[PATCH] D68360: PR41162 Implement LKK remainder and divisibility algorithms [urem]

[Dávid Bolvanský via Phabricator via llvm-commits]

xbolva00 added a comment.

Still some regressions.



================
Comment at: llvm/test/CodeGen/X86/urem-vector-lkk.ll:55
 ; SSE:       # %bb.0:
-; SSE-NEXT:    movdqa {{.*#+}} xmm1 = [44151,44151,44151,44151,44151,44151,44151,44151]
-; SSE-NEXT:    pmulhuw %xmm0, %xmm1
-; SSE-NEXT:    psrlw $6, %xmm1
-; SSE-NEXT:    pmullw {{.*}}(%rip), %xmm1
-; SSE-NEXT:    psubw %xmm1, %xmm0
+; SSE-NEXT:    pmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; SSE-NEXT:    pmulld {{.*}}(%rip), %xmm0
----------------
this is worse


================
Comment at: llvm/test/CodeGen/X86/vector-idiv-udiv-128.ll:629
+; AVX1-NEXT:    vpmuludq %xmm4, %xmm1, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
+; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0,1],xmm3[2,3],xmm1[4,5],xmm3[6,7]
----------------
regression


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D68360/new/

https://reviews.llvm.org/D68360