On Fri, Jul 31, 2015 at 06:46:20PM +0530, Goel, Akash wrote:
> Since the suggestion is to reuse the gtt_pwrite_fast function only
> for non-shmem backed objects, can we also relax the Tiling
> constraint here, apart from removing the PIN_NONBLOCK flag. As
> anyways there is a put_fence already
On Tue, Sep 15, 2015 at 02:03:27PM +0530, ankitprasad.r.sha...@intel.com wrote:
> @@ -1090,17 +1184,17 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void
> *data,
> goto out;
> }
>
> - /* prime objects have no backing filp to GEM pread/pwrite
> - * pages from.
>
On Tue, 2015-09-15 at 10:54 +0100, Chris Wilson wrote:
> On Tue, Sep 15, 2015 at 02:03:27PM +0530, ankitprasad.r.sha...@intel.com
> wrote:
> > @@ -1090,17 +1184,17 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void
> > *data,
> > goto out;
> > }
> >
> > - /* prime objects
On Fri, 2015-07-31 at 18:46 +0530, Goel, Akash wrote:
>
> On 7/22/2015 8:09 PM, Chris Wilson wrote:
> > On Wed, Jul 22, 2015 at 07:21:49PM +0530, ankitprasad.r.sha...@intel.com
> > wrote:
> >> static int
> >> i915_gem_shmem_pread(struct drm_device *dev,
> >> struct
On 7/22/2015 8:09 PM, Chris Wilson wrote:
On Wed, Jul 22, 2015 at 07:21:49PM +0530, ankitprasad.r.sha...@intel.com wrote:
static int
i915_gem_shmem_pread(struct drm_device *dev,
struct drm_i915_gem_object *obj,
@@ -754,17 +850,20 @@ i915_gem_pread_ioctl(struct
Hi,
On 07/22/2015 02:51 PM, ankitprasad.r.sha...@intel.com wrote:
From: Ankitprasad Sharma ankitprasad.r.sha...@intel.com
This patch adds support for extending the pread/pwrite functionality
for objects not backed by shmem. The access will be made through
gtt interface.
This will cover prime
On Wed, Jul 22, 2015 at 04:46:00PM +0100, Tvrtko Ursulin wrote:
But also, why it is even necessary to drop the mutex here?
Locking inversion with our own pagefault handler. Happens since at least
mesa can return gtt mmaps to gl clients, which can then in turn try to
upload that with pwrite
On 07/22/2015 05:05 PM, Daniel Vetter wrote:
On Wed, Jul 22, 2015 at 04:46:00PM +0100, Tvrtko Ursulin wrote:
But also, why it is even necessary to drop the mutex here?
Locking inversion with our own pagefault handler. Happens since at least
mesa can return gtt mmaps to gl clients, which can
On Wed, Jul 22, 2015 at 07:21:49PM +0530, ankitprasad.r.sha...@intel.com wrote:
static int
i915_gem_shmem_pread(struct drm_device *dev,
struct drm_i915_gem_object *obj,
@@ -754,17 +850,20 @@ i915_gem_pread_ioctl(struct drm_device *dev, void *data,
goto out;
On Thu, Jul 02, 2015 at 11:42:44AM +0100, Tvrtko Ursulin wrote:
I am not super familiar with low level mapping business. But it
looks correct to me. Just one question would be if there are any
downsides to WC mapping? If in the read case it would be any
advantage not to ask for WC?
Read-side
On 07/01/2015 10:25 AM, ankitprasad.r.sha...@intel.com wrote:
From: Ankitprasad Sharma ankitprasad.r.sha...@intel.com
This patch adds support for extending the pread/pwrite functionality
for objects not backed by shmem. The access will be made through
gtt interface.
This will cover prime
On 07/02/2015 12:00 PM, Chris Wilson wrote:
+ /* This is a slow read/write as it tries to read from
+* and write to user memory which may result into page
+* faults
+*/
+ ret = slow_user_access(dev_priv-gtt.mappable,
On Thu, Jul 02, 2015 at 12:27:42PM +0100, Tvrtko Ursulin wrote:
On 07/02/2015 12:00 PM, Chris Wilson wrote:
+ /* This is a slow read/write as it tries to read from
+ * and write to user memory which may result into page
+ * faults
+ */
+ ret
Tested-By: Intel Graphics QA PRTS (Patch Regression Test System Contact:
shuang...@intel.com)
Task id: 6697
-Summary-
Platform Delta drm-intel-nightly Series Applied
ILK
On Wed, Jul 01, 2015 at 02:55:47PM +0530, ankitprasad.r.sha...@intel.com wrote:
- ret = i915_gem_shmem_pread(dev, obj, args, file);
+ /* pread for non shmem backed objects */
+ if (!obj-base.filp) {
+ if (obj-tiling_mode == I915_TILING_NONE)
+ ret =
15 matches
Mail list logo