From: "Maciej W. Rozycki" <macro@ds2.pg.gda.pl>
To: binutils@sourceware.cygnus.com
Subject: Problems with symbol versioning on MIPS/ELF
Date: Tue, 12 Oct 1999 10:17:00 -0000 [thread overview]
Message-ID: <Pine.GSO.3.96.991012184750.15112C-100000@delta.DS2.net> (raw)
Hi,
I am getting following errors from ld:
mipsel-linux-as -o symver.o symver.s
mipsel-linux-ld -shared --version-script=symver.map -o symver.so symver.o
mipsel-linux-ld: bfd assertion fail elf32-mips.c:5281
mipsel-linux-ld: bfd assertion fail elf32-mips.c:5155
mipsel-linux-ld: bfd assertion fail elf32-mips.c:5158
symver.o: In function `sym1':
symver.o(.text+0x4): relocation truncated to fit: R_MIPS_GOT16 _DYNAMIC
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
mipsel-linux-ld: bfd assertion fail elf32-mips.c:8169
make: *** [symver.so] Error 1
symvers.s:
.abicalls
.text
.set noreorder
.globl sym1
.type sym1,@function
.cprestore 16
sym1:
la $5,_DYNAMIC
jal sym2
nop
symver.map:
VER_1.0 {
global:
sym1;
local:
*;
};
Also if the jal opcode is absent from the above code, ld segfaults:
Program received signal SIGSEGV, Segmentation fault.
0x807fb96 in mips_elf_global_got_index (abfd=0x810e9a8, h=0x8109a70)
at elf32-mips.c:5155
5155 BFD_ASSERT (h->dynindx >= g->global_gotsym->dynindx);
(gdb) print g->global_gotsym
$14 = (struct elf_link_hash_entry *) 0x0
It looks like not all BFD structures get updated properly for symbols
that get forced to be local by a version script. I tried to track the
problem down and created a following patch that fullfills the first
assertion.
diff -u --recursive --new-file --exclude-from=.exclude binutils.macro/bfd/elflink.h binutils/bfd/elflink.h
--- binutils.macro/bfd/elflink.h Fri Oct 8 16:41:22 1999
+++ binutils/bfd/elflink.h Mon Oct 11 22:45:11 1999
@@ -3597,6 +3597,7 @@
h->elf_link_hash_flags &=~
ELF_LINK_HASH_NEEDS_PLT;
h->dynindx = -1;
+ --elf_hash_table (info)->dynsymcount;
h->plt.offset = (bfd_vma) -1;
/* FIXME: The name of the symbol has
already been recorded in the dynamic
@@ -3711,6 +3712,7 @@
h->elf_link_hash_flags |= ELF_LINK_FORCED_LOCAL;
h->elf_link_hash_flags &=~ ELF_LINK_HASH_NEEDS_PLT;
h->dynindx = -1;
+ --elf_hash_table (info)->dynsymcount;
h->plt.offset = (bfd_vma) -1;
/* FIXME: The name of the symbol has already
been recorded in the dynamic string table
@@ -3735,6 +3737,7 @@
h->elf_link_hash_flags |= ELF_LINK_FORCED_LOCAL;
h->elf_link_hash_flags &=~ ELF_LINK_HASH_NEEDS_PLT;
h->dynindx = -1;
+ --elf_hash_table (info)->dynsymcount;
h->plt.offset = (bfd_vma) -1;
/* FIXME: The name of the symbol has already been
recorded in the dynamic string table section. */
Unfortunately I don't know how to make mips_elf_local_relocation_p()
return a valid result for relocations that refer to symbols forced to be
local. Any help would be appreciated.
Maciej
--
+ Maciej W. Rozycki, Technical University of Gdansk, Poland +
+--------------------------------------------------------------+
+ e-mail: macro@ds2.pg.gda.pl, PGP key available +
reply other threads:[~1999-10-12 10:17 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Pine.GSO.3.96.991012184750.15112C-100000@delta.DS2.net \
--to=macro@ds2.pg.gda.pl \
--cc=binutils@sourceware.cygnus.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).