thr3ads.net - llvm dev - [LLVMdev] Debug information under windows [Oct 2012]

If this information is useful, please help other people find it:
Share via:

Daniel Kłobuszewski

2012-Oct-26 14:55 UTC

[LLVMdev] Debug information under windows

Hello,

Recently I found binaries produced with LLVM impossible to debug under 
Windows. This was probably related to the following bug: 
http://llvm.org/bugs/show_bug.cgi?id=13636

Asm generated from .ll files revealed that some offsets to debug 
information were incorrect: they were absolute instead of relative to 
their sections.

Following patch seemed to have repaired the problem, so I'm just leaving 
it here. Hopefully this will help someone.

Regards,
Daniel


diff -r 547972237a05 -r 6ed9378f04a3 lib/MC/MCAsmInfoCOFF.cpp
--- a/lib/MC/MCAsmInfoCOFF.cpp	Wed Oct 10 20:39:45 2012 +0200
+++ b/lib/MC/MCAsmInfoCOFF.cpp	Thu Oct 18 12:41:27 2012 +0200
@@ -38,6 +38,10 @@
    HasMicrosoftFastStdCallMangling = true;

    SupportsDataRegions = false;
+
+  DwarfUsesLabelOffsetForRanges = false;
+  DwarfRequiresRelocationForSectionOffset = false;
+  DwarfUsesRelocationsForStringPool = false;
  }

  void MCAsmInfoMicrosoft::anchor() { }

Rafael Espíndola

2012-Oct-26 20:46 UTC

head link

[LLVMdev] Debug information under windows

Can you add some tests?

On 26 October 2012 10:55, Daniel Kłobuszewski
<daniel.klobuszewski at ibs.org.pl> wrote:> Hello,
>
> Recently I found binaries produced with LLVM impossible to debug under
> Windows. This was probably related to the following bug:
> http://llvm.org/bugs/show_bug.cgi?id=13636
>
> Asm generated from .ll files revealed that some offsets to debug
information
> were incorrect: they were absolute instead of relative to their sections.
>
> Following patch seemed to have repaired the problem, so I'm just
leaving it
> here. Hopefully this will help someone.
>
> Regards,
> Daniel
>
>
> diff -r 547972237a05 -r 6ed9378f04a3 lib/MC/MCAsmInfoCOFF.cpp
> --- a/lib/MC/MCAsmInfoCOFF.cpp  Wed Oct 10 20:39:45 2012 +0200
> +++ b/lib/MC/MCAsmInfoCOFF.cpp  Thu Oct 18 12:41:27 2012 +0200
> @@ -38,6 +38,10 @@
>    HasMicrosoftFastStdCallMangling = true;
>
>    SupportsDataRegions = false;
> +
> +  DwarfUsesLabelOffsetForRanges = false;
> +  DwarfRequiresRelocationForSectionOffset = false;
> +  DwarfUsesRelocationsForStringPool = false;
>  }
>
>  void MCAsmInfoMicrosoft::anchor() { }
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev

Emil Djupfeldt

2012-Oct-26 22:09 UTC

head link

[LLVMdev] Debug information under windows

Hi,

I'm having some trouble applying this patch. Which svn version is it 
against? Because from what I can tell, the Dwarf... are not defined in 
this class or any of it's base classes in HEAD.

regards,
Emil

On 2012-10-26 16:55, Daniel Kłobuszewski wrote:> Hello,
>
> Recently I found binaries produced with LLVM impossible to debug under 
> Windows. This was probably related to the following bug: 
> http://llvm.org/bugs/show_bug.cgi?id=13636
>
> Asm generated from .ll files revealed that some offsets to debug 
> information were incorrect: they were absolute instead of relative to 
> their sections.
>
> Following patch seemed to have repaired the problem, so I'm just 
> leaving it here. Hopefully this will help someone.
>
> Regards,
> Daniel
>
>
> diff -r 547972237a05 -r 6ed9378f04a3 lib/MC/MCAsmInfoCOFF.cpp
> --- a/lib/MC/MCAsmInfoCOFF.cpp Wed Oct 10 20:39:45 2012 +0200
> +++ b/lib/MC/MCAsmInfoCOFF.cpp Thu Oct 18 12:41:27 2012 +0200
> @@ -38,6 +38,10 @@
> HasMicrosoftFastStdCallMangling = true;
>
> SupportsDataRegions = false;
> +
> + DwarfUsesLabelOffsetForRanges = false;
> + DwarfRequiresRelocationForSectionOffset = false;
> + DwarfUsesRelocationsForStringPool = false;
> }
>
> void MCAsmInfoMicrosoft::anchor() { }
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev

Daniel Kłobuszewski

2012-Oct-29 09:16 UTC

head link

[LLVMdev] Debug information under windows

I was manually running generated binaries under gdb and using objdump to 
analyze their debug symbols. I do not have a script for that.

Basically what needs to be checked is whether or not offsets to 
.debug_<something> sections (e.g. DW_FORM_strp, DW_AT_ranges) are valid.

W dniu 2012-10-26 22:46, Rafael Espíndola pisze:> Can you add some tests?
>
> On 26 October 2012 10:55, Daniel Kłobuszewski
> <daniel.klobuszewski at ibs.org.pl> wrote:
>> Hello,
>>
>> Recently I found binaries produced with LLVM impossible to debug under
>> Windows. This was probably related to the following bug:
>> http://llvm.org/bugs/show_bug.cgi?id=13636
>>
>> Asm generated from .ll files revealed that some offsets to debug
information
>> were incorrect: they were absolute instead of relative to their
sections.
>>
>> Following patch seemed to have repaired the problem, so I'm just
leaving it
>> here. Hopefully this will help someone.
>>
>> Regards,
>> Daniel
>>
>>
>> diff -r 547972237a05 -r 6ed9378f04a3 lib/MC/MCAsmInfoCOFF.cpp
>> --- a/lib/MC/MCAsmInfoCOFF.cpp  Wed Oct 10 20:39:45 2012 +0200
>> +++ b/lib/MC/MCAsmInfoCOFF.cpp  Thu Oct 18 12:41:27 2012 +0200
>> @@ -38,6 +38,10 @@
>>     HasMicrosoftFastStdCallMangling = true;
>>
>>     SupportsDataRegions = false;
>> +
>> +  DwarfUsesLabelOffsetForRanges = false;
>> +  DwarfRequiresRelocationForSectionOffset = false;
>> +  DwarfUsesRelocationsForStringPool = false;
>>   }
>>
>>   void MCAsmInfoMicrosoft::anchor() { }
>>
>> _______________________________________________
>> LLVM Developers mailing list
>> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev

Daniel Kłobuszewski

2012-Nov-19 09:57 UTC

head link

[LLVMdev] Debug information under windows

W dniu 2012-10-26 16:55, Daniel Kłobuszewski pisze:> Hello,
>
> Recently I found binaries produced with LLVM impossible to debug under
> Windows. This was probably related to the following bug:
> http://llvm.org/bugs/show_bug.cgi?id=13636
>
> Asm generated from .ll files revealed that some offsets to debug
> information were incorrect: they were absolute instead of relative to
> their sections.
>
> Following patch seemed to have repaired the problem, so I'm just
leaving
> it here. Hopefully this will help someone.
>
> Regards,
> Daniel
>
>
> diff -r 547972237a05 -r 6ed9378f04a3 lib/MC/MCAsmInfoCOFF.cpp
> --- a/lib/MC/MCAsmInfoCOFF.cpp    Wed Oct 10 20:39:45 2012 +0200
> +++ b/lib/MC/MCAsmInfoCOFF.cpp    Thu Oct 18 12:41:27 2012 +0200
> @@ -38,6 +38,10 @@
>     HasMicrosoftFastStdCallMangling = true;
>
>     SupportsDataRegions = false;
> +
> +  DwarfUsesLabelOffsetForRanges = false;
> +  DwarfRequiresRelocationForSectionOffset = false;
> +  DwarfUsesRelocationsForStringPool = false;
>   }
>
>   void MCAsmInfoMicrosoft::anchor() { }
>
Hello again,

"fix" from my previous message was helpful only for one compilation 
unit, but was working incorrectly in case when more CUs were linked 
together.

Since the bug I've linked seems to still be unresolved, I'm sending my 
solution to it. Basically all that needed to be done was to emit 
.secrel32 instructions in few places instead of sections offsets in 
label difference form.

Here is how I've been building executable for tests (under MinGW console):

llc test.bc -filetype=asm -march=x86 -x86-asm-syntax=att 
-mtriple=i686-pc-mingw32 -disable-fp-elim
llc test_main.bc -filetype=asm -march=x86 -x86-asm-syntax=att 
-mtriple=i686-pc-mingw32 -disable-fp-elim
gcc -g -c test.s -o test.o
gcc -g -c test_main.s -o test_main.o
gcc -g test_main.o test.o -o test

Someone more familiar with LLVM code than me should probably review my 
proposed patch. My source revision was 3.1 LLVM release.

Regards,
Daniel
-------------- next part --------------
diff -r 547972237a05 -r d28fa29d4b44 lib/CodeGen/AsmPrinter/AsmPrinter.cpp
--- a/lib/CodeGen/AsmPrinter/AsmPrinter.cpp	Wed Oct 10 20:39:45 2012 +0200
+++ b/lib/CodeGen/AsmPrinter/AsmPrinter.cpp	Mon Nov 19 10:31:49 2012 +0100
@@ -1398,6 +1398,13 @@
 void AsmPrinter::EmitLabelPlusOffset(const MCSymbol *Label, uint64_t Offset,
                                       unsigned Size)
   const {
+  // if COFF .secrel32 is available, use it
+  if (MAI->getDwarfSectionOffsetDirective()) {
+    MCSymbol *SetLabel = GetTempSymbol("set", SetCounter++);
+    OutStreamer.EmitAssignment(SetLabel, MCConstantExpr::Create(Offset,
OutContext));
+    OutStreamer.EmitCOFFSecRel32(SetLabel);
+    return;
+  }
 
   // Emit Label+Offset
   const MCExpr *Plus diff -r 547972237a05 -r d28fa29d4b44
lib/CodeGen/AsmPrinter/DIE.cpp
--- a/lib/CodeGen/AsmPrinter/DIE.cpp	Wed Oct 10 20:39:45 2012 +0200
+++ b/lib/CodeGen/AsmPrinter/DIE.cpp	Mon Nov 19 10:31:49 2012 +0100
@@ -257,6 +257,11 @@
 /// EmitValue - Emit delta value.
 ///
 void DIEDelta::EmitValue(AsmPrinter *AP, unsigned Form) const {
+  if (AP->MAI->getDwarfSectionOffsetDirective()) {
+    // FIXME: Can DIEDelta be used in other contexts than section offset?
+    AP->EmitSectionOffset(LabelHi, LabelLo);
+    return;
+  }
   AP->EmitLabelDifference(LabelHi, LabelLo, SizeOf(AP, Form));
 }
 
diff -r 547972237a05 -r d28fa29d4b44 lib/CodeGen/AsmPrinter/DwarfDebug.cpp
--- a/lib/CodeGen/AsmPrinter/DwarfDebug.cpp	Wed Oct 10 20:39:45 2012 +0200
+++ b/lib/CodeGen/AsmPrinter/DwarfDebug.cpp	Mon Nov 19 10:31:49 2012 +0100
@@ -567,9 +567,14 @@
   NewCU->addUInt(Die, dwarf::DW_AT_low_pc, dwarf::DW_FORM_addr, 0);
   // DW_AT_stmt_list is a offset of line number information for this
   // compile unit in debug_line section.
-  if (Asm->MAI->doesDwarfRequireRelocationForSectionOffset())
-    NewCU->addLabel(Die, dwarf::DW_AT_stmt_list, dwarf::DW_FORM_data4,
+  if (Asm->MAI->doesDwarfRequireRelocationForSectionOffset()) {
+    if (Asm->MAI->getDwarfSectionOffsetDirective()) // COFF
+      NewCU->addDelta(Die, dwarf::DW_AT_stmt_list, dwarf::DW_FORM_data4,
+                    Asm->GetTempSymbol("section_line"), 0); //
will expand to secrel32 on COFF. TODO: separate "DieCOFFLabel" class
or something
+    else
+      NewCU->addLabel(Die, dwarf::DW_AT_stmt_list, dwarf::DW_FORM_data4,
                     Asm->GetTempSymbol("section_line"));
+  }
   else
     NewCU->addUInt(Die, dwarf::DW_AT_stmt_list, dwarf::DW_FORM_data4, 0);
 
@@ -1638,7 +1643,7 @@
     }
     case dwarf::DW_AT_location: {
       if (DIELabel *L = dyn_cast<DIELabel>(Values[i]))
-        Asm->EmitLabelDifference(L->getValue(), DwarfDebugLocSectionSym,
4);
+        Asm->EmitSectionOffset(L->getValue(), DwarfDebugLocSectionSym);
       else
         Values[i]->EmitValue(Asm, Form);
       break;
diff -r 547972237a05 -r d28fa29d4b44 lib/MC/MCAsmInfoCOFF.cpp
--- a/lib/MC/MCAsmInfoCOFF.cpp	Wed Oct 10 20:39:45 2012 +0200
+++ b/lib/MC/MCAsmInfoCOFF.cpp	Mon Nov 19 10:31:49 2012 +0100
@@ -38,6 +38,8 @@
   HasMicrosoftFastStdCallMangling = true;
 
   SupportsDataRegions = false;
+
+  DwarfUsesRelocationsForStringPool = false;
 }
 
 void MCAsmInfoMicrosoft::anchor() { }

Nathan Jeffords

2012-Nov-20 17:01 UTC

head link

[LLVMdev] Debug information under windows

For reference, I have attached a patch I created some time ago to solve
this issue. My implementation differs in that I created a new type to
represent the section offset instead of special casing DIEDelta. When it
was originally committed it broke one of the build bots so it was
uncommitted. I didn't have the resources to investigate the issue so I have
been maintaining the patch on my own build.


On Mon, Nov 19, 2012 at 1:57 AM, Daniel Kłobuszewski <
daniel.klobuszewski at ibs.org.pl> wrote:
> W dniu 2012-10-26 16:55, Daniel Kłobuszewski pisze:
>
>  Hello,
>>
>> Recently I found binaries produced with LLVM impossible to debug under
>> Windows. This was probably related to the following bug:
>>
http://llvm.org/bugs/show_bug.**cgi?id=13636<http://llvm.org/bugs/show_bug.cgi?id=13636>
>>
>> Asm generated from .ll files revealed that some offsets to debug
>> information were incorrect: they were absolute instead of relative to
>> their sections.
>>
>> Following patch seemed to have repaired the problem, so I'm just
leaving
>> it here. Hopefully this will help someone.
>>
>> Regards,
>> Daniel
>>
>>
>> diff -r 547972237a05 -r 6ed9378f04a3 lib/MC/MCAsmInfoCOFF.cpp
>> --- a/lib/MC/MCAsmInfoCOFF.cpp    Wed Oct 10 20:39:45 2012 +0200
>> +++ b/lib/MC/MCAsmInfoCOFF.cpp    Thu Oct 18 12:41:27 2012 +0200
>> @@ -38,6 +38,10 @@
>>     HasMicrosoftFastStdCallManglin**g = true;
>>
>>     SupportsDataRegions = false;
>> +
>> +  DwarfUsesLabelOffsetForRanges = false;
>> +  DwarfRequiresRelocationForSect**ionOffset = false;
>> +  DwarfUsesRelocationsForStringP**ool = false;
>>   }
>>
>>   void MCAsmInfoMicrosoft::anchor() { }
>>
>>
> Hello again,
>
> "fix" from my previous message was helpful only for one
compilation unit,
> but was working incorrectly in case when more CUs were linked together.
>
> Since the bug I've linked seems to still be unresolved, I'm sending
my
> solution to it. Basically all that needed to be done was to emit .secrel32
> instructions in few places instead of sections offsets in label difference
> form.
>
> Here is how I've been building executable for tests (under MinGW
console):
>
> llc test.bc -filetype=asm -march=x86 -x86-asm-syntax=att
> -mtriple=i686-pc-mingw32 -disable-fp-elim
> llc test_main.bc -filetype=asm -march=x86 -x86-asm-syntax=att
> -mtriple=i686-pc-mingw32 -disable-fp-elim
> gcc -g -c test.s -o test.o
> gcc -g -c test_main.s -o test_main.o
> gcc -g test_main.o test.o -o test
>
> Someone more familiar with LLVM code than me should probably review my
> proposed patch. My source revision was 3.1 LLVM release.
>
> Regards,
> Daniel
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>
>-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20121120/64ae4900/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: windebug.patch
Type: application/octet-stream
Size: 7321 bytes
Desc: not available
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20121120/64ae4900/attachment.obj>

Possibly Parallel Threads

Search for more reasonably related threads

llvm dev - Oct 2012 - [LLVMdev] Debug information under windows

[LLVMdev] Debug information under windows

[LLVMdev] Debug information under windows

[LLVMdev] Debug information under windows

[LLVMdev] Debug information under windows

[LLVMdev] Debug information under windows

[LLVMdev] Debug information under windows

Possibly Parallel Threads