linux-kernel - Re: [PATCH] mm: warn about allocations which stall for too long

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]

Date:   Fri, 23 Sep 2016 16:44:26 +0800
From:   "Hillf Danton" <hillf.zj@...baba-inc.com>
To:     "'Michal Hocko'" <mhocko@...nel.org>
Cc:     <linux-mm@...ck.org>,
        "'Andrew Morton'" <akpm@...ux-foundation.org>,
        "'Johannes Weiner'" <hannes@...xchg.org>,
        "'Mel Gorman'" <mgorman@...e.de>,
        "'Tetsuo Handa'" <penguin-kernel@...ove.SAKURA.ne.jp>,
        "'LKML'" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] mm: warn about allocations which stall for too long

On Friday, September 23, 2016 4:32 PM, Michal Hocko wrote
> On Fri 23-09-16 16:29:36, Hillf Danton wrote:
> [...]
> > > @@ -3659,6 +3661,15 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
> > >  	else
> > >  		no_progress_loops++;
> > >
> > > +	/* Make sure we know about allocations which stall for too long */
> > > +	if (!(gfp_mask & __GFP_NOWARN) && time_after(jiffies, alloc_start + stall_timeout)) {
> > > +		pr_warn("%s: page alloction stalls for %ums: order:%u mode:%#x(%pGg)\n",
> > > +				current->comm, jiffies_to_msecs(jiffies-alloc_start),
> >
> > Better if pid is also printed.
> 
> I've tried to be consistent with warn_alloc_failed and that doesn't
> print pid either. Maybe both of them should. Dunno
> 
With pid imho we can distinguish two tasks with same name in a simpler way. 

> > > +				order, gfp_mask, &gfp_mask);
> > > +		stall_timeout += 10 * HZ;
> >
> > Alternatively	 alloc_start = jiffies;
> 
> Then we would lose the cumulative time in the output which is imho
> helpful because you cannot tell whether the new warning is a new request
> or the old one still looping.
> 
Fair.

thanks
Hillf