r283061 - [PowerPC] Enable soft-float for PPC64, and +soft-float -> -hard-float

Hal Finkel via cfe-commits cfe-commits at lists.llvm.org
Sat Oct 1 19:10:45 PDT 2016


Author: hfinkel
Date: Sat Oct  1 21:10:45 2016
New Revision: 283061

URL: http://llvm.org/viewvc/llvm-project?rev=283061&view=rev
Log:
[PowerPC] Enable soft-float for PPC64, and +soft-float -> -hard-float

Enable soft-float support on PPC64, as the backend now supports it. Also, the
backend now uses -hard-float instead of +soft-float, so set the target features
accordingly.

Fixes PR26970.

Added:
    cfe/trunk/test/CodeGen/ppc64-soft-float.c
Modified:
    cfe/trunk/lib/CodeGen/TargetInfo.cpp
    cfe/trunk/lib/Driver/Tools.cpp
    cfe/trunk/test/Driver/ppc-features.cpp

Modified: cfe/trunk/lib/CodeGen/TargetInfo.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/TargetInfo.cpp?rev=283061&r1=283060&r2=283061&view=diff
==============================================================================
--- cfe/trunk/lib/CodeGen/TargetInfo.cpp (original)
+++ cfe/trunk/lib/CodeGen/TargetInfo.cpp Sat Oct  1 21:10:45 2016
@@ -3899,6 +3899,7 @@ private:
   static const unsigned GPRBits = 64;
   ABIKind Kind;
   bool HasQPX;
+  bool IsSoftFloatABI;
 
   // A vector of float or double will be promoted to <4 x f32> or <4 x f64> and
   // will be passed in a QPX register.
@@ -3929,8 +3930,10 @@ private:
   }
 
 public:
-  PPC64_SVR4_ABIInfo(CodeGen::CodeGenTypes &CGT, ABIKind Kind, bool HasQPX)
-      : ABIInfo(CGT), Kind(Kind), HasQPX(HasQPX) {}
+  PPC64_SVR4_ABIInfo(CodeGen::CodeGenTypes &CGT, ABIKind Kind, bool HasQPX,
+                     bool SoftFloatABI)
+      : ABIInfo(CGT), Kind(Kind), HasQPX(HasQPX),
+        IsSoftFloatABI(SoftFloatABI) {}
 
   bool isPromotableTypeForABI(QualType Ty) const;
   CharUnits getParamTypeAlignment(QualType Ty) const;
@@ -3978,8 +3981,10 @@ class PPC64_SVR4_TargetCodeGenInfo : pub
 
 public:
   PPC64_SVR4_TargetCodeGenInfo(CodeGenTypes &CGT,
-                               PPC64_SVR4_ABIInfo::ABIKind Kind, bool HasQPX)
-      : TargetCodeGenInfo(new PPC64_SVR4_ABIInfo(CGT, Kind, HasQPX)) {}
+                               PPC64_SVR4_ABIInfo::ABIKind Kind, bool HasQPX,
+                               bool SoftFloatABI)
+      : TargetCodeGenInfo(new PPC64_SVR4_ABIInfo(CGT, Kind, HasQPX,
+                                                 SoftFloatABI)) {}
 
   int getDwarfEHStackPointer(CodeGen::CodeGenModule &M) const override {
     // This is recovered from gcc output.
@@ -4197,8 +4202,11 @@ bool PPC64_SVR4_ABIInfo::isHomogeneousAg
   if (const BuiltinType *BT = Ty->getAs<BuiltinType>()) {
     if (BT->getKind() == BuiltinType::Float ||
         BT->getKind() == BuiltinType::Double ||
-        BT->getKind() == BuiltinType::LongDouble)
+        BT->getKind() == BuiltinType::LongDouble) {
+      if (IsSoftFloatABI)
+        return false;
       return true;
+    }
   }
   if (const VectorType *VT = Ty->getAs<VectorType>()) {
     if (getContext().getTypeSize(VT) == 128 || IsQPXVectorTy(Ty))
@@ -8107,8 +8115,10 @@ const TargetCodeGenInfo &CodeGenModule::
       if (getTarget().getABI() == "elfv2")
         Kind = PPC64_SVR4_ABIInfo::ELFv2;
       bool HasQPX = getTarget().getABI() == "elfv1-qpx";
+      bool IsSoftFloat = CodeGenOpts.FloatABI == "soft";
 
-      return SetCGInfo(new PPC64_SVR4_TargetCodeGenInfo(Types, Kind, HasQPX));
+      return SetCGInfo(new PPC64_SVR4_TargetCodeGenInfo(Types, Kind, HasQPX,
+                                                        IsSoftFloat));
     } else
       return SetCGInfo(new PPC64TargetCodeGenInfo(Types));
   case llvm::Triple::ppc64le: {
@@ -8117,8 +8127,10 @@ const TargetCodeGenInfo &CodeGenModule::
     if (getTarget().getABI() == "elfv1" || getTarget().getABI() == "elfv1-qpx")
       Kind = PPC64_SVR4_ABIInfo::ELFv1;
     bool HasQPX = getTarget().getABI() == "elfv1-qpx";
+    bool IsSoftFloat = CodeGenOpts.FloatABI == "soft";
 
-    return SetCGInfo(new PPC64_SVR4_TargetCodeGenInfo(Types, Kind, HasQPX));
+    return SetCGInfo(new PPC64_SVR4_TargetCodeGenInfo(Types, Kind, HasQPX,
+                                                      IsSoftFloat));
   }
 
   case llvm::Triple::nvptx:

Modified: cfe/trunk/lib/Driver/Tools.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Driver/Tools.cpp?rev=283061&r1=283060&r2=283061&view=diff
==============================================================================
--- cfe/trunk/lib/Driver/Tools.cpp (original)
+++ cfe/trunk/lib/Driver/Tools.cpp Sat Oct  1 21:10:45 2016
@@ -1591,15 +1591,8 @@ static void getPPCTargetFeatures(const D
   handleTargetFeaturesGroup(Args, Features, options::OPT_m_ppc_Features_Group);
 
   ppc::FloatABI FloatABI = ppc::getPPCFloatABI(D, Args);
-  if (FloatABI == ppc::FloatABI::Soft &&
-      !(Triple.getArch() == llvm::Triple::ppc64 ||
-        Triple.getArch() == llvm::Triple::ppc64le))
-    Features.push_back("+soft-float");
-  else if (FloatABI == ppc::FloatABI::Soft &&
-           (Triple.getArch() == llvm::Triple::ppc64 ||
-            Triple.getArch() == llvm::Triple::ppc64le))
-    D.Diag(diag::err_drv_invalid_mfloat_abi)
-        << "soft float is not supported for ppc64";
+  if (FloatABI == ppc::FloatABI::Soft)
+    Features.push_back("-hard-float");
 
   // Altivec is a bit weird, allow overriding of the Altivec feature here.
   AddTargetFeature(Args, Features, options::OPT_faltivec,

Added: cfe/trunk/test/CodeGen/ppc64-soft-float.c
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/test/CodeGen/ppc64-soft-float.c?rev=283061&view=auto
==============================================================================
--- cfe/trunk/test/CodeGen/ppc64-soft-float.c (added)
+++ cfe/trunk/test/CodeGen/ppc64-soft-float.c Sat Oct  1 21:10:45 2016
@@ -0,0 +1,171 @@
+// RUN: %clang_cc1 -msoft-float -mfloat-abi soft -triple powerpc64le-unknown-linux-gnu -emit-llvm -o - %s | FileCheck -check-prefix=CHECK -check-prefix=CHECK-LE %s
+// RUN: %clang_cc1 -msoft-float -mfloat-abi soft -triple powerpc64-unknown-linux-gnu -emit-llvm -o - %s | FileCheck -check-prefix=CHECK -check-prefix=CHECK-BE %s
+
+// Test float returns and params.
+
+// CHECK: define float @func_p1(float %x)
+float func_p1(float x) { return x; }
+
+// CHECK: define double @func_p2(double %x)
+double func_p2(double x) { return x; }
+
+// CHECK: define ppc_fp128 @func_p3(ppc_fp128 %x)
+long double func_p3(long double x) { return x; }
+
+// Test homogeneous float aggregate passing and returning.
+
+struct f1 { float f[1]; };
+struct f2 { float f[2]; };
+struct f3 { float f[3]; };
+struct f4 { float f[4]; };
+struct f5 { float f[5]; };
+struct f6 { float f[6]; };
+struct f7 { float f[7]; };
+struct f8 { float f[8]; };
+struct f9 { float f[9]; };
+
+struct fab { float a; float b; };
+struct fabc { float a; float b; float c; };
+
+struct f2a2b { float a[2]; float b[2]; };
+
+// CHECK-LE: define i32 @func_f1(float inreg %x.coerce)
+// CHECK-BE: define void @func_f1(%struct.f1* noalias sret %agg.result, float inreg %x.coerce)
+struct f1 func_f1(struct f1 x) { return x; }
+
+// CHECK-LE: define i64 @func_f2(i64 %x.coerce)
+// CHECK-BE: define void @func_f2(%struct.f2* noalias sret %agg.result, i64 %x.coerce)
+struct f2 func_f2(struct f2 x) { return x; }
+
+// CHECK-LE: define { i64, i64 } @func_f3([2 x i64] %x.coerce)
+// CHECK-BE: define void @func_f3(%struct.f3* noalias sret %agg.result, [2 x i64] %x.coerce)
+struct f3 func_f3(struct f3 x) { return x; }
+
+// CHECK-LE: define { i64, i64 } @func_f4([2 x i64] %x.coerce)
+// CHECK-BE: define void @func_f4(%struct.f4* noalias sret %agg.result, [2 x i64] %x.coerce)
+struct f4 func_f4(struct f4 x) { return x; }
+
+// CHECK: define void @func_f5(%struct.f5* noalias sret %agg.result, [3 x i64] %x.coerce)
+struct f5 func_f5(struct f5 x) { return x; }
+
+// CHECK: define void @func_f6(%struct.f6* noalias sret %agg.result, [3 x i64] %x.coerce)
+struct f6 func_f6(struct f6 x) { return x; }
+
+// CHECK: define void @func_f7(%struct.f7* noalias sret %agg.result, [4 x i64] %x.coerce)
+struct f7 func_f7(struct f7 x) { return x; }
+
+// CHECK: define void @func_f8(%struct.f8* noalias sret %agg.result, [4 x i64] %x.coerce)
+struct f8 func_f8(struct f8 x) { return x; }
+
+// CHECK: define void @func_f9(%struct.f9* noalias sret %agg.result, [5 x i64] %x.coerce)
+struct f9 func_f9(struct f9 x) { return x; }
+
+// CHECK-LE: define i64 @func_fab(i64 %x.coerce)
+// CHECK-BE: define void @func_fab(%struct.fab* noalias sret %agg.result, i64 %x.coerce)
+struct fab func_fab(struct fab x) { return x; }
+
+// CHECK-LE: define { i64, i64 } @func_fabc([2 x i64] %x.coerce)
+// CHECK-BE: define void @func_fabc(%struct.fabc* noalias sret %agg.result, [2 x i64] %x.coerce)
+struct fabc func_fabc(struct fabc x) { return x; }
+
+// CHECK-LE: define { i64, i64 } @func_f2a2b([2 x i64] %x.coerce)
+// CHECK-BE: define void @func_f2a2b(%struct.f2a2b* noalias sret %agg.result, [2 x i64] %x.coerce)
+struct f2a2b func_f2a2b(struct f2a2b x) { return x; }
+
+// CHECK-LABEL: @call_f1
+// CHECK-BE: %[[TMP0:[^ ]+]] = alloca %struct.f1, align 4
+// CHECK: %[[TMP:[^ ]+]] = load float, float* getelementptr inbounds (%struct.f1, %struct.f1* @global_f1, i32 0, i32 0, i32 0), align 4
+// CHECK-LE: call i32 @func_f1(float inreg %[[TMP]])
+// CHECK-BE: call void @func_f1(%struct.f1* sret %[[TMP0]], float inreg %[[TMP]])
+struct f1 global_f1;
+void call_f1(void) { global_f1 = func_f1(global_f1); }
+
+// CHECK-LABEL: @call_f2
+// CHECK-BE: %[[TMP0:[^ ]+]] = alloca %struct.f2, align 4
+// CHECK: %[[TMP:[^ ]+]] = load i64, i64* bitcast (%struct.f2* @global_f2 to i64*), align 4
+// CHECK-LE: call i64 @func_f2(i64 %[[TMP]])
+// CHECK-BE: call void @func_f2(%struct.f2* sret %[[TMP0]], i64 %[[TMP]])
+struct f2 global_f2;
+void call_f2(void) { global_f2 = func_f2(global_f2); }
+
+// CHECK-LABEL: @call_f3
+// CHECK-BE: %[[TMP0:[^ ]+]] = alloca %struct.f3, align 4
+// CHECK: %[[TMP1:[^ ]+]] = alloca [2 x i64]
+// CHECK: %[[TMP2:[^ ]+]] = bitcast [2 x i64]* %[[TMP1]] to i8*
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %[[TMP2]], i8* bitcast (%struct.f3* @global_f3 to i8*), i64 12, i32 4, i1 false)
+// CHECK: %[[TMP3:[^ ]+]] = load [2 x i64], [2 x i64]* %[[TMP1]]
+// CHECK-LE: call { i64, i64 } @func_f3([2 x i64] %[[TMP3]])
+// CHECK-BE: call void @func_f3(%struct.f3* sret %[[TMP0]], [2 x i64] %[[TMP3]])
+struct f3 global_f3;
+void call_f3(void) { global_f3 = func_f3(global_f3); }
+
+// CHECK-LABEL: @call_f4
+// CHECK-BE: %[[TMP0:[^ ]+]] = alloca %struct.f4, align 4
+// CHECK: %[[TMP:[^ ]+]] = load [2 x i64], [2 x i64]* bitcast (%struct.f4* @global_f4 to [2 x i64]*), align 4
+// CHECK-LE: call { i64, i64 } @func_f4([2 x i64] %[[TMP]])
+// CHECK-BE: call void @func_f4(%struct.f4* sret %[[TMP0]], [2 x i64] %[[TMP]])
+struct f4 global_f4;
+void call_f4(void) { global_f4 = func_f4(global_f4); }
+
+// CHECK-LABEL: @call_f5
+// CHECK: %[[TMP0:[^ ]+]] = alloca %struct.f5, align 4
+// CHECK: %[[TMP1:[^ ]+]] = alloca [3 x i64]
+// CHECK: %[[TMP2:[^ ]+]] = bitcast [3 x i64]* %[[TMP1]] to i8*
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %[[TMP2]], i8* bitcast (%struct.f5* @global_f5 to i8*), i64 20, i32 4, i1 false)
+// CHECK: %[[TMP3:[^ ]+]] = load [3 x i64], [3 x i64]* %[[TMP1]]
+// CHECK: call void @func_f5(%struct.f5* sret %[[TMP0]], [3 x i64] %[[TMP3]])
+struct f5 global_f5;
+void call_f5(void) { global_f5 = func_f5(global_f5); }
+
+// CHECK-LABEL: @call_f6
+// CHECK: %[[TMP0:[^ ]+]] = alloca %struct.f6, align 4
+// CHECK: %[[TMP:[^ ]+]] = load [3 x i64], [3 x i64]* bitcast (%struct.f6* @global_f6 to [3 x i64]*), align 4
+// CHECK: call void @func_f6(%struct.f6* sret %[[TMP0]], [3 x i64] %[[TMP]])
+struct f6 global_f6;
+void call_f6(void) { global_f6 = func_f6(global_f6); }
+
+// CHECK-LABEL: @call_f7
+// CHECK: %[[TMP0:[^ ]+]] = alloca %struct.f7, align 4
+// CHECK: %[[TMP1:[^ ]+]] = alloca [4 x i64], align 8
+// CHECK: %[[TMP2:[^ ]+]] = bitcast [4 x i64]* %[[TMP1]] to i8*
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %[[TMP2]], i8* bitcast (%struct.f7* @global_f7 to i8*), i64 28, i32 4, i1 false)
+// CHECK: %[[TMP3:[^ ]+]] = load [4 x i64], [4 x i64]* %[[TMP1]], align 8
+// CHECK: call void @func_f7(%struct.f7* sret %[[TMP0]], [4 x i64] %[[TMP3]])
+struct f7 global_f7;
+void call_f7(void) { global_f7 = func_f7(global_f7); }
+
+// CHECK-LABEL: @call_f8
+// CHECK: %[[TMP0:[^ ]+]] = alloca %struct.f8, align 4
+// CHECK: %[[TMP:[^ ]+]] = load [4 x i64], [4 x i64]* bitcast (%struct.f8* @global_f8 to [4 x i64]*), align 4
+// CHECK: call void @func_f8(%struct.f8* sret %[[TMP0]], [4 x i64] %[[TMP]])
+struct f8 global_f8;
+void call_f8(void) { global_f8 = func_f8(global_f8); }
+
+// CHECK-LABEL: @call_f9
+// CHECK: %[[TMP1:[^ ]+]] = alloca [5 x i64]
+// CHECK: %[[TMP2:[^ ]+]] = bitcast [5 x i64]* %[[TMP1]] to i8*
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %[[TMP2]], i8* bitcast (%struct.f9* @global_f9 to i8*), i64 36, i32 4, i1 false)
+// CHECK: %[[TMP3:[^ ]+]] = load [5 x i64], [5 x i64]* %[[TMP1]]
+// CHECK: call void @func_f9(%struct.f9* sret %{{[^ ]+}}, [5 x i64] %[[TMP3]])
+struct f9 global_f9;
+void call_f9(void) { global_f9 = func_f9(global_f9); }
+
+// CHECK-LABEL: @call_fab
+// CHECK: %[[TMP0:[^ ]+]] = alloca %struct.fab, align 4
+// CHECK: %[[TMP:[^ ]+]] = load i64, i64* bitcast (%struct.fab* @global_fab to i64*), align 4
+// CHECK-LE: %call = call i64 @func_fab(i64 %[[TMP]])
+// CHECK-BE: call void @func_fab(%struct.fab* sret %[[TMP0]], i64 %[[TMP]])
+struct fab global_fab;
+void call_fab(void) { global_fab = func_fab(global_fab); }
+
+// CHECK-LABEL: @call_fabc
+// CHECK-BE: %[[TMPX:[^ ]+]] = alloca %struct.fabc, align 4
+// CHECK: %[[TMP0:[^ ]+]] = alloca [2 x i64], align 8
+// CHECK: %[[TMP2:[^ ]+]] = bitcast [2 x i64]* %[[TMP0]] to i8*
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %[[TMP2]], i8* bitcast (%struct.fabc* @global_fabc to i8*), i64 12, i32 4, i1 false)
+// CHECK: %[[TMP3:[^ ]+]] = load [2 x i64], [2 x i64]* %[[TMP0]], align 8
+// CHECK-LE: %call = call { i64, i64 } @func_fabc([2 x i64] %[[TMP3]])
+// CHECK-BE: call void @func_fabc(%struct.fabc* sret %[[TMPX]], [2 x i64] %[[TMP3]])
+struct fabc global_fabc;
+void call_fabc(void) { global_fabc = func_fabc(global_fabc); }
+

Modified: cfe/trunk/test/Driver/ppc-features.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Driver/ppc-features.cpp?rev=283061&r1=283060&r2=283061&view=diff
==============================================================================
--- cfe/trunk/test/Driver/ppc-features.cpp (original)
+++ cfe/trunk/test/Driver/ppc-features.cpp Sat Oct  1 21:10:45 2016
@@ -14,27 +14,27 @@
 
 // check -msoft-float option for ppc32
 // RUN: %clang -target powerpc-unknown-linux-gnu %s -msoft-float -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-SOFTFLOAT %s
-// CHECK-SOFTFLOAT: "-target-feature" "+soft-float"
+// CHECK-SOFTFLOAT: "-target-feature" "-hard-float"
 
 // check -mfloat-abi=soft option for ppc32
 // RUN: %clang -target powerpc-unknown-linux-gnu %s -mfloat-abi=soft -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-FLOATABISOFT %s
-// CHECK-FLOATABISOFT: "-target-feature" "+soft-float"
+// CHECK-FLOATABISOFT: "-target-feature" "-hard-float"
 
 // check -mhard-float option for ppc32
 // RUN: %clang -target powerpc-unknown-linux-gnu %s -mhard-float -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-HARDFLOAT %s
-// CHECK-HARDFLOAT-NOT: "-target-feature" "+soft-float"
+// CHECK-HARDFLOAT-NOT: "-target-feature" "-hard-float"
 
 // check -mfloat-abi=hard option for ppc32
 // RUN: %clang -target powerpc-unknown-linux-gnu %s -mfloat-abi=hard -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-FLOATABIHARD %s
-// CHECK-FLOATABIHARD-NOT: "-target-feature" "+soft-float"
+// CHECK-FLOATABIHARD-NOT: "-target-feature" "-hard-float"
 
 // check combine -mhard-float -msoft-float option for ppc32
 // RUN: %clang -target powerpc-unknown-linux-gnu %s -mhard-float -msoft-float -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-HARDSOFT %s
-// CHECK-HARDSOFT: "-target-feature" "+soft-float"
+// CHECK-HARDSOFT: "-target-feature" "-hard-float"
 
 // check combine -msoft-float -mhard-float option for ppc32
 // RUN: %clang -target powerpc-unknown-linux-gnu %s -msoft-float -mhard-float -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-SOFTHARD %s
-// CHECK-SOFTHARD-NOT: "-target-feature" "+soft-float"
+// CHECK-SOFTHARD-NOT: "-target-feature" "-hard-float"
 
 // check -mfloat-abi=x option
 // RUN: %clang -target powerpc-unknown-linux-gnu %s -mfloat-abi=x -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-ERRMSG %s
@@ -42,19 +42,19 @@
 
 // check -msoft-float option for ppc64
 // RUN: %clang -target powerpc64-unknown-linux-gnu %s -msoft-float -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-SOFTFLOAT64 %s
-// CHECK-SOFTFLOAT64: error: invalid float ABI 'soft float is not supported for ppc64'
+// CHECK-SOFTFLOAT64: "-target-feature" "-hard-float"
 
 // check -mfloat-abi=soft option for ppc64
 // RUN: %clang -target powerpc64-unknown-linux-gnu %s -mfloat-abi=soft -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-FLOATABISOFT64 %s
-// CHECK-FLOATABISOFT64: error: invalid float ABI 'soft float is not supported for ppc64'
+// CHECK-FLOATABISOFT64: "-target-feature" "-hard-float"
 
 // check -msoft-float option for ppc64
 // RUN: %clang -target powerpc64le-unknown-linux-gnu %s -msoft-float -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-SOFTFLOAT64le %s
-// CHECK-SOFTFLOAT64le: error: invalid float ABI 'soft float is not supported for ppc64'
+// CHECK-SOFTFLOAT64le: "-target-feature" "-hard-float"
 
 // check -mfloat-abi=soft option for ppc64
 // RUN: %clang -target powerpc64le-unknown-linux-gnu %s -mfloat-abi=soft -### -o %t.o 2>&1 | FileCheck --check-prefix=CHECK-FLOATABISOFT64le %s
-// CHECK-FLOATABISOFT64le: error: invalid float ABI 'soft float is not supported for ppc64'
+// CHECK-FLOATABISOFT64le: "-target-feature" "-hard-float"
 
 // CHECK: invalid argument '-faltivec' only allowed with 'ppc/ppc64/ppc64le'
 




More information about the cfe-commits mailing list