[Bug binutils/4791] Update to latest gnustandards (Was: etc/standards.texi: @strong{Note...} produces a spurious cross-reference in Info)
--- Additional Comments From nickc at redhat dot com 2007-08-13 09:46 --- I have imported the current version of the standards.texi file from the ww.gnu.org website and checked it into the sourceware repository. Cheers Nick etc/ChangeLog 2007-08-13 Nick Clifton <[EMAIL PROTECTED]> * standards.texi: Import 22 July 2007 version. * fdl.texi: Import version 1.2. * make-stds.texi: Import 2006 version. -- What|Removed |Added Status|WAITING |RESOLVED Resolution||FIXED http://sourceware.org/bugzilla/show_bug.cgi?id=4791 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/4918] New: ___tls_get_addr call requires @PLT even without -fpic
On Linux/x86 and Linux/x86-64, I got
[EMAIL PROTECTED] tls-5]$ cat x.c
#include
__thread int foo __attribute__((tls_model("global-dynamic"))) = 100;
int
main ()
{
printf ("foo: %d\n", foo);
return 0;
}
[EMAIL PROTECTED] tls-5]$ gcc -o x x.c
/usr/local/bin/ld: BFD (Linux/GNU Binutils) 2.18.50.0.1.20070811 assertion fail
/net/gnu-13/export/linux/src/binutils/binutils/bfd/elf64-x86-64.c:2568
[EMAIL PROTECTED] tls-5]$ ./x
foo: 100
The assert comes from
/* GD->LE transition.
.byte 0x66; leaq [EMAIL PROTECTED](%rip), %rdi
.word 0x; rex64; call [EMAIL PROTECTED]
Change it into:
movq %fs:0, %rax
leaq [EMAIL PROTECTED](%rax), %rax */
BFD_ASSERT (rel->r_offset >= 4);
for (i = 0; i < 4; i++)
BFD_ASSERT (bfd_get_8 (input_bfd,
contents + rel->r_offset - 4 + i)
== tlsgd[i]);
BFD_ASSERT (rel->r_offset + 12 <= input_section->size);
for (i = 0; i < 4; i++)
BFD_ASSERT (bfd_get_8 (input_bfd,
contents + rel->r_offset + 4 + i)
== tlsgd[i+4]); BFD_ASSERT (rel
+ 1 < relend);
BFD_ASSERT (ELF64_R_TYPE (rel[1].r_info) == R_X86_64_PLT32);
Why is R_X86_64_PLT32 required here?
--
Summary: ___tls_get_addr call requires @PLT even without -fpic
Product: binutils
Version: 2.19 (HEAD)
Status: NEW
Severity: normal
Priority: P2
Component: ld
AssignedTo: unassigned at sources dot redhat dot com
ReportedBy: hjl at lucon dot org
CC: bug-binutils at gnu dot org
http://sourceware.org/bugzilla/show_bug.cgi?id=4918
--- You are receiving this mail because: ---
You are on the CC list for the bug, or are watching someone who is.
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/4918] ___tls_get_addr call requires @PLT even without -fpic
-- What|Removed |Added CC||jakub at redhat dot com http://sourceware.org/bugzilla/show_bug.cgi?id=4918 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/4918] ___tls_get_addr call requires @PLT for global-dynamic even without -fpic
-- What|Removed |Added Summary|___tls_get_addr call|___tls_get_addr call |requires @PLT even without -|requires @PLT for global- |fpic|dynamic even without -fpic http://sourceware.org/bugzilla/show_bug.cgi?id=4918 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
gas: additional macro functionality (patch attached)
Hi! When I was trying to implement a direct threaded code interpreter (forth look alike) with gas for the msp430 microcontroller, I missed the possibility to reference a) the individual invocation count of each macro and b) the previous invocation count for each macro. Problem: I need to build a single linked list at assemble time. This is traditionally done with somethink like: .set LINK, 0 .macro HEAD foo, bar ... .word LINK .setLINK,. .endm This does not work with gas assembling msp430 code on a PPC Mac, because LINK only gets resoved at relocation time, so all LINK fields are the same :-( The presence of the \@ ideom hinted to me, that I could easily tweak gas to yield the information I need. So I made two changes:: 1) "macro_entry" gets another field "count", giving the individual invocation count 2) The ideoms \#, \< and \> yield the current, previous amd next invocation count when used in a macro. So I can write the linked list code like: .macro HEAD len,name,link,flags=0 lnk\#: .byte (\flags << 5) | \len nam\@: .ascii "\name" .orgnam\@ + 3 ; pad short names .word lnk\< .endm Tested, works :-) I would love to see my code incorporated in the next gas release, since I consider it useful not only to me. If you need to have some documentation writen to do this, i'll be happy to provide it. Attached are the diff's for macro.h and macro.c Best regards Olaf -- /~\ The ASCII | Olaf P?schel, OLMOS Technologies GmbH, Germany \ / Ribbon Campaign | Senefelderstra?e 11, 38124 Braunschweig X Against HTML| Phone: +49-531-22020-0 Fax: +49-531-22020-99 / \ Email! | OLMOS supports signed and/or encrypted mail. gas-2.17_macro.c.diff Description: Binary data gas-2.17_macro.h.diff Description: Binary data ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
