[Mesa-dev] [PATCH 4/4] radeon/ac: use llvm.amdgcn.sffbh intrinsic instead of AMDGPU.flbit.i32

Dave Airlie airlied at gmail.com
Thu Feb 16 03:48:30 UTC 2017


From: Dave Airlie <airlied at redhat.com>

Use the newer intrinsic.

Signed-off-by: Dave Airlie <airlied at redhat.com>
---
 src/amd/common/ac_llvm_build.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/amd/common/ac_llvm_build.c b/src/amd/common/ac_llvm_build.c
index 180fd24..dda2c0b 100644
--- a/src/amd/common/ac_llvm_build.c
+++ b/src/amd/common/ac_llvm_build.c
@@ -769,7 +769,8 @@ ac_emit_imsb(struct ac_llvm_context *ctx,
 	     LLVMValueRef arg,
 	     LLVMTypeRef dst_type)
 {
-	LLVMValueRef msb = ac_emit_llvm_intrinsic(ctx, "llvm.AMDGPU.flbit.i32",
+	const char *intr_name = (HAVE_LLVM < 0x0400) ? "llvm.AMDGPU.flbit.i32" : "llvm.amdgcn.sffbh";
+	LLVMValueRef msb = ac_emit_llvm_intrinsic(ctx, intr_name,
 						  dst_type, &arg, 1,
 						  AC_FUNC_ATTR_READNONE);
 
-- 
2.7.4



More information about the mesa-dev mailing list