[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190908083535.GD16220@zn.tnic>
Date: Sun, 8 Sep 2019 10:35:35 +0200
From: Borislav Petkov <bp@...en8.de>
To: "Hawa, Hanna" <hhhawa@...zon.com>
Cc: Robert Richter <rrichter@...vell.com>,
"mchehab@...nel.org" <mchehab@...nel.org>,
"james.morse@....com" <james.morse@....com>,
"linux-edac@...r.kernel.org" <linux-edac@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"dwmw@...zon.co.uk" <dwmw@...zon.co.uk>,
"benh@...zon.com" <benh@...zon.com>,
"ronenk@...zon.com" <ronenk@...zon.com>,
"talel@...zon.com" <talel@...zon.com>,
"jonnyc@...zon.com" <jonnyc@...zon.com>,
"hanochu@...zon.com" <hanochu@...zon.com>
Subject: Re: [PATCH 1/1] edac: Add an API for edac device to report for
multiple errors
On Sun, Sep 08, 2019 at 10:16:02AM +0200, Borislav Petkov wrote:
> On Sun, Sep 08, 2019 at 10:58:31AM +0300, Hawa, Hanna wrote:
> > > Better use WARN_ON_ONCE() to avoid flooding.
> >
> > In case of two drivers using this function with wrong error count, only the
> > first WARN_ON_ONCE will catch in this case, and other will miss other wrong
> > usage of other edac device drivers.
>
> The idea is to catch any driver using a 0 error count and fix it, not to
> flood dmesg. You want _ONCE.
... and you want to return early too, i.e.,
if (WARN_ON_ONCE(!error_count))
return;
Frankly, I'd even remove all the warning functionality and simply do
if (!error_count)
return;
but let's see how much it screams first.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
Powered by blists - more mailing lists