This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH 2/4] Consolidate off_t/off64_t syscall argument passing
- From: Adhemerval Zanella <adhemerval dot zanella at linaro dot org>
- To: libc-alpha at sourceware dot org, Adhemerval Zanella <adhemerval dot zanella at linaro dot com>
- Date: Mon, 11 Apr 2016 10:07:31 -0300
- Subject: Re: [PATCH 2/4] Consolidate off_t/off64_t syscall argument passing
- Authentication-results: sourceware.org; auth=none
- References: <1456431555-14182-1-git-send-email-adhemerval dot zanella at linaro dot org> <1456431555-14182-3-git-send-email-adhemerval dot zanella at linaro dot org> <20160410041412 dot GT6588 at vapier dot lan> <570B95EE dot 2090302 at linaro dot org>
On 11-04-2016 09:17, Adhemerval Zanella wrote:
>
>
> On 10-04-2016 01:14, Mike Frysinger wrote:
>> On 25 Feb 2016 17:19, Adhemerval Zanella wrote:
>>> --- a/sysdeps/unix/sysv/linux/sysdep.h
>>> +++ b/sysdeps/unix/sysv/linux/sysdep.h
>>>
>>> +/* Provide a dummy argument that can be used to force register
>>> + alignment for register pairs if required by the syscall ABI. */
>>> +#ifdef __ASSUME_ALIGNED_REGISTER_PAIRS
>>> +#define __ALIGNMENT_ARG 0,
>>> +#define __ALIGNMENT_COUNT(a,b) b
>>> +#else
>>> +#define __ALIGNMENT_ARG
>>> +#define __ALIGNMENT_COUNT(a,b) a
>>> +#endif
>>
>> while not a new issue, should the return be parenthesized ?
>> i.e. (b) and (a) ?
>> -mike
>>
>
> I will change that, thanks.
In fact we can not change it: some functions uses the define as
__ALIGNMENT_COUNT (3, 4) and some ports uses to correct select
the LOAD_ARGS_{3,4} in syscall generation (nios2). Using
parenthesis leads it to evaluate as LOAD_ARGS_(3) and this
issuing a build issue (this can be fixed I think, but it is
not the idea of this patch).