From: Sakari Ailus <sakari.ailus@iki.fi>
To: Todor Tomov <todor.tomov@linaro.org>
Cc: mchehab@kernel.org, hans.verkuil@cisco.com,
javier@osg.samsung.com, s.nawrocki@samsung.com,
linux-media@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-arm-msm@vger.kernel.org
Subject: Re: [PATCH v3 22/23] camss: Use optimal clock frequency rates
Date: Thu, 20 Jul 2017 18:23:13 +0300 [thread overview]
Message-ID: <20170720152313.joqfo7dqg5gbxrfl@valkosipuli.retiisi.org.uk> (raw)
In-Reply-To: <1500287629-23703-23-git-send-email-todor.tomov@linaro.org>
Hi Todor,
On Mon, Jul 17, 2017 at 01:33:48PM +0300, Todor Tomov wrote:
> Use standard V4L2 control to get pixel clock rate from a sensor
> linked in the media controller pipeline. Then calculate clock
> rates on CSIPHY, CSID and VFE to use the lowest possible.
>
> If the currnet pixel clock rate of the sensor cannot be read then
> use the highest possible. This case covers also the CSID test
> generator usage.
>
> If VFE is already powered on by another pipeline, check that the
> current VFE clock rate is high enough for the new pipeline.
> If not return busy error code as VFE clock rate cannot be changed
> while VFE is running.
>
> Signed-off-by: Todor Tomov <todor.tomov@linaro.org>
> ---
> .../media/platform/qcom/camss-8x16/camss-csid.c | 324 ++++++++++++++-------
> .../media/platform/qcom/camss-8x16/camss-csid.h | 2 +-
> .../media/platform/qcom/camss-8x16/camss-csiphy.c | 112 +++++--
> .../media/platform/qcom/camss-8x16/camss-csiphy.h | 2 +-
> .../media/platform/qcom/camss-8x16/camss-ispif.c | 23 +-
> .../media/platform/qcom/camss-8x16/camss-ispif.h | 4 +-
> drivers/media/platform/qcom/camss-8x16/camss-vfe.c | 289 +++++++++++++++---
> drivers/media/platform/qcom/camss-8x16/camss-vfe.h | 2 +-
> drivers/media/platform/qcom/camss-8x16/camss.c | 51 +++-
> drivers/media/platform/qcom/camss-8x16/camss.h | 17 +-
> 10 files changed, 634 insertions(+), 192 deletions(-)
>
> diff --git a/drivers/media/platform/qcom/camss-8x16/camss-csid.c b/drivers/media/platform/qcom/camss-8x16/camss-csid.c
> index 2bf3415..5c0e359 100644
> --- a/drivers/media/platform/qcom/camss-8x16/camss-csid.c
> +++ b/drivers/media/platform/qcom/camss-8x16/camss-csid.c
> @@ -68,122 +68,224 @@ static const struct {
> u8 data_type;
> u8 decode_format;
> u8 uncompr_bpp;
> + u8 spp; /* bus samples per pixel */
> } csid_input_fmts[] = {
> {
> MEDIA_BUS_FMT_UYVY8_2X8,
> MEDIA_BUS_FMT_UYVY8_2X8,
> DATA_TYPE_YUV422_8BIT,
> DECODE_FORMAT_UNCOMPRESSED_8_BIT,
> - 16
> + 8,
> + 2
If your original patch had added the comma on the previous line, this would
be simply adding a line per array entry.
> },
> {
> MEDIA_BUS_FMT_VYUY8_2X8,
> MEDIA_BUS_FMT_VYUY8_2X8,
> DATA_TYPE_YUV422_8BIT,
> DECODE_FORMAT_UNCOMPRESSED_8_BIT,
> - 16
> + 8,
> + 2
> },
> {
> MEDIA_BUS_FMT_YUYV8_2X8,
> MEDIA_BUS_FMT_YUYV8_2X8,
> DATA_TYPE_YUV422_8BIT,
> DECODE_FORMAT_UNCOMPRESSED_8_BIT,
> - 16
> + 8,
> + 2
> },
> {
> MEDIA_BUS_FMT_YVYU8_2X8,
> MEDIA_BUS_FMT_YVYU8_2X8,
> DATA_TYPE_YUV422_8BIT,
> DECODE_FORMAT_UNCOMPRESSED_8_BIT,
> - 16
> + 8,
> + 2
> },
> {
> MEDIA_BUS_FMT_SBGGR8_1X8,
> MEDIA_BUS_FMT_SBGGR8_1X8,
> DATA_TYPE_RAW_8BIT,
> DECODE_FORMAT_UNCOMPRESSED_8_BIT,
> - 8
> + 8,
> + 1
> },
> {
> MEDIA_BUS_FMT_SGBRG8_1X8,
> MEDIA_BUS_FMT_SGBRG8_1X8,
> DATA_TYPE_RAW_8BIT,
> DECODE_FORMAT_UNCOMPRESSED_8_BIT,
> - 8
> + 8,
> + 1
> },
> {
> MEDIA_BUS_FMT_SGRBG8_1X8,
> MEDIA_BUS_FMT_SGRBG8_1X8,
> DATA_TYPE_RAW_8BIT,
> DECODE_FORMAT_UNCOMPRESSED_8_BIT,
> - 8
> + 8,
> + 1
> },
> {
> MEDIA_BUS_FMT_SRGGB8_1X8,
> MEDIA_BUS_FMT_SRGGB8_1X8,
> DATA_TYPE_RAW_8BIT,
> DECODE_FORMAT_UNCOMPRESSED_8_BIT,
> - 8
> + 8,
> + 1
> },
> {
> MEDIA_BUS_FMT_SBGGR10_1X10,
> MEDIA_BUS_FMT_SBGGR10_1X10,
> DATA_TYPE_RAW_10BIT,
> DECODE_FORMAT_UNCOMPRESSED_10_BIT,
> - 10
> + 10,
> + 1
> },
> {
> MEDIA_BUS_FMT_SGBRG10_1X10,
> MEDIA_BUS_FMT_SGBRG10_1X10,
> DATA_TYPE_RAW_10BIT,
> DECODE_FORMAT_UNCOMPRESSED_10_BIT,
> - 10
> + 10,
> + 1
> },
> {
> MEDIA_BUS_FMT_SGRBG10_1X10,
> MEDIA_BUS_FMT_SGRBG10_1X10,
> DATA_TYPE_RAW_10BIT,
> DECODE_FORMAT_UNCOMPRESSED_10_BIT,
> - 10
> + 10,
> + 1
> },
> {
> MEDIA_BUS_FMT_SRGGB10_1X10,
> MEDIA_BUS_FMT_SRGGB10_1X10,
> DATA_TYPE_RAW_10BIT,
> DECODE_FORMAT_UNCOMPRESSED_10_BIT,
> - 10
> + 10,
> + 1
> },
> {
> MEDIA_BUS_FMT_SBGGR12_1X12,
> MEDIA_BUS_FMT_SBGGR12_1X12,
> DATA_TYPE_RAW_12BIT,
> DECODE_FORMAT_UNCOMPRESSED_12_BIT,
> - 12
> + 12,
> + 1
> },
> {
> MEDIA_BUS_FMT_SGBRG12_1X12,
> MEDIA_BUS_FMT_SGBRG12_1X12,
> DATA_TYPE_RAW_12BIT,
> DECODE_FORMAT_UNCOMPRESSED_12_BIT,
> - 12
> + 12,
> + 1
> },
> {
> MEDIA_BUS_FMT_SGRBG12_1X12,
> MEDIA_BUS_FMT_SGRBG12_1X12,
> DATA_TYPE_RAW_12BIT,
> DECODE_FORMAT_UNCOMPRESSED_12_BIT,
> - 12
> + 12,
> + 1
> },
> {
> MEDIA_BUS_FMT_SRGGB12_1X12,
> MEDIA_BUS_FMT_SRGGB12_1X12,
> DATA_TYPE_RAW_12BIT,
> DECODE_FORMAT_UNCOMPRESSED_12_BIT,
> - 12
> + 12,
> + 1
> }
> };
>
> /*
> + * csid_get_uncompressed - map media bus format to uncompressed media bus format
> + * @code: media bus format code
> + *
> + * Return uncompressed media bus format code
> + */
> +static u32 csid_get_uncompressed(u32 code)
> +{
> + unsigned int i;
> +
> + for (i = 0; i < ARRAY_SIZE(csid_input_fmts); i++)
> + if (code == csid_input_fmts[i].code)
> + break;
> +
> + return csid_input_fmts[i].uncompressed;
> +}
I think my suggestion of returning an array entry (a struct) was a good
one. :-)
> +
> +/*
> + * csid_get_data_type - map media bus format to data type
> + * @code: media bus format code
> + *
> + * Return data type code
> + */
> +static u8 csid_get_data_type(u32 code)
> +{
> + unsigned int i;
> +
> + for (i = 0; i < ARRAY_SIZE(csid_input_fmts); i++)
> + if (code == csid_input_fmts[i].code)
> + break;
> +
> + return csid_input_fmts[i].data_type;
> +}
> +
> +/*
> + * csid_get_decode_format - map media bus format to decode format
> + * @code: media bus format code
> + *
> + * Return decode format code
> + */
> +static u8 csid_get_decode_format(u32 code)
> +{
> + unsigned int i;
> +
> + for (i = 0; i < ARRAY_SIZE(csid_input_fmts); i++)
> + if (code == csid_input_fmts[i].code)
> + break;
> +
> + return csid_input_fmts[i].decode_format;
> +}
> +
> +/*
> + * csid_get_bpp - map media bus format to bits per pixel
> + * @code: media bus format code
> + *
> + * Return number of bits per pixel
> + */
> +static u8 csid_get_bpp(u32 code)
> +{
> + unsigned int i;
> +
> + for (i = 0; i < ARRAY_SIZE(csid_input_fmts); i++)
> + if (code == csid_input_fmts[i].uncompressed)
> + break;
> +
> + return csid_input_fmts[i].uncompr_bpp;
> +}
> +
> +/*
> + * csid_get_spp - map media bus format to bus samples per pixel
> + * @code: media bus format code
> + *
> + * Return number of bus samples per pixel
> + */
> +static u8 csid_get_spp(u32 code)
> +{
> + unsigned int i;
> +
> + for (i = 0; i < ARRAY_SIZE(csid_input_fmts); i++)
> + if (code == csid_input_fmts[i].uncompressed)
> + break;
> +
> + return csid_input_fmts[i].spp;
> +}
> +
> +/*
> * csid_isr - CSID module interrupt handler
> * @irq: Interrupt line
> * @dev: CSID device
--
Regards,
Sakari Ailus
e-mail: sakari.ailus@iki.fi XMPP: sailus@retiisi.org.uk
next prev parent reply other threads:[~2017-07-20 15:23 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-17 10:33 [PATCH v3 00/23] Qualcomm 8x16 Camera Subsystem driver Todor Tomov
2017-07-17 10:33 ` [PATCH v3 01/23] [media] media: Make parameter of media_entity_remote_pad() const Todor Tomov
2017-07-17 10:33 ` [PATCH v3 02/23] [media] v4l2-mediabus: Add helper functions Todor Tomov
2017-07-17 10:33 ` [PATCH v3 03/23] v4l: Add packed Bayer raw12 pixel formats Todor Tomov
2017-07-17 10:33 ` [PATCH v3 04/23] dt-bindings: media: Binding document for Qualcomm Camera subsystem driver Todor Tomov
2017-07-20 10:13 ` Sakari Ailus
2017-08-04 11:54 ` Todor Tomov
2017-08-05 8:19 ` Sakari Ailus
2017-07-20 10:15 ` Sakari Ailus
2017-07-24 16:43 ` Rob Herring
2017-07-17 10:33 ` [PATCH v3 05/23] MAINTAINERS: Add " Todor Tomov
2017-07-17 10:33 ` [PATCH v3 06/23] doc: media/v4l-drivers: Add Qualcomm Camera Subsystem driver document Todor Tomov
2017-07-17 10:33 ` [PATCH v3 07/23] media: camss: Add CSIPHY files Todor Tomov
2017-07-20 12:15 ` Sakari Ailus
2017-07-17 10:33 ` [PATCH v3 08/23] media: camss: Add CSID files Todor Tomov
2017-07-20 14:13 ` Sakari Ailus
2017-07-17 10:33 ` [PATCH v3 09/23] media: camss: Add ISPIF files Todor Tomov
2017-07-20 14:51 ` Sakari Ailus
2017-07-17 10:33 ` [PATCH v3 10/23] media: camss: Add VFE files Todor Tomov
2017-07-20 14:59 ` Sakari Ailus
2017-07-25 14:02 ` Todor Tomov
2017-08-04 18:02 ` Sakari Ailus
2017-08-07 6:49 ` Todor Tomov
2017-07-17 10:33 ` [PATCH v3 11/23] media: camss: Add files which handle the video device nodes Todor Tomov
2017-07-17 10:33 ` [PATCH v3 12/23] media: camms: Add core files Todor Tomov
2017-07-19 10:44 ` Hans Verkuil
2017-07-17 10:33 ` [PATCH v3 13/23] media: camss: Enable building Todor Tomov
2017-07-19 13:17 ` kbuild test robot
2017-07-17 10:33 ` [PATCH v3 14/23] camss: vfe: Format conversion support using PIX interface Todor Tomov
2017-07-20 15:11 ` Sakari Ailus
2017-07-17 10:33 ` [PATCH v3 15/23] doc: media/v4l-drivers: Qualcomm Camera Subsystem - PIX Interface Todor Tomov
2017-07-20 15:13 ` Sakari Ailus
2017-07-17 10:33 ` [PATCH v3 16/23] camss: vfe: Support for frame padding Todor Tomov
2017-07-20 15:17 ` Sakari Ailus
2018-09-03 15:38 ` Todor Tomov
2018-09-13 15:04 ` Sakari Ailus
2017-07-17 10:33 ` [PATCH v3 17/23] camss: vfe: Add interface for scaling Todor Tomov
2017-07-20 15:20 ` Sakari Ailus
2017-07-25 15:36 ` Todor Tomov
2017-08-04 17:59 ` Sakari Ailus
2017-07-17 10:33 ` [PATCH v3 18/23] camss: vfe: Configure scaler module in VFE Todor Tomov
2017-07-17 10:33 ` [PATCH v3 19/23] camss: vfe: Add interface for cropping Todor Tomov
2017-07-17 10:33 ` [PATCH v3 20/23] camss: vfe: Configure crop module in VFE Todor Tomov
2017-07-17 10:33 ` [PATCH v3 21/23] doc: media/v4l-drivers: Qualcomm Camera Subsystem - Scale and crop Todor Tomov
2017-07-17 10:33 ` [PATCH v3 22/23] camss: Use optimal clock frequency rates Todor Tomov
2017-07-19 15:59 ` kbuild test robot
2017-07-21 7:55 ` Todor Tomov
2017-07-20 15:23 ` Sakari Ailus [this message]
2017-07-17 10:33 ` [PATCH v3 23/23] doc: media/v4l-drivers: Qualcomm Camera Subsystem - Media graph Todor Tomov
2017-07-19 10:54 ` [PATCH v3 00/23] Qualcomm 8x16 Camera Subsystem driver Hans Verkuil
2017-07-21 7:39 ` Todor Tomov
2017-07-31 9:20 ` Hans Verkuil
2017-07-20 15:25 ` Sakari Ailus
2017-07-21 7:50 ` Todor Tomov
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=20170720152313.joqfo7dqg5gbxrfl@valkosipuli.retiisi.org.uk \
--to=sakari.ailus@iki.fi \
--cc=hans.verkuil@cisco.com \
--cc=javier@osg.samsung.com \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=mchehab@kernel.org \
--cc=s.nawrocki@samsung.com \
--cc=todor.tomov@linaro.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).