Re: [LTP] [PATCH v3 1/3] lib: adding .supported_archs field in tst_test structure

From: Li Wang <liwang@redhat.com>
To: Cyril Hrubis <chrubis@suse.cz>
Cc: LTP List <ltp@lists.linux.it>
Subject: Re: [LTP] [PATCH v3 1/3] lib: adding .supported_archs field in tst_test structure
Date: Tue, 9 Nov 2021 15:52:39 +0800	[thread overview]
Message-ID: <CAEemH2fB1kyM9OYNgHKf-4MxZf+ifXvVry4UpWO=Ute2wG8Z2w@mail.gmail.com> (raw)
In-Reply-To: <YYkZXtBn+ZKOXVzB@yuki>

[-- Attachment #1.1: Type: text/plain, Size: 1068 bytes --]

> > diff --git a/include/tst_arch.h b/include/tst_arch.h
> > new file mode 100644
> > index 000000000..784c3093b
> > --- /dev/null
> > +++ b/include/tst_arch.h
> > @@ -0,0 +1,39 @@
> > +/* SPDX-License-Identifier: GPL-2.0-or-later
> > + * Copyright (c) 2021 Li Wang <liwang@redhat.com>
> > + */
> > +
> > +#ifndef TST_ARCH_H__
> > +#define TST_ARCH_H__
> > +
> > +enum tst_arch_type {
> > +     TST_I386,
> > +     TST_X86,
>
> Why do we have both i386 and X86 here? Isn't __i386__ synonymous for
> __x86__ does gcc even define __x86__?
>

My fault, I just copy that from testcase defines and didn't check
if __x86__ validatable or not.

>
> I doubt that we care about the differencies between i386, i586 and i686
> at all. I would just keep TST_X86 in the list for any 32bit intel
> compatible hardware.
>

I prefer to only keep TST_I386 and TST_X86_64 for use. Becuase
so far I didn't see there is a requirement on i[4|5|6]86 in LTP at all.
And, we can add that if we really need it in the future.

The rest of the comments looks good, thanks.

-- 
Regards,
Li Wang

[-- Attachment #1.2: Type: text/html, Size: 2119 bytes --]

[-- Attachment #2: Type: text/plain, Size: 60 bytes --]

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp