linux-scsi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: Ric Wheeler <rwheeler@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>,
	Sumit Saxena <sumit.saxena@broadcom.com>,
	linux-scsi@vger.kernel.org, martin.petersen@oracle.com,
	thenzl@redhat.com, jejb@linux.vnet.ibm.com,
	kashyap.desai@broadcom.com, Christoph Hellwig <hch@infradead.org>,
	"Martin K. Petersen" <mkp@mkp.net>,
	Jeff Moyer <jmoyer@redhat.com>, Gris Ge <fge@redhat.com>,
	Ewan Milne <emilne@redhat.com>, Jens Axboe <axboe@kernel.dk>
Subject: Re: [PATCH 4/7] megaraid_sas: Send SYNCHRONIZE_CACHE command to firmware
Date: Mon, 17 Oct 2016 08:55:06 -0700	[thread overview]
Message-ID: <20161017155506.GA25521@infradead.org> (raw)
In-Reply-To: <9e5fbcef-bd72-71be-76f6-f41e26b5cb3c@redhat.com>

On Mon, Oct 17, 2016 at 09:01:29AM -0400, Ric Wheeler wrote:
> This must go in - without this fix, there is no data integrity for any file system.

megaraid always had odd ideas on cache flushing, and this might be
a opportunity to write down all the assumptions and document them.

> In effect, this driver by default has been throwing away SYNCHRONIZE_CACHE
> commands even when acting in JBOD/non-RAID mode.

That would explain some issues we've seen with megaraid hardware, but
it seems a bit too shocking to be true.

Looking over the patch I disagree with the module option - we must do
the right thing by default, which is sending SYNCHRONIZE_CACHE commands
if the WCE bit is set.  If there are controllers where this is harmful
for RAID mode and we can't fix the firmware in time we'll need to make
special exceptions for this case in the driver based on the PCI ID
and knowing what we talk to instead of leaving it to the user.

> * having T10 & T13 report the existence of a volatile write cache - this is
> different than WCE set, some devices have a write cache and are
> battery/flash backed.

T10 is pretty clear now the WCE should only be set for a non-voltile
cache.  For a while they had odd NV bits to allow flushing a
non-volatile cache, but in the latest revisions all that is gone.

  parent reply	other threads:[~2016-10-17 15:55 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-17 10:24 [PATCH 0/7] megaraid_sas: Updates for scsi-next Sumit Saxena
2016-10-17 10:24 ` [PATCH 1/7] megaraid_sas: For SRIOV enabled firmware, ensure VF driver waits for 30secs before reset Sumit Saxena
2016-10-17 11:29   ` Hannes Reinecke
2016-10-17 12:43   ` Tomas Henzl
2016-10-17 10:24 ` [PATCH 2/7] megaraid_sas: Send correct PhysArm to FW for R1 VD downgrade Sumit Saxena
2016-10-17 11:29   ` Hannes Reinecke
2016-10-17 12:50   ` Tomas Henzl
2016-10-17 10:24 ` [PATCH 3/7] megaraid_sas: Do not fire DCMDs during PCI shutdown/detach Sumit Saxena
2016-10-17 11:31   ` Hannes Reinecke
2016-10-17 12:19     ` Sumit Saxena
2016-10-17 10:24 ` [PATCH 4/7] megaraid_sas: Send SYNCHRONIZE_CACHE command to firmware Sumit Saxena
2016-10-17 11:34   ` Hannes Reinecke
2016-10-17 12:13     ` Sumit Saxena
2016-10-17 13:01     ` Ric Wheeler
2016-10-17 13:31       ` Sumit Saxena
2016-10-17 15:55       ` Christoph Hellwig [this message]
2016-10-17 16:08         ` Ric Wheeler
2016-10-17 16:23           ` Ewan D. Milne
2016-10-17 16:20       ` James Bottomley
2016-10-17 16:27         ` Ric Wheeler
2016-10-17 17:19           ` James Bottomley
2016-10-17 17:30             ` Ric Wheeler
2016-10-17 17:36             ` Kashyap Desai
2016-10-17 17:51               ` James Bottomley
2016-10-18 13:56                 ` Tomas Henzl
2016-10-19  9:50                   ` Ching Huang
2016-10-19 12:55                     ` Tomas Henzl
2016-10-19 18:07                       ` Raghava Aditya Renukunta
2016-10-21 16:30                         ` Tomas Henzl
2016-10-25  2:02                       ` Martin K. Petersen
     [not found]                         ` <CAKiknE_MM88eVON1g_qy7wbb2fkxdAs6O0SRSzN-RbQqSvx1eA@mail.gmail.com>
2016-10-27  2:07                           ` Martin K. Petersen
2016-10-18 18:47                 ` Sumit Saxena
2016-10-17 16:29         ` Ric Wheeler
2016-10-17 13:13   ` Tomas Henzl
2016-10-17 13:28     ` Sumit Saxena
2016-10-17 13:57       ` Tomas Henzl
2016-10-17 14:25         ` Sumit Saxena
2016-10-18 13:07         ` Ric Wheeler
2016-10-18 13:22           ` Sumit Saxena
2016-10-17 10:24 ` [PATCH 5/7] megaraid_sas: driver version upgrade Sumit Saxena
2016-10-17 11:35   ` Hannes Reinecke
2016-10-17 12:20     ` Sumit Saxena
2016-10-17 10:24 ` [PATCH 6/7] MAINTAINERS: Update megaraid maintainers list Sumit Saxena
2016-10-17 11:37   ` Hannes Reinecke
2016-10-17 10:24 ` [PATCH 7/7] megaraid_sas: Do not set MPI2_TYPE_CUDA for JBOD FP path for FW which does not support JBOD sequence map Sumit Saxena
2016-10-17 11:37   ` Hannes Reinecke
2016-10-17 13:23   ` Tomas Henzl

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=20161017155506.GA25521@infradead.org \
    --to=hch@infradead.org \
    --cc=axboe@kernel.dk \
    --cc=emilne@redhat.com \
    --cc=fge@redhat.com \
    --cc=hare@suse.de \
    --cc=jejb@linux.vnet.ibm.com \
    --cc=jmoyer@redhat.com \
    --cc=kashyap.desai@broadcom.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=mkp@mkp.net \
    --cc=rwheeler@redhat.com \
    --cc=sumit.saxena@broadcom.com \
    --cc=thenzl@redhat.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).