Re: [Intel-gfx] [PATCH] drm/i915: Update WaFlushCoherentL3CacheLinesAtContextSwitch

2015-07-06 Thread Dave Gordon
On 06/07/15 15:33, Daniel Vetter wrote: On Mon, Jul 06, 2015 at 02:16:54PM +0100, Dave Gordon wrote: On 06/07/15 13:38, Daniel Vetter wrote: On Mon, Jul 06, 2015 at 12:52:51PM +0100, Dave Gordon wrote: On 03/07/15 16:42, Chris Wilson wrote: On Fri, Jul 03, 2015 at 02:27:31PM +0100, Arun

Re: [Intel-gfx] [PATCH 2/3] igt/gem_pread: Support to verify pread/pwrite for non-shmem backed obj

2015-07-06 Thread Dave Gordon
On 03/07/15 10:37, Tvrtko Ursulin wrote: On 07/01/2015 10:26 AM, ankitprasad.r.sha...@intel.com wrote: From: Ankitprasad Sharma This patch adds support to verify pread/pwrite for non-shmem backed objects. It also shows the pread/pwrite speed. It also tests speeds for pread with and without u

Re: [Intel-gfx] [PATCH] drm/i915: RMW register cycles considered evil

2015-07-06 Thread Dave Gordon
On 06/07/15 13:50, Ville Syrjälä wrote: On Mon, Jul 06, 2015 at 02:42:02PM +0200, Daniel Vetter wrote: Especially for workarounds which is stuff that's almost impossible to verify: The initial state from the firmware on boot-up and after resume could be different, which will hide bugs when we do

Re: [Intel-gfx] [PATCH] drm/i915: Update WaFlushCoherentL3CacheLinesAtContextSwitch

2015-07-06 Thread Dave Gordon
On 06/07/15 13:38, Daniel Vetter wrote: On Mon, Jul 06, 2015 at 12:52:51PM +0100, Dave Gordon wrote: On 03/07/15 16:42, Chris Wilson wrote: On Fri, Jul 03, 2015 at 02:27:31PM +0100, Arun Siluvery wrote: In this WA we need to set GEN8_L3SQCREG4[21:21] and reset it after PIPE_CONTROL

Re: [Intel-gfx] [PATCH 02/15] drm/i915: Embedded microcontroller (uC) firmware loading support

2015-07-06 Thread Dave Gordon
On 24/06/15 11:29, Daniel Vetter wrote: On Fri, Jun 19, 2015 at 09:43:11AM +0100, Dave Gordon wrote: On 18/06/15 15:49, Daniel Vetter wrote: On Thu, Jun 18, 2015 at 01:11:34PM +0100, Dave Gordon wrote: On 17/06/15 13:05, Daniel Vetter wrote: On Mon, Jun 15, 2015 at 07:36:20PM +0100, Dave

Re: [Intel-gfx] [PATCH] drm/i915: Update WaFlushCoherentL3CacheLinesAtContextSwitch

2015-07-06 Thread Dave Gordon
implementation uses MI_LOAD_REGISTER_MEM which is currently only used by command parser and was using a default length of 0. This is now updated with correct length and moved to appropriate place. Cc: Chris Wilson Cc: Dave Gordon Signed-off-by: Arun Siluvery --- drivers/gpu/drm/i915

[Intel-gfx] [PATCH 13/15] drm/i915: Debugfs interface for GuC submission statistics

2015-07-03 Thread Dave Gordon
This provides a means of reading status and counts relating to GuC actions and submissions. v2: Remove surplus blank line in output [Chris Wilson] Signed-off-by: Dave Gordon Signed-off-by: Alex Dai --- drivers/gpu/drm/i915/i915_debugfs.c | 40 + 1 file

[Intel-gfx] [PATCH 15/15] drm/i915: Enable GuC submission, where supported

2015-07-03 Thread Dave Gordon
Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_params.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c index a8f48dd..479c97b 100644 --- a/drivers/gpu/drm/i915/i915_params.c +++ b/drivers

[Intel-gfx] [PATCH 12/15] drm/i915: Integrate GuC-based command submission

2015-07-03 Thread Dave Gordon
tions [Chris Wilson] Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_debugfs.c| 2 +- drivers/gpu/drm/i915/i915_guc_submission.c | 48 +++--- drivers/gpu/drm/i915/intel_guc.h | 1 + drivers/gpu/drm/i915/intel_

[Intel-gfx] [PATCH 11/15] drm/i915: Interrupt routing for GuC submission

2015-07-03 Thread Dave Gordon
Turn on interrupt steering to route necessary interrupts to GuC. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_reg.h | 11 +-- drivers/gpu/drm/i915/intel_guc_loader.c | 51 + 2 files changed, 60

[Intel-gfx] [PATCH 04/15] drm/i915: Add GuC-related header files

2015-07-03 Thread Dave Gordon
f-by: Dave Gordon --- drivers/gpu/drm/i915/i915_guc_reg.h | 102 +++ drivers/gpu/drm/i915/intel_guc_fwif.h | 236 ++ 2 files changed, 338 insertions(+) create mode 100644 drivers/gpu/drm/i915/i915_guc_reg.h create mode 100644 drivers/gpu/drm/

[Intel-gfx] [PATCH 10/15] drm/i915: Implementation of GuC client

2015-07-03 Thread Dave Gordon
ris Wilson] Rationalise type declarations [Chris Wilson] Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_guc_submission.c | 661 + drivers/gpu/drm/i915/intel_guc.h | 43 ++ drivers/gpu/drm/i915/intel_guc_loader.c

[Intel-gfx] [PATCH 03/15] drm/i915: Add GuC-related module parameters

2015-07-03 Thread Dave Gordon
From: Alex Dai Two new module parameters: "enable_guc_submission" which will turn on submission of batchbuffers via the GuC (when implemented), and "guc_log_level" which controls the level of debugging logged by the GuC and captured by the host. Signed-off-by: Alex Dai

[Intel-gfx] [PATCH 00/15 v3] Batch submission via GuC

2015-07-03 Thread Dave Gordon
drm/i915: GuC submission setup, phase 1 drm/i915: Enable GuC firmware log drm/i915: Integrate GuC-based command submission Documentation/drm: kerneldoc for GuC Dave Gordon (8): drm/i915: Add i915_gem_object_create_from_data() drm/i915: Embedded microcontroller (uC) firmware loading sup

[Intel-gfx] [PATCH 14/15] Documentation/drm: kerneldoc for GuC

2015-07-03 Thread Dave Gordon
From: Alex Dai Add overview design of GuC, plus some key points related to the implementation. Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- Documentation/DocBook/drm.tmpl | 19 +++ 1 file changed, 19 insertions(+) diff --git a/Documentation/DocBook/drm.tmpl b

[Intel-gfx] [PATCH 08/15] drm/i915: GuC submission setup, phase 1

2015-07-03 Thread Dave Gordon
and the GuC's debug log. v2: Remove redundant initialisation [Chris Wilson] Defer adding struct members until needed [Chris Wilson] Local functions should pass dev_priv rather than dev [Chris Wilson] Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gp

[Intel-gfx] [PATCH 06/15] drm/i915: Debugfs interface to read GuC load status

2015-07-03 Thread Dave Gordon
-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_debugfs.c | 45 + 1 file changed, 45 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 71ba519..65d40f1 100644

[Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_create_from_data()

2015-07-03 Thread Dave Gordon
VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_drv.h | 2 ++ drivers/gpu/drm/i915/i915_gem.c | 40 2 files changed, 42 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index

[Intel-gfx] [PATCH 02/15] drm/i915: Embedded microcontroller (uC) firmware loading support

2015-07-03 Thread Dave Gordon
ach uC-specific loader, thus reducing code duplication and testing effort. v2: Local functions should pass dev_priv rather than dev [Chris Wilson] Various other improvements per Chris Wilson's review comments v3: Defeatured version without async prefetch [Daniel Vetter] Signed

[Intel-gfx] [PATCH 05/15] drm/i915: GuC-specific firmware loader

2015-07-03 Thread Dave Gordon
ommon firmware loading mechanism by Dave Gordon as well as new firmware layout etc. v2: Various improvements per review comments by Chris Wilson v3: Removed 'wait' parameter to intel_guc_ucode_load() as prefetch is no longer supported in the common firmware loader, per

[Intel-gfx] [PATCH 09/15] drm/i915: Enable GuC firmware log

2015-07-03 Thread Dave Gordon
From: Alex Dai Allocate a GEM object to hold GuC log data. A debugfs interface (i915_guc_log_dump) is provided to print out the log content. v2: Add struct members at point of use [Chris Wilson] Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915

[Intel-gfx] [PATCH 07/15] drm/i915: Expose two LRC functions for GuC submission mode

2015-07-03 Thread Dave Gordon
previously part of an internal patch, but which was omitted from the version recently committed to drm-intel-nightly: 7a01a0a drm/i915/lrc: Update PDPx registers with lri commands So we reinstate this change here. Issue: VIZ-4884 Signed-off-by: Dave Gordon --- drivers/gpu/drm/i

Re: [Intel-gfx] [PATCH 2/2 i-g-t] lib/igt.cocci: Add 64-bit and float compare functions

2015-07-03 Thread Dave Gordon
On 01/07/15 14:02, Daniel Vetter wrote: On Tue, Jun 30, 2015 at 11:14:54AM -0300, Paulo Zanoni wrote: 2015-06-30 10:54 GMT-03:00 Chris Wilson : On Tue, Jun 30, 2015 at 02:41:09PM +0100, Michel Thierry wrote: @@ -1109,7 +1109,7 @@ static void setup_sink_crc(void) set_mode_for_params(&pri

[Intel-gfx] [PATCH 2/2] drm/i915: Defer late hardware initialisation until first open

2015-06-30 Thread Dave Gordon
more complete environment than the driver_load stage presents. This can be important for embedded programmable devices that need firmware loaded from a file before they can be used. Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_drv.h |1 + drivers/gpu/drm/i915/i915_

[Intel-gfx] [PATCH 1/2] drm/i915: Split late "for_each_ring" loop from i915_gem_init_hw()

2015-06-30 Thread Dave Gordon
containing LRI instructions to set up context-specific registers. Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_drv.h |1 + drivers/gpu/drm/i915/i915_gem.c | 15 ++- 2 files changed, 15 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu

Re: [Intel-gfx] [PATCH] drm/i915: Preallocate request before access of the ring

2015-06-29 Thread Dave Gordon
On 29/06/15 12:39, Jani Nikula wrote: > On Wed, 06 May 2015, Daniel Vetter wrote: >> On Thu, Apr 30, 2015 at 01:54:41PM +0100, Dave Gordon wrote: >>> On 29/04/15 17:10, yu@intel.com wrote: >>>> From: Alex Dai >>>> >>>> This is to

Re: [Intel-gfx] [PATCH] drm/i915: Reserve space improvements

2015-06-26 Thread Dave Gordon
On 25/06/15 19:38, Tomas Elf wrote: > On 24/06/2015 18:03, john.c.harri...@intel.com wrote: >> From: John Harrison >> >> An earlier patch was added to reserve space in the ring buffer for the >> commands issued during 'add_request()'. The initial version was >> pessimistic in the way it handled bu

Re: [Intel-gfx] [PATCH 12/18] drm/i915/gtt: Fill scratch page

2015-06-26 Thread Dave Gordon
On 25/06/15 18:51, Chris Wilson wrote: > On Thu, Jun 25, 2015 at 06:35:14PM +0300, Mika Kuoppala wrote: >> During review of dynamic page tables series, I was able >> to hit a lite restore bug with execlists. I assume that >> due to incorrect pd, the batch run out of legit address space >> and into

Re: [Intel-gfx] [PATCH v4 4/6] drm/i915/gen8: Add WaFlushCoherentL3CacheLinesAtContextSwitch workaround

2015-06-26 Thread Dave Gordon
On 16/06/15 21:38, Chris Wilson wrote: > On Tue, Jun 16, 2015 at 08:25:23PM +0100, Arun Siluvery wrote: >> +/* WaFlushCoherentL3CacheLinesAtContextSwitch:bdw */ >> +if (IS_BROADWELL(ring->dev)) { >> +struct drm_i915_private *dev_priv = ring->dev->dev_private; > > dev_priv = to_

Re: [Intel-gfx] [PATCH libdrm] intel: Add EXEC_OBJECT_SUPPORTS_48BADDRESS flag.

2015-06-26 Thread Dave Gordon
On 23/06/15 13:21, Michel Thierry wrote: > Gen8+ supports 48-bit virtual addresses, but some objects must always be > allocated inside the 32-bit address range. > > In specific, any resource used with flat/heapless (0x-0xf000) > General State Heap (GSH) or Intruction State Heap (ISH) m

Re: [Intel-gfx] [PATCH] sna: Correct misspelled cacheing->caching in comments

2015-06-26 Thread Dave Gordon
On 24/06/15 11:04, Chris Wilson wrote: > On Tue, Jun 23, 2015 at 07:45:42AM +0200, Sedat Dilek wrote: >> I have seen this typo once and added an entry to codespell's dictionary.txt >> file. >> >> $ diff -uprN /usr/share/codespell/dictionary.txt.orig >> /usr/share/codespell/dictionary.txt >> --- /

Re: [Intel-gfx] [PATCH 09/17 v2] drm/i915: Expose two LRC functions for GuC submission mode

2015-06-26 Thread Dave Gordon
On 26/06/15 08:31, Chris Wilson wrote: > On Thu, Jun 25, 2015 at 01:57:16PM -0700, Yu Dai wrote: >> >> On 06/25/2015 07:40 AM, Dave Gordon wrote: >>> GuC submission is basically execlist submission, but with the GuC >>> handling the actual writes to the ELSP and

[Intel-gfx] [PATCH 17/17 v2] drm/i915: Enable GuC submission, where supported

2015-06-25 Thread Dave Gordon
Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_params.c |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c index a8f48dd..479c97b 100644 --- a/drivers/gpu/drm/i915/i915_params.c +++ b

[Intel-gfx] [PATCH 13/17 v2] drm/i915: Interrupt routing for GuC submission

2015-06-25 Thread Dave Gordon
Turn on interrupt steering to route necessary interrupts to GuC. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_reg.h | 11 +-- drivers/gpu/drm/i915/intel_guc_loader.c | 51 +++ 2 files changed, 60

[Intel-gfx] [PATCH 16/17 v2] Documentation/drm: kerneldoc for GuC

2015-06-25 Thread Dave Gordon
From: Alex Dai Add overview design of GuC, plus some key points related to the implementation. Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- Documentation/DocBook/drm.tmpl | 19 +++ 1 file changed, 19 insertions(+) diff --git a/Documentation/DocBook/drm.tmpl b

[Intel-gfx] [PATCH 15/17 v2] drm/i915: Debugfs interface for GuC submission statistics

2015-06-25 Thread Dave Gordon
This provides a means of reading status and counts relating to GuC actions and submissions. v2: Remove surplus blank line in output [Chris Wilson] Signed-off-by: Dave Gordon Signed-off-by: Alex Dai --- drivers/gpu/drm/i915/i915_debugfs.c | 40 +++ 1 file

[Intel-gfx] [PATCH 03/17 v2] drm/i915: Add GuC-related module parameters

2015-06-25 Thread Dave Gordon
From: Alex Dai Two new module parameters: "enable_guc_submission" which will turn on submission of batchbuffers via the GuC (when implemented), and "guc_log_level" which controls the level of debugging logged by the GuC and captured by the host. Signed-off-by: Alex Dai

[Intel-gfx] [PATCH 11/17 v2] drm/i915: Enable GuC firmware log

2015-06-25 Thread Dave Gordon
From: Alex Dai Allocate a GEM object to hold GuC log data. A debugfs interface (i915_guc_log_dump) is provided to print out the log content. v2: Add struct members at point of use [Chris Wilson] Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915

[Intel-gfx] [PATCH 14/17 v2] drm/i915: Integrate GuC-based command submission

2015-06-25 Thread Dave Gordon
tions [Chris Wilson] Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_debugfs.c|2 +- drivers/gpu/drm/i915/i915_guc_submission.c | 48 +--- drivers/gpu/drm/i915/intel_guc.h |1 + drivers/gpu/drm

[Intel-gfx] [PATCH 07/17 v2] drm/i915: Split late for_each_ring loop from i915_gem_init_hw()

2015-06-25 Thread Dave Gordon
the inline loop with a call. This will allow a future patch to add a call from another locations (for now, there are no other calls). Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_drv.h |1 + drivers/gpu/drm/i915/i915_gem.c | 85 ++- 2 files

[Intel-gfx] [PATCH 10/17 v2] drm/i915: GuC submission setup, phase 1

2015-06-25 Thread Dave Gordon
and the GuC's debug log. v2: Remove redundant initialisation [Chris Wilson] Defer adding struct members until needed [Chris Wilson] Local functions should pass dev_priv rather than dev [Chris Wilson] Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gp

[Intel-gfx] [PATCH 08/17 v2] drm/i915: Defer default hardware context initialisation until first open

2015-06-25 Thread Dave Gordon
erefore be in non-GuC mode, and will be allowed to complete even if the GuC cannot be loaded or used. Issue: VIZ-4884 Signed-off-by: Dave Gordon Signed-off-by: Alex Dai --- drivers/gpu/drm/i915/i915_drv.h |2 ++ drivers/gpu/drm/i915/i915_gem.c |8 - drivers/gp

[Intel-gfx] [PATCH 12/17 v2] drm/i915: Implementation of GuC client

2015-06-25 Thread Dave Gordon
ris Wilson] Rationalise type declarations [Chris Wilson] Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_guc_submission.c | 661 drivers/gpu/drm/i915/intel_guc.h | 43 ++ drivers/gpu/drm/i915/intel_guc_loader.c

[Intel-gfx] [PATCH 06/17 v2] drm/i915: Debugfs interface to read GuC load status

2015-06-25 Thread Dave Gordon
-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_debugfs.c | 45 +++ 1 file changed, 45 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index f3b8062..8f53f69 100644

[Intel-gfx] [PATCH 05/17 v2] drm/i915: GuC-specific firmware loader

2015-06-25 Thread Dave Gordon
ified firmware loading mechanism by Dave Gordon as well as new firmware layout etc. v2: Various improvements per review comments by Chris Wilson Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/Makefile |3 + drivers/gpu/drm/i915/i915_

[Intel-gfx] [PATCH 01/17 v2] drm/i915: Add i915_gem_object_create_from_data()

2015-06-25 Thread Dave Gordon
VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_drv.h |2 ++ drivers/gpu/drm/i915/i915_gem.c | 39 +++ 2 files changed, 41 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv

[Intel-gfx] [PATCH 09/17 v2] drm/i915: Expose two LRC functions for GuC submission mode

2015-06-25 Thread Dave Gordon
, changing their names to better describe what they do (they're related to logical ring contexts rather than to execlists per se). v2: Replaces previous "drm/i915: Move execlists defines from .c to .h" Issue: VIZ-4884 Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/int

[Intel-gfx] [PATCH 04/17 v2] drm/i915: Add GuC-related header files

2015-06-25 Thread Dave Gordon
f-by: Dave Gordon --- drivers/gpu/drm/i915/i915_guc_reg.h | 102 ++ drivers/gpu/drm/i915/intel_guc_fwif.h | 236 + 2 files changed, 338 insertions(+) create mode 100644 drivers/gpu/drm/i915/i915_guc_reg.h create mode 100644 drivers/gpu/drm/

[Intel-gfx] [PATCH 02/17 v2] drm/i915: Embedded microcontroller (uC) firmware loading support

2015-06-25 Thread Dave Gordon
used by each uC-specific loader, thus reducing code duplication and testing effort. v2: Local functions should pass dev_priv rather than dev [Chris Wilson] Various other improvements per Chris Wilson's review comments Signed-off-by: Dave Gordon Signed-off-by: Alex Dai --- drivers/g

[Intel-gfx] [PATCH 00/17 v2] Batch submission via GuC

2015-06-25 Thread Dave Gordon
mware log drm/i915: Integrate GuC-based command submission Documentation/drm: kerneldoc for GuC Dave Gordon (10): drm/i915: Add i915_gem_object_create_from_data() drm/i915: Embedded microcontroller (uC) firmware loading support drm/i915: Add GuC-related header files drm/i915: Split lat

Re: [Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_write() to i915_gem.c

2015-06-25 Thread Dave Gordon
On 24/06/15 10:32, Daniel Vetter wrote: > On Thu, Jun 18, 2015 at 07:28:26PM +0100, Dave Gordon wrote: >> On 18/06/15 15:31, Daniel Vetter wrote: >>> On Thu, Jun 18, 2015 at 12:49:55PM +0100, Dave Gordon wrote: >>>> On 17/06/15 13:02, Daniel Vetter wrote: >>&g

Re: [Intel-gfx] [PATCH 00/15] Batch submission via GuC

2015-06-25 Thread Dave Gordon
On 17/06/15 13:43, Daniel Vetter wrote: > On Mon, Jun 15, 2015 at 07:36:18PM +0100, Dave Gordon wrote: >> This patch series enables command submission via the GuC. In this mode, >> instead of the host CPU driving the execlist port directly, it hands >> over work items to the

Re: [Intel-gfx] [PATCH 14/15] drm/i915: Debugfs interface for GuC submission statistics

2015-06-24 Thread Dave Gordon
On 16/06/15 10:28, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:32PM +0100, Dave Gordon wrote: >> This provides a means of reading status and counts relating >> to GuC actions and submissions. > > Anything that ends to ease debugging also tends to ease > postmort

Re: [Intel-gfx] [PATCH 04/15] drm/i915: Add GuC-related header files

2015-06-24 Thread Dave Gordon
On 15/06/15 21:20, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:22PM +0100, Dave Gordon wrote: >> From: Alex Dai >> >> intel_guc_api.h contains the subset of the GuC interface that we >> will need for submission of commands through the GuC. These MUST

Re: [Intel-gfx] [PATCH 04/15] drm/i915: Add GuC-related header files

2015-06-23 Thread Dave Gordon
On 17/06/15 16:01, Dave Gordon wrote: > On 15/06/15 21:20, Chris Wilson wrote: >> On Mon, Jun 15, 2015 at 07:36:22PM +0100, Dave Gordon wrote: >>> From: Alex Dai >>> >>> intel_guc_api.h contains the subset of the GuC interface that we >>> will ne

Re: [Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_write() to i915_gem.c

2015-06-23 Thread Dave Gordon
On 22/06/15 13:37, Chris Wilson wrote: > On Mon, Jun 22, 2015 at 12:59:00PM +0100, Dave Gordon wrote: >> On 19/06/15 09:44, Chris Wilson wrote: >>> On Thu, Jun 18, 2015 at 07:07:46PM +0100, Dave Gordon wrote: >>>> On 18/06/15 13:10, Chris Wilson wrote: >>&g

Re: [Intel-gfx] [PATCH 12/15] drm/i915: Interrupt routing for GuC submission

2015-06-23 Thread Dave Gordon
On 17/06/15 13:41, Daniel Vetter wrote: > On Wed, Jun 17, 2015 at 02:22:19PM +0200, Daniel Vetter wrote: >> On Wed, Jun 17, 2015 at 09:20:44AM +0100, Dave Gordon wrote: >>> On 16/06/15 10:24, Chris Wilson wrote: >>>> On Mon, Jun 15, 2015 at 07:36:30PM +0100, Dave Go

Re: [Intel-gfx] [PATCH] drm/i915: Avoid GPU stalls from kswapd

2015-06-22 Thread Dave Gordon
On 22/06/15 16:35, Daniel Vetter wrote: > On Mon, Jun 22, 2015 at 03:18:12PM +0100, Chris Wilson wrote: >> On Mon, Jun 22, 2015 at 04:11:01PM +0200, Daniel Vetter wrote: >>> On Fri, Jun 19, 2015 at 02:04:16PM +0100, Chris Wilson wrote: Exclude active GPU pages from the purview of the backgroun

Re: [Intel-gfx] [PATCH 08/15] drm/i915: Move execlists defines from .c to .h

2015-06-22 Thread Dave Gordon
On 17/06/15 08:59, Chris Wilson wrote: > On Wed, Jun 17, 2015 at 08:31:59AM +0100, Dave Gordon wrote: >> On 16/06/15 10:37, Chris Wilson wrote: >>> On Mon, Jun 15, 2015 at 07:36:26PM +0100, Dave Gordon wrote: >>>> From: "Michael H. Nguyen" >>>>

Re: [Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_write() to i915_gem.c

2015-06-22 Thread Dave Gordon
On 19/06/15 09:44, Chris Wilson wrote: > On Thu, Jun 18, 2015 at 07:07:46PM +0100, Dave Gordon wrote: >> On 18/06/15 13:10, Chris Wilson wrote: >>> On Thu, Jun 18, 2015 at 12:49:55PM +0100, Dave Gordon wrote: >>>> On 17/06/15 13:02, Daniel Vetter wrote: >>>

Re: [Intel-gfx] [PATCH 13/15] drm/i915: Integrate GuC-based command submission

2015-06-19 Thread Dave Gordon
On 16/06/15 10:22, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:31PM +0100, Dave Gordon wrote: >> From: Alex Dai >> >> GuC-based submission is mostly the same as execlist mode, up to >> intel_logical_ring_advance_and_submit(), where the context being >>

Re: [Intel-gfx] [PATCH 11/15] drm/i915: Implementation of GuC client

2015-06-19 Thread Dave Gordon
On 15/06/15 22:55, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:29PM +0100, Dave Gordon wrote: >> +/* Get valid workqueue item and return it back to offset */ >> +static int guc_get_workqueue_space(struct i915_guc_client *gc, u32 *offset) >> +{ >> +st

Re: [Intel-gfx] [PATCH 09/15] drm/i915: GuC submission setup, phase 1

2015-06-19 Thread Dave Gordon
On 19/06/15 18:02, Dave Gordon wrote: > On 15/06/15 22:32, Chris Wilson wrote: [snip] >> Try to keep comments to explain why rather than what. Most of the >> comments here fall into the "i++; // postincrement i" category. >> -Chris > > Most of the "

Re: [Intel-gfx] [PATCH 09/15] drm/i915: GuC submission setup, phase 1

2015-06-19 Thread Dave Gordon
On 15/06/15 22:32, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:27PM +0100, Dave Gordon wrote: >> +static struct drm_i915_gem_object *gem_allocate_guc_obj(struct drm_device >> *dev, >> +u32 size) >> +{ >>

Re: [Intel-gfx] [PATCH 05/15] drm/i915: GuC-specific firmware loader

2015-06-19 Thread Dave Gordon
On 18/06/15 21:12, Chris Wilson wrote: > On Thu, Jun 18, 2015 at 10:53:10AM -0700, Yu Dai wrote: >> >> >> On 06/15/2015 01:30 PM, Chris Wilson wrote: >>> On Mon, Jun 15, 2015 at 07:36:23PM +0100, Dave Gordon wrote: >>>> + /* Set the source

Re: [Intel-gfx] [PATCH 07/15] drm/i915: Defer default hardware context initialisation until first open

2015-06-19 Thread Dave Gordon
On 16/06/15 10:35, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:25PM +0100, Dave Gordon wrote: >> +static int i915_gem_context_first_open(struct drm_device *dev) >> +{ >> +struct drm_i915_private *dev_priv = dev->dev_private; >> +int ret; >> + &

Re: [Intel-gfx] [PATCH 07/15] drm/i915: Defer default hardware context initialisation until first open

2015-06-19 Thread Dave Gordon
On 17/06/15 13:18, Daniel Vetter wrote: > On Mon, Jun 15, 2015 at 07:36:25PM +0100, Dave Gordon wrote: >> In order to fully initialise the default contexts, we have to execute >> batchbuffer commands on the GPU engines. But in the case of GuC-based >> batch submission, we ca

Re: [Intel-gfx] [PATCH 02/15] drm/i915: Embedded microcontroller (uC) firmware loading support

2015-06-19 Thread Dave Gordon
On 18/06/15 15:49, Daniel Vetter wrote: > On Thu, Jun 18, 2015 at 01:11:34PM +0100, Dave Gordon wrote: >> On 17/06/15 13:05, Daniel Vetter wrote: >>> On Mon, Jun 15, 2015 at 07:36:20PM +0100, Dave Gordon wrote: >>>> Current devices may contain one or more programma

Re: [Intel-gfx] [PATCH 06/15] drm/i915: Debugfs interface to read GuC load status

2015-06-19 Thread Dave Gordon
On 16/06/15 10:40, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:24PM +0100, Dave Gordon wrote: >> From: Alex Dai >> >> The new node provides access to the status of the common uC loader >> code and the GuC-specific loader; also the scratch registers used >

Re: [Intel-gfx] [PATCH 05/15] drm/i915: GuC-specific firmware loader

2015-06-18 Thread Dave Gordon
On 15/06/15 21:30, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:23PM +0100, Dave Gordon wrote: >> +/* We can't enable contexts until all firmware is loaded */ >> +ret = intel_guc_ucode_load(dev, false); > > Pardon. I know context initialisation is

Re: [Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_write() to i915_gem.c

2015-06-18 Thread Dave Gordon
On 18/06/15 15:31, Daniel Vetter wrote: > On Thu, Jun 18, 2015 at 12:49:55PM +0100, Dave Gordon wrote: >> On 17/06/15 13:02, Daniel Vetter wrote: >>> On Wed, Jun 17, 2015 at 08:23:40AM +0100, Dave Gordon wrote: >>>> On 15/06/15 21:09, Chris Wilson wrote: >>&g

Re: [Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_write() to i915_gem.c

2015-06-18 Thread Dave Gordon
On 18/06/15 13:10, Chris Wilson wrote: > On Thu, Jun 18, 2015 at 12:49:55PM +0100, Dave Gordon wrote: >> On 17/06/15 13:02, Daniel Vetter wrote: >>> Domain handling is required for all gem objects, and the resulting bugs if >>> you don't for one-off objects are

Re: [Intel-gfx] [PATCH 02/15] drm/i915: Embedded microcontroller (uC) firmware loading support

2015-06-18 Thread Dave Gordon
On 17/06/15 13:05, Daniel Vetter wrote: > On Mon, Jun 15, 2015 at 07:36:20PM +0100, Dave Gordon wrote: >> Current devices may contain one or more programmable microcontrollers >> that need to have a firmware image (aka "binary blob") loaded from an >> externa

Re: [Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_write() to i915_gem.c

2015-06-18 Thread Dave Gordon
On 17/06/15 13:02, Daniel Vetter wrote: > On Wed, Jun 17, 2015 at 08:23:40AM +0100, Dave Gordon wrote: >> On 15/06/15 21:09, Chris Wilson wrote: >>> On Mon, Jun 15, 2015 at 07:36:19PM +0100, Dave Gordon wrote: >>>> From: Alex Dai >>>> >>>>

Re: [Intel-gfx] [RFC 03/11] drm/i915: Add reset stats entry point for per-engine reset.

2015-06-18 Thread Dave Gordon
On 16/06/15 14:54, Chris Wilson wrote: > On Tue, Jun 16, 2015 at 03:48:09PM +0200, Daniel Vetter wrote: >> On Mon, Jun 08, 2015 at 06:33:59PM +0100, Chris Wilson wrote: >>> On Mon, Jun 08, 2015 at 06:03:21PM +0100, Tomas Elf wrote: In preparation for per-engine reset add way for setting contex

Re: [Intel-gfx] (no subject)

2015-06-18 Thread Dave Gordon
On 17/06/15 12:04, Daniel Vetter wrote: > On Fri, Jun 12, 2015 at 09:25:36PM +0100, Dave Gordon wrote: >> Updated version split into two. The first tidies up the _ring_prepare() >> functions and removes the corner case where we might have had to wait >> twice; the second is a

Re: [Intel-gfx] [PATCH 04/15] drm/i915: Add GuC-related header files

2015-06-17 Thread Dave Gordon
On 15/06/15 21:20, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:22PM +0100, Dave Gordon wrote: >> From: Alex Dai >> >> intel_guc_api.h contains the subset of the GuC interface that we >> will need for submission of commands through the GuC. These MUST

Re: [Intel-gfx] [PATCH 12/15] drm/i915: Interrupt routing for GuC submission

2015-06-17 Thread Dave Gordon
On 16/06/15 10:24, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:30PM +0100, Dave Gordon wrote: >> +static void direct_interrupts_to_guc(struct drm_i915_private *dev_priv) >> +{ >> +struct intel_engine_cs *ring; >> +int i, irqs; >> + >> +/

Re: [Intel-gfx] [PATCH 08/15] drm/i915: Move execlists defines from .c to .h

2015-06-17 Thread Dave Gordon
On 16/06/15 10:37, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:26PM +0100, Dave Gordon wrote: >> From: "Michael H. Nguyen" >> >> Move defines from intel_lrc.c to i915_reg.h so they are accessible >> to the GuC submission code; and expose a previous

Re: [Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_write() to i915_gem.c

2015-06-17 Thread Dave Gordon
On 15/06/15 21:09, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:36:19PM +0100, Dave Gordon wrote: >> From: Alex Dai >> >> i915_gem_object_write() is a generic function to copy data from a plain >> linear buffer to a paged gem object. >> >> We will nee

Re: [Intel-gfx] [RFC] drm/i915: prevent out of range pt in the PDE macros (take 2)

2015-06-16 Thread Dave Gordon
On 15/06/15 11:53, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 11:33:37AM +0100, Dave Gordon wrote: >> On 13/06/15 09:28, Chris Wilson wrote: >>> On Fri, Jun 12, 2015 at 06:30:56PM -0300, Paulo Zanoni wrote: >>>> From: Paulo Zanoni >>>> >>

Re: [Intel-gfx] [PATCH] drm/i915: Reinstate order of operations in {intel, logical}_ring_begin()

2015-06-16 Thread Dave Gordon
On 15/06/15 21:41, Chris Wilson wrote: > On Mon, Jun 15, 2015 at 07:11:37PM +0100, Dave Gordon wrote: >>> It still applies. If you submit say 1024 interrupted execbuffers they >> >> What is an interrupted execbuffer? AFAICT we hold the struct_mutex while >> stuffing

Re: [Intel-gfx] [PATCH] drm/i915/skl: replace csr_mutex by completion in csr firmware loading

2015-06-15 Thread Dave Gordon
On 15/06/15 11:07, Daniel Vetter wrote: > On Thu, Jun 04, 2015 at 03:36:32PM +0100, Dave Gordon wrote: >> On 04/06/15 06:59, Sagar Arun Kamble wrote: >>> >>> Hi Daniel, >>> >>> We already are grabbing RPM reference before start of DMC FW load and >&g

[Intel-gfx] [PATCH 15/15] Documentation/drm: kerneldoc for GuC

2015-06-15 Thread Dave Gordon
From: Alex Dai Add overview design of GuC, plus some key points related to the implementation. Signed-off-by: Alex Dai --- Documentation/DocBook/drm.tmpl | 19 +++ 1 file changed, 19 insertions(+) diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl i

[Intel-gfx] [PATCH 16/15] drm/i915: Enable GuC submission, where supported

2015-06-15 Thread Dave Gordon
Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_params.c |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c index 5134095..926a6df 100644 --- a/drivers/gpu/drm/i915/i915_params.c +++ b

[Intel-gfx] [PATCH 13/15] drm/i915: Integrate GuC-based command submission

2015-06-15 Thread Dave Gordon
e GuC, the GuC firmware requires the ring-tail-offset to be represented as an 11-bit value, expressed in QWords. Therefore, the ringbuffer size must be reduced to the representable range (4 pages). Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm

[Intel-gfx] [PATCH 14/15] drm/i915: Debugfs interface for GuC submission statistics

2015-06-15 Thread Dave Gordon
This provides a means of reading status and counts relating to GuC actions and submissions. Signed-off-by: Dave Gordon Signed-off-by: Alex Dai --- drivers/gpu/drm/i915/i915_debugfs.c | 41 +++ 1 file changed, 41 insertions(+) diff --git a/drivers/gpu/drm/i915

[Intel-gfx] [PATCH 06/15] drm/i915: Debugfs interface to read GuC load status

2015-06-15 Thread Dave Gordon
From: Alex Dai The new node provides access to the status of the common uC loader code and the GuC-specific loader; also the scratch registers used for communicatio between the i915 driver and the GuC firmware. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers

[Intel-gfx] [PATCH 08/15] drm/i915: Move execlists defines from .c to .h

2015-06-15 Thread Dave Gordon
From: "Michael H. Nguyen" Move defines from intel_lrc.c to i915_reg.h so they are accessible to the GuC submission code; and expose a previously static function in the execlist code which will also be required for GuC submission. Issue: VIZ-4884 Signed-off-by: Michael H. Nguyen Signed-off-by: A

[Intel-gfx] [PATCH 12/15] drm/i915: Interrupt routing for GuC submission

2015-06-15 Thread Dave Gordon
Turn on interrupt steering to route necessary interrupts to GuC. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_reg.h | 11 +-- drivers/gpu/drm/i915/intel_guc_loader.c | 51 +++ 2 files changed, 60

[Intel-gfx] [PATCH 11/15] drm/i915: Implementation of GuC client

2015-06-15 Thread Dave Gordon
Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_guc_submission.c | 668 drivers/gpu/drm/i915/intel_guc.h |5 + drivers/gpu/drm/i915/intel_guc_loader.c| 10 + 3 files changed, 683 insertions(+) diff --git a/drivers/gpu/drm/i915

[Intel-gfx] [PATCH 10/15] drm/i915: Enable GuC firmware log

2015-06-15 Thread Dave Gordon
From: Alex Dai Allocate a GEM object to hold GuC log data. A debugfs interface (i915_guc_log_dump) is provided to print out the log content. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_debugfs.c| 29 +++ drivers

[Intel-gfx] [PATCH 09/15] drm/i915: GuC submission setup, phase 1

2015-06-15 Thread Dave Gordon
and the GuC's debug log. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/Makefile |3 +- drivers/gpu/drm/i915/i915_guc_submission.c | 122 drivers/gpu/drm/i915/intel_guc.h |4 + d

[Intel-gfx] [PATCH 05/15] drm/i915: GuC-specific firmware loader

2015-06-15 Thread Dave Gordon
ified firmware loading mechanism by Dave Gordon as well as new firmware layout etc. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/Makefile |3 + drivers/gpu/drm/i915/i915_dma.c |4 + drivers/gpu/drm/i915/i915_drv.h |

[Intel-gfx] [PATCH 07/15] drm/i915: Defer default hardware context initialisation until first open

2015-06-15 Thread Dave Gordon
erefore be in non-GuC mode, and will be allowed to complete even if the GuC cannot be loaded or used. Issue: VIZ-4884 Signed-off-by: Dave Gordon Signed-off-by: Alex Dai --- drivers/gpu/drm/i915/i915_drv.h |2 ++ drivers/gpu/drm/i915/i915_gem.c |9 +- drivers/gp

[Intel-gfx] [PATCH 02/15] drm/i915: Embedded microcontroller (uC) firmware loading support

2015-06-15 Thread Dave Gordon
used by each uC-specific loader, thus reducing code duplication and testing effort. Signed-off-by: Dave Gordon Signed-off-by: Alex Dai --- drivers/gpu/drm/i915/Makefile |3 + drivers/gpu/drm/i915/intel_uc_loader.c | 312 drivers/gpu/drm/i915/intel_uc_

[Intel-gfx] [PATCH 04/15] drm/i915: Add GuC-related header files

2015-06-15 Thread Dave Gordon
setting it running. Some of these also need to be kept in sync with the firmware. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915/i915_reg.h |4 +- drivers/gpu/drm/i915/intel_guc.h | 169 + drivers/gpu/drm/i915

[Intel-gfx] [PATCH 03/15] drm/i915: Add GuC-related module parameters

2015-06-15 Thread Dave Gordon
From: Alex Dai Two new module parameters: "enable_guc_submission" which will turn on submission of batchbuffers via the GuC (when implemented), and "guc_log_level" which controls the level of debugging logged by the GuC and captured by the host. Signed-off-by: Alex Dai

[Intel-gfx] [PATCH 01/15] drm/i915: Add i915_gem_object_write() to i915_gem.c

2015-06-15 Thread Dave Gordon
From: Alex Dai i915_gem_object_write() is a generic function to copy data from a plain linear buffer to a paged gem object. We will need this for the microcontroller firmware loading support code. Issue: VIZ-4884 Signed-off-by: Alex Dai Signed-off-by: Dave Gordon --- drivers/gpu/drm/i915

[Intel-gfx] [PATCH 00/15] Batch submission via GuC

2015-06-15 Thread Dave Gordon
log drm/i915: Implementation of GuC client drm/i915: Integrate GuC-based command submission Documentation/drm: kerneldoc for GuC Dave Gordon (5): drm/i915: Embedded microcontroller (uC) firmware loading support drm/i915: Defer default hardware context initialisation until first drm/i915: Interru

<    5   6   7   8   9   10   11   12   >