From: Chris Wilson <chris@chris-wilson.co.uk>
To: Daniel Vetter <daniel@ffwll.ch>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 13/28] drm/i915: Avoid clearing preallocated regions from the GTT
Date: Tue, 01 May 2012 18:35:01 +0100 [thread overview]
Message-ID: <1335893703_109806@CP5-2952> (raw)
In-Reply-To: <20120501173010.GB4832@phenom.ffwll.local>
On Tue, 1 May 2012 19:30:10 +0200, Daniel Vetter <daniel@ffwll.ch> wrote:
> On Tue, Apr 24, 2012 at 03:47:42PM +0100, Chris Wilson wrote:
> > + /* Clear any non-preallocated blocks */
> > + list_for_each_entry(entry, &dev_priv->mm.gtt_space.hole_stack, hole_stack) {
> > + unsigned long hole_start = entry->start + entry->size;
> > + unsigned long hole_end = list_entry(entry->node_list.next,
> > + struct drm_mm_node,
> > + node_list)->start;
> > +
> > + DRM_DEBUG_KMS("clearing unused GTT space: [%lx, %lx]\n",
> > + hole_start, hole_end);
> > +
> > + intel_gtt_clear_range(hole_start / PAGE_SIZE,
> > + (hole_end-hole_start) / PAGE_SIZE);
> > + }
>
> Assuming I'm reading the code correctly this does not clear the unmanaged
> guard page at the end of the gtt.
Right, missed. The simplest fix would seem to be do that as an explicit
step afterwards. Bonus marks for commentary.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
next prev parent reply other threads:[~2012-05-01 17:35 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-04-24 14:47 Allocate objects from stolen memory in preparation for BIOS takeover Chris Wilson
2012-04-24 14:47 ` [PATCH 01/28] drm/i915: Remove the list of pinned inactive objects Chris Wilson
2012-04-24 14:47 ` [PATCH 02/28] drm/i915: Remove the deferred-free list Chris Wilson
2012-04-24 14:47 ` [PATCH 03/28] drm/i915: Track unbound pages Chris Wilson
2012-04-24 14:47 ` [PATCH 04/28] drm/i915: Only pwrite through the GTT if there is space in the aperture Chris Wilson
2012-04-24 14:47 ` [PATCH 05/28] drm/i915: Try harder to allocate an mmap_offset Chris Wilson
2012-04-24 14:47 ` [PATCH 06/28] drm/i915: Pin backing pages for pwrite Chris Wilson
2012-04-24 14:47 ` [PATCH 07/28] drm/i915: Pin pages for pread Chris Wilson
2012-04-24 14:47 ` [PATCH 08/28] drm/i915/debugfs: Show (count, size) of purgeable objects in i915_gem_objects Chris Wilson
2012-04-24 14:47 ` [PATCH 09/28] drm: Introduce drm_mm_create_block() Chris Wilson
2012-04-24 14:47 ` [PATCH 10/28] drm/i915: Split the stolen handling for GEM out of i915_dma.c Chris Wilson
2012-04-24 14:47 ` [PATCH 11/28] drm/i915: Fix detection of stolen base for gen2 Chris Wilson
2012-04-24 14:47 ` [PATCH 12/28] drm/i915: Move GEM initialisation from i915_dma.c to i915_gem.c Chris Wilson
2012-04-24 14:47 ` [PATCH 13/28] drm/i915: Avoid clearing preallocated regions from the GTT Chris Wilson
2012-05-01 17:30 ` Daniel Vetter
2012-05-01 17:35 ` Chris Wilson [this message]
2012-04-24 14:47 ` [PATCH 14/28] drm/i915: Always use the scatterlists if available for GTT insertion Chris Wilson
2012-04-24 14:47 ` [PATCH 15/28] drm/i915: Tidy handling of the scatterlist by ppgtt Chris Wilson
2012-04-24 14:47 ` [PATCH 16/28] drm/i915: Delay allocation of stolen space for FBC Chris Wilson
2012-04-24 14:47 ` [PATCH 17/28] drm/i915: Extract general object init routine Chris Wilson
2012-04-24 14:47 ` [PATCH 18/28] drm/i915: Allow objects to be created with no backing pages, but stolen space Chris Wilson
2012-04-24 14:47 ` [PATCH 19/28] drm/i915: Support readback of stolen objects upon error Chris Wilson
2012-04-24 14:47 ` [PATCH 20/28] drm/i915: Handle stolen objects in pwrite Chris Wilson
2012-04-24 14:47 ` [PATCH 21/28] drm/i915: Handle stolen objects for pread Chris Wilson
2012-04-24 14:47 ` [PATCH 22/28] drm/i915: Introduce i915_gem_object_create_stolen() Chris Wilson
2012-04-24 14:47 ` [PATCH 23/28] drm/i915: Allocate fbcon from stolen memory Chris Wilson
2012-04-24 14:47 ` [PATCH 24/28] drm/i915: Allocate ringbuffers " Chris Wilson
2012-04-24 14:47 ` [PATCH 25/28] drm/i915: Allocate overlay registers " Chris Wilson
2012-04-24 14:47 ` [PATCH 26/28] drm/i915: Introduce i915_gem_object_create_stolen_for_preallocated Chris Wilson
2012-04-24 14:47 ` [PATCH 27/28] drm/i915: Split the framebuffer_info creation into a separate routine Chris Wilson
2012-04-24 14:47 ` [PATCH 28/28] drm/i915: Wrap the preallocated BIOS framebuffer and preserve for KMS fbcon Chris Wilson
2012-04-24 17:22 ` [PATCH] [-1/28] drm/i915: Remove i915_gem_evict_inactive() Chris Wilson
2012-04-25 9:47 ` Daniel Vetter
2012-04-29 16:43 ` [PATCH] drm/i915: Fix location of stolen memory register for SandyBridge+ Chris Wilson
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=1335893703_109806@CP5-2952 \
--to=chris@chris-wilson.co.uk \
--cc=daniel@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