gcc/gdb/as/ld issue

[Andy Chittenden]

Further to my diagnostic that discovered that the BINCL optimisation in
_bfd_link_section_stabs was too aggressive based on length of symbol, I've
put together a patch (see attached) that compares the value of the symbols
as well as the length of the symbols. It turns out, in our environment, that
even though the "hash" of the symbol is the same, the value of the symbols
is different more often than than the length of the symbol. Obviously, I
need to take out the fprintfs but apart from that, would this patch be OK to
fix this problem?

-- 
Andy, BlueArc Engineering 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: stabs.diff
Type: application/octet-stream
Size: 2581 bytes
Desc: not available
URL: <https://sourceware.org/pipermail/binutils/attachments/20040322/ef0c1d3d/attachment.obj>