[PATCH-bfd] i386-mingw32-ld crash on x86_64 linux

[Dave Korn]

Dave Korn wrote:
> Alan Modra wrote:
>> So, something like the following.  I won't commit this as it's
>> untested, and I'm not sure whether the ext->e.e_name[0] == 0 case
>> ought to be handled.
> 
>   Hmm.  That case will arise if the symbol name is long, and I guess it could
> be since we do allow long section names.  I'll whip up something based on your
> patch and run my standard set of testsuites on it.

  I remembered that we had a function to handle all the name-related
complications for us already, so I just used it.  It should be OK to call from
within _bfd_XXi_swap_sym_in, since it is often called immediately afterward in
callers of _bfd_XXi_swap_sym_in anyway.

  I'm running this through a test cycle.  Peter, would you care to test if it
fixes your specific problem?

bfd/ChangeLog

	* peXXigen.c (_bfd_XXi_swap_sym_in):  Fix name handling w.r.t
	long names and non-NUL-terminated strings.

    cheers,
      DaveK
-------------- next part --------------
A non-text attachment was scrubbed...
Name: curr-testing-patch.diff
Type: text/x-c
Size: 1446 bytes
Desc: not available
URL: <https://sourceware.org/pipermail/binutils/attachments/20090418/a8171ee9/attachment.bin>