This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: [PATCH v2] arm: properly range check immediate operands of VSHL and VQSHL
- From: "Jan Beulich" <JBeulich at suse dot com>
- To: "Richard Earnshaw" <rearnsha at arm dot com>
- Cc: "paul at codesourcery dot com" <paul at codesourcery dot com>, "nickc at redhat dot com" <nickc at redhat dot com>, "binutils at sourceware dot org" <binutils at sourceware dot org>
- Date: Wed, 10 Apr 2013 14:14:14 +0100
- Subject: Re: [PATCH v2] arm: properly range check immediate operands of VSHL and VQSHL
- References: <516301B602000078000CB677 at nat28 dot tlf dot novell dot com> <51655ECD02000078000CC0BD at nat28 dot tlf dot novell dot com> <51655CD0 dot 90506 at arm dot com>
>>> On 10.04.13 at 14:36, Richard Earnshaw <rearnsha@arm.com> wrote:
> On 10/04/13 11:45, Jan Beulich wrote:
>> These two, other than VQSHLU, didn't have their immediates properly range
>> checked so far.
>>
>> gas/
>> 2013-04-10 Jan Beulich <jbeulich@suse.com>
>>
>> * gas/config/tc-arm.c (do_neon_shl_imm): Check immediate range.
>> (do_neon_qshl_imm): Likewise.
>>
>> gas/testsuite/
>> 2013-04-10 Jan Beulich <jbeulich@suse.com>
>>
>> * gas/arm/neon-addressing-bad.s: Add test for invalid VSHL,
>> VQSHL, and VQSHLU immediates.
>> * gas/arm/neon-addressing-bad.l: Update accordingly.
>>
>
> neon-addressing-bad is really checking addressing modes for
> loads/stores. I'd prefer these new checks to be in a separate file;
> perhaps neon-dataproc-bad.
>
> OK with that change.
Actually I specifically put them there, both because of the other
new tests you had me put there (which also weren't addressing
mode ones), and because neon-addressing-bad.d has
# name: Bad operand in Advanced SIMD Neon instructions
Please clarify.
Jan