From: "Ira W. Snyder" <iws@ovro•caltech.edu>
To: Dan Williams <dan.j.williams@intel•com>
Cc: "R58472@freescale•com" <R58472@freescale•com>,
"B04825@freescale•com" <B04825@freescale•com>,
"linuxppc-dev@ozlabs•org" <linuxppc-dev@ozlabs•org>,
"scottwood@freescale•com" <scottwood@freescale•com>,
"Dipen.Dudhat@freescale•com" <Dipen.Dudhat@freescale•com>,
"Maneesh.Gupta@freescale•com" <Maneesh.Gupta@freescale•com>,
"herbert@gondor•apana.org.au" <herbert@gondor•apana.org.au>
Subject: Re: [PATCH 8/8] fsldma: major cleanups and fixes
Date: Tue, 2 Feb 2010 13:16:57 -0800 [thread overview]
Message-ID: <20100202211656.GA2609@ovro.caltech.edu> (raw)
In-Reply-To: <4B6892F4.9070906@intel.com>
On Tue, Feb 02, 2010 at 02:02:44PM -0700, Dan Williams wrote:
> Ira W. Snyder wrote:
> > Fix locking. Use two queues in the driver, one for pending transacions, and
> > one for transactions which are actually running on the hardware. Call
> > dma_run_dependencies() on descriptor cleanup so that the async_tx API works
> > correctly.
>
> I notice that fsldma diverges from other dma drivers in that the
> callback is performed with interrupts disabled. MD/raid5 currently
> assumes that interrupts are enabled in its callback routines (see
> ops_complete_biofill()'s use of spin_lock_irq()). On top of these
> changes can we align fsldma to the other raid offload drivers (mv_xor,
> iop-adma, ioatdma) and provide callbacks with irq's enabled?
>
> I'll proceed with applying these patches as they obviously improve
> things, but you will hit the irq problem when performing reads to a
> degraded array.
>
In the fsldma driver, all callbacks are run from tasklet (softirq)
context. That's under local_irq_disable(), right? Hardirq's certainly
aren't disabled there.
Is a DMAEngine user expected to call the device_is_tx_complete()
function until it returns that the DMA has completed?
If so, it is pretty easy to switch to a workqueue instead of a tasklet
to handle the callbacks. The cost is increased latency until the
callbacks are processed.
Would you want a driver-wide singlethreaded workqueue? A driver-wide
multi-threaded workqueue? A workqueue per-device? A workqueue
per-channel? This starts to get excessive, IMO.
Ira
next prev parent reply other threads:[~2010-02-02 21:17 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-06 23:33 [PATCH 0/8 v2] fsldma: cleanup driver and fix async_tx compatibility Ira W. Snyder
2010-01-06 23:33 ` [PATCH 1/8] fsldma: reduce kernel text size Ira W. Snyder
2010-01-06 23:34 ` [PATCH 2/8] fsldma: remove unused structure members Ira W. Snyder
2010-01-06 23:34 ` [PATCH 3/8] fsldma: rename struct fsl_dma_chan to struct fsldma_chan Ira W. Snyder
2010-01-06 23:34 ` [PATCH 4/8] fsldma: rename dest to dst for uniformity Ira W. Snyder
2010-01-06 23:34 ` [PATCH 5/8] fsldma: clean up the OF subsystem routines Ira W. Snyder
2010-01-06 23:34 ` [PATCH 6/8] fsldma: simplify IRQ probing and handling Ira W. Snyder
2010-01-06 23:34 ` [PATCH 7/8] fsldma: rename fsl_chan to chan Ira W. Snyder
2010-01-06 23:34 ` [PATCH 8/8] fsldma: major cleanups and fixes Ira W. Snyder
2010-02-02 21:02 ` Dan Williams
2010-02-02 21:16 ` Ira W. Snyder [this message]
2010-02-02 21:23 ` Dan Williams
2010-02-02 21:36 ` Ira W. Snyder
-- strict thread matches above, loose matches on Subject: below --
2010-01-01 6:10 fsldma: cleanup driver and fix async_tx compatibility Ira W. Snyder
2010-01-01 6:10 ` [PATCH 8/8] fsldma: major cleanups and fixes Ira W. Snyder
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=20100202211656.GA2609@ovro.caltech.edu \
--to=iws@ovro$(echo .)caltech.edu \
--cc=B04825@freescale$(echo .)com \
--cc=Dipen.Dudhat@freescale$(echo .)com \
--cc=Maneesh.Gupta@freescale$(echo .)com \
--cc=R58472@freescale$(echo .)com \
--cc=dan.j.williams@intel$(echo .)com \
--cc=herbert@gondor$(echo .)apana.org.au \
--cc=linuxppc-dev@ozlabs$(echo .)org \
--cc=scottwood@freescale$(echo .)com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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