[PATCH] D25717: [x86][inline-asm][clang][fixup] accept 'v' constraint

2016-11-27 Thread coby via Phabricator via cfe-commits
coby abandoned this revision.
coby added a comment.

deprecated


Repository:
  rL LLVM

https://reviews.llvm.org/D25717



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits


[PATCH] D25717: [x86][inline-asm][clang][fixup] accept 'v' constraint

2016-10-18 Thread coby via cfe-commits
coby created this revision.
coby added a reviewer: m_zuckerman.
coby added a subscriber: cfe-commits.
coby set the repository for this revision to rL LLVM.

In respect to https://reviews.llvm.org/D25004 (reverted)


Repository:
  rL LLVM

https://reviews.llvm.org/D25717

Files:
  lib/Basic/Targets.cpp
  test/CodeGen/x86-inline-asm-v-constraint.c


Index: lib/Basic/Targets.cpp
===
--- lib/Basic/Targets.cpp
+++ lib/Basic/Targets.cpp
@@ -3949,6 +3949,7 @@
   case 'u': // Second from top of floating point stack.
   case 'q': // Any register accessible as [r]l: a, b, c, and d.
   case 'y': // Any MMX register.
+  case 'v': // Any {X,Y,Z}MM register (Arch & context dependent)
   case 'x': // Any SSE register.
   case 'Q': // Any register accessible as [r]h: a, b, c, and d.
   case 'R': // "Legacy" registers: ax, bx, cx, dx, di, si, sp, bp.
@@ -3989,6 +3990,7 @@
   case 't':
   case 'u':
 return Size <= 128;
+  case 'v':
   case 'x':
 if (SSELevel >= AVX512F)
   // 512-bit zmm registers can be used if target supports AVX512F.
Index: test/CodeGen/x86-inline-asm-v-constraint.c
===
--- test/CodeGen/x86-inline-asm-v-constraint.c
+++ test/CodeGen/x86-inline-asm-v-constraint.c
@@ -0,0 +1,30 @@
+// RUN: %clang_cc1 %s -triple x86_64-unknown-linux-gnu -emit-llvm -target-cpu 
x86-64 -o - | FileCheck %s --check-prefix SSE
+// RUN: %clang_cc1 %s -triple x86_64-unknown-linux-gnu -emit-llvm -target-cpu 
skylake -D AVX -o - | FileCheck %s --check-prefixes AVX,SSE
+// RUN: %clang_cc1 %s -triple x86_64-unknown-linux-gnu -emit-llvm -target-cpu 
skylake-avx512 -D AVX512 -D AVX -o - | FileCheck %s --check-prefixes 
AVX512,AVX,SSE
+// RUN: %clang_cc1 %s -triple x86_64-unknown-linux-gnu -emit-llvm -target-cpu 
knl -D AVX -D AVX512 -o - | FileCheck %s --check-prefixes AVX512,AVX,SSE
+
+typedef float __m128 __attribute__ ((vector_size (16)));
+typedef float __m256 __attribute__ ((vector_size (32)));
+typedef float __m512 __attribute__ ((vector_size (64)));
+
+// SSE: call <4 x float> asm "vmovhlps $1, $2, $0", 
"=v,v,v,~{dirflag},~{fpsr},~{flags}"
+__m128 testXMM(__m128 _xmm0, long _l) {
+  __asm__("vmovhlps %1, %2, %0" :"=v"(_xmm0) : "v"(_l), "v"(_xmm0));
+  return _xmm0;
+}
+
+// AVX: call <8 x float> asm "vmovsldup $1, $0", 
"=v,v,~{dirflag},~{fpsr},~{flags}"
+__m256 testYMM(__m256 _ymm0) {
+#ifdef AVX
+  __asm__("vmovsldup %1, %0" :"=v"(_ymm0) : "v"(_ymm0));
+#endif
+  return _ymm0;
+}
+
+// AVX512: call <16 x float> asm "vpternlogd $$0, $1, $2, $0", 
"=v,v,v,~{dirflag},~{fpsr},~{flags}"
+__m512 testZMM(__m512 _zmm0, __m512 _zmm1) {
+#ifdef AVX512
+  __asm__("vpternlogd $0, %1, %2, %0" :"=v"(_zmm0) : "v"(_zmm1), "v"(_zmm0));
+#endif
+  return _zmm0;
+}


Index: lib/Basic/Targets.cpp
===
--- lib/Basic/Targets.cpp
+++ lib/Basic/Targets.cpp
@@ -3949,6 +3949,7 @@
   case 'u': // Second from top of floating point stack.
   case 'q': // Any register accessible as [r]l: a, b, c, and d.
   case 'y': // Any MMX register.
+  case 'v': // Any {X,Y,Z}MM register (Arch & context dependent)
   case 'x': // Any SSE register.
   case 'Q': // Any register accessible as [r]h: a, b, c, and d.
   case 'R': // "Legacy" registers: ax, bx, cx, dx, di, si, sp, bp.
@@ -3989,6 +3990,7 @@
   case 't':
   case 'u':
 return Size <= 128;
+  case 'v':
   case 'x':
 if (SSELevel >= AVX512F)
   // 512-bit zmm registers can be used if target supports AVX512F.
Index: test/CodeGen/x86-inline-asm-v-constraint.c
===
--- test/CodeGen/x86-inline-asm-v-constraint.c
+++ test/CodeGen/x86-inline-asm-v-constraint.c
@@ -0,0 +1,30 @@
+// RUN: %clang_cc1 %s -triple x86_64-unknown-linux-gnu -emit-llvm -target-cpu x86-64 -o - | FileCheck %s --check-prefix SSE
+// RUN: %clang_cc1 %s -triple x86_64-unknown-linux-gnu -emit-llvm -target-cpu skylake -D AVX -o - | FileCheck %s --check-prefixes AVX,SSE
+// RUN: %clang_cc1 %s -triple x86_64-unknown-linux-gnu -emit-llvm -target-cpu skylake-avx512 -D AVX512 -D AVX -o - | FileCheck %s --check-prefixes AVX512,AVX,SSE
+// RUN: %clang_cc1 %s -triple x86_64-unknown-linux-gnu -emit-llvm -target-cpu knl -D AVX -D AVX512 -o - | FileCheck %s --check-prefixes AVX512,AVX,SSE
+
+typedef float __m128 __attribute__ ((vector_size (16)));
+typedef float __m256 __attribute__ ((vector_size (32)));
+typedef float __m512 __attribute__ ((vector_size (64)));
+
+// SSE: call <4 x float> asm "vmovhlps $1, $2, $0", "=v,v,v,~{dirflag},~{fpsr},~{flags}"
+__m128 testXMM(__m128 _xmm0, long _l) {
+  __asm__("vmovhlps %1, %2, %0" :"=v"(_xmm0) : "v"(_l), "v"(_xmm0));
+  return _xmm0;
+}
+
+// AVX: call <8 x float> asm "vmovsldup $1, $0", "=v,v,~{dirflag},~{fpsr},~{flags}"
+__m256 testYMM(__m256 _ymm0) {
+#ifdef AVX
+  __asm__("vmovsldup %1, %0" :"=v"(_ymm0) : "v"(_ymm0));
+#endif
+  return _ymm0;