Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1281192 > unrolled thread

Re: [PATCH] Improve Atheros ethernet driver not to do order 4 GFP_ATOMIC allocation

Started byDavid Miller <davem@davemloft.net>
First post2015-12-01 21:40 +0100
Last post2015-12-03 09:20 +0100
Articles 3 — 3 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH] Improve Atheros ethernet driver not to do order 4  GFP_ATOMIC allocation David Miller <davem@davemloft.net> - 2015-12-01 21:40 +0100
    Re: [PATCH] Improve Atheros ethernet driver not to do order 4  GFP_ATOMIC allocation Pavel Machek <pavel@ucw.cz> - 2015-12-03 08:50 +0100
    Re: [PATCH] Improve Atheros ethernet driver not to do order 4  GFP_ATOMIC allocation Michal Hocko <mhocko@kernel.org> - 2015-12-03 09:20 +0100

#1281192 — Re: [PATCH] Improve Atheros ethernet driver not to do order 4 GFP_ATOMIC allocation

FromDavid Miller <davem@davemloft.net>
Date2015-12-01 21:40 +0100
SubjectRe: [PATCH] Improve Atheros ethernet driver not to do order 4 GFP_ATOMIC allocation
Message-ID<qAZSq-622-19@gated-at.bofh.it>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--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]


#1282761

FromPavel Machek <pavel@ucw.cz>
Date2015-12-03 08:50 +0100
Message-ID<qBwOl-1XG-9@gated-at.bofh.it>
In reply to#1281192
On Wed 2015-12-02 22:43:31, Chris Snook wrote:
> On Tue, Dec 1, 2015 at 12:35 PM David Miller <davem@davemloft.net> wrote:
> 
> > From: Michal Hocko <mhocko@kernel.org>
> > Date: Mon, 30 Nov 2015 14:21:29 +0100
> >
> > > On Sat 28-11-15 15:51:13, Pavel Machek wrote:
> > >>
> > >> atl1c driver is doing order-4 allocation with GFP_ATOMIC
> > >> priority. That often breaks  networking after resume. Switch to
> > >> GFP_KERNEL. Still not ideal, but should be significantly better.
> > >
> > > It is not clear why GFP_KERNEL can replace GFP_ATOMIC safely neither
> > > from the changelog nor from the patch context.
> >
> > Earlier in the function we do a GFP_KERNEL kmalloc so:
> >
> > ¯\_(ツ)_/¯
> >
> > It should be fine.
> >
> 
> AFAICT, the people who benefit from GFP_ATOMIC are the people running all
> their storage over NFS/iSCSI who are suspending their machines while
> they're so busy they don't have any clean order 4 pagecache to drop, and
> want the machine to panic rather than hang. The people who benefit
>from

iSCSI on machine that suspends... is that a joke or complicated way of
saying that noone benefits? And code uses... both GFP_ATOMIC and
GFP_KERNEL so that both sides are equally unhappy? :-).

Do you want to test the patch, update the subject line and send it to
Davem, or should I do it?

Do you see a way to split the allocation? Not even order 4 GFP_KERNEL
allocation is a nice thing to do...

									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]


#1282774

FromMichal Hocko <mhocko@kernel.org>
Date2015-12-03 09:20 +0100
Message-ID<qBxho-2o6-25@gated-at.bofh.it>
In reply to#1281192
On Wed 02-12-15 22:43:31, Chris Snook wrote:
> On Tue, Dec 1, 2015 at 12:35 PM David Miller <davem@davemloft.net> wrote:
> 
> > From: Michal Hocko <mhocko@kernel.org>
> > Date: Mon, 30 Nov 2015 14:21:29 +0100
> >
> > > On Sat 28-11-15 15:51:13, Pavel Machek wrote:
> > >>
> > >> atl1c driver is doing order-4 allocation with GFP_ATOMIC
> > >> priority. That often breaks  networking after resume. Switch to
> > >> GFP_KERNEL. Still not ideal, but should be significantly better.
> > >
> > > It is not clear why GFP_KERNEL can replace GFP_ATOMIC safely neither
> > > from the changelog nor from the patch context.
> >
> > Earlier in the function we do a GFP_KERNEL kmalloc so:
> >
> > ¯\_(ツ)_/¯
> >
> > It should be fine.
> >
> 
> AFAICT, the people who benefit from GFP_ATOMIC are the people running all
> their storage over NFS/iSCSI who are suspending their machines while
> they're so busy they don't have any clean order 4 pagecache to drop, and
> want the machine to panic rather than hang.

Why would GFP_KERNEL order-4 allocation hang? It will fail if there are
not >=4 order pages available even after reclaim and/or compaction.
GFP_ATOMIC allocations should be used only when an access to memory
reserves is really required. If the allocation just doesn't want to
invoke direct reclaim then GFP_NOWAIT is a more suitable alternative.

-- 
Michal Hocko
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web