From: Jesse Barnes <jbarnes@virtuousgeek.org>
To: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 1/2] drm/i915: clock readout support for DDI v2
Date: Tue, 21 Jan 2014 12:38:44 -0800 [thread overview]
Message-ID: <20140121123844.115a877e@jbarnes-desktop> (raw)
In-Reply-To: <20140121113656.GA9454@intel.com>
On Tue, 21 Jan 2014 13:36:56 +0200
Ville Syrjälä <ville.syrjala@linux.intel.com> wrote:
> > +static int intel_ddi_calc_wrpll_link(u32 wrpll)
> > +{
> > + int n, p, r;
> > +
> > + r = wrpll & WRPLL_DIVIDER_REF_MASK;
> > + p = (wrpll & WRPLL_DIVIDER_POST_MASK) >> WRPLL_DIVIDER_POST_SHIFT;
> > + n = (wrpll & WRPLL_DIVIDER_FB_MASK) >> WRPLL_DIVIDER_FB_SHIFT;
> > +
> > + return (LC_FREQ * n) / (p * r);
>
> This is assuming the WRPLL will use the LCPLL as reference. Ideally we
> should read out the ref clock settings too.
I don't think I see that in this config, but I've added code to look
for that. Not sure if I got the ref freq right either; I think it's
135MHz in the PCH case...
> > + case PORT_CLK_SEL_SPLL:
> > + link_clock = 135000;
>
> SPLL could also output 810 MHz.
And even 2700 MHz. Fixed.
> > + break;
> > + default:
> > + WARN(1, "bad port clock sel\n");
> > + return;
> > + }
>
> Could do the port_clock = link_clock * 2; here, and then pass port clock
> to intel_dotclock_calculate() and avoid having to multiply crtc_clock by
> 2 afterwards.
Yeah that tidies things up nicely. Fixed.
> As a side note, I must say it's a bit annoying that the DDI PLL code is
> different to the rest of our PLL code. Hz vs. kHz etc. Makes it a bit
> harder to figure out what it's doing.
Yeah that could be converted, mixing them up definitely isn't ideal.
> > +
> > + if (pipe_config->has_pch_encoder)
> > + pipe_config->adjusted_mode.crtc_clock =
> > + intel_dotclock_calculate(link_clock,
> > + &pipe_config->fdi_m_n);
> > + else
>
> else if (has_dp_encoder)
>
> > + pipe_config->adjusted_mode.crtc_clock =
> > + intel_dotclock_calculate(link_clock,
> > + &pipe_config->dp_m_n);
>
> else
> .crtc_clock = link_clock; // or port_clock if you take my suggestion above
Fixed.
> > + intel_cpu_transcoder_get_m_n(crtc, pipe_config->cpu_transcoder,
> > + &pipe_config->dp_m_n);
>
> Not needed. We already do intel_dp_get_m_n() in intel_ddi_get_config().
Ah right, just above. Fixed.
Thanks a lot for the review.
--
Jesse Barnes, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2014-01-21 20:38 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-01-20 22:18 [PATCH 1/2] drm/i915: clock readout support for DDI v2 Jesse Barnes
2014-01-20 22:18 ` [PATCH 2/2] drm/i915: always check clocks when comparing pipe configs Jesse Barnes
2014-01-21 11:36 ` [PATCH 1/2] drm/i915: clock readout support for DDI v2 Ville Syrjälä
2014-01-21 20:38 ` Jesse Barnes [this message]
2014-01-21 20:42 ` [PATCH] drm/i915: clock readout support for DDI v3 Jesse Barnes
2014-01-22 8:56 ` Ville Syrjälä
2014-01-22 9:46 ` 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=20140121123844.115a877e@jbarnes-desktop \
--to=jbarnes@virtuousgeek.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=ville.syrjala@linux.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