From: Mel Gorman <mgorman@suse.de>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-MM <linux-mm@kvack.org>,
Linux-Netdev <netdev@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
David Miller <davem@davemloft.net>, Neil Brown <neilb@suse.de>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Mike Christie <michaelc@cs.wisc.edu>,
Eric B Munson <emunson@mgebm.net>
Subject: Re: [PATCH 05/16] mm: allow PF_MEMALLOC from softirq context
Date: Wed, 2 May 2012 17:24:03 +0100 [thread overview]
Message-ID: <20120502162403.GE11435@suse.de> (raw)
In-Reply-To: <20120501150813.657cd5c0.akpm@linux-foundation.org>
On Tue, May 01, 2012 at 03:08:13PM -0700, Andrew Morton wrote:
> On Mon, 16 Apr 2012 13:16:52 +0100
> Mel Gorman <mgorman@suse.de> wrote:
>
> > This is needed to allow network softirq packet processing to make
> > use of PF_MEMALLOC.
>
> hm, why? You just added __GFP_MEMALLOC so we don't need to futz with
> PF_MEMALLOC?
>
The number of call sites is a problem. In patch 12, PF_MEMALLOC is set
where required. For example it is set in __netif_receive_skb() before it
calls packet_type->func() which is a per-protocol receive function such
as net/ipv4/ip_input.c#ip_rcv(). To use __GFP_MEMALLOC, every allocation
on this path would need to check the skb and set the flag as appropriate
for every protocol. This would make a mess and seeing as it is needed for
every allocation it makes more sense to set PF_MEMALLOC.
> > Currently softirq context cannot use PF_MEMALLOC due to it not being
> > associated with a task, and therefore not having task flags to fiddle
> > with - thus the gfp to alloc flag mapping ignores the task flags when
> > in interrupts (hard or soft) context.
> >
> > Allowing softirqs to make use of PF_MEMALLOC therefore requires some
> > trickery. We basically borrow the task flags from whatever process
> > happens to be preempted by the softirq.
> >
> > So we modify the gfp to alloc flags mapping to not exclude task flags
> > in softirq context, and modify the softirq code to save, clear and
> > restore the PF_MEMALLOC flag.
> >
> > The save and clear, ensures the preempted task's PF_MEMALLOC flag
> > doesn't leak into the softirq. The restore ensures a softirq's
> > PF_MEMALLOC flag cannot leak back into the preempted process.
> >
> > ...
> >
> > --- a/include/linux/sched.h
> > +++ b/include/linux/sched.h
> > @@ -1913,6 +1913,13 @@ static inline void rcu_copy_process(struct task_struct *p)
> >
> > #endif
> >
> > +static inline void tsk_restore_flags(struct task_struct *p,
> > + unsigned long pflags, unsigned long mask)
>
> The naming is poor.
>
> p -> "tsk" or "task"
> pflags -> "old_flags"
> mask -> "flags"
>
I went with orig_flags instead of old_flags so it reads as "restore the
original task flags".
> > +{
> > + p->flags &= ~mask;
> > + p->flags |= pflags & mask;
> > +}
> > +
> > #ifdef CONFIG_SMP
> > extern void do_set_cpus_allowed(struct task_struct *p,
> > const struct cpumask *new_mask);
> > diff --git a/kernel/softirq.c b/kernel/softirq.c
> > index 671f959..d349caa 100644
> > --- a/kernel/softirq.c
> > +++ b/kernel/softirq.c
> > @@ -210,6 +210,8 @@ asmlinkage void __do_softirq(void)
> > __u32 pending;
> > int max_restart = MAX_SOFTIRQ_RESTART;
> > int cpu;
> > + unsigned long pflags = current->flags;
>
> "old_flags"
>
> > + current->flags &= ~PF_MEMALLOC;
>
> The line before this one would be a suitable place for a comment!
>
/*
* Mask out PF_MEMALLOC s current task context is borrowed for the
* softirq. A softirq handled such as network RX might set PF_MEMALLOC
* again if the socket is related to swap
*/
?
> > pending = local_softirq_pending();
> > account_system_vtime(current);
> > @@ -265,6 +267,7 @@ restart:
> >
> > account_system_vtime(current);
> > __local_bh_enable(SOFTIRQ_OFFSET);
> > + tsk_restore_flags(current, pflags, PF_MEMALLOC);
> > }
> >
> > ...
> >
--
Mel Gorman
SUSE Labs
next prev parent reply other threads:[~2012-05-02 16:24 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-04-16 12:16 [PATCH 00/16] Swap-over-NBD without deadlocking V9 Mel Gorman
2012-04-16 12:16 ` [PATCH 01/16] mm: Serialize access to min_free_kbytes Mel Gorman
2012-04-23 23:50 ` David Rientjes
2012-04-16 12:16 ` [PATCH 02/16] mm: sl[au]b: Add knowledge of PFMEMALLOC reserve pages Mel Gorman
2012-04-23 23:51 ` David Rientjes
2012-04-25 15:05 ` Mel Gorman
2012-04-16 12:16 ` [PATCH 03/16] mm: slub: Optimise the SLUB fast path to avoid pfmemalloc checks Mel Gorman
2012-04-16 12:16 ` [PATCH 04/16] mm: Introduce __GFP_MEMALLOC to allow access to emergency reserves Mel Gorman
2012-04-16 12:16 ` [PATCH 05/16] mm: allow PF_MEMALLOC from softirq context Mel Gorman
2012-05-01 22:08 ` Andrew Morton
2012-05-02 16:24 ` Mel Gorman [this message]
2012-04-16 12:16 ` [PATCH 06/16] mm: Ignore mempolicies when using ALLOC_NO_WATERMARK Mel Gorman
2012-04-16 12:16 ` [PATCH 07/16] net: Introduce sk_allocation() to allow addition of GFP flags depending on the individual socket Mel Gorman
2012-04-16 12:16 ` [PATCH 08/16] netvm: Allow the use of __GFP_MEMALLOC by specific sockets Mel Gorman
2012-04-16 12:16 ` [PATCH 09/16] netvm: Allow skb allocation to use PFMEMALLOC reserves Mel Gorman
2012-04-16 12:16 ` [PATCH 10/16] netvm: Propagate page->pfmemalloc to skb Mel Gorman
2012-04-16 12:16 ` [PATCH 11/16] netvm: Propagate page->pfmemalloc from netdev_alloc_page " Mel Gorman
2012-04-16 12:16 ` [PATCH 12/16] netvm: Set PF_MEMALLOC as appropriate during SKB processing Mel Gorman
2012-04-16 12:17 ` [PATCH 13/16] mm: Micro-optimise slab to avoid a function call Mel Gorman
2012-04-16 12:17 ` [PATCH 14/16] nbd: Set SOCK_MEMALLOC for access to PFMEMALLOC reserves Mel Gorman
2012-04-16 12:17 ` [PATCH 15/16] mm: Throttle direct reclaimers if PF_MEMALLOC reserves are low and swap is backed by network storage Mel Gorman
2012-05-01 22:24 ` Andrew Morton
2012-05-02 16:24 ` Mel Gorman
2012-04-16 12:17 ` [PATCH 16/16] mm: Account for the number of times direct reclaimers get throttled Mel Gorman
2012-05-01 22:28 ` [PATCH 00/16] Swap-over-NBD without deadlocking V9 Andrew Morton
2012-05-03 15:00 ` Mel Gorman
2012-05-03 17:06 ` David Miller
2012-05-04 10:16 ` Mel Gorman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20120502162403.GE11435@suse.de \
--to=mgorman@suse.de \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=davem@davemloft.net \
--cc=emunson@mgebm.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=michaelc@cs.wisc.edu \
--cc=neilb@suse.de \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).