[llvm-commits] [llvm] r46251 - /llvm/trunk/include/llvm/Support/Compiler.h
Author: asl
Date: Tue Jan 22 05:01:23 2008
New Revision: 46251
URL: http://llvm.org/viewvc/llvm-project?rev=46251&view=rev
Log:
Silence a warning. Should we turn this into configure-time check?
Modified:
llvm/trunk/include/llvm/Support/Compiler.h
Modified: llvm/trunk/include/llvm/Support/Compiler.h
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/include/llvm/Support/Compiler.h?rev=46251&r1=46250&r2=46251&view=diff
==
--- llvm/trunk/include/llvm/Support/Compiler.h (original)
+++ llvm/trunk/include/llvm/Support/Compiler.h Tue Jan 22 05:01:23 2008
@@ -17,7 +17,7 @@
// The VISIBILITY_HIDDEN macro, used for marking classes with the GCC-specific
// visibility("hidden") attribute.
-#if __GNUC__ >= 4
+#if (__GNUC__ >= 4) && !defined(__MINGW32__)
#define VISIBILITY_HIDDEN __attribute__ ((visibility("hidden")))
#else
#define VISIBILITY_HIDDEN
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm] r46250 - /llvm/trunk/utils/TableGen/Record.h
Author: asl
Date: Tue Jan 22 05:00:07 2008
New Revision: 46250
URL: http://llvm.org/viewvc/llvm-project?rev=46250&view=rev
Log:
Add interator interface to DAGInit also
Modified:
llvm/trunk/utils/TableGen/Record.h
Modified: llvm/trunk/utils/TableGen/Record.h
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/utils/TableGen/Record.h?rev=46250&r1=46249&r2=46250&view=diff
==
--- llvm/trunk/utils/TableGen/Record.h (original)
+++ llvm/trunk/utils/TableGen/Record.h Tue Jan 22 05:00:07 2008
@@ -898,6 +898,28 @@
virtual Init *resolveReferences(Record &R, const RecordVal *RV);
virtual std::string getAsString() const;
+
+ typedef std::vector::iterator arg_iterator;
+ typedef std::vector::const_iterator const_arg_iterator;
+ typedef std::vector::iterator name_iterator;
+ typedef std::vector::const_iterator const_name_iterator;
+
+ inline arg_iteratorarg_begin() { return Args.begin(); }
+ inline const_arg_iterator arg_begin() const { return Args.begin(); }
+ inline arg_iteratorarg_end () { return Args.end(); }
+ inline const_arg_iterator arg_end () const { return Args.end(); }
+
+ inline size_t arg_size () const { return Args.size(); }
+ inline boolarg_empty() const { return Args.empty(); }
+
+ inline name_iterator name_begin() { return ArgNames.begin(); }
+ inline const_name_iterator name_begin() const { return ArgNames.begin(); }
+ inline name_iterator name_end () { return ArgNames.end(); }
+ inline const_name_iterator name_end () const { return ArgNames.end(); }
+
+ inline size_t name_size () const { return ArgNames.size(); }
+ inline boolname_empty() const { return ArgNames.empty(); }
+
};
//===--===//
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
Re: [llvm-commits] [llvm] r46251 - /llvm/trunk/include/llvm/Support/Compiler.h
On Jan 22, 2008, at 3:01 AM, Anton Korobeynikov wrote: > Silence a warning. Should we turn this into configure-time check? Probably. There's a custom one in gcc already (gcc/configure.ac). -eric ___ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] a question about type conversion propagation and elimination
I am a newbie to LLVM, so I have to say sorry if I asked the question in the
wrong place.
In some cases when I generate LLVM IR from machine assembly(with limited type
information) I have to convert the pointers to I32, after the standard mem2reg
pass there still are things like:
inttoptr i32 %1 to i8*
While in fact this can be replaced by the I8* operations, because the I8* can
be unsigned compared, and get rid of the redundant Instrs, Is there any pass
can do this, or is there any tricks to avoid these redundant conversions while
retain the meaning of the program?
define i8* @tag_on(i8* %ptr, i8* %end, i8 %tag) {
Entry:
ptrtoint i8* %end to i32; :0 [#uses=2]
ptrtoint i8* %ptr to i32; :1 [#uses=4]
icmp ugt i32 %0, %1 ; :2 [#uses=1]
br i1 %2, label %B1, label %B2
B1: ; preds = %Entry
inttoptr i32 %1 to i8* ; :3 [#uses=1]
store i8 1, i8* %3
add i32 %1, 1 ; :4 [#uses=1]
br label %B2
B2: ; preds = %Entry, %B1
%eax.0 = phi i32 [ %4, %B1 ], [ %1, %Entry ];
[#uses=4]
icmp ugt i32 %0, %eax.0 ; :5 [#uses=1]
br i1 %5, label %B3, label %exit
B3: ; preds = %B2
inttoptr i32 %eax.0 to i8* ; :6 [#uses=1]
store i8 1, i8* %6
add i32 %eax.0, 1 ; :7 [#uses=1]
br label %exit
exit: ; preds = %B2, %B3
%eax.1 = phi i32 [ %7, %B3 ], [ %eax.0, %B2 ] ;
[#uses=1]
inttoptr i32 %eax.1 to i8* ; :8 [#uses=1]
ret i8* %8
}
humeafo
2008-01-22
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
Re: [llvm-commits] a question about type conversion propagation and elimination
On Jan 22, 2008, at 2:10 AM, humeafo wrote:
I am a newbie to LLVM, so I have to say sorry if I asked the
question in the wrong place.
In some cases when I generate LLVM IR from machine assembly(with
limited type information) I have to convert the pointers to I32,
after the standard mem2reg pass there still are things like:
inttoptr i32 %1 to i8*
While in fact this can be replaced by the I8* operations, because
the I8* can be unsigned compared, and get rid of the redundant
Instrs, Is there any pass
can do this, or is there any tricks to avoid these redundant
conversions while retain the meaning of the program?
Sure, the instcombine pass will do this. Make sure to add the correct
target data string though. For example, if I add the datalayout
string for x86-32 darwin:
target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-
i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-
f80:128:128"
Instcombine produces:
define i8* @tag_on(i8* %ptr, i8* %end, i8 %tag) {
Entry:
icmp ugt i8* %end, %ptr ; :0 [#uses=1]
br i1 %0, label %B1, label %B2
B1: ; preds = %Entry
store i8 1, i8* %ptr
%ctg2 = getelementptr i8* %ptr, i32 1 ; [#uses=1]
br label %B2
B2: ; preds = %B1, %Entry
%eax.0.in = phi i8* [ %ctg2, %B1 ], [ %ptr, %Entry ];
[#uses=4]
icmp ult i8* %eax.0.in, %end; :1 [#uses=1]
br i1 %1, label %B3, label %exit
B3: ; preds = %B2
store i8 1, i8* %eax.0.in
%ctg21 = getelementptr i8* %eax.0.in, i32 1 ;
[#uses=1]
br label %exit
exit: ; preds = %B3, %B2
%eax.1.in = phi i8* [ %ctg21, %B3 ], [ %eax.0.in, %B2 ] ;
[#uses=1]
ret i8* %eax.1.in
}
Note that it eliminated all the ptrtoint casts (in this case) and
infered some getelementptrs.
I'm sending this to the llvmdev list, which is a better place for high-
level questions than llvm-commits.
-Chris
define i8* @tag_on(i8* %ptr, i8* %end, i8 %tag) {
Entry:
ptrtoint i8* %end to i32; :0 [#uses=2]
ptrtoint i8* %ptr to i32; :1 [#uses=4]
icmp ugt i32 %0, %1 ; :2 [#uses=1]
br i1 %2, label %B1, label %B2
B1: ; preds = %Entry
inttoptr i32 %1 to i8* ; :3 [#uses=1]
store i8 1, i8* %3
add i32 %1, 1 ; :4 [#uses=1]
br label %B2
B2: ; preds = %Entry, %B1
%eax.0 = phi i32 [ %4, %B1 ], [ %1, %Entry ];
[#uses=4]
icmp ugt i32 %0, %eax.0 ; :5 [#uses=1]
br i1 %5, label %B3, label %exit
B3: ; preds = %B2
inttoptr i32 %eax.0 to i8* ; :6 [#uses=1]
store i8 1, i8* %6
add i32 %eax.0, 1 ; :7 [#uses=1]
br label %exit
exit: ; preds = %B2, %B3
%eax.1 = phi i32 [ %7, %B3 ], [ %eax.0, %B2 ] ;
[#uses=1]
inttoptr i32 %eax.1 to i8* ; :8 [#uses=1]
ret i8* %8
}
humeafo
2008-01-22
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm] r46254 - /llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
Author: lattner
Date: Tue Jan 22 13:09:33 2008
New Revision: 46254
URL: http://llvm.org/viewvc/llvm-project?rev=46254&view=rev
Log:
Simplify SelectionDAG::getNode so that a big switch stmt is not #ifndef
NDEBUG. This is in response to a really nasty bug I introduced that
Dale tracked down, hopefully this won't happen in the future.
Many thanks Dale.
Modified:
llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
Modified: llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAG.cpp?rev=46254&r1=46253&r2=46254&view=diff
==
--- llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAG.cpp (original)
+++ llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAG.cpp Tue Jan 22 13:09:33
2008
@@ -1859,15 +1859,34 @@
SDOperand SelectionDAG::getNode(unsigned Opcode, MVT::ValueType VT,
SDOperand N1, SDOperand N2) {
-#ifndef NDEBUG
+ ConstantSDNode *N1C = dyn_cast(N1.Val);
+ ConstantSDNode *N2C = dyn_cast(N2.Val);
switch (Opcode) {
+ default: break;
case ISD::TokenFactor:
assert(VT == MVT::Other && N1.getValueType() == MVT::Other &&
N2.getValueType() == MVT::Other && "Invalid token factor!");
+// Fold trivial token factors.
+if (N1.getOpcode() == ISD::EntryToken) return N2;
+if (N2.getOpcode() == ISD::EntryToken) return N1;
break;
case ISD::AND:
+assert(MVT::isInteger(VT) && N1.getValueType() == N2.getValueType() &&
+ N1.getValueType() == VT && "Binary operator types must match!");
+// (X & 0) -> 0. This commonly occurs when legalizing i64 values, so it's
+// worth handling here.
+if (N2C && N2C->getValue() == 0)
+ return N2;
+break;
case ISD::OR:
case ISD::XOR:
+assert(MVT::isInteger(VT) && N1.getValueType() == N2.getValueType() &&
+ N1.getValueType() == VT && "Binary operator types must match!");
+// (X ^| 0) -> X. This commonly occurs when legalizing i64 values, so it's
+// worth handling here.
+if (N2C && N2C->getValue() == 0)
+ return N1;
+break;
case ISD::UDIV:
case ISD::UREM:
case ISD::MULHU:
@@ -1879,8 +1898,6 @@
case ISD::MUL:
case ISD::SDIV:
case ISD::SREM:
-assert(MVT::isInteger(N1.getValueType()) && "Should use F* for FP ops");
-// fall through.
case ISD::FADD:
case ISD::FSUB:
case ISD::FMUL:
@@ -1912,6 +1929,7 @@
"Cannot FP_ROUND_INREG integer types");
assert(MVT::getSizeInBits(EVT) <= MVT::getSizeInBits(VT) &&
"Not rounding down!");
+if (cast(N2)->getVT() == VT) return N1; // Not actually
rounding.
break;
}
case ISD::FP_ROUND:
@@ -1919,33 +1937,83 @@
MVT::isFloatingPoint(N1.getValueType()) &&
MVT::getSizeInBits(VT) <= MVT::getSizeInBits(N1.getValueType()) &&
isa(N2) && "Invalid FP_ROUND!");
+if (N1.getValueType() == VT) return N1; // noop conversion.
break;
case ISD::AssertSext:
- case ISD::AssertZext:
- case ISD::SIGN_EXTEND_INREG: {
+ case ISD::AssertZext: {
MVT::ValueType EVT = cast(N2)->getVT();
assert(VT == N1.getValueType() && "Not an inreg extend!");
assert(MVT::isInteger(VT) && MVT::isInteger(EVT) &&
"Cannot *_EXTEND_INREG FP types");
assert(MVT::getSizeInBits(EVT) <= MVT::getSizeInBits(VT) &&
"Not extending!");
+break;
}
+ case ISD::SIGN_EXTEND_INREG: {
+MVT::ValueType EVT = cast(N2)->getVT();
+assert(VT == N1.getValueType() && "Not an inreg extend!");
+assert(MVT::isInteger(VT) && MVT::isInteger(EVT) &&
+ "Cannot *_EXTEND_INREG FP types");
+assert(MVT::getSizeInBits(EVT) <= MVT::getSizeInBits(VT) &&
+ "Not extending!");
+if (EVT == VT) return N1; // Not actually extending
- default: break;
- }
-#endif
-
- ConstantSDNode *N1C = dyn_cast(N1.Val);
- ConstantSDNode *N2C = dyn_cast(N2.Val);
- if (N1C) {
-if (Opcode == ISD::SIGN_EXTEND_INREG) {
+if (N1C) {
int64_t Val = N1C->getValue();
unsigned FromBits = MVT::getSizeInBits(cast(N2)->getVT());
Val <<= 64-FromBits;
Val >>= 64-FromBits;
return getConstant(Val, VT);
}
+break;
+ }
+ case ISD::EXTRACT_VECTOR_ELT:
+assert(N2C && "Bad EXTRACT_VECTOR_ELT!");
+
+// EXTRACT_VECTOR_ELT of CONCAT_VECTORS is often formed while lowering is
+// expanding copies of large vectors from registers.
+if (N1.getOpcode() == ISD::CONCAT_VECTORS &&
+N1.getNumOperands() > 0) {
+ unsigned Factor =
+MVT::getVectorNumElements(N1.getOperand(0).getValueType());
+ return getNode(ISD::EXTRACT_VECTOR_ELT, VT,
+ N1.getOperand(N2C->getValue() / Factor),
+ getConstant(N2C->getValue() % Factor, N2.getValueType()));
+}
+
+// EXTRACT_VECTOR_ELT of BUILD_VECTOR is often formed while lowering is
+// exp
Re: [llvm-commits] [llvm] r46220 - /llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c
This fails with llvm-gcc-4.0, darwin 8, x86.
../../src/gcc/llvm-types.cpp:82: failed assertion `(!TYPE_SIZE(Tr)
|| !Ty->isSized() || !isInt64(TYPE_SIZE(Tr), true) || getInt64
(TYPE_SIZE(Tr), true) == getTargetData().getABITypeSizeInBits(Ty)) &&
"LLVM type size doesn't match GCC type size!"'
2008-01-21-PackedBitFields.c is also failing for me.
-Tanya
On Jan 21, 2008, at 2:15 PM, Devang Patel wrote:
> Author: dpatel
> Date: Mon Jan 21 16:15:58 2008
> New Revision: 46220
>
> URL: http://llvm.org/viewvc/llvm-project?rev=46220&view=rev
> Log:
> New test.
>
> Added:
> llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c
>
> Added: llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CFrontend/
> 2008-01-21-PackedStructField.c?rev=46220&view=auto
>
> ==
>
> --- llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c (added)
> +++ llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c Mon
> Jan 21 16:15:58 2008
> @@ -0,0 +1,5 @@
> +// RUN: %llvmgcc %s -S -o -
> +
> +struct X { long double b; unsigned char c; double __attribute__
> ((packed)) d; };
> +struct X x = { 3.0L, 5, 3.0 };
> +
>
>
> ___
> llvm-commits mailing list
> llvm-commits@cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm-gcc-4.2] r46261 - /llvm-gcc-4.2/trunk/gcc/objc/objc-act.c
Author: void Date: Tue Jan 22 17:06:58 2008 New Revision: 46261 URL: http://llvm.org/viewvc/llvm-project?rev=46261&view=rev Log: We were getting this error: embedding a directive within macro arguments is not portable build_tree_list is a macro... Modified: llvm-gcc-4.2/trunk/gcc/objc/objc-act.c Modified: llvm-gcc-4.2/trunk/gcc/objc/objc-act.c URL: http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/objc/objc-act.c?rev=46261&r1=46260&r2=46261&view=diff == --- llvm-gcc-4.2/trunk/gcc/objc/objc-act.c (original) +++ llvm-gcc-4.2/trunk/gcc/objc/objc-act.c Tue Jan 22 17:06:58 2008 @@ -10867,14 +10867,14 @@ /* LLVM LOCAL begin make initializer size match type size */ /* APPLE LOCAL ObjC new abi */ - initlist = build_tree_list (NULL_TREE, build_int_cst ( #ifdef OBJCPLUS - NULL_TREE, + initlist = build_tree_list (NULL_TREE, + build_int_cst (NULL_TREE, init_val)); #else - newabi ? NULL_TREE : ptr_type_node, + initlist = build_tree_list (NULL_TREE, + build_int_cst (newabi ? NULL_TREE : ptr_type_node, +init_val)); #endif - -init_val)); /* LLVM LOCAL end */ initlist = tree_cons (NULL_TREE, build_int_cst (NULL_TREE, size), initlist); initlist = tree_cons (NULL_TREE, list, initlist); ___ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm] r46262 - in /llvm/trunk: lib/Target/X86/X86CallingConv.td test/CodeGen/X86/sse-varargs.ll
Author: evancheng
Date: Tue Jan 22 17:26:53 2008
New Revision: 46262
URL: http://llvm.org/viewvc/llvm-project?rev=46262&view=rev
Log:
SSE varargs arguments are passed in memory.
Added:
llvm/trunk/test/CodeGen/X86/sse-varargs.ll
Modified:
llvm/trunk/lib/Target/X86/X86CallingConv.td
Modified: llvm/trunk/lib/Target/X86/X86CallingConv.td
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86CallingConv.td?rev=46262&r1=46261&r2=46262&view=diff
==
--- llvm/trunk/lib/Target/X86/X86CallingConv.td (original)
+++ llvm/trunk/lib/Target/X86/X86CallingConv.td Tue Jan 22 17:26:53 2008
@@ -190,8 +190,8 @@
CCIfType<[f80], CCAssignToStack<0, 4>>,
// The first 4 vector arguments are passed in XMM registers.
- CCIfType<[v16i8, v8i16, v4i32, v2i64, v4f32, v2f64],
- CCAssignToReg<[XMM0, XMM1, XMM2, XMM3]>>,
+ CCIfNotVarArg>>,
// Other vectors get 16-byte stack slots that are 16-byte aligned.
CCIfType<[v16i8, v8i16, v4i32, v2i64, v4f32, v2f64], CCAssignToStack<16,
16>>,
Added: llvm/trunk/test/CodeGen/X86/sse-varargs.ll
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/sse-varargs.ll?rev=46262&view=auto
==
--- llvm/trunk/test/CodeGen/X86/sse-varargs.ll (added)
+++ llvm/trunk/test/CodeGen/X86/sse-varargs.ll Tue Jan 22 17:26:53 2008
@@ -0,0 +1,9 @@
+; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse2 | grep xmm | grep esp
+
+define i32 @t() nounwind {
+entry:
+ tail call void (i32, ...)* @foo( i32 1, <4 x i32> < i32 10, i32 11, i32
12, i32 13 > ) nounwind
+ ret i32 0
+}
+
+declare void @foo(i32, ...)
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm] r46263 - /llvm/trunk/lib/CodeGen/StrongPHIElimination.cpp
Author: resistor
Date: Tue Jan 22 17:58:54 2008
New Revision: 46263
URL: http://llvm.org/viewvc/llvm-project?rev=46263&view=rev
Log:
Fix an iterator invalidation issue.
Modified:
llvm/trunk/lib/CodeGen/StrongPHIElimination.cpp
Modified: llvm/trunk/lib/CodeGen/StrongPHIElimination.cpp
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/CodeGen/StrongPHIElimination.cpp?rev=46263&r1=46262&r2=46263&view=diff
==
--- llvm/trunk/lib/CodeGen/StrongPHIElimination.cpp (original)
+++ llvm/trunk/lib/CodeGen/StrongPHIElimination.cpp Tue Jan 22 17:58:54 2008
@@ -751,11 +751,17 @@
// FIXME: Insert last-minute copies
// Remove PHIs
- for (MachineFunction::iterator I = Fn.begin(), E = Fn.end(); I != E; ++I)
+ std::vector phis;
+ for (MachineFunction::iterator I = Fn.begin(), E = Fn.end(); I != E; ++I) {
for (MachineBasicBlock::iterator BI = I->begin(), BE = I->end();
BI != BE; ++BI)
if (BI->getOpcode() == TargetInstrInfo::PHI)
-BI->eraseFromParent();
+phis.push_back(BI);
+ }
+
+ for (std::vector::iterator I = phis.begin(), E = phis.end();
+ I != E; ++I)
+(*I)->eraseFromParent();
return false;
}
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm-gcc-4.0] r46264 - in /llvm-gcc-4.0/trunk/gcc: llvm-internal.h llvm-types.cpp
Author: dpatel
Date: Tue Jan 22 18:07:02 2008
New Revision: 46264
URL: http://llvm.org/viewvc/llvm-project?rev=46264&view=rev
Log:
Backport patches(r46210 and r46212) from llvm-gcc-4.2
Modified:
llvm-gcc-4.0/trunk/gcc/llvm-internal.h
llvm-gcc-4.0/trunk/gcc/llvm-types.cpp
Modified: llvm-gcc-4.0/trunk/gcc/llvm-internal.h
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.0/trunk/gcc/llvm-internal.h?rev=46264&r1=46263&r2=46264&view=diff
==
--- llvm-gcc-4.0/trunk/gcc/llvm-internal.h (original)
+++ llvm-gcc-4.0/trunk/gcc/llvm-internal.h Tue Jan 22 18:07:02 2008
@@ -157,7 +157,7 @@
private:
const Type *ConvertRECORD(tree_node *type, tree_node *orig_type);
const Type *ConvertUNION(tree_node *type, tree_node *orig_type);
- void DecodeStructFields(tree_node *Field, StructTypeConversionInfo &Info);
+ bool DecodeStructFields(tree_node *Field, StructTypeConversionInfo &Info);
void DecodeStructBitField(tree_node *Field, StructTypeConversionInfo &Info);
};
Modified: llvm-gcc-4.0/trunk/gcc/llvm-types.cpp
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.0/trunk/gcc/llvm-types.cpp?rev=46264&r1=46263&r2=46264&view=diff
==
--- llvm-gcc-4.0/trunk/gcc/llvm-types.cpp (original)
+++ llvm-gcc-4.0/trunk/gcc/llvm-types.cpp Tue Jan 22 18:07:02 2008
@@ -1180,6 +1180,8 @@
ExtraBitsAvailable = E;
}
+ bool isPacked() { return Packed; }
+
void markAsPacked() {
Packed = true;
}
@@ -1427,37 +1429,9 @@
void addNewBitField(unsigned Size, unsigned FirstUnallocatedByte);
- void convertToPacked();
-
void dump() const;
};
-// LLVM disagrees as to where to put field natural field ordering.
-// ordering. Therefore convert to a packed struct.
-void StructTypeConversionInfo::convertToPacked() {
- assert (!Packed && "Packing a packed struct!");
- Packed = true;
-
- // Fill the padding that existed from alignment restrictions
- // with byte arrays to ensure the same layout when converting
- // to a packed struct.
- for (unsigned x = 1; x < ElementOffsetInBytes.size(); ++x) {
-if (ElementOffsetInBytes[x-1] + ElementSizeInBytes[x-1]
-< ElementOffsetInBytes[x]) {
- uint64_t padding = ElementOffsetInBytes[x]
-- ElementOffsetInBytes[x-1] - ElementSizeInBytes[x-1];
- const Type *Pad = Type::Int8Ty;
- Pad = ArrayType::get(Pad, padding);
- ElementOffsetInBytes.insert(ElementOffsetInBytes.begin() + x,
- ElementOffsetInBytes[x-1] +
- ElementSizeInBytes[x-1]);
- ElementSizeInBytes.insert(ElementSizeInBytes.begin() + x, padding);
- Elements.insert(Elements.begin() + x, Pad);
- PaddingElement.insert(PaddingElement.begin() + x, true);
-}
- }
-}
-
// Add new element which is a bit field. Size is not the size of bit filed,
// but size of bits required to determine type of new Field which will be
// used to access this bit field.
@@ -1545,17 +1519,18 @@
/// DecodeStructFields - This method decodes the specified field, if it is a
/// FIELD_DECL, adding or updating the specified StructTypeConversionInfo to
-/// reflect it.
-void TypeConverter::DecodeStructFields(tree Field,
+/// reflect it. Return tree if field is decode correctly. Otherwise return
+/// false.
+bool TypeConverter::DecodeStructFields(tree Field,
StructTypeConversionInfo &Info) {
if (TREE_CODE(Field) != FIELD_DECL ||
TREE_CODE(DECL_FIELD_OFFSET(Field)) != INTEGER_CST)
-return;
+return true;
// Handle bit-fields specially.
if (DECL_BIT_FIELD_TYPE(Field)) {
DecodeStructBitField(Field, Info);
-return;
+return true;
}
Info.allFieldsAreNotBitFields();
@@ -1573,12 +1548,19 @@
if (!Info.ResizeLastElementIfOverlapsWith(StartOffsetInBytes, Field, Ty)) {
// LLVM disagrees as to where this field should go in the natural field
// ordering. Therefore convert to a packed struct and try again.
-Info.convertToPacked();
-DecodeStructFields(Field, Info);
+return false;
+ }
+ else if (TYPE_USER_ALIGN(TREE_TYPE(Field))
+ && DECL_ALIGN_UNIT(Field) != Info.getTypeAlignment(Ty)
+ && !Info.isPacked()) {
+// If Field has user defined alignment and it does not match Ty alignment
+// then convert to a packed struct and try again.
+return false;
} else
// At this point, we know that adding the element will happen at the right
// offset. Add it.
Info.addElement(Ty, StartOffsetInBytes, Info.getTypeSize(Ty));
+ return true;
}
/// DecodeStructBitField - This method decodes the specified bit-field, adding
@@ -1735,8 +1717,24 @@
// Convert over all of the elements of the struct.
- for (tree Field = TYPE_FIELDS(type); Field; Field = TREE_CHAIN(Field))
-DecodeStru
[llvm-commits] [llvm-gcc-4.2] r46265 - /llvm-gcc-4.2/trunk/gcc/llvm-backend.cpp
Author: johannes
Date: Tue Jan 22 18:48:04 2008
New Revision: 46265
URL: http://llvm.org/viewvc/llvm-project?rev=46265&view=rev
Log:
Fix handling of CONST_DECL with section information.
Modified:
llvm-gcc-4.2/trunk/gcc/llvm-backend.cpp
Modified: llvm-gcc-4.2/trunk/gcc/llvm-backend.cpp
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/llvm-backend.cpp?rev=46265&r1=46264&r2=46265&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/llvm-backend.cpp (original)
+++ llvm-gcc-4.2/trunk/gcc/llvm-backend.cpp Tue Jan 22 18:48:04 2008
@@ -919,15 +919,6 @@
GV->setSection(Section);
#endif
}
-#ifdef LLVM_IMPLICIT_TARGET_GLOBAL_VAR_SECTION
-else if (TREE_CODE(decl) == CONST_DECL) {
- if (const char *Section =
- LLVM_IMPLICIT_TARGET_GLOBAL_VAR_SECTION(decl)) {
-GV->setSection(Section);
- }
-}
-#endif
-
// Set the alignment for the global if one of the following condition is
met
// 1) DECL_ALIGN_UNIT does not match alignment as per ABI specification
@@ -947,8 +938,16 @@
// Add annotate attributes for globals
if (DECL_ATTRIBUTES(decl))
AddAnnotateAttrsToGlobal(GV, decl);
- }
+#ifdef LLVM_IMPLICIT_TARGET_GLOBAL_VAR_SECTION
+ } else if (TREE_CODE(decl) == CONST_DECL) {
+if (const char *Section =
+LLVM_IMPLICIT_TARGET_GLOBAL_VAR_SECTION(decl)) {
+ GV->setSection(Section);
+}
+#endif
+ }
+
if (TheDebugInfo) TheDebugInfo->EmitGlobalVariable(GV, decl);
TREE_ASM_WRITTEN(decl) = 1;
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm-gcc-4.2] r46266 - /llvm-gcc-4.2/trunk/gcc/config/darwin.c
Author: johannes
Date: Tue Jan 22 18:48:48 2008
New Revision: 46266
URL: http://llvm.org/viewvc/llvm-project?rev=46266&view=rev
Log:
Cosmetic changes to placate gcc testsuite.
Modified:
llvm-gcc-4.2/trunk/gcc/config/darwin.c
Modified: llvm-gcc-4.2/trunk/gcc/config/darwin.c
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/config/darwin.c?rev=46266&r1=46265&r2=46266&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/config/darwin.c (original)
+++ llvm-gcc-4.2/trunk/gcc/config/darwin.c Tue Jan 22 18:48:48 2008
@@ -1527,11 +1527,11 @@
if (!strcmp(IDENTIFIER_POINTER(typename), "__builtin_ObjCString")) {
if (flag_next_runtime)
-return "__OBJC,__cstring_object,regular,no_dead_strip";
+return "__OBJC, __cstring_object,regular,no_dead_strip";
else
-return "__OBJC,__string_object,no_dead_strip";
+return "__OBJC, __string_object,no_dead_strip";
} else if (!strcmp(IDENTIFIER_POINTER(typename), "__builtin_CFString")) {
- return "__DATA,__cfstring";
+ return "__DATA, __cfstring";
} else {
return 0;
}
@@ -1580,7 +1580,7 @@
return "__OBJC,__module_info,regular,no_dead_strip";
else if (!strncmp (name, "IMAGE_INFO", 10))
return (flag_objc_abi == 1 ?
-"__OBJC,__image_info,regular" /*,no_dead_strip";*/ :
+"__OBJC, __image_info,regular" /*,no_dead_strip";*/ :
"__DATA, __objc_imageinfo, regular, no_dead_strip");
else if (!strncmp (name, "PROTOCOL_INSTANCE_METHODS_", 26))
return "__OBJC,__cat_inst_meth,regular,no_dead_strip";
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm] r46267 - in /llvm/trunk/lib/Target: ARM/ARMAsmPrinter.cpp PowerPC/PPCAsmPrinter.cpp X86/X86AsmPrinter.cpp
Author: johannes
Date: Tue Jan 22 18:58:14 2008
New Revision: 46267
URL: http://llvm.org/viewvc/llvm-project?rev=46267&view=rev
Log:
Honor explicit section information on Darwin.
Modified:
llvm/trunk/lib/Target/ARM/ARMAsmPrinter.cpp
llvm/trunk/lib/Target/PowerPC/PPCAsmPrinter.cpp
llvm/trunk/lib/Target/X86/X86AsmPrinter.cpp
Modified: llvm/trunk/lib/Target/ARM/ARMAsmPrinter.cpp
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/ARM/ARMAsmPrinter.cpp?rev=46267&r1=46266&r2=46267&view=diff
==
--- llvm/trunk/lib/Target/ARM/ARMAsmPrinter.cpp (original)
+++ llvm/trunk/lib/Target/ARM/ARMAsmPrinter.cpp Tue Jan 22 18:58:14 2008
@@ -908,6 +908,10 @@
std::string SectionName = ".section " + I->getSection();
SectionName += ",\"aw\",%progbits";
SwitchToDataSection(SectionName.c_str());
+ } else if (I->hasSection() && Subtarget->isTargetDarwin()) {
+// Honor all section names on Darwin; ObjC uses this
+std::string SectionName = ".section " + I->getSection();
+SwitchToDataSection(SectionName.c_str());
} else {
if (C->isNullValue() && !NoZerosInBSS && TAI->getBSSSection())
SwitchToDataSection(I->isThreadLocal() ? TAI->getTLSBSSSection() :
Modified: llvm/trunk/lib/Target/PowerPC/PPCAsmPrinter.cpp
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/PowerPC/PPCAsmPrinter.cpp?rev=46267&r1=46266&r2=46267&view=diff
==
--- llvm/trunk/lib/Target/PowerPC/PPCAsmPrinter.cpp (original)
+++ llvm/trunk/lib/Target/PowerPC/PPCAsmPrinter.cpp Tue Jan 22 18:58:14 2008
@@ -958,8 +958,11 @@
break;
}
}
-
-if (!I->isConstant())
+if (I->hasSection()) {
+ // Honor all section names on Darwin; ObjC uses this
+ std::string SectionName = ".section " + I->getSection();
+ SwitchToDataSection(SectionName.c_str());
+} else if (!I->isConstant())
SwitchToDataSection(TAI->getDataSection(), I);
else {
// Read-only data.
Modified: llvm/trunk/lib/Target/X86/X86AsmPrinter.cpp
URL:
http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86AsmPrinter.cpp?rev=46267&r1=46266&r2=46267&view=diff
==
--- llvm/trunk/lib/Target/X86/X86AsmPrinter.cpp (original)
+++ llvm/trunk/lib/Target/X86/X86AsmPrinter.cpp Tue Jan 22 18:58:14 2008
@@ -275,7 +275,10 @@
assert(!Subtarget->isTargetDarwin());
SectionName += ",\"aw\",@progbits";
}
-
+SwitchToDataSection(SectionName.c_str());
+ } else if (I->hasSection() && Subtarget->isTargetDarwin()) {
+// Honor all section names on Darwin; ObjC uses this
+std::string SectionName = ".section " + I->getSection();
SwitchToDataSection(SectionName.c_str());
} else {
if (C->isNullValue() && !NoZerosInBSS && TAI->getBSSSection())
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm-gcc-4.2] r46269 - in /llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg: const-cfstring-4-64bit.mm const-cfstring-4.mm const-str-10.mm const-str-11.mm const-str-9.mm objc-gc-section-1.m
Author: johannes
Date: Tue Jan 22 19:40:15 2008
New Revision: 46269
URL: http://llvm.org/viewvc/llvm-project?rev=46269&view=rev
Log:
Enhance some more tests to take functionally
equivalent llvm syntax.
Modified:
llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-cfstring-4-64bit.mm
llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-cfstring-4.mm
llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-10.mm
llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-11.mm
llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-9.mm
llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/objc-gc-section-1.mm
llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/objc-gc-section-2.mm
llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/objc2-protocol-ext-1.mm
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-cfstring-4-64bit.mm
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/obj-c%2B%2B.dg/const-cfstring-4-64bit.mm?rev=46269&r1=46268&r2=46269&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-cfstring-4-64bit.mm
(original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-cfstring-4-64bit.mm Tue
Jan 22 19:40:15 2008
@@ -8,4 +8,5 @@
CFStringRef appKey = (CFStringRef) @"com.apple.soundpref";
/* { dg-final { scan-assembler ".section __DATA, __cfstring" } } */
-/* { dg-final { scan-assembler
".quad\t___CFConstantStringClassReference\n\t.long\t1992\n\t.space
4\n\t.quad\t.*\n\t.quad\t19\n\t.data" } } */
+/* LLVM LOCAL accept llvm syntax */
+/* { dg-final { scan-assembler
".quad\t___CFConstantStringClassReference\n\t.long\t1992\n\t.space(
|\t)4\n\t.quad\t.*\n\t.quad\t19\n" } } */
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-cfstring-4.mm
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/obj-c%2B%2B.dg/const-cfstring-4.mm?rev=46269&r1=46268&r2=46269&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-cfstring-4.mm (original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-cfstring-4.mm Tue Jan 22
19:40:15 2008
@@ -10,4 +10,5 @@
CFStringRef appKey = (CFStringRef) @"com.apple.soundpref";
/* { dg-final { scan-assembler ".section __DATA, __cfstring" } } */
-/* { dg-final { scan-assembler
".long\t___CFConstantStringClassReference\n\t.long\t1992\n\t.long\t.*\n\t.long\t19\n\t.data"
} } */
+/* LLVM LOCAL accept llvm syntax */
+/* { dg-final { scan-assembler
".long\t___CFConstantStringClassReference\n\t.long\t1992\n\t.long\t.*\n\t.long\t19\n"
} } */
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-10.mm
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/obj-c%2B%2B.dg/const-str-10.mm?rev=46269&r1=46268&r2=46269&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-10.mm (original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-10.mm Tue Jan 22
19:40:15 2008
@@ -31,4 +31,5 @@
const NSConstantString *appKey = @"MyApp";
/* { dg-final { scan-assembler ".section __OBJC, __cstring_object" } } */
-/* { dg-final { scan-assembler
".long\t__NSConstantStringClassReference\n\t.long\t.*\n\t.long\t5\n\t.data" } }
*/
+/* LLVM LOCAL accept llvm syntax */
+/* { dg-final { scan-assembler
".long\t__NSConstantStringClassReference\n\t.long\t.*\n\t.long\t5\n" } } */
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-11.mm
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/obj-c%2B%2B.dg/const-str-11.mm?rev=46269&r1=46268&r2=46269&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-11.mm (original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-11.mm Tue Jan 22
19:40:15 2008
@@ -30,4 +30,5 @@
const XStr *appKey = @"MyApp";
/* { dg-final { scan-assembler ".section __OBJC, __cstring_object" } } */
-/* { dg-final { scan-assembler
".long\t__XStrClassReference\n\t.long\t.*\n\t.long\t5\n\t.data" } } */
+/* LLVM LOCAL accept llvm syntax */
+/* { dg-final { scan-assembler
".long\t__XStrClassReference\n\t.long\t.*\n\t.long\t5\n" } } */
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-9.mm
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/obj-c%2B%2B.dg/const-str-9.mm?rev=46269&r1=46268&r2=46269&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-9.mm (original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/obj-c++.dg/const-str-9.mm Tue Jan 22
19:40:15 2008
@@ -23,4 +23,5 @@
const NSConstantString *appKey = @"MyApp";
/* { dg-final { scan-assembler ".section __OBJC, __cstring_object" } } */
-/* { dg-final { scan-assembler
".long\t__NSConstantStringCl
[llvm-commits] [llvm-gcc-4.2] r46270 - in /llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg: const-cfstring-4-64bit.m const-cfstring-4.m const-str-10-64bit.m const-str-10.m const-str-11-64bit.m const-str-11.m
Author: johannes
Date: Tue Jan 22 19:42:09 2008
New Revision: 46270
URL: http://llvm.org/viewvc/llvm-project?rev=46270&view=rev
Log:
Yet more test enhancements to take functionally
equivalent llvm syntax.
Modified:
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4-64bit.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10-64bit.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-11-64bit.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-11.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-9-64bit.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-9.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/image-info.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/objc2-alignment-test-1.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/objc2-const-str-64bit-1.m
llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/objc2-protocol-ext-1.m
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4-64bit.m
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4-64bit.m?rev=46270&r1=46269&r2=46270&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4-64bit.m (original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4-64bit.m Tue Jan
22 19:42:09 2008
@@ -8,4 +8,5 @@
static CFStringRef appKey = (CFStringRef) @"com.apple.soundpref";
/* { dg-final { scan-assembler ".section __DATA, __cfstring" } } */
-/* { dg-final { scan-assembler
".quad\t___CFConstantStringClassReference\n\t.long\t1992\n\t.space
4\n\t.quad\t.*\n\t.quad\t19\n\t.data" } } */
+/* LLVM LOCAL accept llvm syntax */
+/* { dg-final { scan-assembler
".quad\t___CFConstantStringClassReference\n\t.long\t1992\n\t.space(
|\t)4\n\t.quad\t.*\n\t.quad\t19\n" } } */
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4.m
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4.m?rev=46270&r1=46269&r2=46270&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4.m (original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-cfstring-4.m Tue Jan 22
19:42:09 2008
@@ -10,4 +10,5 @@
static CFStringRef appKey = (CFStringRef) @"com.apple.soundpref";
/* { dg-final { scan-assembler ".section __DATA, __cfstring" } } */
-/* { dg-final { scan-assembler
".long\t___CFConstantStringClassReference\n\t.long\t1992\n\t.long\t.*\n\t.long\t19\n\t.data"
} } */
+/* LLVM LOCAL accept llvm syntax */
+/* { dg-final { scan-assembler
".long\t___CFConstantStringClassReference\n\t.long\t1992\n\t.long\t.*\n\t.long\t19\n"
} } */
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10-64bit.m
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10-64bit.m?rev=46270&r1=46269&r2=46270&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10-64bit.m (original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10-64bit.m Tue Jan 22
19:42:09 2008
@@ -30,4 +30,5 @@
const NSConstantString *appKey = @"MyApp";
/* { dg-final { scan-assembler ".section __OBJC, __cstring_object" } } */
-/* { dg-final { scan-assembler
".quad\t__NSConstantStringClassReference\n\t.quad\t.*\n\t.long\t5\n\t.space
4\n\t.data" } } */
+/* LLVM LOCAL accept llvm syntax */
+/* { dg-final { scan-assembler
".quad\t__NSConstantStringClassReference\n\t.quad\t.*\n\t.long\t5\n\t.space(
|\t)4\n" } } */
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10.m
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10.m?rev=46270&r1=46269&r2=46270&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10.m (original)
+++ llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-10.m Tue Jan 22 19:42:09
2008
@@ -33,4 +33,5 @@
const NSConstantString *appKey = @"MyApp";
/* { dg-final { scan-assembler ".section __OBJC, __cstring_object" } } */
-/* { dg-final { scan-assembler
".long\t__NSConstantStringClassReference\n\t.long\t.*\n\t.long\t5\n\t.data" } }
*/
+/* LLVM LOCAL accept llvm syntax */
+/* { dg-final { scan-assembler
".long\t__NSConstantStringClassReference\n\t.long\t.*\n\t.long\t5\n" } } */
Modified: llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-11-64bit.m
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-11-64bit.m?rev=46270&r1=46269&r2=46270&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/testsuite/objc.dg/const-str-11-64bit.m (original)
+++ llvm-gcc-4.2/trunk/
[llvm-commits] [llvm-gcc-4.2] r46271 - in /llvm-gcc-4.2/trunk/gcc: config/i386/i386.c toplev.c
Author: evancheng
Date: Tue Jan 22 20:49:24 2008
New Revision: 46271
URL: http://llvm.org/viewvc/llvm-project?rev=46271&view=rev
Log:
Need to call init_regs() from backend_init() to ensure TYPE_MODE for
user-defined vector types are set up correctly; remove the workaround from
i386.c.
Modified:
llvm-gcc-4.2/trunk/gcc/config/i386/i386.c
llvm-gcc-4.2/trunk/gcc/toplev.c
Modified: llvm-gcc-4.2/trunk/gcc/config/i386/i386.c
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/config/i386/i386.c?rev=46271&r1=46270&r2=46271&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/config/i386/i386.c (original)
+++ llvm-gcc-4.2/trunk/gcc/config/i386/i386.c Tue Jan 22 20:49:24 2008
@@ -3270,16 +3270,9 @@
int offset = tree_low_cst (BINFO_OFFSET (base_binfo), 0) * 8;
tree type = BINFO_TYPE (base_binfo);
-/* LLVM local */
-#ifdef ENABLE_LLVM
- num = classify_argument (type_natural_mode (type),
- type, subclasses,
- (offset + bit_offset) % 256);
-#else
num = classify_argument (TYPE_MODE (type),
type, subclasses,
(offset + bit_offset) % 256);
-#endif
if (!num)
return 0;
for (i = 0; i < num; i++)
@@ -3315,18 +3308,10 @@
}
else
{
-/* LLVM local */
-#ifdef ENABLE_LLVM
- num = classify_argument (type_natural_mode (TREE_TYPE
(field)),
- TREE_TYPE (field), subclasses,
- (int_bit_position (field)
- + bit_offset) % 256);
-#else
num = classify_argument (TYPE_MODE (TREE_TYPE (field)),
TREE_TYPE (field), subclasses,
(int_bit_position (field)
+ bit_offset) % 256);
-#endif
if (!num)
return 0;
for (i = 0; i < num; i++)
@@ -3345,14 +3330,8 @@
/* Arrays are handled as small records. */
{
int num;
-/* LLVM local */
-#ifdef ENABLE_LLVM
- num = classify_argument (type_natural_mode (TREE_TYPE (type)),
-TREE_TYPE (type), subclasses, bit_offset);
-#else
num = classify_argument (TYPE_MODE (TREE_TYPE (type)),
TREE_TYPE (type), subclasses, bit_offset);
-#endif
if (!num)
return 0;
@@ -3384,16 +3363,9 @@
if (TREE_TYPE (field) == error_mark_node)
continue;
-/* LLVM local */
-#ifdef ENABLE_LLVM
- num = classify_argument (type_natural_mode (TREE_TYPE
(field)),
- TREE_TYPE (field), subclasses,
- bit_offset);
-#else
num = classify_argument (TYPE_MODE (TREE_TYPE (field)),
TREE_TYPE (field), subclasses,
bit_offset);
-#endif
if (!num)
return 0;
for (i = 0; i < num; i++)
Modified: llvm-gcc-4.2/trunk/gcc/toplev.c
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/toplev.c?rev=46271&r1=46270&r2=46271&view=diff
==
--- llvm-gcc-4.2/trunk/gcc/toplev.c (original)
+++ llvm-gcc-4.2/trunk/gcc/toplev.c Tue Jan 22 20:49:24 2008
@@ -2052,6 +2052,9 @@
/* LLVM LOCAL begin */
#ifdef ENABLE_LLVM
+ /* note: stor-layout.c:layout_type() uses have_regs_of_mode[] to initialize
+ TYPE_MODE field for user specified vector types. */
+ init_regs ();
/* note: we need init_emit_once above to initialize ptr_mode and friends. */
llvm_initialize_backend();
return;
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm] r46272 - /llvm/branches/release_22/lib/Target/Mips/MipsRegisterInfo.cpp
Author: tbrethou
Date: Tue Jan 22 21:56:34 2008
New Revision: 46272
URL: http://llvm.org/viewvc/llvm-project?rev=46272&view=rev
Log:
Merge from mainline.
Trivial patch to fix two warnings
Modified:
llvm/branches/release_22/lib/Target/Mips/MipsRegisterInfo.cpp
Modified: llvm/branches/release_22/lib/Target/Mips/MipsRegisterInfo.cpp
URL:
http://llvm.org/viewvc/llvm-project/llvm/branches/release_22/lib/Target/Mips/MipsRegisterInfo.cpp?rev=46272&r1=46271&r2=46272&view=diff
==
--- llvm/branches/release_22/lib/Target/Mips/MipsRegisterInfo.cpp (original)
+++ llvm/branches/release_22/lib/Target/Mips/MipsRegisterInfo.cpp Tue Jan 22
21:56:34 2008
@@ -292,6 +292,9 @@
FPOffset = NumBytes;
RAOffset = 0;
NumBytes += 4;
+ } else {
+// No calls and no fp.
+RAOffset = FPOffset = 0;
}
MFI->setObjectOffset(MFI->CreateStackObject(4,4), FPOffset);
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
Re: [llvm-commits] [llvm] r46220 - /llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c
On Jan 22, 2008, at 3:02 PM, Tanya Lattner wrote:
> This fails with llvm-gcc-4.0, darwin 8, x86.
>
> ../../src/gcc/llvm-types.cpp:82: failed assertion `(!TYPE_SIZE(Tr)
> || !Ty->isSized() || !isInt64(TYPE_SIZE(Tr), true) || getInt64
> (TYPE_SIZE(Tr), true) == getTargetData().getABITypeSizeInBits(Ty)) &&
> "LLVM type size doesn't match GCC type size!"'
>
> 2008-01-21-PackedBitFields.c is also failing for me.
Does it work with 4.2? If llvm 2.2 will be the last release that
includes llvm-gcc 4.0, I don't think this is a problem.
-chris
>
>
> -Tanya
>
> On Jan 21, 2008, at 2:15 PM, Devang Patel wrote:
>
>> Author: dpatel
>> Date: Mon Jan 21 16:15:58 2008
>> New Revision: 46220
>>
>> URL: http://llvm.org/viewvc/llvm-project?rev=46220&view=rev
>> Log:
>> New test.
>>
>> Added:
>>llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c
>>
>> Added: llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CFrontend/
>> 2008-01-21-PackedStructField.c?rev=46220&view=auto
>>
>> =
>> =
>>
>> --- llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c (added)
>> +++ llvm/trunk/test/CFrontend/2008-01-21-PackedStructField.c Mon
>> Jan 21 16:15:58 2008
>> @@ -0,0 +1,5 @@
>> +// RUN: %llvmgcc %s -S -o -
>> +
>> +struct X { long double b; unsigned char c; double __attribute__
>> ((packed)) d; };
>> +struct X x = { 3.0L, 5, 3.0 };
>> +
>>
>>
>> ___
>> llvm-commits mailing list
>> llvm-commits@cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
> ___
> llvm-commits mailing list
> llvm-commits@cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
Re: [llvm-commits] [llvm] r46187 - in /llvm/trunk: include/llvm/ADT/FoldingSet.h lib/Support/FoldingSet.cpp
On Jan 18, 2008, at 8:22 PM, Ted Kremenek wrote: > Author: kremenek > Date: Fri Jan 18 22:22:50 2008 > New Revision: 46187 > > URL: http://llvm.org/viewvc/llvm-project?rev=46187&view=rev > Log: > Made 'FoldingSetNodeID' a proper class instead of a nested class in > 'FoldingSetNodeImpl' (previously 'FoldingSetNodeID' was a typedef of > 'FoldingSetNodeImpl::NodeID'). > > Why? Clients can now easily forward declare 'FoldingSetNodeID' > without having > to include FoldingSet.h. Whoa, craziness! Thanks Ted, -Chris ___ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
Re: [llvm-commits] [llvm-gcc-4.2] r46164 - in /llvm-gcc-4.2/trunk/gcc/config/i386: llvm-i386-target.h llvm-i386.cpp
On Jan 19, 2008, at 12:47 AM, Duncan Sands wrote: >> i32 / i64 all integer structs are not passed byval. > > What about integer arrays? This is an optimization, not a correctness issue. Can you give an example that you mean? -Chris ___ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
[llvm-commits] [llvm-gcc-4.2] r46273 - in /llvm-gcc-4.2/branches/release_22: configure configure.in
Author: tbrethou
Date: Tue Jan 22 23:47:49 2008
New Revision: 46273
URL: http://llvm.org/viewvc/llvm-project?rev=46273&view=rev
Log:
Disable bootstrap.
Modified:
llvm-gcc-4.2/branches/release_22/configure
llvm-gcc-4.2/branches/release_22/configure.in
Modified: llvm-gcc-4.2/branches/release_22/configure
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/branches/release_22/configure?rev=46273&r1=46272&r2=46273&view=diff
==
--- llvm-gcc-4.2/branches/release_22/configure (original)
+++ llvm-gcc-4.2/branches/release_22/configure Tue Jan 22 23:47:49 2008
@@ -45,7 +45,7 @@
--with-build-sysroot=sysroot
use sysroot as the system root during the build"
ac_help="$ac_help
- --enable-bootstrap Enable bootstrapping [yes if native build]"
+ --enable-bootstrap Enable bootstrapping [no]"
ac_help="$ac_help
--enable-serial-[{host,target,build}-]configure
Force sequential configuration of
@@ -3172,7 +3172,7 @@
# Default behavior. Enable bootstrap if we have a compiler
# and we are in a native configuration.
yes:$build:$build:default)
-enable_bootstrap=yes ;;
+enable_bootstrap=no ;;
*:*:*:default)
enable_bootstrap=no ;;
Modified: llvm-gcc-4.2/branches/release_22/configure.in
URL:
http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/branches/release_22/configure.in?rev=46273&r1=46272&r2=46273&view=diff
==
--- llvm-gcc-4.2/branches/release_22/configure.in (original)
+++ llvm-gcc-4.2/branches/release_22/configure.in Tue Jan 22 23:47:49 2008
@@ -1865,7 +1865,7 @@
# not to nest @if/@endif pairs, because configure will not warn you at all.
AC_ARG_ENABLE([bootstrap],
-[ --enable-bootstrap Enable bootstrapping [yes if native build]],,
+[ --enable-bootstrap Enable bootstrapping [no]],,
enable_bootstrap=default)
# Issue errors and warnings for invalid/strange bootstrap combinations.
@@ -1880,7 +1880,7 @@
# Default behavior. Enable bootstrap if we have a compiler
# and we are in a native configuration.
yes:$build:$build:default)
-enable_bootstrap=yes ;;
+enable_bootstrap=no ;;
*:*:*:default)
enable_bootstrap=no ;;
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
Re: [llvm-commits] [llvm-gcc-4.2] r46164 - in /l lvm-gcc-4.2/trunk/gcc/config/i386: llvm-i386-ta rget.h llvm-i386.cpp
> > What about integer arrays? > > This is an optimization, not a correctness issue. Can you give an > example that you mean? According to the x86-64 ABI, arrays of integers that are not more than 8 bytes long are passed in integer registers. Likewise for structs. So it looks like a correctness issue as well as an optimization :) CIao, Duncan. ___ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
Re: [llvm-commits] [llvm-gcc-4.2] r46164 - in /llvm-gcc-4.2/trunk/gcc/config/i386: llvm-i386-target.h llvm-i386.cpp
On Jan 22, 2008, at 10:03 PM, Duncan Sands wrote: >>> What about integer arrays? >> >> This is an optimization, not a correctness issue. Can you give an >> example that you mean? > > According to the x86-64 ABI, arrays of integers that are not more > than 8 bytes long are passed in integer registers. Likewise for > structs. So it looks like a correctness issue as well as an > optimization :) Okay, well we already get many other x86-64 issues wrong already, but Evan is chipping away at it. How do you pass an array by value in C? Example please, -Chris ___ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
Re: [llvm-commits] [llvm-gcc-4.2] r46164 - in /l lvm-gcc-4.2/trunk/gcc/config/i386: llvm-i386-ta rget.h llvm-i386.cpp
> Okay, well we already get many other x86-64 issues wrong already, but
> Evan is chipping away at it. How do you pass an array by value in C?
> Example please,
I find the x86-64 ABI hard to interpret, but it seems to say that
aggregates are classified recursively, so it looks like a struct
containing a small integer array should be passed in integer registers.
Also, it is easy in Ada: the compiler passes small arrays by value,
for example:
package Arr is
type ByValArray is array (1 .. 2) of Short_Integer;
function PassArrayByVal (A : ByValArray) return Short_Integer;
end;
package body Arr is
function PassArrayByVal (A : ByValArray) return Short_Integer is
begin
return A(1) + A(2);
end;
end;
->
define i16 @arr__passarraybyval(i32 %a.0.0) signext nounwind {
(Here presumably the i32 was synthesized by llvm-convert from the
two element array of shorts passed by value).
Can you please clarify the roles of llvm-gcc and the code generators
in getting ABI compatibility. When generating IR for x86-64, llvm-gcc
sometimes chops by-value structs into pieces, and sometimes passes the
struct as a byval parameter. Since it chops up all-integer structs,
and this corresponds more or less to what the ABI says, I assumed this
was an attempt to get ABI correctness. Especially as the code generators
don't seem to bother themselves with following the details of the ABI (yet),
and just push byval parameters onto the stack. Since you say
>> This is an optimization, not a correctness issue
I guess this means that the plan is to teach the codegenerators how to
pass any aggregate byval in an ABI conformant way (not the case right now),
but still do some chopping up in the front-end to help the optimizers.
Of course this chopping up needs to be done carefully so the final result
squirted out by the codegenerators (once they are ABI conformant) is the
same as if the chopping had not been done... Is this chopping really a
big win? Is it not possible to get an equivalent level of optimization
by enhancing alias analysis?
Ciao,
Duncan.
___
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
