dri-devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
To: Tomi Valkeinen <tomi.valkeinen@ti.com>
Cc: dri-devel@lists.freedesktop.org
Subject: Re: [PATCH v3 10/20] drm: omapdrm: Only commit planes on active CRTCs
Date: Tue, 13 Dec 2016 15:12:17 +0200	[thread overview]
Message-ID: <36161054.r6L69h47Mf@avalon> (raw)
In-Reply-To: <14d64377-3662-89cd-d910-021a04d7fc9a@ti.com>

Hi Tomi,

On Tuesday 13 Dec 2016 09:58:34 Tomi Valkeinen wrote:
> On 13/12/16 00:53, Laurent Pinchart wrote:
> > On Tuesday 20 Sep 2016 16:51:11 Tomi Valkeinen wrote:
> >> On 19/09/16 15:27, Laurent Pinchart wrote:
> >>> The DRM core supports skipping plane update for inactive CRTCs for
> >>> hardware that don't need it or can't cope with it. That's our case, so
> >>> use the DRM core infrastructure instead of reinventing it.
> >> 
> >> I don't follow this desc. What is omapdrm reinventing? At least this
> >> patch does not remove any of the "reinvention".
> > 
> > There used to be one, but it got removed when I rebased the patch series.
> > I'll reword the commit message.
> > 
> >> What does DRM_PLANE_COMMIT_ACTIVE_ONLY do? Skips plane HW configuration
> >> for planes on crtcs that are disabled? The plane HW config will still be
> >> done when the crtc is about to be enabled, right?
> > 
> > It skips plane update (atomic_begin, atomic_disable, atomic_flush) for
> > disabled CRTCs. The CRTC .begin() operation is still called for those
> > CRTCs, only plane update is skipped.
> 
> I didn't catch that. Skips atomic_begin, but .begin is still called?

Sorry, I meant .disable().

> > Now that I wrote that, I'm not quite sure this change is right. It looks
> > like disabling a plane is shadowed, and without an atomic_flush call the
> > GO bit will never be set. However, the problem predates this patch, as
> > the GO bit will only be set if dispc_mgr_is_enabled() returns true, which
> > shouldn't be the case for disabled CRTCs.
> > 
> > How is this supposed to work, how is plane disable supposed to be
> > synchronized at the hardware level ?
> 
> If a crtc is currently disabled (in the HW), all the new settings
> (including plane enable/disable bit) can be programmed and when the crtc
> is enabled, those settings are taken into use. GO bit is not necessary.
> 
> When the crtc is active, you need to set the GO bit to get the new
> settings into use at next vblank.
> 
> If a crtc is to be disabled, we can just disable the crtc, GO bit is not
> necessary.

Thanks for the clarification.

-- 
Regards,

Laurent Pinchart

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2016-12-13 13:11 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-19 12:27 [PATCH v3 00/20] OMAP DRM fixes and improvements Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 01/20] drm: omapdrm: fb: Limit number of planes per framebuffer to two Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 02/20] drm: omapdrm: fb: Use format information provided by the DRM core Laurent Pinchart
2016-09-20 12:47   ` Tomi Valkeinen
2016-12-12 21:41     ` Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 03/20] drm: omapdrm: fb: Simplify objects lookup when creating framebuffer Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 04/20] drm: omapdrm: fb: Simplify mode command checks " Laurent Pinchart
2016-09-20 12:55   ` Tomi Valkeinen
2016-09-19 12:27 ` [PATCH v3 05/20] drm: omapdrm: fb: Turn framebuffer creation error messages into debug Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 06/20] drm: omapdrm: Handle FIFO underflow IRQs internally Laurent Pinchart
2016-09-20 13:17   ` Tomi Valkeinen
2016-09-20 13:27   ` Tomi Valkeinen
2016-12-12 21:59     ` Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 07/20] drm: omapdrm: Handle CRTC error IRQs directly Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 08/20] drm: omapdrm: Handle OCP error IRQ directly Laurent Pinchart
2016-09-20 13:34   ` Tomi Valkeinen
2016-12-12 22:09     ` Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 09/20] drm: omapdrm: Replace DSS manager state check with omapdrm CRTC state Laurent Pinchart
2016-09-20 13:44   ` Tomi Valkeinen
2016-12-12 22:16     ` Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 10/20] drm: omapdrm: Only commit planes on active CRTCs Laurent Pinchart
2016-09-20 13:51   ` Tomi Valkeinen
2016-12-12 22:53     ` Laurent Pinchart
2016-12-13  7:58       ` Tomi Valkeinen
2016-12-13 13:12         ` Laurent Pinchart [this message]
2016-09-19 12:27 ` [PATCH v3 11/20] drm: omapdrm: Check DSS manager state in the enable/disable helpers Laurent Pinchart
2016-09-20 13:57   ` Tomi Valkeinen
2016-12-12 23:07     ` Laurent Pinchart
2016-12-13  8:15       ` Tomi Valkeinen
2016-12-13 23:56         ` Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 12/20] drm: omapdrm: Prevent processing the same event multiple times Laurent Pinchart
2016-09-27 12:24   ` Tomi Valkeinen
2016-12-12 10:29     ` Laurent Pinchart
2016-09-27 15:05   ` Daniel Kurtz
2016-12-12 10:21     ` Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 13/20] drm: omapdrm: Use a spinlock to protect the CRTC pending flag Laurent Pinchart
2016-09-27 12:27   ` Tomi Valkeinen
2016-09-19 12:27 ` [PATCH v3 14/20] drm: omapdrm: Keep vblank interrupt enabled while CRTC is active Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 15/20] drm: omapdrm: Don't expose the omap_irq_(un)register() functions Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 16/20] drm: omapdrm: Remove unused parameter from omap_drm_irq handler Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 17/20] drm: omapdrm: Don't call DISPC power handling in IRQ wait functions Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 18/20] drm: omapdrm: Make pipe2vbl function static Laurent Pinchart
2016-12-12 10:41   ` Tomi Valkeinen
2016-12-12 23:17     ` Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 19/20] drm: omapdrm: Simplify IRQ wait implementation Laurent Pinchart
2016-09-19 12:27 ` [PATCH v3 20/20] drm: omapdrm: Remove global variables Laurent Pinchart

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=36161054.r6L69h47Mf@avalon \
    --to=laurent.pinchart@ideasonboard.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=tomi.valkeinen@ti.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