All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 16/18] drm/i915: Overcome display engine stride limits via GTT remapping
Date: Thu, 19 Jul 2018 22:20:07 +0300	[thread overview]
Message-ID: <20180719192007.GF5565@intel.com> (raw)
In-Reply-To: <153202687272.4229.12486120875611529948@skylake-alporthouse-com>

On Thu, Jul 19, 2018 at 08:01:12PM +0100, Chris Wilson wrote:
> Quoting Ville Syrjala (2018-07-19 19:22:12)
> > From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > 
> > The display engine stride limits are getting in our way. On SKL+
> > we are limited to 8k pixels, which is easily exceeded with three
> > 4k displays. To overcome this limitation we can remap the pages
> > in the GTT to provide the display engine with a view of memory
> > with a smaller stride.
> > 
> > The code is mostly already there as We already play tricks with
> > the plane surface address and x/y offsets.
> > 
> > A few caveats apply:
> > * linear buffers need the fb stride to be page aligned, as
> >   otherwise the remapped lines wouldn't start at the same
> >   spot
> > * compressed buffers can't be remapped due to the new
> >   ccs hash mode causing the virtual address of the pages
> >   to affect the interpretation of the compressed data. IIRC
> >   the old hash was limited to the low 12 bits so if we were
> >   using that mode we could remap. As it stands we just refuse
> >   to remapp with compressed fbs.
> > * no remapping gen2/3 as we'd need a fence for the remapped
> >   vma, which we currently don't have
> 
> But... you just forbade us getting a fence, there's nothing that
> actually would stop the fence register assignment from working?

Hmm. I thought we still had some kind of fence==object relationship.
Or maybe I'm just thinking about the uapi. I should probably read the
actual code instead of relying on ancient/invalid knowledge :)

-- 
Ville Syrjälä
Intel
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2018-07-19 19:20 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-19 18:21 [PATCH 00/18] drm/i915: GTT remapping for display Ville Syrjala
2018-07-19 18:21 ` [PATCH 01/18] drm/i915: Fix glk/cnl display w/a #1175 Ville Syrjala
2018-07-20 10:55   ` Imre Deak
2018-07-19 18:21 ` [PATCH 02/18] drm/i915: s/tile_offset/aligned_offset/ Ville Syrjala
2018-08-22 21:55   ` Souza, Jose
2018-07-19 18:21 ` [PATCH 03/18] drm/i915: Add .max_stride() plane hook Ville Syrjala
2018-08-22 22:03   ` Souza, Jose
2018-08-22 22:19     ` Souza, Jose
2018-08-23 14:48       ` Ville Syrjälä
2018-08-23 14:52     ` Ville Syrjälä
2018-07-19 18:22 ` [PATCH 04/18] drm/i915: Use pipe A primary plane .max_stride() as the global stride limit Ville Syrjala
2018-08-22 22:22   ` Souza, Jose
2018-07-19 18:22 ` [PATCH 05/18] drm/i915: Rename the plane_state->main/aux to plane_state->color_plane[] Ville Syrjala
2018-08-22 23:02   ` Souza, Jose
2018-07-19 18:22 ` [PATCH 06/18] drm/i915: Store the final plane stride in plane_state Ville Syrjala
2018-07-20 11:06   ` [PATCH v2 " Ville Syrjala
2018-08-22 23:44     ` Souza, Jose
2018-07-19 18:22 ` [PATCH 07/18] drm/i915: Store ggtt_view " Ville Syrjala
2018-08-24 20:13   ` Souza, Jose
2018-07-19 18:22 ` [PATCH 08/18] drm/i915: s/int plane/int color_plane/ Ville Syrjala
2018-08-23  1:14   ` Rodrigo Vivi
2018-08-24  0:05     ` Souza, Jose
2018-07-19 18:22 ` [PATCH 09/18] drm/i915: Nuke plane->can_scale/min_downscale Ville Syrjala
2018-08-24  0:32   ` Souza, Jose
2018-07-19 18:22 ` [PATCH 10/18] drm/i915: Extract per-platform plane->check() functions Ville Syrjala
2018-08-24  1:01   ` Souza, Jose
2018-08-24 12:03     ` Ville Syrjälä
2018-07-19 18:22 ` [PATCH 11/18] drm/i915: Move skl plane fb related checks into a better place Ville Syrjala
2018-08-24 19:56   ` Souza, Jose
2018-08-27 11:48     ` Ville Syrjälä
2018-07-19 18:22 ` [PATCH 12/18] drm/i915: Move display w/a #1175 Ville Syrjala
2018-08-23  1:09   ` Rodrigo Vivi
2018-07-19 18:22 ` [PATCH 13/18] drm/i915: Move chv rotation checks to plane->check() Ville Syrjala
2018-08-24 20:04   ` Souza, Jose
2018-07-19 18:22 ` [PATCH 14/18] drm/i915: Extract intel_cursor_check_surface() Ville Syrjala
2018-08-24 20:08   ` Souza, Jose
2018-07-19 18:22 ` [PATCH 15/18] drm/i915: Add a new "remapped" gtt_view Ville Syrjala
2018-07-19 18:59   ` Chris Wilson
2018-07-19 19:33     ` Ville Syrjälä
2018-07-19 19:46       ` Chris Wilson
2018-07-19 19:55         ` Ville Syrjälä
2018-07-19 20:16         ` Ville Syrjälä
2018-07-19 20:25           ` Chris Wilson
2018-07-19 18:22 ` [PATCH 16/18] drm/i915: Overcome display engine stride limits via GTT remapping Ville Syrjala
2018-07-19 19:01   ` Chris Wilson
2018-07-19 19:20     ` Ville Syrjälä [this message]
2018-07-19 18:22 ` [PATCH 17/18] drm/i915: Bump gen4+ fb stride limit to 256KiB Ville Syrjala
2018-08-24 20:49   ` Souza, Jose
2018-07-19 18:22 ` [PATCH 18/18] drm/i915: Bump gen4+ fb size limits to 32kx32k Ville Syrjala
2018-08-24 20:49   ` Souza, Jose
2018-07-19 18:52 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: GTT remapping for display Patchwork
2018-07-19 19:00 ` ✗ Fi.CI.SPARSE: " Patchwork
2018-07-19 19:15 ` ✓ Fi.CI.BAT: success " Patchwork
2018-07-20  0:02 ` ✗ Fi.CI.IGT: failure " Patchwork
2018-07-20 11:15 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: GTT remapping for display (rev2) Patchwork
2018-07-20 11:23 ` ✗ Fi.CI.SPARSE: " Patchwork
2018-07-20 11:37 ` ✓ Fi.CI.BAT: success " Patchwork
2018-07-21 14:16 ` ✓ Fi.CI.IGT: " Patchwork

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=20180719192007.GF5565@intel.com \
    --to=ville.syrjala@linux.intel.com \
    --cc=chris@chris-wilson.co.uk \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.