From: Jani Nikula <jani.nikula@intel.com>
To: Shobhit Kumar <shobhit.kumar@intel.com>,
intel-gfx <intel-gfx@lists.freedesktop.org>
Cc: vijayakumar.balakrishnan@intel.com, yogesh.mohan.marimuthu@intel.com
Subject: Re: [PATCH v2 3/7] drm/i915: Compute dsi_clk from pixel clock
Date: Fri, 15 Nov 2013 10:40:23 +0200 [thread overview]
Message-ID: <87li0qgirs.fsf@intel.com> (raw)
In-Reply-To: <87zjp6gmd6.fsf@intel.com>
On Fri, 15 Nov 2013, Jani Nikula <jani.nikula@intel.com> wrote:
> On Sat, 09 Nov 2013, Shobhit Kumar <shobhit.kumar@intel.com> wrote:
>> Pixel clock based calculation is recommended in the MIPI host controller
>> documentation
>>
>> v2: Based on review comments from Jani and Ville
>> - Use dsi_clk in KHz rather than converting in Hz and back to MHz
>> - RR formula is retained though not used but return dsi_clk in KHz now
>> - Moved the m-n-p changes into a separate patch
>> - Removed the parameter check for intel_dsi->dsi_clock_freq. This will be
>> bought back in if needed when appropriate panel drivers are done
>>
>> Signed-off-by: Vijayakumar Balakrishnan <vijayakumar.balakrishnan@intel.com>
>> Signed-off-by: Shobhit Kumar <shobhit.kumar@intel.com>
>> ---
>> drivers/gpu/drm/i915/intel_dsi_pll.c | 46 +++++++++++++++++++++++++++++-----
>> 1 file changed, 40 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
>> index 44279b2..9f3e6b0 100644
>> --- a/drivers/gpu/drm/i915/intel_dsi_pll.c
>> +++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
>> @@ -50,6 +50,8 @@ static const u32 lfsr_converts[] = {
>> 71, 35 /* 91 - 92 */
>> };
>>
>> +#ifdef DSI_CLK_FROM_RR
>> +
>> static u32 dsi_rr_formula(const struct drm_display_mode *mode,
>> int pixel_format, int video_mode_format,
>> int lane_count, bool eotp)
>> @@ -121,7 +123,7 @@ static u32 dsi_rr_formula(const struct drm_display_mode *mode,
>>
>> /* the dsi clock is divided by 2 in the hardware to get dsi ddr clock */
>> dsi_bit_clock_hz = bytes_per_x_frames_x_lanes * 8;
>> - dsi_clk = dsi_bit_clock_hz / (1000 * 1000);
>> + dsi_clk = dsi_bit_clock_hz / 1000;
>>
>> if (eotp && video_mode_format == VIDEO_MODE_BURST)
>> dsi_clk *= 2;
>> @@ -129,6 +131,38 @@ static u32 dsi_rr_formula(const struct drm_display_mode *mode,
>> return dsi_clk;
>> }
>>
>> +#else
>> +
>> +/* Get DSI clock from pixel clock */
>> +static u32 dsi_clk_from_pclk(const struct drm_display_mode *mode,
>> + int pixel_format, int lane_count)
>> +{
>> + u32 dsi_clk_khz;
>> + u32 bpp;
>> +
>> + switch (pixel_format) {
>> + default:
>> + case VID_MODE_FORMAT_RGB888:
>> + case VID_MODE_FORMAT_RGB666_LOOSE:
>> + bpp = 24;
>> + break;
>> + case VID_MODE_FORMAT_RGB666:
>> + bpp = 18;
>> + break;
>> + case VID_MODE_FORMAT_RGB565:
>> + bpp = 16;
>> + break;
>> + }
>> +
>> + /* DSI data rate = pixel clock * bits per pixel / lane count
>> + pixel clock is converted from KHz to Hz */
>> + dsi_clk_khz = DIV_ROUND_CLOSEST(mode->clock * bpp, lane_count);
>> +
>> + return dsi_clk_khz;
>> +}
>> +
>> +#endif
>> +
>> #ifdef MNP_FROM_TABLE
>>
>> struct dsi_clock_table {
>> @@ -200,13 +234,14 @@ static int dsi_calc_mnp(u32 dsi_clk, struct dsi_mnp *dsi_mnp)
>> u32 calc_p;
>> u32 m_seed;
>>
>> - if (dsi_clk < 300 || dsi_clk > 1150) {
>> + /* dsi_clk is expected in KHZ */
>> + if (dsi_clk < 300000 || dsi_clk > 1150000) {
>> DRM_ERROR("DSI CLK Out of Range\n");
>> return -ECHRNG;
>> }
>>
>> ref_clk = 25000;
>> - target_dsi_clk = dsi_clk * 1000;
>> + target_dsi_clk = dsi_clk;
>
> The *other* dsi_calc_mnp() (with MNP_FROM_TABLE defined) remains
> unchanged, and bitrots. Either /= 1000 the input there, or better, if
> you don't see that the other function will be needed, just remove it.
Other than that,
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
>
> BR,
> Jani.
>
>
>> error = 0xFFFFFFFF;
>> calc_m = 0;
>> calc_p = 0;
>> @@ -251,9 +286,8 @@ static void vlv_configure_dsi_pll(struct intel_encoder *encoder)
>> struct dsi_mnp dsi_mnp;
>> u32 dsi_clk;
>>
>> - dsi_clk = dsi_rr_formula(mode, intel_dsi->pixel_format,
>> - intel_dsi->video_mode_format,
>> - intel_dsi->lane_count, !intel_dsi->eot_disable);
>> + dsi_clk = dsi_clk_from_pclk(mode, intel_dsi->pixel_format,
>> + intel_dsi->lane_count);
>>
>> ret = dsi_calc_mnp(dsi_clk, &dsi_mnp);
>> if (ret) {
>> --
>> 1.7.9.5
>>
>> _______________________________________________
>> Intel-gfx mailing list
>> Intel-gfx@lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
>
> --
> Jani Nikula, Intel Open Source Technology Center
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
--
Jani Nikula, Intel Open Source Technology Center
next prev parent reply other threads:[~2013-11-15 8:43 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-09 9:49 [PATCH v2 0/7] drm/i915: Baytrail MIPI DSI support Updated Shobhit Kumar
2013-11-09 9:49 ` [PATCH v2 1/7] drm/i915: Add more dev ops for MIPI sub encoder Shobhit Kumar
2013-11-15 8:29 ` Jani Nikula
2013-11-09 9:49 ` [PATCH v2 2/7] drm/i915: Use FLISDSI interface for band gap reset Shobhit Kumar
2013-11-15 9:10 ` Jani Nikula
2013-11-09 9:49 ` [PATCH v2 3/7] drm/i915: Compute dsi_clk from pixel clock Shobhit Kumar
2013-11-15 7:22 ` Jani Nikula
2013-11-15 8:40 ` Jani Nikula [this message]
2013-11-09 9:49 ` [PATCH v2 4/7] drm/i915: Try harder to get best m, n, p values with minimal error Shobhit Kumar
2013-11-15 7:19 ` Jani Nikula
2013-11-09 9:49 ` [PATCH v2 5/7] drm/i915: Reorganize the DSI enable/disable sequence Shobhit Kumar
2013-11-15 8:27 ` Jani Nikula
2013-11-15 8:55 ` Daniel Vetter
2013-11-20 1:39 ` Shobhit Kumar
2013-12-06 11:20 ` Shobhit Kumar
2013-12-06 11:25 ` Shobhit Kumar
2013-11-09 9:49 ` [PATCH v2 6/7] drm/i915: Remove redundant DSI PLL enabling Shobhit Kumar
2013-11-15 8:41 ` Jani Nikula
2013-11-09 9:49 ` [PATCH v2 7/7] drm/i915: Parametrize the dphy and other spec specific parameters Shobhit Kumar
2013-11-15 7:52 ` Jani Nikula
2013-11-15 8:42 ` Jani Nikula
2013-11-09 10:28 ` [PATCH v2 0/7] drm/i915: Baytrail MIPI DSI support Updated Daniel Vetter
2013-11-11 8:50 ` [Intel-gfx] " Thierry Reding
2013-11-11 10:28 ` Shobhit Kumar
[not found] ` <52A7F4A0.6050902@intel.com>
[not found] ` <CAKMK7uGU=R3j1TDgLZzUKtztrY6P_akzHHeWEQy_Jw7DdQpiTg@mail.gmail.com>
[not found] ` <87r49j60ym.fsf@intel.com>
[not found] ` <52A86FF2.5050200@intel.com>
[not found] ` <CAKMK7uGA-ENZRQySGVrDkBy7dTOigkKpwpTn63rfGM+UAGvPZA@mail.gmail.com>
2013-12-11 14:25 ` Daniel Vetter
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=87li0qgirs.fsf@intel.com \
--to=jani.nikula@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=shobhit.kumar@intel.com \
--cc=vijayakumar.balakrishnan@intel.com \
--cc=yogesh.mohan.marimuthu@intel.com \
/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