All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Hellstrom <thomas@shipmail.org>
To: "Marek Olšák" <maraeo@gmail.com>
Cc: dri-devel@lists.freedesktop.org
Subject: Re: [RFC] drm/ttm: add minimum residency constraint for bo eviction
Date: Thu, 29 Nov 2012 10:18:07 +0100	[thread overview]
Message-ID: <50B7284F.9090708@shipmail.org> (raw)
In-Reply-To: <CAAxE2A43OTdppDDtY710BbSBA+oB0y15YptsksN=sDt2=-5+2w@mail.gmail.com>

On 11/28/2012 10:51 PM, Marek Olšák wrote:
> I think the problem with Radeon/TTM is much deeper. Let me demonstrate
> it on the following example.
>
> Unigine Heaven needs about 385MB of space for static resources, that's
> only 75% of my 512MB card. Yet, TTM is not capable of getting all of
> that into VRAM. If I allow GTT placements, I get 20 fps, which is the
> old Mesa behavior. If I force VRAM placements, I get 3 fps, because we
> validate buffers 10 times per frame and there's probably a lot of
> buffer evictions during each validation.
>

Marek,
Did you look at the total amount of referenced buffers in the ring 
including vertex buffers?

Depending on how hard you throttle, I guess vertex / index buffer data 
referenced by the
ring commands may well exceed the VRAM limitation.

/Thomas

  parent reply	other threads:[~2012-11-29  9:18 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-11-28 15:58 [RFC] drm/ttm: add minimum residency constraint for bo eviction j.glisse
2012-11-28 15:58 ` [PATCH] " j.glisse
2012-11-28 23:18   ` Thomas Hellstrom
2012-11-28 23:24     ` Jerome Glisse
2012-11-28 23:44       ` Alan Swanson
2012-11-29  0:01         ` Jerome Glisse
2012-11-29  2:15         ` Marek Olšák
2012-11-29  8:04           ` Thomas Hellstrom
2012-11-29 12:52             ` Marek Olšák
2012-11-29 20:33               ` Thomas Hellstrom
2012-11-29 21:58                 ` Marek Olšák
2012-11-30  8:38                   ` Thomas Hellstrom
2012-11-30  9:39                   ` Asynchronous eviction [WAS Re: [PATCH] drm/ttm: add minimum residency constraint for bo eviction] Thomas Hellstrom
2012-11-30 16:30                     ` Jerome Glisse
2012-11-30 17:08                       ` Thomas Hellstrom
2012-11-30 17:18                         ` Jerome Glisse
2012-11-30 17:43                           ` Thomas Hellstrom
2012-11-30 18:07                             ` Jerome Glisse
2012-11-30 18:31                               ` Thomas Hellstrom
2012-11-30 19:25                                 ` Jerome Glisse
2012-11-30 20:35                                   ` Thomas Hellstrom
2012-11-30 21:07                                     ` Jerome Glisse
2012-11-30 21:36                                       ` Thomas Hellstrom
2012-11-30 22:02                                         ` Jerome Glisse
2012-11-29  8:41       ` [PATCH] drm/ttm: add minimum residency constraint for bo eviction Thomas Hellstrom
2012-11-29 15:50         ` Jerome Glisse
2012-11-28 21:51 ` [RFC] " Marek Olšák
2012-11-28 23:18   ` Jerome Glisse
2012-11-29  9:18   ` Thomas Hellstrom [this message]
2012-11-29  9:28     ` Michel Dänzer
2012-11-29  9:48       ` Thomas Hellstrom
2012-11-29 19:20     ` Marek Olšák
2012-11-29 19:36       ` Jerome Glisse
2012-11-29 20:40       ` Thomas Hellstrom

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=50B7284F.9090708@shipmail.org \
    --to=thomas@shipmail.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=maraeo@gmail.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 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.