linux-kernel - Re: [PATCH] f2fs: introduce DEFAULT_IO_TIMEOUT

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <61e43dcb781c9e880fac95b525830fd384de122a.camel@dubeyko.com>
Date:   Thu, 26 Dec 2019 16:08:40 +0300
From:   Vyacheslav Dubeyko <slava@...eyko.com>
To:     Geert Uytterhoeven <geert@...ux-m68k.org>
Cc:     Chao Yu <yuchao0@...wei.com>, Jaegeuk Kim <jaegeuk@...nel.org>,
        linux-f2fs-devel@...ts.sourceforge.net,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Chao Yu <chao@...nel.org>,
        Linux FS Devel <linux-fsdevel@...r.kernel.org>
Subject: Re: [PATCH] f2fs: introduce DEFAULT_IO_TIMEOUT_JIFFIES

Hi Geert,

On Thu, 2019-12-26 at 11:43 +0100, Geert Uytterhoeven wrote:
> Hi Vyacheslav,
> 
> On Wed, Dec 25, 2019 at 10:58 AM Vyacheslav Dubeyko <
> slava@...eyko.com> wrote:
> > On Mon, 2019-12-23 at 09:41 +0100, Geert Uytterhoeven wrote:
> > > On Mon, Dec 23, 2019 at 5:01 AM Chao Yu <yuchao0@...wei.com>
> > > wrote:
> > > > As Geert Uytterhoeven reported:
> > > > 
> > > > for parameter HZ/50 in congestion_wait(BLK_RW_ASYNC, HZ/50);
> > > > 
> > > > On some platforms, HZ can be less than 50, then unexpected 0
> > > > timeout
> > > > jiffies will be set in congestion_wait().
> > 
> > It looks like that HZ could have various value on diferent
> > platforms.
> > So, why does it need to divide HZ on 50? Does it really necessary?
> > Could it be used HZ only without the division operation?
> 
> A timeout of HZ means 1 second.
> HZ/50 means 20 ms, but has the risk of being zero, if HZ < 50.
> 
> If you want to use a timeout of 20 ms, you best use
> msecs_to_jiffies(20),
> as that takes care of the special cases, and never returns 0.
> 

The msecs_to_jiffies(20) looks much better for my taste. Maybe, we
could use this as solution of the issue?

Thanks,
Viacheslav Dubeyko.

> Gr{oetje,eeting}s,
> 
>                         Geert
>