From: Craig Topper <craig.topper@gmail.com>
Date: Sat, 31 Jan 2015 06:31:23 +0000 (+0000)
Subject: [X86] Add range checking to the immediate arguments of many of the SSE/AVX builtins.
X-Git-Tag: llvmorg-3.7.0-rc1~13531
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=1601525f1c66bf88bfd5e1030b4fecedab769327;p=platform%2Fupstream%2Fllvm.git

[X86] Add range checking to the immediate arguments of many of the SSE/AVX builtins.

llvm-svn: 227674
---

diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index cfa6111..f11ead8 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -878,12 +878,18 @@ bool Sema::CheckX86BuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall) {
   switch (BuiltinID) {
   default: return false;
   case X86::BI_mm_prefetch: i = 1; l = 0; u = 3; break;
-  case X86::BI__builtin_ia32_cmpps:
-  case X86::BI__builtin_ia32_cmpss:
-  case X86::BI__builtin_ia32_cmppd:
-  case X86::BI__builtin_ia32_cmpsd:
-  case X86::BI__builtin_ia32_cmpps512_mask:
-  case X86::BI__builtin_ia32_cmppd512_mask: i = 2; l = 0; u = 31; break;
+  case X86::BI__builtin_ia32_vextractf128_pd256:
+  case X86::BI__builtin_ia32_vextractf128_ps256:
+  case X86::BI__builtin_ia32_vextractf128_si256:
+  case X86::BI__builtin_ia32_extract128i256: i = 1, l = 0, u = 1; break;
+  case X86::BI__builtin_ia32_vinsertf128_pd256:
+  case X86::BI__builtin_ia32_vinsertf128_ps256:
+  case X86::BI__builtin_ia32_vinsertf128_si256:
+  case X86::BI__builtin_ia32_insert128i256:
+  case X86::BI__builtin_ia32_blendpd: i = 2, l = 0; u = 1; break;
+  case X86::BI__builtin_ia32_blendps:
+  case X86::BI__builtin_ia32_blendpd256:
+  case X86::BI__builtin_ia32_sha1rnds4: i = 2, l = 0; u = 3; break;
   case X86::BI__builtin_ia32_cmpb128_mask:
   case X86::BI__builtin_ia32_cmpw128_mask:
   case X86::BI__builtin_ia32_cmpd128_mask:
@@ -908,6 +914,20 @@ bool Sema::CheckX86BuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall) {
   case X86::BI__builtin_ia32_ucmpw512_mask:
   case X86::BI__builtin_ia32_ucmpd512_mask:
   case X86::BI__builtin_ia32_ucmpq512_mask: i = 2; l = 0; u = 7; break;
+  case X86::BI__builtin_ia32_roundps:
+  case X86::BI__builtin_ia32_roundpd:
+  case X86::BI__builtin_ia32_roundps256:
+  case X86::BI__builtin_ia32_roundpd256: i = 1, l = 0; u = 15; break;
+  case X86::BI__builtin_ia32_roundss:
+  case X86::BI__builtin_ia32_roundsd: i = 2, l = 0; u = 15; break;
+  case X86::BI__builtin_ia32_cmpps:
+  case X86::BI__builtin_ia32_cmpss:
+  case X86::BI__builtin_ia32_cmppd:
+  case X86::BI__builtin_ia32_cmpsd:
+  case X86::BI__builtin_ia32_cmpps256:
+  case X86::BI__builtin_ia32_cmppd256:
+  case X86::BI__builtin_ia32_cmpps512_mask:
+  case X86::BI__builtin_ia32_cmppd512_mask: i = 2; l = 0; u = 31; break;
   }
   return SemaBuiltinConstantArgRange(TheCall, i, l, u);
 }
diff --git a/clang/test/CodeGen/builtins-x86.c b/clang/test/CodeGen/builtins-x86.c
index e9ae834..2eb0b5b 100644
--- a/clang/test/CodeGen/builtins-x86.c
+++ b/clang/test/CodeGen/builtins-x86.c
@@ -403,16 +403,16 @@ void f0() {
   tmp_V4f = __builtin_ia32_vpermilvarps(tmp_V4f, tmp_V4i);
   tmp_V4d = __builtin_ia32_vpermilvarpd256(tmp_V4d, tmp_V4LLi);
   tmp_V8f = __builtin_ia32_vpermilvarps256(tmp_V8f, tmp_V8i);
-  tmp_V4d = __builtin_ia32_blendpd256(tmp_V4d, tmp_V4d, 0x7);
+  tmp_V4d = __builtin_ia32_blendpd256(tmp_V4d, tmp_V4d, 0x1);
   tmp_V8f = __builtin_ia32_blendps256(tmp_V8f, tmp_V8f, 0x7);
   tmp_V4d = __builtin_ia32_blendvpd256(tmp_V4d, tmp_V4d, tmp_V4d);
   tmp_V8f = __builtin_ia32_blendvps256(tmp_V8f, tmp_V8f, tmp_V8f);
   tmp_V8f = __builtin_ia32_dpps256(tmp_V8f, tmp_V8f, 0x7);
   tmp_V4d = __builtin_ia32_cmppd256(tmp_V4d, tmp_V4d, 0);
   tmp_V8f = __builtin_ia32_cmpps256(tmp_V8f, tmp_V8f, 0);
-  tmp_V2d = __builtin_ia32_vextractf128_pd256(tmp_V4d, 0x7);
-  tmp_V4f = __builtin_ia32_vextractf128_ps256(tmp_V8f, 0x7);
-  tmp_V4i = __builtin_ia32_vextractf128_si256(tmp_V8i, 0x7);
+  tmp_V2d = __builtin_ia32_vextractf128_pd256(tmp_V4d, 0x1);
+  tmp_V4f = __builtin_ia32_vextractf128_ps256(tmp_V8f, 0x1);
+  tmp_V4i = __builtin_ia32_vextractf128_si256(tmp_V8i, 0x1);
   tmp_V4d = __builtin_ia32_cvtdq2pd256(tmp_V4i);
   tmp_V8f = __builtin_ia32_cvtdq2ps256(tmp_V8i);
   tmp_V4f = __builtin_ia32_cvtpd2ps256(tmp_V4d);
@@ -424,9 +424,9 @@ void f0() {
   tmp_V4d = __builtin_ia32_vperm2f128_pd256(tmp_V4d, tmp_V4d, 0x7);
   tmp_V8f = __builtin_ia32_vperm2f128_ps256(tmp_V8f, tmp_V8f, 0x7);
   tmp_V8i = __builtin_ia32_vperm2f128_si256(tmp_V8i, tmp_V8i, 0x7);
-  tmp_V4d = __builtin_ia32_vinsertf128_pd256(tmp_V4d, tmp_V2d, 0x7);
-  tmp_V8f = __builtin_ia32_vinsertf128_ps256(tmp_V8f, tmp_V4f, 0x7);
-  tmp_V8i = __builtin_ia32_vinsertf128_si256(tmp_V8i, tmp_V4i, 0x7);
+  tmp_V4d = __builtin_ia32_vinsertf128_pd256(tmp_V4d, tmp_V2d, 0x1);
+  tmp_V8f = __builtin_ia32_vinsertf128_ps256(tmp_V8f, tmp_V4f, 0x1);
+  tmp_V8i = __builtin_ia32_vinsertf128_si256(tmp_V8i, tmp_V4i, 0x1);
   tmp_V4d = __builtin_ia32_sqrtpd256(tmp_V4d);
   tmp_V8f = __builtin_ia32_sqrtps256(tmp_V8f);
   tmp_V8f = __builtin_ia32_rsqrtps256(tmp_V8f);
@@ -497,7 +497,7 @@ void f0() {
   tmp_V2f = __builtin_ia32_pswapdsf(tmp_V2f);
   tmp_V2i = __builtin_ia32_pswapdsi(tmp_V2i);
 
-  tmp_V4i = __builtin_ia32_sha1rnds4(tmp_V4i, tmp_V4i, imm_i);
+  tmp_V4i = __builtin_ia32_sha1rnds4(tmp_V4i, tmp_V4i, imm_i_0_4);
   tmp_V4i = __builtin_ia32_sha1nexte(tmp_V4i, tmp_V4i);
   tmp_V4i = __builtin_ia32_sha1msg1(tmp_V4i, tmp_V4i);
   tmp_V4i = __builtin_ia32_sha1msg2(tmp_V4i, tmp_V4i);
diff --git a/clang/test/CodeGen/sha-builtins.c b/clang/test/CodeGen/sha-builtins.c
index 181dba1..2b11ded 100644
--- a/clang/test/CodeGen/sha-builtins.c
+++ b/clang/test/CodeGen/sha-builtins.c
@@ -7,7 +7,7 @@
 
 __m128i test_sha1rnds4(__m128i a, __m128i b) {
   // CHECK: call <4 x i32> @llvm.x86.sha1rnds4
-  return _mm_sha1rnds4_epu32(a, b, 8);
+  return _mm_sha1rnds4_epu32(a, b, 3);
 }
 __m128i test_sha1nexte(__m128i a, __m128i b) {
   // CHECK: call <4 x i32> @llvm.x86.sha1nexte