From: Joel Fernandes <joelf-l0cyMroinI0@public.gmane.org>
To: Peter Ujfalusi <peter.ujfalusi-l0cyMroinI0@public.gmane.org>,
dan.j.williams-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
vinod.koul-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
nsekhar-l0cyMroinI0@public.gmane.org
Cc: davinci-linux-open-source-VycZQUHpC/PFrsHnngEfi1aTQe2KTcn/@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
mporter-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org,
dmaengine-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org
Subject: Re: [PATCH v2 08/14] DMA: edma: Use different eventq for cyclic channels
Date: Thu, 10 Apr 2014 11:36:30 -0500 [thread overview]
Message-ID: <5346C88E.8010504@ti.com> (raw)
In-Reply-To: <1396357575-30585-9-git-send-email-peter.ujfalusi-l0cyMroinI0@public.gmane.org>
On 04/01/2014 08:06 AM, Peter Ujfalusi wrote:
> To improve latency with cyclic DMA operation it is preferred to
> use different eventq/tc than the default which is used by all
> other drivers (mmc, spi, i2c, etc).
> When preparing the cyclic dma ask for non default queue for the
> channel which is going to be used with cyclic mode.
>
> Signed-off-by: Peter Ujfalusi <peter.ujfalusi-l0cyMroinI0@public.gmane.org>
> ---
> drivers/dma/edma.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
> index 1dd9e8806975..10048b40fac8 100644
> --- a/drivers/dma/edma.c
> +++ b/drivers/dma/edma.c
> @@ -628,6 +628,9 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
> edesc->pset[i].opt |= TCINTEN;
> }
>
> + /* Use different eventq/tc for cyclic DMA to reduce latency */
> + edma_request_non_default_queue(echan->ch_num);
> +
> return vchan_tx_prep(&echan->vchan, &edesc->vdesc, tx_flags);
> }
>
>
Is there any way to guarantee that the non-default queue is of the
highest priority, or in other words default queue is of lowest priority.
I know you set queue 1 as default because by default 0 is higher
priority. And then assigning non-default queue.
When assigning default to Queue 1, it would be good to also call
assign_priority_to_queue and set QUEPRI to 7 for Queue 1. Since 0, 2 and
4 are all non-defaults.
Thanks,
-Joel
WARNING: multiple messages have this Message-ID (diff)
From: joelf@ti.com (Joel Fernandes)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v2 08/14] DMA: edma: Use different eventq for cyclic channels
Date: Thu, 10 Apr 2014 11:36:30 -0500 [thread overview]
Message-ID: <5346C88E.8010504@ti.com> (raw)
In-Reply-To: <1396357575-30585-9-git-send-email-peter.ujfalusi@ti.com>
On 04/01/2014 08:06 AM, Peter Ujfalusi wrote:
> To improve latency with cyclic DMA operation it is preferred to
> use different eventq/tc than the default which is used by all
> other drivers (mmc, spi, i2c, etc).
> When preparing the cyclic dma ask for non default queue for the
> channel which is going to be used with cyclic mode.
>
> Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
> ---
> drivers/dma/edma.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
> index 1dd9e8806975..10048b40fac8 100644
> --- a/drivers/dma/edma.c
> +++ b/drivers/dma/edma.c
> @@ -628,6 +628,9 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
> edesc->pset[i].opt |= TCINTEN;
> }
>
> + /* Use different eventq/tc for cyclic DMA to reduce latency */
> + edma_request_non_default_queue(echan->ch_num);
> +
> return vchan_tx_prep(&echan->vchan, &edesc->vdesc, tx_flags);
> }
>
>
Is there any way to guarantee that the non-default queue is of the
highest priority, or in other words default queue is of lowest priority.
I know you set queue 1 as default because by default 0 is higher
priority. And then assigning non-default queue.
When assigning default to Queue 1, it would be good to also call
assign_priority_to_queue and set QUEPRI to 7 for Queue 1. Since 0, 2 and
4 are all non-defaults.
Thanks,
-Joel
WARNING: multiple messages have this Message-ID (diff)
From: Joel Fernandes <joelf@ti.com>
To: Peter Ujfalusi <peter.ujfalusi@ti.com>,
<dan.j.williams@intel.com>, <vinod.koul@intel.com>,
<nsekhar@ti.com>
Cc: <dmaengine@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
<linux-arm-kernel@lists.infradead.org>,
<linux-omap@vger.kernel.org>,
<davinci-linux-open-source@linux.davincidsp.com>,
<mporter@linaro.org>
Subject: Re: [PATCH v2 08/14] DMA: edma: Use different eventq for cyclic channels
Date: Thu, 10 Apr 2014 11:36:30 -0500 [thread overview]
Message-ID: <5346C88E.8010504@ti.com> (raw)
In-Reply-To: <1396357575-30585-9-git-send-email-peter.ujfalusi@ti.com>
On 04/01/2014 08:06 AM, Peter Ujfalusi wrote:
> To improve latency with cyclic DMA operation it is preferred to
> use different eventq/tc than the default which is used by all
> other drivers (mmc, spi, i2c, etc).
> When preparing the cyclic dma ask for non default queue for the
> channel which is going to be used with cyclic mode.
>
> Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
> ---
> drivers/dma/edma.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
> index 1dd9e8806975..10048b40fac8 100644
> --- a/drivers/dma/edma.c
> +++ b/drivers/dma/edma.c
> @@ -628,6 +628,9 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
> edesc->pset[i].opt |= TCINTEN;
> }
>
> + /* Use different eventq/tc for cyclic DMA to reduce latency */
> + edma_request_non_default_queue(echan->ch_num);
> +
> return vchan_tx_prep(&echan->vchan, &edesc->vdesc, tx_flags);
> }
>
>
Is there any way to guarantee that the non-default queue is of the
highest priority, or in other words default queue is of lowest priority.
I know you set queue 1 as default because by default 0 is higher
priority. And then assigning non-default queue.
When assigning default to Queue 1, it would be good to also call
assign_priority_to_queue and set QUEPRI to 7 for Queue 1. Since 0, 2 and
4 are all non-defaults.
Thanks,
-Joel
next prev parent reply other threads:[~2014-04-10 16:36 UTC|newest]
Thread overview: 132+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-01 13:06 [PATCH v2 00/14] dma: edma: Fixes for cyclic (audio) operation Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 02/14] dma: edma: Correct the handling of src/dst_maxburst == 0 Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 04/14] dma: edma: Set DMA_CYCLIC capability flag Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 05/14] arm: common: edma: Select event queue 1 as default when booted with DT Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
[not found] ` <1396357575-30585-6-git-send-email-peter.ujfalusi-l0cyMroinI0@public.gmane.org>
2014-04-10 16:23 ` Joel Fernandes
2014-04-10 16:23 ` Joel Fernandes
2014-04-10 16:23 ` Joel Fernandes
2014-04-11 8:17 ` Sekhar Nori
2014-04-11 8:17 ` Sekhar Nori
2014-04-11 8:17 ` Sekhar Nori
[not found] ` <5347A4FD.1030803-l0cyMroinI0@public.gmane.org>
2014-04-11 8:50 ` Peter Ujfalusi
2014-04-11 8:50 ` Peter Ujfalusi
2014-04-11 8:50 ` Peter Ujfalusi
[not found] ` <5347ACDE.7040407-l0cyMroinI0@public.gmane.org>
2014-04-11 8:56 ` Sekhar Nori
2014-04-11 8:56 ` Sekhar Nori
2014-04-11 8:56 ` Sekhar Nori
[not found] ` <5347AE49.5020109-l0cyMroinI0@public.gmane.org>
2014-04-11 9:38 ` Peter Ujfalusi
2014-04-11 9:38 ` Peter Ujfalusi
2014-04-11 9:38 ` Peter Ujfalusi
2014-04-11 9:42 ` Vinod Koul
2014-04-11 9:42 ` Vinod Koul
2014-04-11 9:42 ` Vinod Koul
[not found] ` <20140411094217.GA32284-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-04-11 10:19 ` Sekhar Nori
2014-04-11 10:19 ` Sekhar Nori
2014-04-11 10:19 ` Sekhar Nori
2014-04-11 11:32 ` Peter Ujfalusi
2014-04-11 11:32 ` Peter Ujfalusi
2014-04-11 11:32 ` Peter Ujfalusi
2014-04-11 11:31 ` Vinod Koul
2014-04-11 11:31 ` Vinod Koul
[not found] ` <20140411113154.GB32284-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-04-11 12:23 ` Peter Ujfalusi
2014-04-11 12:23 ` Peter Ujfalusi
2014-04-11 12:23 ` Peter Ujfalusi
2014-04-11 12:46 ` Vinod Koul
2014-04-11 12:46 ` Vinod Koul
2014-04-11 12:46 ` Vinod Koul
[not found] ` <20140411124641.GC32284-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-04-14 11:56 ` Peter Ujfalusi
2014-04-14 11:56 ` Peter Ujfalusi
2014-04-14 11:56 ` Peter Ujfalusi
[not found] ` <534BCCD3.9060805-l0cyMroinI0@public.gmane.org>
2014-04-14 12:12 ` Sekhar Nori
2014-04-14 12:12 ` Sekhar Nori
2014-04-14 12:12 ` Sekhar Nori
[not found] ` <534BD0B5.5000004-l0cyMroinI0@public.gmane.org>
2014-04-14 12:41 ` Peter Ujfalusi
2014-04-14 12:41 ` Peter Ujfalusi
2014-04-14 12:41 ` Peter Ujfalusi
[not found] ` <534BD788.3050406-l0cyMroinI0@public.gmane.org>
2014-04-14 14:32 ` Sekhar Nori
2014-04-14 14:32 ` Sekhar Nori
2014-04-14 14:32 ` Sekhar Nori
[not found] ` <534BF181.6060503-l0cyMroinI0@public.gmane.org>
2014-04-16 12:59 ` Peter Ujfalusi
2014-04-16 12:59 ` Peter Ujfalusi
2014-04-16 12:59 ` Peter Ujfalusi
[not found] ` <534E7EB0.9000601-l0cyMroinI0@public.gmane.org>
2014-04-16 16:05 ` Joel Fernandes
2014-04-16 16:05 ` Joel Fernandes
2014-04-16 16:05 ` Joel Fernandes
[not found] ` <534EAA49.7030702-l0cyMroinI0@public.gmane.org>
2014-04-24 9:07 ` Peter Ujfalusi
2014-04-24 9:07 ` Peter Ujfalusi
2014-04-24 9:07 ` Peter Ujfalusi
2014-04-11 20:16 ` Joel Fernandes
2014-04-11 20:16 ` Joel Fernandes
2014-04-11 20:16 ` Joel Fernandes
2014-04-01 13:06 ` [PATCH v2 11/14] dma: edma: Reduce debug print verbosity for non verbose debugging Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 12/14] dma: edma: Prefix debug prints where the text were identical in prep callbacks Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 13/14] dma: edma: Add channel number to debug prints Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 14/14] dma: edma: Print the direction value as well when it is not supported Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
[not found] ` <1396357575-30585-1-git-send-email-peter.ujfalusi-l0cyMroinI0@public.gmane.org>
2014-04-01 13:06 ` [PATCH v2 01/14] platform_data: edma: Be precise with the paRAM struct Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 03/14] dma: edma: Add support for DMA_PAUSE/RESUME operation Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-11 16:43 ` Vinod Koul
2014-04-11 16:43 ` Vinod Koul
[not found] ` <20140411164327.GD32284-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-04-11 20:51 ` Joel Fernandes
2014-04-11 20:51 ` Joel Fernandes
2014-04-11 20:51 ` Joel Fernandes
2014-04-01 13:06 ` [PATCH v2 06/14] arm: common: edma: Save the number of event queues/TCs Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 07/14] arm: common: edma: API to request non default queue for a channel Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-11 8:43 ` Sekhar Nori
2014-04-11 8:43 ` Sekhar Nori
2014-04-11 8:43 ` Sekhar Nori
2014-04-01 13:06 ` [PATCH v2 08/14] DMA: edma: Use different eventq for cyclic channels Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
[not found] ` <1396357575-30585-9-git-send-email-peter.ujfalusi-l0cyMroinI0@public.gmane.org>
2014-04-10 16:36 ` Joel Fernandes [this message]
2014-04-10 16:36 ` Joel Fernandes
2014-04-10 16:36 ` Joel Fernandes
[not found] ` <5346C88E.8010504-l0cyMroinI0@public.gmane.org>
2014-04-11 16:47 ` Vinod Koul
2014-04-11 16:47 ` Vinod Koul
2014-04-11 16:47 ` Vinod Koul
[not found] ` <20140411164755.GE32284-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2014-04-11 20:56 ` Joel Fernandes
2014-04-11 20:56 ` Joel Fernandes
2014-04-11 20:56 ` Joel Fernandes
2014-04-01 13:06 ` [PATCH v2 09/14] dma: edma: Implement device_slave_caps callback Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` [PATCH v2 10/14] dma: edma: Simplify direction configuration in edma_config_pset() Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-01 13:06 ` Peter Ujfalusi
2014-04-10 22:40 ` Joel Fernandes
2014-04-10 22:40 ` Joel Fernandes
2014-04-10 22:40 ` Joel Fernandes
[not found] ` <53471DC5.4090001-l0cyMroinI0@public.gmane.org>
2014-04-11 6:39 ` Peter Ujfalusi
2014-04-11 6:39 ` Peter Ujfalusi
2014-04-11 6:39 ` Peter Ujfalusi
[not found] ` <53478E0E.8040709-l0cyMroinI0@public.gmane.org>
2014-04-11 19:57 ` Joel Fernandes
2014-04-11 19:57 ` Joel Fernandes
2014-04-11 19:57 ` Joel Fernandes
2014-04-10 22:52 ` [PATCH v2 00/14] dma: edma: Fixes for cyclic (audio) operation Joel Fernandes
2014-04-10 22:52 ` Joel Fernandes
2014-04-10 22:52 ` Joel Fernandes
2014-04-11 16:52 ` Vinod Koul
2014-04-11 16:52 ` Vinod Koul
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=5346C88E.8010504@ti.com \
--to=joelf-l0cymroini0@public.gmane.org \
--cc=dan.j.williams-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
--cc=davinci-linux-open-source-VycZQUHpC/PFrsHnngEfi1aTQe2KTcn/@public.gmane.org \
--cc=dmaengine-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=mporter-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org \
--cc=nsekhar-l0cyMroinI0@public.gmane.org \
--cc=peter.ujfalusi-l0cyMroinI0@public.gmane.org \
--cc=vinod.koul-ral2JQCrhuEAvxtiuMwx3w@public.gmane.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.