Since llvm.SI.tid is used in several places, it would be better to add a helper function that returns tid.
Also Mesa needs to support LLVM 3.6 too. Marek On Sat, Apr 16, 2016 at 4:39 AM, Tom Stellard <thomas.stell...@amd.com> wrote: > We're trying to move to more of the new style intrinsics with include > the correct target name, and map directly to ISA instructions. > --- > src/gallium/drivers/radeonsi/si_shader.c | 12 ++++++++++-- > 1 file changed, 10 insertions(+), 2 deletions(-) > > diff --git a/src/gallium/drivers/radeonsi/si_shader.c > b/src/gallium/drivers/radeonsi/si_shader.c > index c26960b..377ff26 100644 > --- a/src/gallium/drivers/radeonsi/si_shader.c > +++ b/src/gallium/drivers/radeonsi/si_shader.c > @@ -4117,14 +4117,22 @@ static void si_llvm_emit_ddxy( > LLVMValueRef indices[2]; > LLVMValueRef store_ptr, load_ptr0, load_ptr1; > LLVMValueRef tl, trbl, result[4]; > + LLVMValueRef tid_args[2]; > unsigned swizzle[4]; > unsigned c; > int idx; > unsigned mask; > > + tid_args[0] = lp_build_const_int32(gallivm, 0xffffffff); > + tid_args[1] = bld_base->uint_bld.zero; > + tid_args[1] = lp_build_intrinsic(gallivm->builder, > + "llvm.amdgcn.mbcnt.lo", ctx->i32, > + tid_args, 2, LLVMReadNoneAttribute); > + > indices[0] = bld_base->uint_bld.zero; > - indices[1] = lp_build_intrinsic(gallivm->builder, "llvm.SI.tid", > ctx->i32, > - NULL, 0, LLVMReadNoneAttribute); > + indices[1] = lp_build_intrinsic(gallivm->builder, > + "llvm.amdgcn.mbcnt.hi", ctx->i32, > + tid_args, 2, LLVMReadNoneAttribute); > store_ptr = LLVMBuildGEP(gallivm->builder, ctx->lds, > indices, 2, ""); > > -- > 2.1.0 > > _______________________________________________ > mesa-dev mailing list > mesa-dev@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/mesa-dev _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
