================ @@ -2641,6 +2641,28 @@ def int_amdgcn_perm : // GFX9 Intrinsics //===----------------------------------------------------------------------===// +/// This is a general-purpose intrinsic for all operations that take a pointer +/// a base location in LDS, and a data size and use it to perform a gather to LDS. +/// This allows abstracting over both global pointers (address space 1) and +/// the buffer-resource-wrapper pointers (address space 7 and 9). +/// TODO: add support for address space 5 and scratch_load_lds. +class AMDGPULoadToLDS : + Intrinsic < + [], + [llvm_anyptr_ty, // Base pointer to load from. Varies per lane. + LLVMQualPointerType<3>, // LDS base pointer to store to. Must be wave-uniform. + llvm_i32_ty, // Data byte size: 1/2/4 (/12/16 for gfx950) + llvm_i32_ty, // imm offset (applied to both input and LDS address) ---------------- krzysz00 wrote:
*big shrug* Is this something y'all want to try and fix here? (Also, procedural history on the buffer intrinsics points me to https://reviews.llvm.org/D124884 ( @rampitec ), which leaves it rather unclear why the immarg was added ... I think it's because for LDS, unlike the other buffer instructions, you can't do voffset => (actual voffset + imm)) https://github.com/llvm/llvm-project/pull/137425 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
