linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jody McIntyre <scjody@sun.com>
To: Dan Williams <dan.j.williams@intel.com>
Cc: Neil Brown <neilb@suse.de>,
	linux-raid@vger.kernel.org, maan@systemlinux.org,
	linux-kernel@vger.kernel.org, yur@emcraft.com, hpa@zytor.com
Subject: Re: [PATCH v2 05/11] md/raid5: add scribble region for buffer lists
Date: Mon, 08 Jun 2009 13:25:35 -0400	[thread overview]
Message-ID: <20090608172535.GA28448@clouds> (raw)
In-Reply-To: <e9c3a7c20906051219x122f443dtc7836f0c16ae9f27@mail.gmail.com>

On Fri, Jun 05, 2009 at 12:19:07PM -0700, Dan Williams wrote:
> One of the design goals was to prevent the occurrence of the
> softlockup watchdog events which seem to trigger on large raid6
> resyncs.  A per-cpu scheme would still require preempt_disable() while
> the calculation is active, so perhaps we just need a call to
> cond_resched() in raid5d to appease the scheduler.

FWIW we added this to the patches shipped with Lustre:

Index: linux-2.6.18-128.1.1/drivers/md/raid5.c
===================================================================
--- linux-2.6.18-128.1.1.orig/drivers/md/raid5.c
+++ linux-2.6.18-128.1.1/drivers/md/raid5.c
@@ -2987,6 +2987,8 @@ static void raid5d (mddev_t *mddev)
 		handle_stripe(sh, conf->spare_page);
 		release_stripe(sh);
 
+		cond_resched();
+
 		spin_lock_irq(&conf->device_lock);
 	}
 	PRINTK("%d stripes handled\n", handled);

I thought most of these issues were gone in more recent kernels, but we
haven't tested RAID on anything other than RHEL 4+5 extensively (Lustre
doesn't support sufficiently new kernels yet.)

Cheers,
Jody

  reply	other threads:[~2009-06-08 17:25 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-05-19  0:59 [PATCH v2 00/11] Asynchronous raid6 acceleration (part 1 of 3) Dan Williams
2009-05-19  0:59 ` [PATCH v2 01/11] async_tx: rename zero_sum to val Dan Williams
     [not found]   ` <f12847240905200110x63b22601idbbdf3369984fa9a@mail.gmail.com>
2009-05-29 13:41     ` Sosnowski, Maciej
2009-06-03 18:12       ` Dan Williams
2009-05-19  0:59 ` [PATCH v2 02/11] async_tx: kill ASYNC_TX_DEP_ACK flag Dan Williams
     [not found]   ` <f12847240905250320w74897dabo6576b4b48bd19c0c@mail.gmail.com>
2009-05-29 13:41     ` Sosnowski, Maciej
2009-06-03 18:42       ` Dan Williams
2009-05-19  0:59 ` [PATCH v2 03/11] async_tx: structify submission arguments, add scribble Dan Williams
2009-05-20  8:06   ` Andre Noll
2009-05-20 18:19     ` Dan Williams
     [not found]   ` <f12847240905250321v774c4e8dscd7a466cd2e61168@mail.gmail.com>
2009-05-29 13:41     ` Sosnowski, Maciej
2009-06-03 19:05       ` Dan Williams
2009-05-19  0:59 ` [PATCH v2 04/11] async_xor: permit callers to pass in a 'dma/page scribble' region Dan Williams
2009-05-20  8:08   ` Andre Noll
2009-05-20 18:35     ` Dan Williams
2009-05-20 19:09       ` Andre Noll
2009-05-22  8:29         ` Andre Noll
2009-05-22 17:25           ` Dan Williams
2009-05-25  7:55             ` Andre Noll
     [not found]   ` <f12847240905250320w523fc657w3bca47f23442f46e@mail.gmail.com>
2009-05-29 13:41     ` Sosnowski, Maciej
2009-05-19  0:59 ` [PATCH v2 05/11] md/raid5: add scribble region for buffer lists Dan Williams
2009-05-20  8:09   ` Andre Noll
2009-05-20 19:05     ` Dan Williams
2009-06-04  6:11   ` Neil Brown
2009-06-05 19:19     ` Dan Williams
2009-06-08 17:25       ` Jody McIntyre [this message]
2009-05-19  0:59 ` [PATCH v2 06/11] async_tx: add sum check flags Dan Williams
     [not found]   ` <f12847240905200111p54382735v6941b52825cf4d7e@mail.gmail.com>
2009-05-29 13:41     ` Sosnowski, Maciej
2009-05-19  1:00 ` [PATCH v2 07/11] async_tx: kill needless module_{init|exit} Dan Williams
     [not found]   ` <f12847240905250323o21113fb9xbc4c16eea07b215@mail.gmail.com>
2009-05-29 13:42     ` Sosnowski, Maciej
2009-05-19  1:00 ` [PATCH v2 08/11] async_tx: add support for asynchronous GF multiplication Dan Williams
2009-05-22  8:29   ` Andre Noll
2009-06-03 22:11     ` Dan Williams
     [not found]   ` <f12847240905200111q37457b29lb9e30879e251888@mail.gmail.com>
2009-05-29 13:42     ` Sosnowski, Maciej
2009-06-03 22:16       ` Dan Williams
2009-05-19  1:00 ` [PATCH v2 09/11] async_tx: add support for asynchronous RAID6 recovery operations Dan Williams
2009-05-22  8:29   ` Andre Noll
2009-05-22 18:39     ` Dan Williams
     [not found]   ` <f12847240905250323q2e14efd6q69022a62cc7fd01f@mail.gmail.com>
2009-05-29 13:42     ` Sosnowski, Maciej
2009-05-19  1:00 ` [PATCH v2 10/11] dmatest: add pq support Dan Williams
     [not found]   ` <f12847240905250324t1a55b757hc8cd06d6b9663efe@mail.gmail.com>
2009-05-29 13:42     ` Sosnowski, Maciej
2009-05-19  1:00 ` [PATCH v2 11/11] async_tx: raid6 recovery self test Dan Williams
2009-05-22  8:29   ` Andre Noll
2009-06-03 21:42     ` Dan Williams
     [not found]   ` <f12847240905250324k2b4a0c7as9ac9b084d3707ce5@mail.gmail.com>
2009-05-29 13:42     ` Sosnowski, Maciej

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=20090608172535.GA28448@clouds \
    --to=scjody@sun.com \
    --cc=dan.j.williams@intel.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-raid@vger.kernel.org \
    --cc=maan@systemlinux.org \
    --cc=neilb@suse.de \
    --cc=yur@emcraft.com \
    /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).