Re: [Intel-gfx] [PATCH v3 04/14] drm/i915/guc: Move doc near related definitions

On 10/13/2017 4:24 AM, Michal Wajdeczko wrote: After Guc code reorg some documentation was left in wrong place. Move it closer to corresponding definitions. Signed-off-by: Michal Wajdeczko Cc: Joonas Lahtinen Cc: Sagar Arun Kamble

[Intel-gfx] ✓ Fi.CI.IGT: success for extract cork and ring measure code

== Series Details == Series: extract cork and ring measure code URL : https://patchwork.freedesktop.org/series/31858/ State : success == Summary == Test gem_flink_race: Subgroup flink_close: pass -> FAIL (shard-hsw) fdo#102655 fdo#102655

Re: [Intel-gfx] [PATCH v3 02/14] drm/i915/guc: Move GuC boot param initialization out of xfer

Ignore checkpatch related review. Saw the fixes in 03/14 On 10/13/2017 10:43 AM, Sagar Arun Kamble wrote: On 10/13/2017 4:24 AM, Michal Wajdeczko wrote: We want to keep ucode xfer functions separate from other initialization. Once separated, add explicit forcewake. v2: use BLITTER domain

Re: [Intel-gfx] [PATCH v3 02/14] drm/i915/guc: Move GuC boot param initialization out of xfer

On 10/13/2017 4:24 AM, Michal Wajdeczko wrote: We want to keep ucode xfer functions separate from other initialization. Once separated, add explicit forcewake. v2: use BLITTER domain only and add comment (Daniele) Suggested-by: Joonas Lahtinen Signed-off-by:

[Intel-gfx] ✓ Fi.CI.IGT: success for series starting with [v2,1/2] drm/i915: Fix DP_AUX_CH_CTL_TIME_OUT naming

== Series Details == Series: series starting with [v2,1/2] drm/i915: Fix DP_AUX_CH_CTL_TIME_OUT naming URL : https://patchwork.freedesktop.org/series/31856/ State : success == Summary == Test perf: Subgroup blocking: pass -> FAIL (shard-hsw) fdo#102252

[Intel-gfx] ✗ Fi.CI.IGT: failure for drm/i915/edp: Do not do link training fallback or prune modes on EDP (rev2)

== Series Details == Series: drm/i915/edp: Do not do link training fallback or prune modes on EDP (rev2) URL : https://patchwork.freedesktop.org/series/31776/ State : failure == Summary == Test gem_userptr_blits: Subgroup sync-unmap-cycles: dmesg-warn -> PASS

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: enable to read CSB and CSB write pointer from HWSP in GVT-g VM (rev2)

== Series Details == Series: drm/i915: enable to read CSB and CSB write pointer from HWSP in GVT-g VM (rev2) URL : https://patchwork.freedesktop.org/series/31789/ State : success == Summary == Series 31789v2 drm/i915: enable to read CSB and CSB write pointer from HWSP in GVT-g VM

[Intel-gfx] [PATCH v5] drm/i915: enable to read CSB and CSB write pointer from HWSP in GVT-g VM

Let GVT-g VM read the CSB and CSB write pointer from virtual HWSP, not all the host support this feature, need to check the BIT(3) of caps in PVINFO. v3 : Remove unnecessary comments. v4 : Separate VM enable patch with GVT-g implementation patch due to code dependency. v5 : Use inline for GVT

Re: [Intel-gfx] [PATCH v4] drm/i915: enable to read CSB and CSB write pointer from HWSP in GVT-g VM

On 10/12/2017 4:16 PM, Chris Wilson wrote: Quoting Weinan Li (2017-10-12 07:50:08) Let GVT-g VM read the CSB and CSB write pointer from virtual HWSP, not all the host support this feature, need to check the BIT(3) of caps in PVINFO. v3 : Remove unnecessary comments. v4 : Separate VM enable

[Intel-gfx] [PATCH] drm/plane: drop num_overlay_planes (v2)

From: Dave Airlie In order to implement plane leasing we need to count things, just make the code consistent with the counting code currently used for counting crtcs/encoders/connectors and drop the need for num_overlay_planes. v2: don't forget to assign plane_ptr. (keithp)

Re: [Intel-gfx] [PATCH 02/12] drm/i915/guc: Allocate separate shared data object for GuC communication

On 12/10/17 13:35, Michel Thierry wrote: On 09/10/17 15:35, Michel Thierry wrote: On 10/9/2017 11:41 AM, Daniele Ceraolo Spurio wrote: On 09/10/17 07:52, Michał Winiarski wrote: We were using first page of kernel context render state for sharing data with GuC. While it's justified by the

[Intel-gfx] [PATCH] drm/plane: drop num_overlay_planes

From: Dave Airlie In order to implement plane leasing we need to count things, just make the code consistent with the counting code currently used for counting crtcs/encoders/connectors and drop the need for num_overlay_planes. Signed-off-by: Dave Airlie

[Intel-gfx] ✓ Fi.CI.IGT: success for lib/igt_gt: Allow non-default contexts to hang non-render rings (rev5)

== Series Details == Series: lib/igt_gt: Allow non-default contexts to hang non-render rings (rev5) URL : https://patchwork.freedesktop.org/series/31693/ State : success == Summary == Test kms_busy: Subgroup extended-modeset-hang-oldfb-with-reset-render-C: dmesg-warn

[Intel-gfx] ✗ Fi.CI.BAT: warning for lib/igt_gt: Allow non-default contexts to hang non-render rings (rev6)

== Series Details == Series: lib/igt_gt: Allow non-default contexts to hang non-render rings (rev6) URL : https://patchwork.freedesktop.org/series/31693/ State : warning == Summary == IGT patchset tested on top of latest successful build 58616272b23efce1e62a3ee0d37e13de6ffc012f igt/gem_eio:

[Intel-gfx] ✗ Fi.CI.IGT: failure for lib/igt_kms: Rewrite property handling to better match atomic. (rev6)

== Series Details == Series: lib/igt_kms: Rewrite property handling to better match atomic. (rev6) URL : https://patchwork.freedesktop.org/series/31733/ State : failure == Summary == Test kms_atomic: Subgroup test_only: skip -> PASS (shard-hsw) Test perf:

Re: [Intel-gfx] [PATCH i-g-t v6] lib/igt_gt: Allow non-default contexts to hang non-render rings

Quoting Vinay Belgaumkar (2017-10-13 00:22:17) > +static bool has_ctx_exec(int fd, unsigned ring, uint32_t ctx) > +{ > + struct drm_i915_gem_execbuffer2 execbuf; > + struct drm_i915_gem_exec_object2 exec; > + > + /* silly ABI, the kernel thinks everyone who has BSD also has BSD2

[Intel-gfx] ✗ Fi.CI.BAT: failure for Move imported bo pread/pwrite tests

== Series Details == Series: Move imported bo pread/pwrite tests URL : https://patchwork.freedesktop.org/series/31860/ State : failure == Summary == Series 31860 revision 1 was fully merged or fully failed: no git log ___ Intel-gfx mailing list

[Intel-gfx] ✗ Fi.CI.BAT: failure for Firmware code reorg (rev2)

== Series Details == Series: Firmware code reorg (rev2) URL : https://patchwork.freedesktop.org/series/31846/ State : failure == Summary == Series 31846v2 Firmware code reorg https://patchwork.freedesktop.org/api/1.0/series/31846/revisions/2/mbox/ Test chamelium: Subgroup

Re: [Intel-gfx] [PATCH v3 10/14] drm/i915/uc: Add message with firmware url

Quoting Michal Wajdeczko (2017-10-12 23:54:43) > diff --git a/drivers/gpu/drm/i915/intel_uc_fw.h > b/drivers/gpu/drm/i915/intel_uc_fw.h > index 5c01849..1d580fa 100644 > --- a/drivers/gpu/drm/i915/intel_uc_fw.h > +++ b/drivers/gpu/drm/i915/intel_uc_fw.h > @@ -27,6 +27,8 @@ > > struct

[Intel-gfx] [PATCH i-g-t v6] lib/igt_gt: Allow non-default contexts to hang non-render rings

This limitation does not exist in latest kernel. It was removed by this patch- commit f7978a0c581a8a840a28306f8da43e06e7fef3bf v2: Added commit id that removes the limitation(Chris Wilson) V3: Generic way to find if kernel supports this instead of hardcoding gens(Chris Wilson) v4: Optimize the

Re: [Intel-gfx] [PATCH i-g-t 0/8] extract cork and ring measure code

Quoting Daniele Ceraolo Spurio (2017-10-12 23:27:26) > The "cork" bo used to stall execution and the interruptible execbuf > used to measure the ring size are repeated almost identically in > several files. Extracting those to common files helps simplifying > the code and avoiding repetition. If

[Intel-gfx] ✓ Fi.CI.BAT: success for Drop tests for object creation from stolen (rev2)

== Series Details == Series: Drop tests for object creation from stolen (rev2) URL : https://patchwork.freedesktop.org/series/31290/ State : success == Summary == IGT patchset tested on top of latest successful build 58616272b23efce1e62a3ee0d37e13de6ffc012f igt/gem_eio: Check hang/eio

[Intel-gfx] [PATCH] tracing, drm/i915: Remove unused trace events

From: Steven Rostedt (VMware) Commit ce8daef3580 ("drm/i915: Remove dead i915_gem_evict_everything()") removed the only instances of trace_i915_gem_evict_everything. Commit a7b9761d0a ("drm/i915: Split i915_gem_flush_ring() into seperate invalidate/flush funcs") removed the

Re: [Intel-gfx] [PATCH i-g-t 1/8] lib/igt_dummyload: add igt_cork

Quoting Daniele Ceraolo Spurio (2017-10-12 23:27:27) > +igt_cork_t *igt_cork_new(int fd); _new does not imply plugged. > +void igt_cork_signal(igt_cork_t *cork); When have you signaled a cork? > +void igt_cork_free(int fd, igt_cork_t *cork); _free does not imply unplug. -Chris

[Intel-gfx] [PATCH v3 11/14] drm/i915/uc: Unify firmware loading

Firmware loading for GuC and Huc are similar. Move common code into generic function for maximum reuse. v2: change message levels (Chris) Signed-off-by: Michal Wajdeczko Cc: Joonas Lahtinen Cc: Sagar Arun Kamble

[Intel-gfx] [PATCH v3 07/14] drm/i915/guc: Move firmware size check out of generic code

Checking GuC firmware size can be done in GuC specific code right before DMA copy as it is unlikely to fail anyway. v2: rebased Signed-off-by: Michal Wajdeczko Cc: Sagar Arun Kamble Cc: Joonas Lahtinen

[Intel-gfx] [PATCH v3 12/14] drm/i915/guc: Update Guc messages on load failure

In case of GuC firmware loading failure we were reporting DRM_ERROR also for case when GuC loading was not strictly required. Signed-off-by: Michal Wajdeczko Cc: Chris Wilson Cc: Joonas Lahtinen Cc: Sagar

[Intel-gfx] [PATCH v3 10/14] drm/i915/uc: Add message with firmware url

In case of firmware fetch failure we should help user find latest firmware. Suggested-by: Chris Wilson Signed-off-by: Michal Wajdeczko Cc: Chris Wilson --- drivers/gpu/drm/i915/intel_uc_fw.c | 2 ++

[Intel-gfx] [PATCH v3 13/14] drm/i915/huc: Move fw select function

No functional change. Just place the function closer to related definitions. Signed-off-by: Michal Wajdeczko Cc: Joonas Lahtinen Cc: Sagar Arun Kamble Reviewed-by: Chris Wilson ---

Re: [Intel-gfx] [PATCH i-g-t 1/8] lib/igt_dummyload: add igt_cork

Quoting Daniele Ceraolo Spurio (2017-10-12 23:27:27) > +typedef struct igt_cork { > + int device; > + uint32_t handle; > + uint32_t fence; > +} igt_cork_t; Oh, just struct no need for typedef when you are transparent. -Chris ___

[Intel-gfx] [PATCH v3 14/14] HAX enable GuC submission for CI

Also revert ("drm/i915/guc: Assert that we switch between known ggtt->invalidate functions") Signed-off-by: Michal Wajdeczko --- drivers/gpu/drm/i915/i915_gem_gtt.c | 8 ++-- drivers/gpu/drm/i915/i915_params.h | 4 ++-- 2 files changed, 4 insertions(+), 8

[Intel-gfx] [PATCH v3 04/14] drm/i915/guc: Move doc near related definitions

After Guc code reorg some documentation was left in wrong place. Move it closer to corresponding definitions. Signed-off-by: Michal Wajdeczko Cc: Joonas Lahtinen Cc: Sagar Arun Kamble ---

[Intel-gfx] [PATCH v3 09/14] drm/i915/uc: Improve debug messages in firmware fetch

Time to remove less important info and make messages clear and consistent. v2: change some message levels (Chris) v3: restore DRM_WARN (Chris) Signed-off-by: Michal Wajdeczko Cc: Sagar Arun Kamble Cc: Joonas Lahtinen

[Intel-gfx] [PATCH v3 06/14] drm/i915/guc: Reorder functions in intel_guc_fw.c

Functions should be defined in their use order. Signed-off-by: Michal Wajdeczko Cc: Joonas Lahtinen Reviewed-by: Chris Wilson --- drivers/gpu/drm/i915/intel_guc_fw.c | 81

[Intel-gfx] [PATCH v3 08/14] drm/i915/guc: Pick better place for Guc final status message

Guc status message printed right after firmware upload may be too optimistic, as we may fail on subsequent steps. Move that message to the end of intel_uc_init_hw where we know the status for sure. v2: use dev_info (Chris) Signed-off-by: Michal Wajdeczko Cc: Joonas

[Intel-gfx] [PATCH v3 05/14] drm/i915/guc: Rename intel_guc_loader.c to intel_guc_fw.c

Remaining functions in intel_guc_loader.c were focused around GuC firmware. Rename them to match object-verb pattern and rename file itself. Suggested-by: Joonas Lahtinen Signed-off-by: Michal Wajdeczko Cc: Joonas Lahtinen

[Intel-gfx] [PATCH v3 00/14] Firmware code reorg

Suggested-by: Joonas Lahtinen Cc: Chris Wilson v1: was sent with wrong prefix "P v4" v2: apply review comments from Chris/Daniele v3: restore DRM_WARN, add firmware url Michal Wajdeczko (14): drm/i915: Move intel_guc_wopcm_size to

[Intel-gfx] [PATCH v3 01/14] drm/i915: Move intel_guc_wopcm_size to intel_guc.c

Function intel_guc_wopcm_size didn't fit well in the old place. With this move upcoming cleanup will be simpler. Suggested-by: Joonas Lahtinen Signed-off-by: Michal Wajdeczko Cc: Joonas Lahtinen Cc:

[Intel-gfx] [PATCH v3 02/14] drm/i915/guc: Move GuC boot param initialization out of xfer

We want to keep ucode xfer functions separate from other initialization. Once separated, add explicit forcewake. v2: use BLITTER domain only and add comment (Daniele) Suggested-by: Joonas Lahtinen Signed-off-by: Michal Wajdeczko Cc:

[Intel-gfx] [PATCH v3 03/14] drm/i915/guc: Small fixups post code move

Existing code needs some style fixes. To avoid polluting pure move patch, do it now as separate step. Suggested-by: Chris Wilson Signed-off-by: Michal Wajdeczko Cc: Chris Wilson Reviewed-by: Chris Wilson

Re: [Intel-gfx] [PATCH i-g-t 1/8] lib/igt_dummyload: add igt_cork

Quoting Daniele Ceraolo Spurio (2017-10-12 23:27:27) > diff --git a/lib/igt_dummyload.h b/lib/igt_dummyload.h > index 215425f..d20a867 100644 > --- a/lib/igt_dummyload.h > +++ b/lib/igt_dummyload.h > @@ -29,6 +29,7 @@ > #include > > #include "igt_aux.h" > +#include "igt_vgem.h" > >

[Intel-gfx] ✓ Fi.CI.BAT: success for extract cork and ring measure code

== Series Details == Series: extract cork and ring measure code URL : https://patchwork.freedesktop.org/series/31858/ State : success == Summary == IGT patchset tested on top of latest successful build 58616272b23efce1e62a3ee0d37e13de6ffc012f igt/gem_eio: Check hang/eio recovery during

Re: [Intel-gfx] [PATCH i-g-t 0/4] Move imported bo pread/pwrite tests

Quoting Daniele Ceraolo Spurio (2017-10-12 23:30:40) > We already had coverage in prime_vgem.c, but since we're testing > pread/pwrite ioctl it makes more sense to have it in the corresponding > test binaries Redundancy is not an issue (sometimes it is a virtue, the slight variations in test

Re: [Intel-gfx] [PATCH i-g-t 0/4] Move imported bo pread/pwrite tests

On 12/10/17 15:30, Daniele Ceraolo Spurio wrote: We already had coverage in prime_vgem.c, but since we're testing pread/pwrite ioctl it makes more sense to have it in the corresponding test binaries Forgot to mention that this series depends on both the other series I've sent Daniele

Re: [Intel-gfx] [i-g-t 2/2] tests/gem_reset_stats: Add client ban test

Quoting Antonio Argenziano (2017-10-12 23:10:37) > When a client gets a fixed amount of its contexts banned, will be > blocked from creating any new contexts itself. This patch adds a test > for this feature. Fixed by whom? The "feature" is a primitive DoS prevention. Atm it is purely

[Intel-gfx] [PATCH i-g-t 1/4] lib/igt_vgem: extract vgem_create_and_import

The same code to create and import a vgem object is used in a couple of places and a couple more are coming up in the next patches so extract the code into a common function Signed-off-by: Daniele Ceraolo Spurio --- lib/igt_dummyload.c | 10 ++

[Intel-gfx] [PATCH i-g-t 4/4] tests/prime_vgem: drop basic-read/write subtests

the same coverage is now in gem_pread and gem_pwrite Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson --- tests/prime_vgem.c | 56 -- 1 file changed, 56 deletions(-) diff --git

[Intel-gfx] [PATCH i-g-t 0/4] Move imported bo pread/pwrite tests

We already had coverage in prime_vgem.c, but since we're testing pread/pwrite ioctl it makes more sense to have it in the corresponding test binaries Suggested-by: Chris Wilson Cc: Chris Wilson Daniele Ceraolo Spurio (4): lib/igt_vgem:

Re: [Intel-gfx] [PATCH] drm/i915: Allow null render state batchbuffers bigger than one page

On Tue, Oct 10, 2017 at 10:29:41AM +, Chris Wilson wrote: > Quoting Chris Wilson (2017-10-10 11:25:38) > > Quoting Rodrigo Vivi (2017-10-05 05:34:02) > > > On Thu, Aug 24, 2017 at 11:00:27PM +, Rodrigo Vivi wrote: > > > > On Thu, Aug 24, 2017 at 3:39 PM, Oscar Mateo

[Intel-gfx] [PATCH i-g-t 2/4] tests/gem_pread: test reads from a bo not backed by struct page

Using an imported vgem bo we can test reads from an object not backed by struct page. These reads use different paths in the kernel. While at it, extract some common code in an helper function and fix the src vs dst naming (they are the other way around). Suggested-by: Chris Wilson

[Intel-gfx] [PATCH i-g-t 3/4] test/gem_pwrite: test writes to a bo not backed by struct page

Using an imported vgem bo we can test writes to an object not backed by struct page. These reads use different paths in the kernel. Suggested-by: Chris Wilson Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson

[Intel-gfx] [PATCH i-g-t v2 6/6] lib/ioctl_wrappers: drop stolen memory related wrappers

The feature was never merged and there has been no progress in the last year. The wrappers are using a wrong ioctl number (38 is HAS_POOLED_EU and not CREATE_VERSION) and extensions of the gem_create and get_aperture ioctl structures that never made it into the kernel. Tests that were using the

[Intel-gfx] [PATCH i-g-t v2 5/6] tests/gem_concurrent_all: drop stolen memory related subtests

The feature was never merged and there has been no progress in the last year. The tests are currently excluded from compilation with and ifdef. v2: keep create_cpu_require around (Chris) Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson

[Intel-gfx] [PATCH i-g-t v2 1/6] tests/gem_pread: drop stolen memory related subtests

The feature was never merged and there has been no progress in the last year. The subtests are currently skipping on all platforms by checking a field in the get_aperture ioctl structure that doesn't exist in the kernel version of the struct. Signed-off-by: Daniele Ceraolo Spurio

[Intel-gfx] [PATCH i-g-t v2 4/6] tests/gem_stolen: drop gem_stolen.c

The feature was never merged and there has been no progress in the last year. The subtests are currently skipping on all platforms by checking a field in the get_aperture ioctl structure that doesn't exist in the kernel version of the struct. v2: remove test from meson build as well (Petri)

[Intel-gfx] [PATCH i-g-t v2 3/6] tests/gem_create: drop stolen memory related subtest

The feature was never merged and there has been no progress in the last year. The subtest is currently skipping on all platforms by checking a field in the get_aperture ioctl structure that doesn't exist in the kernel version of the struct. Signed-off-by: Daniele Ceraolo Spurio

[Intel-gfx] [PATCH i-g-t v2 2/6] tests/gem_pwrite: drop stolen memory related subtests

The feature was never merged and there has been no progress in the last year. The subtests are currently skipping on all platforms by checking a field in the get_aperture ioctl structure that doesn't exist in the kernel version of the struct. Signed-off-by: Daniele Ceraolo Spurio

[Intel-gfx] [PATCH i-g-t v2 0/6] Drop tests for object creation from stolen

Respin with review fixes for meson build and gem_concurrent_all. I'll send the new tests for pread/pwrite suggested by Chris as a separated series as they depend on the other cleanup I've sent and this series can potentially be merged in the meantime. Daniele Ceraolo Spurio (6):

[Intel-gfx] [PATCH i-g-t 8/8] tests/gem_ringfill: use intel_measure_ring_size

With intel_measure_ring_size added as common function we can use it instead of the local copy Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson --- tests/gem_ringfill.c | 96 +--- 1

[Intel-gfx] [PATCH i-g-t 6/8] tests/gem_wait: use igt_cork

With igt_cork added as common utility we can use it instead of the local copy Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson --- tests/gem_wait.c | 52 1 file changed, 8

[Intel-gfx] [PATCH i-g-t 3/8] tests/gem_exec_schedule: use new common functions

With intel_measure_ring_size and igt_cork added as common utilities we can use them instead of the local copy of those utilities Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson --- tests/gem_exec_schedule.c | 159

[Intel-gfx] [PATCH i-g-t 4/8] tests/gem_exec_fence: use new common functions

With intel_measure_ring_size and igt_cork added as common utilities we can use them instead of the local copy of those utilities Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson --- tests/gem_exec_fence.c | 115

[Intel-gfx] [PATCH i-g-t 0/8] extract cork and ring measure code

The "cork" bo used to stall execution and the interruptible execbuf used to measure the ring size are repeated almost identically in several files. Extracting those to common files helps simplifying the code and avoiding repetition. Cc: Chris Wilson Daniele Ceraolo

[Intel-gfx] [PATCH i-g-t 7/8] tests/gem_exec_await: use intel_measure_ring_size

With intel_measure_ring_size added as common function we can use it instead of the local copy Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson --- tests/gem_exec_await.c | 90 +- 1

[Intel-gfx] [PATCH i-g-t 2/8] lib/igt_gt: add intel_measure_ring_size

The logic to measure the ring size is replicated almost identically in several tests. Adding it as a common function will make the code cleaner. The tests are updated in follow up patches. Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson

[Intel-gfx] [PATCH i-g-t 5/8] tests/gem_exec_latency: use new common functions

With intel_measure_ring_size and igt_cork added as common utilities we can use them instead of the local copy of those utilities Signed-off-by: Daniele Ceraolo Spurio Cc: Chris Wilson --- tests/gem_exec_latency.c | 99

[Intel-gfx] [PATCH i-g-t 1/8] lib/igt_dummyload: add igt_cork

The "cork" bo (imported bo with attached fence) is used in several tests to stall execution. Moving it to a common place makes the codebase cleaner. Note that the actual test updates is done in follow up patches as it is simpler to do in one go after one more common function is added in the next

[Intel-gfx] ✗ Fi.CI.BAT: failure for series starting with [1/2] lib: Move __gem_context_create to common ioctl wrapper library.

== Series Details == Series: series starting with [1/2] lib: Move __gem_context_create to common ioctl wrapper library. URL : https://patchwork.freedesktop.org/series/31857/ State : failure == Summary == Series 31857 revision 1 was fully merged or fully failed: no git log

Re: [Intel-gfx] [i-g-t 2/2] tests/gem_reset_stats: Add client ban test

On 12/10/17 15:10, Antonio Argenziano wrote: When a client gets a fixed amount of its contexts banned, will be blocked from creating any new contexts itself. This patch adds a test for this feature. Signed-off-by: Antonio Argenziano Cc: Michel Thierry

Re: [Intel-gfx] [PATCH i-g-t 1/2] lib: Move __gem_context_create to common ioctl wrapper library.

Quoting Antonio Argenziano (2017-10-12 23:10:36) > +/** > + * gem_require_contexts: > + * @fd: open i915 drm file descriptor > + * > + * This helper will automatically skip the test on platforms where context > + * support is not available. > + */ > +void gem_require_contexts(int fd) > +{ > +

[Intel-gfx] [i-g-t 2/2] tests/gem_reset_stats: Add client ban test

When a client gets a fixed amount of its contexts banned, will be blocked from creating any new contexts itself. This patch adds a test for this feature. Signed-off-by: Antonio Argenziano Cc: Michel Thierry --- tests/gem_ctx_create.c |

[Intel-gfx] [PATCH i-g-t 1/2] lib: Move __gem_context_create to common ioctl wrapper library.

This patch adds a context creation ioctl wrapper that returns the error for the caller to consume. Multiple tests that implemented this already, have been changed to use the new library function. v2: - Add gem_require_contexts() to check for contexts support (Chris) Signed-off-by:

[Intel-gfx] ✓ Fi.CI.BAT: success for series starting with [v2,1/2] drm/i915: Fix DP_AUX_CH_CTL_TIME_OUT naming

== Series Details == Series: series starting with [v2,1/2] drm/i915: Fix DP_AUX_CH_CTL_TIME_OUT naming URL : https://patchwork.freedesktop.org/series/31856/ State : success == Summary == Series 31856v1 series starting with [v2,1/2] drm/i915: Fix DP_AUX_CH_CTL_TIME_OUT naming

Re: [Intel-gfx] [PATCH v2 2/2] drm/i915/bdw: Fix DP_AUX_CH_CTL_TIME_OUT setting

On Thu, Oct 12, 2017 at 09:30:37PM +, James Ausmus wrote: > Per BSpec, 400us is "BDW+ Do not use this setting." - not just PORT_A. > Set BDW to 600us unconditionally. > > v2: > -Split in to two patches (Rodrigo) > > Cc: Jani Nikula > Cc: Ville Syrjälä

Re: [Intel-gfx] [PATCH v2 1/2] drm/i915: Fix DP_AUX_CH_CTL_TIME_OUT naming

On Thu, Oct 12, 2017 at 09:30:36PM +, James Ausmus wrote: > Rename DP_AUX_CH_CTL_TIME_OUT_1600us to DP_AUX_CH_CTL_TIME_OUT_MAX, as > the meaning of the (3 << 26) value varies per platform, but it's always the > maximum timeout for that platform. Pre-CNL it means 1600us, and for CNL > it means

[Intel-gfx] ✗ Fi.CI.BAT: failure for lib: Ask the kernel to quiesce the GPU

== Series Details == Series: lib: Ask the kernel to quiesce the GPU URL : https://patchwork.freedesktop.org/series/31850/ State : failure == Summary == IGT patchset tested on top of latest successful build 58616272b23efce1e62a3ee0d37e13de6ffc012f igt/gem_eio: Check hang/eio recovery during

Re: [Intel-gfx] [PATCH v2 2/2] drm/i915/bdw: Fix DP_AUX_CH_CTL_TIME_OUT setting

On Thu, Oct 12, 2017 at 1:59 PM, Rodrigo Vivi wrote: > On Thu, Oct 12, 2017 at 06:58:20PM +, Ausmus, James wrote: >> On Wed, Oct 4, 2017 at 3:54 PM, Runyan, Arthur J >> wrote: >> > I think the failure was with one particularly slow eDP

[Intel-gfx] [PATCH v2 2/2] drm/i915/bdw: Fix DP_AUX_CH_CTL_TIME_OUT setting

Per BSpec, 400us is "BDW+ Do not use this setting." - not just PORT_A. Set BDW to 600us unconditionally. v2: -Split in to two patches (Rodrigo) Cc: Jani Nikula Cc: Ville Syrjälä Signed-off-by: James Ausmus ---

[Intel-gfx] [PATCH v2 1/2] drm/i915: Fix DP_AUX_CH_CTL_TIME_OUT naming

Rename DP_AUX_CH_CTL_TIME_OUT_1600us to DP_AUX_CH_CTL_TIME_OUT_MAX, as the meaning of the (3 << 26) value varies per platform, but it's always the maximum timeout for that platform. Pre-CNL it means 1600us, and for CNL it means 3200us. v2: -Split in to two patches (Rodrigo) Cc: Rodrigo Vivi

Re: [Intel-gfx] [PATCH v2 02/13] drm/i915/guc: Move GuC boot param initialization out of xfer

On 12/10/17 12:16, Michal Wajdeczko wrote: We want to keep ucode xfer functions separate from other initialization. Once separated, add explicit forcewake. v2: use BLITTER domain only and add comment (Daniele) Suggested-by: Joonas Lahtinen Signed-off-by:

Re: [Intel-gfx] [PATCH] HAX drm/i915: Disable CSR (DMC) for Kabylake

Quoting Rodrigo Vivi (2017-10-12 21:55:24) > On Thu, Oct 12, 2017 at 07:43:04PM +, Chris Wilson wrote: > > Quoting Rodrigo Vivi (2017-10-12 19:04:45) > > > On Thu, Oct 12, 2017 at 10:18:13AM +, Chris Wilson wrote: > > > > --- > > > > > > Why? > > > > Have you looked at the random but

[Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915: Stop asserting on set-wedged vs nop_submit_request ordering

== Series Details == Series: drm/i915: Stop asserting on set-wedged vs nop_submit_request ordering URL : https://patchwork.freedesktop.org/series/31848/ State : failure == Summary == Series 31848v1 drm/i915: Stop asserting on set-wedged vs nop_submit_request ordering

Re: [Intel-gfx] [PATCH v2 2/2] drm/i915/bdw: Fix DP_AUX_CH_CTL_TIME_OUT setting

On Thu, Oct 12, 2017 at 06:58:20PM +, Ausmus, James wrote: > On Wed, Oct 4, 2017 at 3:54 PM, Runyan, Arthur J > wrote: > > I think the failure was with one particularly slow eDP panel, but it is > > safer to apply this to all ports. > > > > Thanks Art. Anyone else

Re: [Intel-gfx] [PATCH] HAX drm/i915: Disable CSR (DMC) for Kabylake

On Thu, Oct 12, 2017 at 07:43:04PM +, Chris Wilson wrote: > Quoting Rodrigo Vivi (2017-10-12 19:04:45) > > On Thu, Oct 12, 2017 at 10:18:13AM +, Chris Wilson wrote: > > > --- > > > > Why? > > Have you looked at the random but frequent mmio death on Kabylake? > Seems rather reminiscent of

[Intel-gfx] [PATCH igt] lib: Ask the kernel to quiesce the GPU

Since the introduction of debugfs/i915_drop_caches, we have offered the ability to wait upon all outstanding batches. This is more efficient and less error prone (one example is the use of context priorities, we have to idle at the lowest in order not to jump over any low priority tasks we want to

Re: [Intel-gfx] [PATCH 03/12] drm/i915/guc: Initialize GuC before restarting engines

On 09/10/17 07:52, Michał Winiarski wrote: Now that we're handling request resubmission the same way as regular submission (from the tasklet), we can move GuC initialization earlier, before restarting the engines. This way, we're no longer being in the state of flux during engine restart - we're

[Intel-gfx] [PATCH] drm/i915: Stop asserting on set-wedged vs nop_submit_request ordering

Since the removal of the stop_machine(), it is allowed and expected for the nop_submit_request() and nop_complete_submit_request() to run in parallel to the i915_gem_set_wedged() processing. As such we can no longer assert that i915_gem_set_wedged() has completed inside the stop_machine prior to

Re: [Intel-gfx] [PATCH 02/12] drm/i915/guc: Allocate separate shared data object for GuC communication

On 09/10/17 15:35, Michel Thierry wrote: On 10/9/2017 11:41 AM, Daniele Ceraolo Spurio wrote: On 09/10/17 07:52, Michał Winiarski wrote: We were using first page of kernel context render state for sharing data with GuC. While it's justified by the fact that those pages are not used (note,

Re: [Intel-gfx] [PATCH i-g-t v5] lib/igt_gt: Allow non-default contexts to hang non-render rings

Quoting Vinay Belgaumkar (2017-10-12 17:47:10) > This limitation does not exist in latest kernel. It was removed by this patch- > > commit f7978a0c581a8a840a28306f8da43e06e7fef3bf > > v2: Added commit id that removes the limitation(Chris Wilson) > V3: Generic way to find if kernel supports this

[Intel-gfx] ✗ Fi.CI.BAT: failure for Firmware code reorg

== Series Details == Series: Firmware code reorg URL : https://patchwork.freedesktop.org/series/31846/ State : failure == Summary == Series 31846v1 Firmware code reorg https://patchwork.freedesktop.org/api/1.0/series/31846/revisions/1/mbox/ Test debugfs_test: Subgroup

Re: [Intel-gfx] [PATCH v2 06/13] drm/i915/guc: Reorder functions in intel_guc_fw.c

Quoting Michal Wajdeczko (2017-10-12 20:16:20) > Functions should be defined in their use order. > > Signed-off-by: Michal Wajdeczko > Cc: Joonas Lahtinen Reviewed-by: Chris Wilson -Chris

Re: [Intel-gfx] [PATCH v2 03/13] drm/i915/guc: Small fixups post code move

Quoting Michal Wajdeczko (2017-10-12 20:16:17) > Existing code needs some style fixes. To avoid polluting > pure move patch, do it now as separate step. > > Suggested-by: Chris Wilson > Signed-off-by: Michal Wajdeczko > Cc: Chris Wilson

Re: [Intel-gfx] [PATCH v2 11/13] drm/i915/guc: Update Guc messages on load failure

Quoting Michal Wajdeczko (2017-10-12 20:16:25) > In case of GuC firmware loading failure we were reporting > DRM_ERROR also for case when GuC loading was not strictly > required. > > Signed-off-by: Michal Wajdeczko > Cc: Chris Wilson > Cc:

[Intel-gfx] ✗ Fi.CI.IGT: failure for series starting with [1/7] tests/perf: Fix a bunch of warnings on 32-bit systems

== Series Details == Series: series starting with [1/7] tests/perf: Fix a bunch of warnings on 32-bit systems URL : https://patchwork.freedesktop.org/series/31818/ State : failure == Summary == Test kms_flip: Subgroup flip-vs-fences-interruptible: pass -> FAIL

Re: [Intel-gfx] [PATCH v2 10/13] drm/i915/uc: Unify firmware loading

Quoting Michal Wajdeczko (2017-10-12 20:16:24) > Firmware loading for GuC and Huc are similar. Move common code > into generic function for maximum reuse. > > v2: change message levels (Chris) > > Signed-off-by: Michal Wajdeczko > Cc: Joonas Lahtinen

Re: [Intel-gfx] [PATCH v2 09/13] drm/i915/uc: Improve debug messages in firmware fetch

Quoting Michal Wajdeczko (2017-10-12 20:16:23) > Time to remove less important info and make messages clear > and consistent. > > v2: change some message levels (Chris) > > Signed-off-by: Michal Wajdeczko > Cc: Sagar Arun Kamble > Cc:

Re: [Intel-gfx] [PATCH v2 08/13] drm/i915/guc: Pick better place for Guc final status message

Quoting Michal Wajdeczko (2017-10-12 20:16:22) > Guc status message printed right after firmware upload may be too > optimistic, as we may fail on subsequent steps. Move that message > to the end of intel_uc_init_hw where we know the status for sure. > > v2: use dev_info (Chris) > >

Re: [Intel-gfx] [PATCH v2 06/13] drm/i915/guc: Reorder functions in intel_guc_fw.c

Quoting Michal Wajdeczko (2017-10-12 20:16:20) > Functions should be defined in their use order. > > Signed-off-by: Michal Wajdeczko > Cc: Joonas Lahtinen Reviewed-by: Chris Wilson -Chris

Re: [Intel-gfx] [PATCH 9/9] drm/i915: Add windowing for primary planes on gen2/3 and chv

On Wed, Oct 11, 2017 at 07:04:55PM +0300, Ville Syrjala wrote: > From: Ville Syrjälä > > Plane B and C (note that we don't actually expose plane C currently) > on gen2/3 have a window generator, as does the primary plane on CHV > pipe B. So let's allow positioning

Re: [Intel-gfx] [PATCH] HAX drm/i915: Disable CSR (DMC) for Kabylake

Quoting Rodrigo Vivi (2017-10-12 19:04:45) > On Thu, Oct 12, 2017 at 10:18:13AM +, Chris Wilson wrote: > > --- > > Why? Have you looked at the random but frequent mmio death on Kabylake? Seems rather reminiscent of earlier DMC bugs. -Chris ___

  1   2   3   4   >