From: prylowski@metasoft.pl (Rafal Prylowski)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH] ep93xx: Implement double buffering for M2M DMA channels
Date: Thu, 22 Mar 2012 11:00:37 +0100 [thread overview]
Message-ID: <4F6AF845.9070209@metasoft.pl> (raw)
In-Reply-To: <4F6A78ED.3050403@gmail.com>
On 2012-03-22 01:57, Ryan Mallon wrote:
> I haven't looked through the patch yet, since I'm waiting on more
> information from Mika and Hartley's testing.
>
> However, the commit log doesn't tell me why we want this change. Is it a
> performance improvement? If so, do you have some numbers that we can
> paste into the commit log?
In principle, using double buffering should be faster than using only
one buffer and disabling/enabling channel each time. But my measurements
doesn't show any significant change.
The real reason for this change is that current code is not 100% reliable
in IDE-DMA (I'm planning to submit ep93xx ide driver to linux-ide soon).
Although using only one buffer and DONE interrupt is simple, surprisingly
we can get interrupt when M2M DMA is in DMA_BUF_ON and DMA_STALL state.
If we disable the channel at this moment, we end with ata timeout.
WARNING: multiple messages have this Message-ID (diff)
From: Rafal Prylowski <prylowski@metasoft.pl>
To: Ryan Mallon <rmallon@gmail.com>
Cc: "linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"vinod.koul@intel.com" <vinod.koul@intel.com>,
Mika Westerberg <mika.westerberg@iki.fi>,
H Hartley Sweeten <hartleys@visionengravers.com>
Subject: Re: [PATCH] ep93xx: Implement double buffering for M2M DMA channels
Date: Thu, 22 Mar 2012 11:00:37 +0100 [thread overview]
Message-ID: <4F6AF845.9070209@metasoft.pl> (raw)
In-Reply-To: <4F6A78ED.3050403@gmail.com>
On 2012-03-22 01:57, Ryan Mallon wrote:
> I haven't looked through the patch yet, since I'm waiting on more
> information from Mika and Hartley's testing.
>
> However, the commit log doesn't tell me why we want this change. Is it a
> performance improvement? If so, do you have some numbers that we can
> paste into the commit log?
In principle, using double buffering should be faster than using only
one buffer and disabling/enabling channel each time. But my measurements
doesn't show any significant change.
The real reason for this change is that current code is not 100% reliable
in IDE-DMA (I'm planning to submit ep93xx ide driver to linux-ide soon).
Although using only one buffer and DONE interrupt is simple, surprisingly
we can get interrupt when M2M DMA is in DMA_BUF_ON and DMA_STALL state.
If we disable the channel at this moment, we end with ata timeout.
next prev parent reply other threads:[~2012-03-22 10:00 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-20 8:09 [PATCH] ep93xx: Implement double buffering for M2M DMA channels Rafal Prylowski
2012-03-20 8:09 ` Rafal Prylowski
2012-03-21 7:07 ` Mika Westerberg
2012-03-21 7:07 ` Mika Westerberg
2012-03-21 7:47 ` Rafal Prylowski
2012-03-21 7:47 ` Rafal Prylowski
2012-03-21 19:33 ` Mika Westerberg
2012-03-21 19:33 ` Mika Westerberg
2012-03-21 17:12 ` H Hartley Sweeten
2012-03-21 17:12 ` H Hartley Sweeten
2012-03-21 19:32 ` Mika Westerberg
2012-03-21 19:32 ` Mika Westerberg
2012-03-22 0:47 ` H Hartley Sweeten
2012-03-22 0:47 ` H Hartley Sweeten
2012-03-22 7:37 ` Mika Westerberg
2012-03-22 7:37 ` Mika Westerberg
2012-03-22 18:52 ` H Hartley Sweeten
2012-03-22 18:52 ` H Hartley Sweeten
2012-03-22 20:03 ` Mika Westerberg
2012-03-22 20:03 ` Mika Westerberg
2012-03-22 21:36 ` H Hartley Sweeten
2012-03-22 21:36 ` H Hartley Sweeten
2012-03-22 23:56 ` H Hartley Sweeten
2012-03-22 23:56 ` H Hartley Sweeten
2012-03-23 7:00 ` Mika Westerberg
2012-03-23 7:00 ` Mika Westerberg
2012-03-22 10:16 ` Rafal Prylowski
2012-03-22 10:16 ` Rafal Prylowski
2012-03-21 19:38 ` Mika Westerberg
2012-03-21 19:38 ` Mika Westerberg
2012-03-23 2:19 ` H Hartley Sweeten
2012-03-23 2:19 ` H Hartley Sweeten
2012-03-23 7:04 ` Mika Westerberg
2012-03-23 7:04 ` Mika Westerberg
2012-03-23 16:09 ` H Hartley Sweeten
2012-03-23 16:09 ` H Hartley Sweeten
2012-03-24 7:32 ` Mika Westerberg
2012-03-24 7:32 ` Mika Westerberg
2012-03-26 6:44 ` Rafal Prylowski
2012-03-26 6:44 ` Rafal Prylowski
2012-03-29 22:33 ` H Hartley Sweeten
2012-03-29 22:33 ` H Hartley Sweeten
2012-04-01 18:49 ` Mika Westerberg
2012-04-01 18:49 ` Mika Westerberg
2012-04-10 17:28 ` Mika Westerberg
2012-04-10 17:28 ` Mika Westerberg
2012-04-10 17:55 ` H Hartley Sweeten
2012-04-10 17:55 ` H Hartley Sweeten
2012-04-11 7:18 ` Rafal Prylowski
2012-04-11 7:18 ` Rafal Prylowski
2012-04-16 18:59 ` H Hartley Sweeten
2012-04-16 18:59 ` H Hartley Sweeten
2012-04-17 7:15 ` Rafal Prylowski
2012-04-17 7:15 ` Rafal Prylowski
2012-04-17 15:46 ` H Hartley Sweeten
2012-04-17 15:46 ` H Hartley Sweeten
2012-04-17 20:51 ` H Hartley Sweeten
2012-04-17 20:51 ` H Hartley Sweeten
2012-04-18 16:41 ` Rafal Prylowski
2012-04-18 16:41 ` Rafal Prylowski
2012-04-18 17:01 ` H Hartley Sweeten
2012-04-18 17:01 ` H Hartley Sweeten
2012-03-22 0:57 ` Ryan Mallon
2012-03-22 0:57 ` Ryan Mallon
2012-03-22 10:00 ` Rafal Prylowski [this message]
2012-03-22 10:00 ` Rafal Prylowski
2012-03-22 13:14 ` Sergei Shtylyov
2012-03-22 13:14 ` Sergei Shtylyov
2012-03-22 14:13 ` Rafal Prylowski
2012-03-22 14:13 ` Rafal Prylowski
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=4F6AF845.9070209@metasoft.pl \
--to=prylowski@metasoft.pl \
--cc=linux-arm-kernel@lists.infradead.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.