Re: [PATCH] D19544: Pass for translating math intrinsics to math library calls.

[Matt via cfe-commits]

mmasten updated this revision to Diff 61111.

http://reviews.llvm.org/D19544

Files:
  include/llvm/Analysis/TargetLibraryInfo.h
  lib/Analysis/TargetLibraryInfo.cpp


Index: lib/Analysis/TargetLibraryInfo.cpp
===================================================================
--- lib/Analysis/TargetLibraryInfo.cpp
+++ lib/Analysis/TargetLibraryInfo.cpp
@@ -23,6 +23,8 @@
                           "No vector functions library"),
                clEnumValN(TargetLibraryInfoImpl::Accelerate, "Accelerate",
                           "Accelerate framework"),
+               clEnumValN(TargetLibraryInfoImpl::SVML, "SVML",
+                          "Intel SVML library"),
                clEnumValEnd));
 
 const char *const TargetLibraryInfoImpl::StandardNames[LibFunc::NumLibFuncs] = {
@@ -1079,6 +1081,92 @@
     addVectorizableFunctions(VecFuncs);
     break;
   }
+  case SVML: {
+    const VecDesc VecFuncs[] = {
+        {"sin", "__svml_sin2", 2},
+        {"sin", "__svml_sin4", 4},
+        {"sin", "__svml_sin8", 8},
+        {"sin", "__svml_sin16", 16},
+        {"sin", "__svml_sin32", 32},
+        {"sinf", "__svml_sinf2", 2},
+        {"sinf", "__svml_sinf4", 4},
+        {"sinf", "__svml_sinf8", 8},
+        {"sinf", "__svml_sinf16", 16},
+        {"sinf", "__svml_sinf32", 32},
+        {"cos", "__svml_cos2", 2},
+        {"cos", "__svml_cos4", 4},
+        {"cos", "__svml_cos8", 8},
+        {"cos", "__svml_cos16", 16},
+        {"cos", "__svml_cos32", 32},
+        {"cosf", "__svml_cosf2", 2},
+        {"cosf", "__svml_cosf4", 4},
+        {"cosf", "__svml_cosf8", 8},
+        {"cosf", "__svml_cosf16", 16},
+        {"cosf", "__svml_cosf32", 32},
+        {"pow", "__svml_pow2", 2},
+        {"pow", "__svml_pow4", 4},
+        {"pow", "__svml_pow8", 8},
+        {"pow", "__svml_pow16", 16},
+        {"pow", "__svml_pow32", 32},
+        {"powf", "__svml_powf2", 2},
+        {"powf", "__svml_powf4", 4},
+        {"powf", "__svml_powf8", 8},
+        {"powf", "__svml_powf16", 16},
+        {"powf", "__svml_powf32", 32},
+        {"llvm.pow.f64", "__svml_pow2", 2},
+        {"llvm.pow.f64", "__svml_pow4", 4},
+        {"llvm.pow.f64", "__svml_pow8", 8},
+        {"llvm.pow.f64", "__svml_pow16", 16},
+        {"llvm.pow.f64", "__svml_pow32", 32},
+        {"llvm.pow.f32", "__svml_powf2", 2},
+        {"llvm.pow.f32", "__svml_powf4", 4},
+        {"llvm.pow.f32", "__svml_powf8", 8},
+        {"llvm.pow.f32", "__svml_powf16", 16},
+        {"llvm.pow.f32", "__svml_powf32", 32},
+        {"exp", "__svml_exp2", 2},
+        {"exp", "__svml_exp4", 4},
+        {"exp", "__svml_exp8", 8},
+        {"exp", "__svml_exp16", 16},
+        {"exp", "__svml_exp32", 32},
+        {"expf", "__svml_expf2", 2},
+        {"expf", "__svml_expf4", 4},
+        {"expf", "__svml_expf8", 8},
+        {"expf", "__svml_expf16", 16},
+        {"expf", "__svml_expf32", 32},
+        {"llvm.exp.f64", "__svml_exp2", 2},
+        {"llvm.exp.f64", "__svml_exp4", 4},
+        {"llvm.exp.f64", "__svml_exp8", 8},
+        {"llvm.exp.f64", "__svml_exp16", 16},
+        {"llvm.exp.f64", "__svml_exp32", 32},
+        {"llvm.exp.f32", "__svml_expf2", 2},
+        {"llvm.exp.f32", "__svml_expf4", 4},
+        {"llvm.exp.f32", "__svml_expf8", 8},
+        {"llvm.exp.f32", "__svml_expf16", 16},
+        {"llvm.exp.f32", "__svml_expf32", 32},
+        {"log", "__svml_log2", 2},
+        {"log", "__svml_log4", 4},
+        {"log", "__svml_log8", 8},
+        {"log", "__svml_log16", 16},
+        {"log", "__svml_log32", 32},
+        {"logf", "__svml_logf2", 2},
+        {"logf", "__svml_logf4", 4},
+        {"logf", "__svml_logf8", 8},
+        {"logf", "__svml_logf16", 16},
+        {"logf", "__svml_logf32", 32},
+        {"llvm.log.f64", "__svml_log2", 2},
+        {"llvm.log.f64", "__svml_log4", 4},
+        {"llvm.log.f64", "__svml_log8", 8},
+        {"llvm.log.f64", "__svml_log16", 16},
+        {"llvm.log.f64", "__svml_log32", 32},
+        {"llvm.log.f32", "__svml_logf2", 2},
+        {"llvm.log.f32", "__svml_logf4", 4},
+        {"llvm.log.f32", "__svml_logf8", 8},
+        {"llvm.log.f32", "__svml_logf16", 16},
+        {"llvm.log.f32", "__svml_logf32", 32},
+    };
+    addVectorizableFunctions(VecFuncs);
+    break;
+  }
   case NoLibrary:
     break;
   }
Index: include/llvm/Analysis/TargetLibraryInfo.h
===================================================================
--- include/llvm/Analysis/TargetLibraryInfo.h
+++ include/llvm/Analysis/TargetLibraryInfo.h
@@ -85,8 +85,9 @@
   /// addVectorizableFunctionsFromVecLib for filling up the tables of
   /// vectorizable functions.
   enum VectorLibrary {
-    NoLibrary, // Don't use any vector library.
-    Accelerate // Use Accelerate framework.
+    NoLibrary,  // Don't use any vector library.
+    Accelerate, // Use Accelerate framework.
+    SVML        // Intel short vector math library.
   };
 
   TargetLibraryInfoImpl();

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