Re: binutils-2.18 and U-Boot

[raptorbrino at aim dot com]

Yah, that fix will fix part of the problem.  The macro
ELF_IS_SECTION_IN_SEGMENT has 3 parts to it, and parts 2
and 3 are susceptible to a wrap condition.  He fixed the
wrap in part3, but it also needs to be fixed in part2.
This patch should fix the wrap in part2.  Look good?

diff -Naur a/include/elf/internal.h b/include/elf/internal.h
--- a/include/elf/internal.h    2007-11-28 05:53:05.000000000 -0800
+++ b/include/elf/internal.h    2007-12-11 16:37:43.906250000 -0800
@@ -284,8 +284,8 @@
      offset within the segment.  */                   \
   && (sec_hdr->sh_type == SHT_NOBITS                  \
       || ((bfd_vma) sec_hdr->sh_offset >= segment->p_offset       \
-      && (sec_hdr->sh_offset + ELF_SECTION_SIZE(sec_hdr, segment)  \
-          <= segment->p_offset + segment->p_filesz)))      \
+      && (sec_hdr->sh_offset - segment->p_offset   \
+         + ELF_SECTION_SIZE(sec_hdr, segment) <= 
segment->p_filesz)))        \
   /* SHF_ALLOC sections must have VMAs within the segment.  Be        \
      careful about segments right at the end of memory.  */       \
   && ((sec_hdr->sh_flags & SHF_ALLOC) == 0                \

By the way, the mainline CVS does not build.  Is that common?


-----Original Message-----
From: Alan Modra <amodra@bigpond.net.au>
To: raptorbrino@aim.com
Cc: binutils@sourceware.org
Sent: Tue, 11 Dec 2007 3:53 pm
Subject: Re: binutils-2.18 and U-Boot


I think this problem has already been fixed in mainline CVS binutils.
See http://sourceware.org/ml/binutils/2007-11/msg00235.html
Please verify and close 5205 if so.

--
Alan Modra
Australia Development Lab, IBM






________________________________________________________________________
More new features than ever.  Check out the new AIM(R) Mail ! - 
http://webmail.aim.com