This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: [2.11 branch]: Failures for armv4l-unknown-linux-gnu
- To: Philip Blundell <philb at gnu dot org>
- Subject: Re: [2.11 branch]: Failures for armv4l-unknown-linux-gnu
- From: Manfred Hollstein <manfred dot h at gmx dot net>
- Date: Sun, 4 Feb 2001 16:43:04 +0100 (CET)
- Cc: binutils at sources dot redhat dot com
- References: <14964.10269.700126.391586@saturn.hollstein.net><E14Onr4-0003HY-00@kings-cross.london.uk.eu.org>
- Reply-To: Manfred Hollstein <manfred dot h at gmx dot net>
On Friday, 2 February 2001, 21:34:54 +0000, philb@gnu.org wrote:
> >regexp_diff match failure
> >regexp "^00000260 <[^>]*> ebfffffe ? bl 00000000 <[^>]*>$"
> >line "00000260 <bar+0x50> ebfffffe bl 00000260 <bar+0x50>"
>
> This looks like some kind of dejagnu lossage. It seems to be picking up your
> old, installed objdump rather than the newly built one in the tree.
No. This is from gas.log after a similar testrun today:
Running /home/gnu/work/GNU/binutils-2.11/gas/testsuite/gas/arm/arm.exp ...
../as-new -marm2 -EL -o dump.o /home/gnu/work/GNU/binutils-2.11/gas/testsuite/gas/arm/inst.s
/home/gnu/work/GNU/binutils-2.11-armv4l-unknown-linux-gnu/gas/testsuite/../../binutils/objdump -dr --prefix-addresses --show-raw-insn dump.o > dump.out
regexp_diff match failure
regexp "^00000260 <[^>]*> ebfffffe ? bl 00000000 <[^>]*>$"
line "00000260 <bar+0x50> ebfffffe bl 00000260 <bar+0x50>"
regexp_diff match failure
regexp "^00000264 <[^>]*> 5bfffffe ? blpl 00000000 <[^>]*>$"
line "00000264 <bar+0x54> 5bfffffe blpl 00000264 <bar+0x54>"
regexp_diff match failure
regexp "^00000268 <[^>]*> eafffffe ? b 00000000 <[^>]*>$"
line "00000268 <bar+0x58> eafffffe b 00000268 <bar+0x58>"
regexp_diff match failure
regexp "^0000026c <[^>]*> dafffffe ? ble 00000000 <[^>]*>$"
line "0000026c <bar+0x5c> dafffffe ble 0000026c <bar+0x5c>"
FAIL: ARM basic instructions
As you can see, it's using the objdump program from my builddir.
> I've managed to reproduce the symptom here, but not consistently.
Hmm.
> The output you're seeing from objdump does seem to be the correct version.
> This was the patch that changed the behaviour; I'm not totally sure I agree
> with it, but that's a different matter.
>
> 2001-01-09 Nick Clifton <nickc@redhat.com>
>
> * arm-dis.c (print_insn): Set pc to zero for instructions with
> a reloc associated with them.
So, does this mean, Nick changed the sources, but forgot to fix the
test cases accordingly?
>
> p.
>
Cheers.
l8er
manfred.