 09:55:30 +02:00

Inputs

…

Lanai

…

LoongArch

[LoongArch] Fix the assertion for atomic store with 'ptr' type

2024-08-19 17:17:36 +08:00

M68k

[LTO] enable ObjCARCContractPass only on optimized build (#101114)

2024-08-09 13:04:25 -07:00

Mips

[Mips] Fix fast isel for i16 bswap. (#103398)

2024-08-16 14:54:51 -07:00

MIR

Reapply "AMDGPU: Move attributor into optimization pipeline (#83131)" and follow up commit "clang/AMDGPU: Defeat attribute optimization in attribute test" (#98851)

2024-07-15 11:51:44 +04:00

MLRegAlloc

…

MSP430

…

NVPTX

[NVPTX] Add elect.sync Intrinsic (#104780)

2024-08-20 20:46:05 +05:30

PowerPC

[PowerPC] Fix mask for __st[d/w/h/b]cx builtins (#104453)

2024-08-22 09:55:46 -04:00

RISCV

[RISCV][GISel] Implement canLowerReturn. (#105465)

2024-08-22 08:12:51 -07:00

SPARC

[SPARC] Remove assertions in printOperand for inline asm operands (#104692)

2024-08-20 20:05:06 +07:00

SPIRV

[SPIR-V] Rework usage of virtual registers' types and classes (#104104)

2024-08-22 09:40:27 +02:00

SystemZ

[SystemZ][z/OS] Continuation of __ptr32 support (#103393)

2024-08-14 13:26:30 -04:00

Thumb

[SelectionDAG] Expand [US]CMP using arithmetic on boolean values instead of selects (#98774)

2024-07-16 20:56:18 +01:00

Thumb2

[ValueTracking] Fix f16 fptosi range for large integers

2024-08-15 18:18:19 +02:00

VE

[DAGCombiner] Freeze maybe poison operands when folding select to logic (#84924)

2024-07-22 17:19:46 +02:00

WebAssembly

[SelectionDAG] Scalarize binary ops of splats before legal types (#100749)

2024-08-15 00:07:00 +08:00

WinCFGuard

…

WinEH

…

X86

[X86] Add BSR/BSF tests to check for implicit zero extension

2024-08-22 16:45:07 +01:00

XCore

…

Xtensa

[Xtensa] Implement lowering Mul/Div/Shift operations. (#99981)

2024-08-22 10:34:21 +02:00