From: Tomas Henzl <thenzl@redhat.com>
To: "Matthew R. Ochs" <mrochs@linux.vnet.ibm.com>,
linux-scsi@vger.kernel.org,
James Bottomley <James.Bottomley@HansenPartnership.com>,
"Nicholas A. Bellinger" <nab@linux-iscsi.org>,
Brian King <brking@linux.vnet.ibm.com>,
Ian Munsie <imunsie@au1.ibm.com>,
Daniel Axtens <dja@ozlabs.au.ibm.com>,
Andrew Donnellan <andrew.donnellan@au1.ibm.com>,
David Laight <David.Laight@ACULAB.COM>
Cc: Michael Neuling <mikey@neuling.org>,
"Manoj N. Kumar" <manoj@linux.vnet.ibm.com>,
linuxppc-dev@lists.ozlabs.org
Subject: Re: [PATCH v6 30/37] cxlflash: Fix to avoid corrupting adapter fops
Date: Fri, 23 Oct 2015 16:00:03 +0200 [thread overview]
Message-ID: <562A3D63.7020707@redhat.com> (raw)
In-Reply-To: <1445458537-60298-1-git-send-email-mrochs@linux.vnet.ibm.com>
On 21.10.2015 22:15, Matthew R. Ochs wrote:
> The fops owned by the adapter can be corrupted in certain scenarios,
> opening a window where certain fops are temporarily NULLed before being
> reset to their proper value. This can potentially lead software to make
> incorrect decisions, leaving the user with the inability to function as
> intended.
>
> An example of this behavior can be observed when there are a number of
> users with a high rate of turn around (attach to LUN, perform an I/O,
> detach from LUN, repeat). Every so often a user is given a valid
> context and adapter file descriptor, but the file associated with the
> descriptor lacks the correct read permission bit (FMODE_CAN_READ) and
> thus the read system call bails before calling the valid read fop.
>
> Background:
>
> The fops is stored in the adapter structure to provide the ability to
> lookup the adapter structure from within the fop handler. CXL services
> use the file's private_data and at present, the CXL context does not
> have a private section. In an effort to limit areas of the cxlflash
> driver with code specific the superpipe function, a design choice was
> made to keep the details of the fops situated away from the legacy
> portions of the driver. This drove the behavior that the adapter fops
> is set at the beginning of the disk attach ioctl handler when there
> are no users present.
>
> The corruption that this fix remedies is due to the fact that the fops
> is initially defaulted to values found within a static structure. When
> the fops is handed down to the CXL services later in the attach path,
> certain services are patched. The fops structure remains correct until
> the user count drops to 0 and the fops is reset, triggering the process
> to repeat again. The user counts are tightly coupled with the creation
> and deletion of the user context. If multiple users perform a disk
> attach at the same time, when the user count is currently 0, some users
> can be in the middle of obtaining a file descriptor and have not yet
> reached the context creation code that [in addition to creating the
> context] increments the user count. Subsequent users coming in to
> perform the attach see that the user count is still 0, and reinitialize
> the fops, temporarily removing the patched fops. The users that are in
> the middle obtaining their file descriptor may then receive an invalid
> descriptor.
>
> The fix simply removes the user count altogether and moves the fops
> initialization to probe time such that it is only performed one time
> for the life of the adapter. In the future, if the CXL services adopt
> a private member for their context, that could be used to store the
> adapter structure reference and cxlflash could revert to a model that
> does not require an embedded fops.
>
> Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
> Signed-off-by: Manoj N. Kumar <manoj@linux.vnet.ibm.com>
> Reviewed-by: Brian King <brking@linux.vnet.ibm.com>
> Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Tomas Henzl <thenzl@redhat.com>
Tomas
next prev parent reply other threads:[~2015-10-23 14:00 UTC|newest]
Thread overview: 69+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-21 20:08 [PATCH v6 00/37] cxlflash: Miscellaneous bug fixes and corrections Matthew R. Ochs
2015-10-21 20:10 ` [PATCH v6 01/37] cxlflash: Fix to avoid invalid port_sel value Matthew R. Ochs
2015-10-21 20:11 ` [PATCH v6 02/37] cxlflash: Replace magic numbers with literals Matthew R. Ochs
2015-10-21 20:11 ` [PATCH v6 03/37] cxlflash: Fix read capacity timeout Matthew R. Ochs
2015-10-21 20:11 ` [PATCH v6 04/37] cxlflash: Fix potential oops following LUN removal Matthew R. Ochs
2015-10-21 20:11 ` [PATCH v6 05/37] cxlflash: Fix data corruption when vLUN used over multiple cards Matthew R. Ochs
2015-10-21 20:11 ` [PATCH v6 06/37] cxlflash: Fix to avoid sizeof(bool) Matthew R. Ochs
2015-10-21 20:11 ` [PATCH v6 07/37] cxlflash: Fix context encode mask width Matthew R. Ochs
2015-10-21 20:11 ` [PATCH v6 08/37] cxlflash: Fix to avoid CXL services during EEH Matthew R. Ochs
2015-10-21 20:12 ` [PATCH v6 09/37] cxlflash: Correct naming of limbo state and waitq Matthew R. Ochs
2015-10-21 20:12 ` [PATCH v6 10/37] cxlflash: Make functions static Matthew R. Ochs
2015-10-21 20:12 ` [PATCH v6 11/37] cxlflash: Refine host/device attributes Matthew R. Ochs
2015-10-23 13:33 ` Tomas Henzl
2015-10-21 20:13 ` [PATCH v6 12/37] cxlflash: Fix to avoid spamming the kernel log Matthew R. Ochs
2015-10-23 13:33 ` Tomas Henzl
2015-10-21 20:13 ` [PATCH v6 13/37] cxlflash: Fix to avoid stall while waiting on TMF Matthew R. Ochs
2015-10-23 13:36 ` Tomas Henzl
2015-10-21 20:13 ` [PATCH v6 14/37] cxlflash: Fix location of setting resid Matthew R. Ochs
2015-10-23 13:37 ` Tomas Henzl
2015-10-21 20:13 ` [PATCH v6 15/37] cxlflash: Fix host link up event handling Matthew R. Ochs
2015-10-23 13:38 ` Tomas Henzl
2015-10-21 20:13 ` [PATCH v6 16/37] cxlflash: Fix async interrupt bypass logic Matthew R. Ochs
2015-10-23 3:40 ` Andrew Donnellan
2015-10-23 13:39 ` Tomas Henzl
2015-10-21 20:13 ` [PATCH v6 17/37] cxlflash: Remove dual port online dependency Matthew R. Ochs
2015-10-23 13:39 ` Tomas Henzl
2015-10-21 20:14 ` [PATCH v6 18/37] cxlflash: Fix AFU version access/storage and add check Matthew R. Ochs
2015-10-23 13:40 ` Tomas Henzl
2015-10-21 20:14 ` [PATCH v6 19/37] cxlflash: Correct usage of scsi_host_put() Matthew R. Ochs
2015-10-23 13:41 ` Tomas Henzl
2015-10-21 20:14 ` [PATCH v6 20/37] cxlflash: Fix to prevent workq from accessing freed memory Matthew R. Ochs
2015-10-23 13:41 ` Tomas Henzl
2015-10-21 20:14 ` [PATCH v6 21/37] cxlflash: Correct behavior in device reset handler following EEH Matthew R. Ochs
2015-10-23 13:42 ` Tomas Henzl
2015-10-21 20:14 ` [PATCH v6 22/37] cxlflash: Remove unnecessary scsi_block_requests Matthew R. Ochs
2015-10-23 13:42 ` Tomas Henzl
2015-10-21 20:14 ` [PATCH v6 23/37] cxlflash: Fix function prolog parameters and return codes Matthew R. Ochs
2015-10-23 13:45 ` Tomas Henzl
2015-10-21 20:14 ` [PATCH v6 24/37] cxlflash: Fix MMIO and endianness errors Matthew R. Ochs
2015-10-23 13:53 ` Tomas Henzl
2015-10-21 20:14 ` [PATCH v6 25/37] cxlflash: Fix to prevent EEH recovery failure Matthew R. Ochs
2015-10-23 13:54 ` Tomas Henzl
2015-10-21 20:15 ` [PATCH v6 26/37] cxlflash: Correct spelling, grammar, and alignment mistakes Matthew R. Ochs
2015-10-23 13:54 ` Tomas Henzl
2015-10-21 20:15 ` [PATCH v6 27/37] cxlflash: Fix to prevent stale AFU RRQ Matthew R. Ochs
2015-10-23 13:55 ` Tomas Henzl
2015-10-21 20:15 ` [PATCH v6 28/37] MAINTAINERS: Add cxlflash driver Matthew R. Ochs
2015-10-21 20:15 ` [PATCH v6 29/37] cxlflash: Fix to double the delay each time Matthew R. Ochs
2015-10-23 13:57 ` Tomas Henzl
2015-10-21 20:15 ` [PATCH v6 30/37] cxlflash: Fix to avoid corrupting adapter fops Matthew R. Ochs
2015-10-23 14:00 ` Tomas Henzl [this message]
2015-10-21 20:15 ` [PATCH v6 31/37] cxlflash: Correct trace string Matthew R. Ochs
2015-10-23 14:00 ` Tomas Henzl
2015-10-21 20:15 ` [PATCH v6 32/37] cxlflash: Fix to avoid potential deadlock on EEH Matthew R. Ochs
2015-10-23 14:01 ` Tomas Henzl
2015-10-21 20:16 ` [PATCH v6 33/37] cxlflash: Fix to avoid leaving dangling interrupt resources Matthew R. Ochs
2015-10-23 14:01 ` Tomas Henzl
2015-10-21 20:16 ` [PATCH v6 34/37] cxlflash: Fix to escalate to LINK_RESET on login timeout Matthew R. Ochs
2015-10-23 14:01 ` Tomas Henzl
2015-10-21 20:16 ` [PATCH v6 35/37] cxlflash: Fix to avoid corrupting port selection mask Matthew R. Ochs
2015-10-22 17:17 ` Manoj Kumar
2015-10-23 3:52 ` Andrew Donnellan
2015-10-21 20:16 ` [PATCH v6 36/37] cxlflash: Fix to avoid lock instrumentation rejection Matthew R. Ochs
2015-10-22 17:34 ` Manoj Kumar
2015-10-23 3:22 ` Andrew Donnellan
2015-10-21 20:16 ` [PATCH v6 37/37] cxlflash: Fix to avoid bypassing context cleanup Matthew R. Ochs
2015-10-22 2:01 ` Andrew Donnellan
2015-10-22 18:05 ` Manoj Kumar
2015-10-27 23:30 ` [PATCH v6 00/37] cxlflash: Miscellaneous bug fixes and corrections Matthew R. Ochs
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=562A3D63.7020707@redhat.com \
--to=thenzl@redhat.com \
--cc=David.Laight@ACULAB.COM \
--cc=James.Bottomley@HansenPartnership.com \
--cc=andrew.donnellan@au1.ibm.com \
--cc=brking@linux.vnet.ibm.com \
--cc=dja@ozlabs.au.ibm.com \
--cc=imunsie@au1.ibm.com \
--cc=linux-scsi@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=manoj@linux.vnet.ibm.com \
--cc=mikey@neuling.org \
--cc=mrochs@linux.vnet.ibm.com \
--cc=nab@linux-iscsi.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).