[Bug gas/16858] weak external reference has wrong value
https://sourceware.org/bugzilla/show_bug.cgi?id=16858 Nick Clifton nickc at redhat dot com changed: What|Removed |Added Attachment #7600|0 |1 is obsolete|| --- Comment #13 from Nick Clifton nickc at redhat dot com --- Created attachment 7603 -- https://sourceware.org/bugzilla/attachment.cgi?id=7603action=edit Revised patch which does not affect non-code symbols -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/16858] weak external reference has wrong value
https://sourceware.org/bugzilla/show_bug.cgi?id=16858 --- Comment #14 from Nick Clifton nickc at redhat dot com --- Hi Bernd, Actually here is a better patch. The previous one also affected non-code weak symbols which was the wrong thing to do. Please try this patch out and let me know if you are happy with it. Cheers Nick -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/16858] weak external reference has wrong value
https://sourceware.org/bugzilla/show_bug.cgi?id=16858 --- Comment #15 from Bernd Edlinger bernd.edlinger at hotmail dot de --- (In reply to Nick Clifton from comment #14) Hi Bernd, Actually here is a better patch. The previous one also affected non-code weak symbols which was the wrong thing to do. Please try this patch out and let me know if you are happy with it. Cheers Nick hmm... both variants of your patch do not affect x86_64, only 32-bit? can you give an example of these non-code weak symbols? -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/16858] weak external reference has wrong value
https://sourceware.org/bugzilla/show_bug.cgi?id=16858 --- Comment #16 from Nick Clifton nickc at redhat dot com --- Hi Bernd, both variants of your patch do not affect x86_64, only 32-bit? Yes. According to my test the x86_64 target is already working. (After the first patch was applied). It was only the 32-bit target that remained broken. can you give an example of these non-code weak symbols? Yes - the linker test ld/testsuite/ld-scripts/weak.* This test breaks if you applied the patch I proposed yesterday, but passes if you apply today's patch. Cheers Nick -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/16952] New: PowerPC LD requests bug report. Issue related to EABI symbols _SDA_BASE_, _SDA2_BASE_
https://sourceware.org/bugzilla/show_bug.cgi?id=16952
Bug ID: 16952
Summary: PowerPC LD requests bug report. Issue related to EABI
symbols _SDA_BASE_, _SDA2_BASE_
Product: binutils
Version: 2.25 (HEAD)
Status: NEW
Severity: minor
Priority: P2
Component: binutils
Assignee: unassigned at sourceware dot org
Reporter: bcbrock at us dot ibm.com
We plan to link an embedded application using the .PPC.EMB.sdata0 facility of
the PowerPC EABI. A simple test failed, and LD requested that we report a bug.
After poking around, the problem seems to be that the linker script must define
_SDA_BASE_ and _SDA2_BASE_, even though these symbols are not explicitly
required by this example.
Source file:
.text
lwz 3, foo@sda21(0)
.data
foo:
.long 0
Linker script:
SECTIONS
{
. = 0x8000;
.text . : { *(.text) }
.PPC.EMB.sdata0 . : { *(.data) }
/* Uncomment to eliminate the bug report
_SDA_BASE_ = .;
_SDA2_BASE_ = .;
*/
}
Messages:
as bug.S -o bug.o
ld bug.o -o bug -Tbug.cmd -Map bug.map
ld: BFD (GNU Binutils) 2.24.51.20140515 internal error, aborting at
../../binutils-gdb/bfd/elflink.c line 8855 in elf_link_output_extsym
ld: Please report this bug.
Thanks,
Bishop Brock
--
You are receiving this mail because:
You are on the CC list for the bug.
___
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/16858] weak external reference has wrong value
https://sourceware.org/bugzilla/show_bug.cgi?id=16858 --- Comment #17 from Bernd Edlinger bernd.edlinger at hotmail dot de --- (In reply to Nick Clifton from comment #16) Hi Nick, yes, now it works for me. Thanks. Maybe you should also add a test case for this mess. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/16946] [v850-elf] Intenal error : tc-v850.c line 3248 in md_assemble
https://sourceware.org/bugzilla/show_bug.cgi?id=16946 Nick Clifton nickc at redhat dot com changed: What|Removed |Added CC||nickc at redhat dot com --- Comment #3 from Nick Clifton nickc at redhat dot com --- Created attachment 7604 -- https://sourceware.org/bugzilla/attachment.cgi?id=7604action=edit Error when ctoff() is used with RH850 ABI -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/16946] [v850-elf] Intenal error : tc-v850.c line 3248 in md_assemble
https://sourceware.org/bugzilla/show_bug.cgi?id=16946 Nick Clifton nickc at redhat dot com changed: What|Removed |Added Status|NEW |ASSIGNED --- Comment #4 from Nick Clifton nickc at redhat dot com --- Hi Masaki-san, You appear to be using an old version of gcc with a newer version of the binutils. The problem is that with the newer binutils the default ABI supported by the assembler is the RH850 ABI, and this does not support the ctoff() pseudo-op. If you were using a newer version of gcc then this would not be a problem - the callt instruction would not be generated. You could work around this problem by compiling with -Wa,-mgcc-abi or -mdisable-callt. I have uploaded a patch which will make the assembler issue an error message instead of generating an internal fault. Please could you try it out and let me know if you have any problems with it. Cheers Nick -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/16946] [v850-elf] Intenal error : tc-v850.c line 3248 in md_assemble
https://sourceware.org/bugzilla/show_bug.cgi?id=16946 --- Comment #5 from Masaki MURANAKA mon...@monami-software.com --- Hello Nick, As you said, I'm using a git head of gcc-4_7 (old). I got better messages after applied your patch: ``` /var/folders/wr/mrhq4cgx2b15l4xwgj12jchmgn/T//ccGtpXOI.s: Assembler messages: /var/folders/wr/mrhq4cgx2b15l4xwgj12jchmgn/T//ccGtpXOI.s:88: Error: callt ctoff(__callt_save_r31c): ctoff() is not supported by the rh850 ABI. Use -mgcc-abi instead /var/folders/wr/mrhq4cgx2b15l4xwgj12jchmgn/T//ccGtpXOI.s:99: Error: callt ctoff(__callt_return_r31c): ctoff() is not supported by the rh850 ABI. Use -mgcc-abi instead make[3]: *** [_absvsi2.o] Error 1 make[2]: *** [multi-do] Error 1 make[1]: *** [all-multi] Error 2 make: *** [all-target-libgcc] Error 2 ``` And I see this issue is caused by ABI. I'll make a local patch for gcc to avoid this issue. Thank you for your suggestion. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
