Re: [PATCH] x86/amd_nb: Check for invalid SMN reads

[Yazen Ghannam <yazen.ghannam@amd.com>]

On 5/10/23 7:35 AM, Borislav Petkov wrote:
> On Wed, Apr 05, 2023 at 03:10:09PM -0400, Yazen Ghannam wrote:
>> What do you think?
> 
> Yes, please put that as a comment over __amd_smn_rw() as to why callers
> should check the retval *and* make both the read and the write
> __must_check and get rid of
> 
>         if (err)
>                 pr_warn("Error %s SMN address 0x%x.\n",
>                         (write ? "writing to" : "reading from"), address);
> 
> which won't be seen in all cases. The __must_check will force the
> callers to do the proper checking which is the only sane thing to do
> with such a variety of bit behaviors.
>

Understood, will do.

Thanks,
Yazen