From: Imre Deak <imre.deak@intel.com>
To: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Intel Graphics Development <intel-gfx@lists.freedesktop.org>
Subject: Re: [PATCH 11/15] drm/i915: compute fdi lane config earlier
Date: Mon, 29 Apr 2013 15:13:00 +0300 [thread overview]
Message-ID: <1367237580.6390.49.camel@intelbox> (raw)
In-Reply-To: <1366363487-15926-12-git-send-email-daniel.vetter@ffwll.ch>
On Fri, 2013-04-19 at 11:24 +0200, Daniel Vetter wrote:
> Now that it's split up, we can easily move it around and precompute
> the fdi lane configuration.
>
> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Imre Deak <imre.deak@intel.com>
> ---
> drivers/gpu/drm/i915/intel_display.c | 71 +++++++++++++++++-------------------
> 1 file changed, 34 insertions(+), 37 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index 7abcd60..8e10e1b 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -3955,6 +3955,37 @@ bool intel_connector_get_hw_state(struct intel_connector *connector)
> return encoder->get_hw_state(encoder, &pipe);
> }
>
> +static void ironlake_fdi_compute_config(struct drm_device *dev,
> + struct intel_crtc_config *pipe_config)
> +{
> + struct drm_display_mode *adjusted_mode = &pipe_config->adjusted_mode;
> + int target_clock, lane, link_bw;
> +
> + /* FDI is a binary signal running at ~2.7GHz, encoding
> + * each output octet as 10 bits. The actual frequency
> + * is stored as a divider into a 100MHz clock, and the
> + * mode pixel clock is stored in units of 1KHz.
> + * Hence the bw of each lane in terms of the mode signal
> + * is:
> + */
> + link_bw = intel_fdi_link_freq(dev) * MHz(100)/KHz(1)/10;
> +
> + if (pipe_config->pixel_target_clock)
> + target_clock = pipe_config->pixel_target_clock;
> + else
> + target_clock = adjusted_mode->clock;
> +
> + lane = ironlake_get_lanes_required(target_clock, link_bw,
> + pipe_config->pipe_bpp);
> +
> + pipe_config->fdi_lanes = lane;
> +
> + if (pipe_config->pixel_multiplier > 1)
> + link_bw *= pipe_config->pixel_multiplier;
> + intel_link_compute_m_n(pipe_config->pipe_bpp, lane, target_clock,
> + link_bw, &pipe_config->fdi_m_n);
> +}
> +
> static bool intel_crtc_compute_config(struct drm_crtc *crtc,
> struct intel_crtc_config *pipe_config)
> {
> @@ -3989,6 +4020,9 @@ static bool intel_crtc_compute_config(struct drm_crtc *crtc,
> pipe_config->pipe_bpp = 8*3;
> }
>
> + if (pipe_config->has_pch_encoder)
> + ironlake_fdi_compute_config(dev, pipe_config);
> +
> return true;
> }
>
> @@ -5480,38 +5514,6 @@ void intel_cpu_transcoder_set_m_n(struct intel_crtc *crtc,
> }
> }
>
> -static void ironlake_fdi_compute_config(struct intel_crtc *intel_crtc)
> -{
> - struct drm_device *dev = intel_crtc->base.dev;
> - struct drm_display_mode *adjusted_mode =
> - &intel_crtc->config.adjusted_mode;
> - int target_clock, lane, link_bw;
> -
> - /* FDI is a binary signal running at ~2.7GHz, encoding
> - * each output octet as 10 bits. The actual frequency
> - * is stored as a divider into a 100MHz clock, and the
> - * mode pixel clock is stored in units of 1KHz.
> - * Hence the bw of each lane in terms of the mode signal
> - * is:
> - */
> - link_bw = intel_fdi_link_freq(dev) * MHz(100)/KHz(1)/10;
> -
> - if (intel_crtc->config.pixel_target_clock)
> - target_clock = intel_crtc->config.pixel_target_clock;
> - else
> - target_clock = adjusted_mode->clock;
> -
> - lane = ironlake_get_lanes_required(target_clock, link_bw,
> - intel_crtc->config.pipe_bpp);
> -
> - intel_crtc->config.fdi_lanes = lane;
> -
> - if (intel_crtc->config.pixel_multiplier > 1)
> - link_bw *= intel_crtc->config.pixel_multiplier;
> - intel_link_compute_m_n(intel_crtc->config.pipe_bpp, lane, target_clock,
> - link_bw, &intel_crtc->config.fdi_m_n);
> -}
> -
> static bool ironlake_needs_fb_cb_tune(struct dpll *dpll, int factor)
> {
> return i9xx_dpll_compute_m(dpll) < factor * dpll->n;
> @@ -5730,10 +5732,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
>
> /* Note, this also computes intel_crtc->fdi_lanes which is used below in
> * ironlake_check_fdi_lanes. */
> - intel_crtc->config.fdi_lanes = 0;
> if (intel_crtc->config.has_pch_encoder) {
> - ironlake_fdi_compute_config(intel_crtc);
> -
> intel_cpu_transcoder_set_m_n(intel_crtc,
> &intel_crtc->config.fdi_m_n);
> }
> @@ -5867,8 +5866,6 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
> intel_set_pipe_timings(intel_crtc, mode, adjusted_mode);
>
> if (intel_crtc->config.has_pch_encoder) {
> - ironlake_fdi_compute_config(intel_crtc);
> -
> intel_cpu_transcoder_set_m_n(intel_crtc,
> &intel_crtc->config.fdi_m_n);
> }
next prev parent reply other threads:[~2013-04-29 12:13 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-19 9:24 [PATCH 00/15] high-bpp fixes and fdi auto dithering Daniel Vetter
2013-04-19 9:24 ` [PATCH 01/15] drm/i915: fixup 12bpc hdmi dotclock handling Daniel Vetter
2013-04-23 15:02 ` Ville Syrjälä
2013-04-23 15:37 ` Daniel Vetter
2013-04-19 9:24 ` [PATCH 02/15] drm/i915: Disable high-bpc on pre-1.4 EDID screens Daniel Vetter
2013-04-23 15:07 ` Ville Syrjälä
2013-04-24 10:54 ` Daniel Vetter
2013-04-19 9:24 ` [PATCH 03/15] drm/i915: force bpp for eDP panels Daniel Vetter
2013-04-19 20:31 ` [PATCH] " Daniel Vetter
2013-04-19 9:24 ` [PATCH 04/15] drm/i915: drop adjusted_mode from *_set_pipeconf functions Daniel Vetter
2013-04-23 15:12 ` Ville Syrjälä
2013-04-19 9:24 ` [PATCH 05/15] drm/i915: implement high-bpc + pipeconf-dither support for g4x/vlv Daniel Vetter
2013-04-19 16:39 ` Jesse Barnes
2013-04-19 18:17 ` [PATCH] " Daniel Vetter
2013-04-19 18:39 ` Jesse Barnes
2013-04-19 19:29 ` Daniel Vetter
2013-04-23 15:27 ` Ville Syrjälä
2013-04-23 20:39 ` Daniel Vetter
2013-04-23 22:27 ` Daniel Vetter
2013-04-24 11:07 ` Ville Syrjälä
2013-04-23 22:30 ` Daniel Vetter
2013-04-24 11:11 ` Ville Syrjälä
2013-04-24 12:57 ` Daniel Vetter
2013-04-24 13:07 ` Ville Syrjälä
2013-04-19 9:24 ` [PATCH 06/15] drm/i915: allow high-bpc modes on DP Daniel Vetter
2013-04-29 10:16 ` Imre Deak
2013-04-19 9:24 ` [PATCH 07/15] drm/i915: Fixup non-24bpp support for VGA screens on Haswell Daniel Vetter
2013-04-24 11:12 ` Ville Syrjälä
2013-04-24 12:50 ` Daniel Vetter
2013-04-19 9:24 ` [PATCH 08/15] drm/i915: move intel_crtc->fdi_lanes to pipe_config Daniel Vetter
2013-04-29 10:17 ` Imre Deak
2013-04-19 9:24 ` [PATCH 09/15] drm/i915: hw state readout support for pipe_config->fdi_lanes Daniel Vetter
2013-04-24 11:23 ` Ville Syrjälä
2013-04-24 12:49 ` Daniel Vetter
2013-04-24 13:30 ` [PATCH] " Daniel Vetter
2013-04-29 10:22 ` Imre Deak
2013-04-29 17:33 ` [PATCH] drm/i915: put the right cpu_transcoder into pipe_config for hw state readout Daniel Vetter
2013-04-29 17:33 ` [PATCH] drm/i915: hw state readout support for pipe_config->fdi_lanes Daniel Vetter
2013-04-19 9:24 ` [PATCH 10/15] drm/i915: split up fdi_set_m_n into computation and hw setup Daniel Vetter
2013-04-24 11:26 ` Ville Syrjälä
2013-04-19 9:24 ` [PATCH 11/15] drm/i915: compute fdi lane config earlier Daniel Vetter
2013-04-29 12:13 ` Imre Deak [this message]
2013-04-19 9:24 ` [PATCH 12/15] drm/i915: Split up ironlake_check_fdi_lanes Daniel Vetter
2013-04-29 12:19 ` Imre Deak
2013-04-19 9:24 ` [PATCH 13/15] drm/i915: move fdi lane configuration checks ahead Daniel Vetter
2013-04-22 10:32 ` Ville Syrjälä
2013-04-22 15:13 ` [PATCH] " Daniel Vetter
2013-04-29 12:31 ` Imre Deak
2013-04-29 17:34 ` Daniel Vetter
2013-04-19 9:24 ` [PATCH 14/15] drm/i915: don't count cpu ports for fdi B/C lane sharing Daniel Vetter
2013-04-29 13:00 ` Imre Deak
2013-04-19 9:24 ` [PATCH 15/15] drm/i915: implement fdi auto-dithering Daniel Vetter
2013-04-29 14:02 ` Imre Deak
2013-04-29 14:43 ` Daniel Vetter
2013-04-29 14:59 ` Imre Deak
2013-04-29 19:35 ` Daniel Vetter
2013-04-19 15:05 ` [PATCH 00/15] high-bpp fixes and fdi auto dithering Chris Wilson
2013-04-25 10:28 ` Jani Nikula
2013-04-29 19:51 ` 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=1367237580.6390.49.camel@intelbox \
--to=imre.deak@intel.com \
--cc=daniel.vetter@ffwll.ch \
--cc=intel-gfx@lists.freedesktop.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