From: Daniel Vetter <daniel@ffwll.ch>
To: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 66/70] drm/i915: Remove obj->pin_mappable
Date: Mon, 13 Apr 2015 14:30:53 +0200 [thread overview]
Message-ID: <20150413123053.GV6092@phenom.ffwll.local> (raw)
In-Reply-To: <552BAA19.4090306@linux.intel.com>
On Mon, Apr 13, 2015 at 12:35:53PM +0100, Tvrtko Ursulin wrote:
>
> On 04/07/2015 05:28 PM, Chris Wilson wrote:
> >The obj->pin_mappable flag only exists for debug purposes and is a
> >hindrance that is mistreated with rotated GGTT views. For debug
> >purposes, it suffices to mark objects with pin_display as being of note.
> >
> >Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> >---
> > drivers/gpu/drm/i915/i915_debugfs.c | 6 +++---
> > drivers/gpu/drm/i915/i915_drv.h | 1 -
> > drivers/gpu/drm/i915/i915_gem.c | 6 +-----
> > 3 files changed, 4 insertions(+), 9 deletions(-)
> >
> >diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
> >index 2e851c6a310c..6508eec3cf60 100644
> >--- a/drivers/gpu/drm/i915/i915_debugfs.c
> >+++ b/drivers/gpu/drm/i915/i915_debugfs.c
> >@@ -166,9 +166,9 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
> > }
> > if (obj->stolen)
> > seq_printf(m, " (stolen: %08llx)", obj->stolen->start);
> >- if (obj->pin_mappable || obj->fault_mappable) {
> >+ if (obj->pin_display || obj->fault_mappable) {
> > char s[3], *t = s;
> >- if (obj->pin_mappable)
> >+ if (obj->pin_display)
> > *t++ = 'p';
> > if (obj->fault_mappable)
> > *t++ = 'f';
> >@@ -464,7 +464,7 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
> > size += i915_gem_obj_ggtt_size(obj);
> > ++count;
> > }
> >- if (obj->pin_mappable) {
> >+ if (obj->pin_display) {
> > mappable_size += i915_gem_obj_ggtt_size(obj);
> > ++mappable_count;
> > }
> >diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> >index eeffefa10612..2c72ee0214b5 100644
> >--- a/drivers/gpu/drm/i915/i915_drv.h
> >+++ b/drivers/gpu/drm/i915/i915_drv.h
> >@@ -1980,7 +1980,6 @@ struct drm_i915_gem_object {
> > * accurate mappable working set.
> > */
> > unsigned int fault_mappable:1;
> >- unsigned int pin_mappable:1;
> > unsigned int pin_display:1;
> >
> > /*
> >diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> >index bd60bb552920..3d4463930267 100644
> >--- a/drivers/gpu/drm/i915/i915_gem.c
> >+++ b/drivers/gpu/drm/i915/i915_gem.c
> >@@ -4445,9 +4445,6 @@ i915_gem_object_do_pin(struct drm_i915_gem_object *obj,
> > WARN_ON(flags & PIN_MAPPABLE && !obj->map_and_fenceable);
> >
> > vma->pin_count++;
> >- if (flags & PIN_MAPPABLE)
> >- obj->pin_mappable |= true;
> >-
> > return 0;
> > }
> >
> >@@ -4487,8 +4484,7 @@ i915_gem_object_ggtt_unpin_view(struct drm_i915_gem_object *obj,
> > WARN_ON(vma->pin_count == 0);
> > WARN_ON(!i915_gem_obj_ggtt_bound_view(obj, view));
> >
> >- if (--vma->pin_count == 0 && view->type == I915_GGTT_VIEW_NORMAL)
> >- obj->pin_mappable = false;
> >+ --vma->pin_count;
> > }
> >
> > bool
> >
>
> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Queued for -next, thanks for the patch.
-Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2015-04-13 12:28 UTC|newest]
Thread overview: 113+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-04-07 15:20 Low hanging fruit take 2 Chris Wilson
2015-04-07 15:20 ` [PATCH 01/70] drm/i915: Cache last obj->pages location for i915_gem_object_get_page() Chris Wilson
2015-04-08 11:16 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 02/70] drm/i915: Fix the flip synchronisation to consider mmioflips Chris Wilson
2015-04-07 15:20 ` [PATCH 03/70] drm/i915: Ensure cache flushes prior to doing CS flips Chris Wilson
2015-04-08 11:23 ` Daniel Vetter
2015-04-08 11:29 ` Chris Wilson
2015-04-07 15:20 ` [PATCH 04/70] drm/i915: Agressive downclocking on Baytrail Chris Wilson
2015-04-07 15:20 ` [PATCH 05/70] drm/i915: Fix computation of last_adjustment for RPS autotuning Chris Wilson
2015-04-07 15:20 ` [PATCH 06/70] drm/i915: Fix race on unreferencing the wrong mmio-flip-request Chris Wilson
2015-04-08 11:30 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 07/70] drm/i915: Boost GPU frequency if we detect outstanding pageflips Chris Wilson
2015-04-08 11:31 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 08/70] drm/i915: Deminish contribution of wait-boosting from clients Chris Wilson
2015-04-07 15:20 ` [PATCH 09/70] drm/i915: Re-enable RPS wait-boosting for all engines Chris Wilson
2015-04-07 15:20 ` [PATCH 10/70] drm/i915: Split i915_gem_batch_pool into its own header Chris Wilson
2015-04-07 15:20 ` [PATCH 11/70] drm/i915: Tidy batch pool logic Chris Wilson
2015-04-07 15:20 ` [PATCH 12/70] drm/i915: Split the batch pool by engine Chris Wilson
2015-04-07 15:20 ` [PATCH 13/70] drm/i915: Free batch pool when idle Chris Wilson
2015-04-07 15:20 ` [PATCH 14/70] drm/i915: Split batch pool into size buckets Chris Wilson
2015-04-07 15:20 ` [PATCH 15/70] drm/i915: Include active flag when describing objects in debugfs Chris Wilson
2015-04-08 11:33 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 16/70] drm/i915: Suppress empty lines from debugfs/i915_gem_objects Chris Wilson
2015-04-08 11:34 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 17/70] drm/i915: Optimistically spin for the request completion Chris Wilson
2015-04-08 11:39 ` Daniel Vetter
2015-04-08 13:43 ` Rantala, Valtteri
2015-04-08 14:15 ` Daniel Vetter
2015-04-13 11:34 ` Tvrtko Ursulin
2015-04-13 12:25 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 18/70] drm/i915: Implement inter-engine read-read optimisations Chris Wilson
2015-04-14 13:51 ` Tvrtko Ursulin
2015-04-14 14:00 ` Chris Wilson
2015-04-07 15:20 ` [PATCH 19/70] drm/i915: Inline check required for object syncing prior to execbuf Chris Wilson
2015-04-07 15:20 ` [PATCH 20/70] drm/i915: Limit ring synchronisation (sw sempahores) RPS boosts Chris Wilson
2015-04-08 11:46 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 21/70] drm/i915: Limit mmio flip " Chris Wilson
2015-04-07 15:20 ` [PATCH 22/70] drm/i915: Reduce frequency of unspecific HSW reg debugging Chris Wilson
2015-04-07 15:20 ` [PATCH 23/70] drm/i915: Record ring->start address in error state Chris Wilson
2015-04-08 11:47 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 24/70] drm/i915: Use simpler form of spin_lock_irq(execlist_lock) Chris Wilson
2015-04-07 15:20 ` [PATCH 25/70] drm/i915: Use the global runtime-pm wakelock for a busy GPU for execlists Chris Wilson
2015-04-07 15:20 ` [PATCH 26/70] drm/i915: Map the execlists context regs once during pinning Chris Wilson
2015-04-07 15:20 ` [PATCH 27/70] drm/i915: Remove vestigal DRI1 ring quiescing code Chris Wilson
2015-04-09 15:02 ` Daniel Vetter
2015-04-09 15:24 ` Chris Wilson
2015-04-09 15:31 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 28/70] drm/i915: Overhaul execlist submission Chris Wilson
2015-04-07 15:20 ` [PATCH 29/70] drm/i915: Move the execlists retirement to the right spot Chris Wilson
2015-04-07 15:20 ` [PATCH 30/70] drm/i915: Map the ringbuffer using WB on LLC machines Chris Wilson
2015-04-07 15:20 ` [PATCH 31/70] drm/i915: Refactor duplicate object vmap functions Chris Wilson
2015-04-07 15:20 ` [PATCH 32/70] drm/i915: Treat ringbuffer writes as write to normal memory Chris Wilson
2015-04-07 15:20 ` [PATCH 33/70] drm/i915: Use a separate slab for requests Chris Wilson
2015-05-22 14:48 ` Robert Beckett
2015-04-07 15:20 ` [PATCH 34/70] drm/i915: Use a separate slab for vmas Chris Wilson
2015-04-10 8:32 ` Daniel Vetter
2015-04-07 15:20 ` [PATCH 35/70] drm/i915: Use the new rq->i915 field where appropriate Chris Wilson
2015-04-07 15:21 ` [PATCH 36/70] drm/i915: Reduce the pointer dance of i915_is_ggtt() Chris Wilson
2015-04-07 15:21 ` [PATCH 37/70] drm/i915: Squash more pointer indirection for i915_is_gtt Chris Wilson
2015-04-07 15:21 ` [PATCH 38/70] drm/i915: Reduce locking in execlist command submission Chris Wilson
2015-04-07 15:21 ` [PATCH 39/70] drm/i915: Reduce more " Chris Wilson
2015-04-07 15:21 ` [PATCH 40/70] drm/i915: Reduce locking in gen8 IRQ handler Chris Wilson
2015-04-07 15:21 ` [PATCH 41/70] drm/i915: Tidy " Chris Wilson
2015-04-10 8:36 ` Daniel Vetter
2015-04-07 15:21 ` [PATCH 42/70] drm/i915: Remove request retirement before each batch Chris Wilson
2015-04-07 15:21 ` [PATCH 43/70] drm/i915: Cache the GGTT offset for the execlists context Chris Wilson
2015-04-07 15:21 ` [PATCH 44/70] drm/i915: Prefer to check for idleness in worker rather than sync-flush Chris Wilson
2015-04-10 8:37 ` Daniel Vetter
2015-04-07 15:21 ` [PATCH 45/70] drm/i915: Remove request->uniq Chris Wilson
2015-04-10 8:38 ` Daniel Vetter
2015-04-07 15:21 ` [PATCH 46/70] drm/i915: Cache the reset_counter for the request Chris Wilson
2015-04-07 15:21 ` [PATCH 47/70] drm/i915: Allocate context objects from stolen Chris Wilson
2015-04-10 8:39 ` Daniel Vetter
2015-04-07 15:21 ` [PATCH 48/70] drm/i915: Introduce an internal allocator for disposable private objects Chris Wilson
2015-04-07 15:21 ` [PATCH 49/70] drm/i915: Do not zero initialise page tables Chris Wilson
2015-04-07 15:21 ` [PATCH 50/70] drm/i915: The argument for postfix is redundant Chris Wilson
2015-04-10 8:53 ` Daniel Vetter
2015-04-10 9:00 ` Chris Wilson
2015-04-10 9:32 ` Daniel Vetter
2015-04-10 9:45 ` Chris Wilson
2015-04-07 15:21 ` [PATCH 51/70] drm/i915: Record the position of the start of the request Chris Wilson
2015-04-07 15:21 ` [PATCH 52/70] drm/i915: Cache the execlist ctx descriptor Chris Wilson
2015-04-07 15:21 ` [PATCH 53/70] drm/i915: Eliminate vmap overhead for cmd parser Chris Wilson
2015-04-07 15:21 ` [PATCH 54/70] drm/i915: Cache last cmd descriptor when parsing Chris Wilson
2015-04-07 15:21 ` [PATCH 55/70] drm/i915: Use WC copies on !llc platforms for the command parser Chris Wilson
2015-04-07 15:21 ` [PATCH 56/70] drm/i915: Cache kmap between relocations Chris Wilson
2015-04-07 15:21 ` [PATCH 57/70] drm/i915: intel_ring_initialized() must be simple and inline Chris Wilson
2015-12-08 15:02 ` [PATCH 0/1] " Dave Gordon
2015-12-08 15:02 ` [PATCH 1/1] " Dave Gordon
2015-12-10 10:24 ` Daniel Vetter
2015-04-07 15:21 ` [PATCH 58/70] drm/i915: Before shrink_all we only need to idle the GPU Chris Wilson
2015-04-07 15:21 ` [PATCH 59/70] drm/i915: Simplify object is-pinned checking for shrinker Chris Wilson
2015-04-07 16:28 ` Chris Wilson
2015-04-07 16:28 ` [PATCH 60/70] drm/i915: Make evict-everything more robust Chris Wilson
2015-04-07 16:28 ` [PATCH 61/70] drm/i915: Make fb_tracking.lock a spinlock Chris Wilson
2015-04-14 14:52 ` Tvrtko Ursulin
2015-04-14 15:05 ` Chris Wilson
2015-04-14 15:15 ` Tvrtko Ursulin
2015-04-07 16:28 ` [PATCH 62/70] drm/i915: Reduce locking inside busy ioctl Chris Wilson
2015-04-07 16:28 ` [PATCH 63/70] drm/i915: Reduce locking inside swfinish ioctl Chris Wilson
2015-04-10 9:14 ` Daniel Vetter
2015-04-15 9:03 ` Chris Wilson
2015-04-15 9:33 ` Daniel Vetter
2015-04-15 9:38 ` Chris Wilson
2015-04-07 16:28 ` [PATCH 64/70] drm/i915: Remove pinned check from madvise ioctl Chris Wilson
2015-04-07 16:28 ` [PATCH 65/70] drm/i915: Reduce locking for gen6+ GT interrupts Chris Wilson
2015-04-07 16:28 ` [PATCH 66/70] drm/i915: Remove obj->pin_mappable Chris Wilson
2015-04-13 11:35 ` Tvrtko Ursulin
2015-04-13 12:30 ` Daniel Vetter [this message]
2015-04-07 16:28 ` [PATCH 67/70] drm/i915: Start passing around i915_vma from execbuffer Chris Wilson
2015-04-07 16:28 ` [PATCH 68/70] drm/i915: Simplify vma-walker for i915_gem_objects Chris Wilson
2015-04-07 16:28 ` [PATCH 69/70] drm/i915: Skip holding an object reference for execbuf preparation Chris Wilson
2015-04-07 16:28 ` [PATCH 70/70] drm/i915: Use vma as the primary token for managing binding 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=20150413123053.GV6092@phenom.ffwll.local \
--to=daniel@ffwll.ch \
--cc=intel-gfx@lists.freedesktop.org \
--cc=tvrtko.ursulin@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