main

GGitHubReland "byte permutes in intra-warp layout conversion" (#7933)

1e6be008创建于 2025年8月22日历史提交

文件	最后提交记录	最后更新时间
allocate_shared_memory.mlir	[AMD] Introduce specialized Allocation pass (#7328)	11 个月前
amdgpu_membar.mlir	[IR] Improve memdesc_index printing to make it obvious what operand is the index (#7876)	9 个月前
async-ops-alias-scopes.mlir	[Gluon] Actually enable scf+cf+arith canonicalizers (#7775)	9 个月前
async_ops_to_llvm.mlir	[BACKEND] bump to llvm/llvm-project@e12cbd8 (#6880)	1 年前
async_ops_to_llvm_invalid.mlir	[AMD] Reject async global to local ops for less than 4 bytes (#6734)	1 年前
buffer_atomic_cas.mlir	[Dialect] Actually enable TMEM layout check and fix all the tests (#7723)	10 个月前
buffer_load_store.mlir	[Dialect] Actually enable TMEM layout check and fix all the tests (#7723)	10 个月前
buffer_load_to_local_to_llvm.mlir	[AMD] Fix alignment calculation for buffer ops offset in AxisAnalysis (#7886)	9 个月前
builtin_func_to_llvm.mlir	[AMD] Add fast_tanhf to libdevice (#7780)	9 个月前
compute-base-ptr.mlir	[AMD] Enable lowerLocalLdSt for AMD path (#7355)	10 个月前
convert_layout.mlir	[LLs] Tree-reduce the xor reduction in LLs codegen (#7816)	9 个月前
dedup-by-constancy.mlir	[Dialect] Actually enable TMEM layout check and fix all the tests (#7723)	10 个月前
ds_transpose.mlir	[AMD] Enable lowerLocalLdSt for AMD path (#7355)	10 个月前
fp_to_fp.mlir	[AMD] Improve register usage in Float8 conversions (#7527)	10 个月前
in_thread_transpose.mlir	[AMD] Optimize transposed GEMM operand cases (#6074)	1 年前
invalid_concat_op.mlir	[AMD] Support register broadcast in slice/concat ops (#7407)	10 个月前
invalid_extractslice_to_llvm.mlir	[AMD] Support register broadcast in slice/concat ops (#7407)	10 个月前
load_store.mlir	[Dialect] Actually enable TMEM layout check and fix all the tests (#7723)	10 个月前
math-denorm-handling.mlir	[Dialect] Actually enable TMEM layout check and fix all the tests (#7723)	10 个月前
mfma-shortcut.mlir	Reland "byte permutes in intra-warp layout conversion" (#7933)	9 个月前
minmax.mlir	[AMD] Eanble NAN propagating MIN/MAX for gfx950 (#6387)	1 年前
tritongpu_to_llvm.mlir	[IR] Improve memdesc_index printing to make it obvious what operand is the index (#7876)	9 个月前
tritongpu_to_llvm_rdna.mlir	[AMD] Use permlanex16 for shuffleXor on rdna (#7269)	11 个月前
tritongpu_wmma_dot_to_llvm.mlir	[AMD] Enable lowerLocalLdSt for AMD path (#7355)	10 个月前
upcast_mxfp.mlir	[Backend] Bump to llvm/llvm-project@bc773632355b (#7881)	9 个月前