[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: arm vmsr bug?
From: |
nick clifton |
Subject: |
Re: arm vmsr bug? |
Date: |
Tue, 20 Mar 2012 11:52:04 +0000 |
User-agent: |
Mozilla/5.0 (X11; Linux i686; rv:10.0.1) Gecko/20120216 Thunderbird/10.0.1 |
Hi Clive,
I think that there's a bug in the generation of ARM vmsr instructions
For future reference, it helps if you can open a bug report for problems
like this. You can use the bugzilla system here to do this:
http://sourceware.org/bugzilla/
in binutils 2.2.
I assume that you mean 2.22. Version 2.2 is *really* old...
vmsr fpexc, r0
I think this is a bug, unless there's some reason for disallowing fpsid& fpexc
that I haven't found....
No. It was just a snafu - the assembler was only supporting the normal
mode version of the VMSR instruction, not the priviledged mode versions
of it.
I am going to apply the patch below to fix this problem, and a similar
one for the VMRS instruction.
Cheers
Nick
gas/ChangeLog
2012-03-20 Nick Clifton <address@hidden>
* config/tc-arm.c (do_vmrs): Accept priviledged mode VFP system
registers.
(do_vmsr): Likewise.
(arm_opcode_insns): Do not default to using the FPSCR register in
the VMRS and VMSR registers.
gas/testsuite/ChangeLog
2012-03-20 Nick Clifton <address@hidden>
* gas/arm/vfp1xD.s: Add tests of the VMSR ad VMRS instructions in
priviledged modes.
* gas/arm/vfp1xD.d: Update expected output.
vmsr.patch
Description: Text Data
- arm vmsr bug?, Clive Davies, 2012/03/08
- Re: arm vmsr bug?,
nick clifton <=