Re: ramoops: is using platform_drivers correct?

[Marco Stornelli]

2011/5/25 Américo Wang <xiyou.wangcong@xxxxxxxxx>:
> On Wed, May 25, 2011 at 1:03 AM, Marco Stornelli
> <marco.stornelli@xxxxxxxxx> wrote:
>> Il 23/05/2011 15:27, Stevie Trujillo ha scritto:
>>>
>>> Hello,
>>>
>>> ramoops (drivers/char/ramoops.c) is for "all" computers right? When I try
>>> to
>>
>> Yes, or at least was my intention. It's true that this drivers is useful
>> when you use an NVRAM, this is typical for the embedded world where the
>> platform driver approach is more diffused. Sure at this point the module
>> parameters are not useful. In addition the platform data struct doesn't
>> define a way to select if to dump only oops. At the end I think a patch it's
>> needed here. I have to look at the code to see if it's possible to use the
>> platform data OR module parameters. I'll submit a patch.
>>
>
> But we have the following code:
>
>        if (reason != KMSG_DUMP_OOPS &&
>            reason != KMSG_DUMP_PANIC &&
>            reason != KMSG_DUMP_KEXEC)
>                return;
>
> which is what you meant by saying "only oops" ?

I meant a way to set dump_oops parameter via platform data.

>
> I still don't think that is a correct way to tell people not to use
> ramoops, we need to document this rather than prevent using
> it in the code.
>

I really don't understand what you mean here.

Marco
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/