[PATCH] D39719: [X86][AVX512] lowering kunpack intrinsic - clang part

jina via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Sun Nov 12 23:45:15 PST 2017

jina.nahias added inline comments.

Comment at: test/CodeGen/avx512f-builtins.c:6231
+  // CHECK: bitcast <16 x i1> %{{.*}} to i16
+  // CHECK: and i32 %{{.*}}, 255
+  // CHECK: shl i32 %{{.*}}, 8
craig.topper wrote:
> Does this really produce kunpackb in the backend? The type promotion here makes me skeptic
the code we get: 
	vpcmpneqd	%zmm1, %zmm0, %k0
	vpcmpneqd	%zmm3, %zmm2, %k1
	kunpckbw	%k1, %k0, %k1
	vpcmpneqd	%zmm5, %zmm4, %k0 {%k1}
	kmovd	%k0, %eax
                                        # kill: %AX<def> %AX<kill> %EAX<kill>


More information about the cfe-commits mailing list