[LLVMbugs] [Bug 16045] New: clang for iOS output very partial __debug_frame section

bugzilla-daemon at llvm.org bugzilla-daemon at llvm.org
Fri May 17 01:33:03 PDT 2013


http://llvm.org/bugs/show_bug.cgi?id=16045

            Bug ID: 16045
           Summary: clang for iOS output very partial __debug_frame
                    section
           Product: new-bugs
           Version: unspecified
          Hardware: Other
                OS: other
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
          Assignee: unassignedbugs at nondot.org
          Reporter: qsr at chromium.org
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

With the following clang version:
% clang -v     
clang version 3.3 (trunk 179138)
Target: x86_64-apple-darwin12.3.0
Thread model: posix


 Given the following file:
---------------------------------------
void foo() {
}

int main() {
  foo();
}
---------------------------------------
I compile it for iOS:

% clang -target arm-apple-darwin7 -g -c /tmp/main.c -o /tmp/main.o

 Then looking at the __debug_frame section:

% otool -s __DWARF __debug_frame /tmp/main.o
/tmp/main.o:
Contents of (__DWARF,__debug_frame) section
0000025c        0000000c ffffffff 7c010001 0000000e 
0000026c        0000000c 00000000 00000000 00000004 
0000027c        0000000c 00000000 00000004 00000018 

 The section is useless to walk the stack.

 Debugger can use the fp to do it, so they don't seem to even care about the
debug_frame section, but is it an expected behavior, or should the
__debug_frame contains the information on how to unwind the stack and retrieve
the register values?

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20130517/dc4b87df/attachment.html>


More information about the llvm-bugs mailing list