We have many functions responsible for allocating different parts of
runtime called from multiple places. Let's stick with keeping
everything in guc_log_register instead.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Michal Wajdeczko
Cc: Sagar Arun K
We plan to decouple log runtime (mapping + relay) from verbosity control.
Let's tidy the code now to reduce the churn in the following patches.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Michal Wajdeczko
Cc: Sagar Arun Kamble
---
drivers/gpu/drm
let).
v2: Assert that context save restore is inhibited, don't assert on ring
alignment. (Chris)
Fixes: 517aaffe0c1b ("drm/i915/execlists: Inhibit context save/restore for the
fake preempt context")
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spur
---
drivers/gpu/drm/i915/i915_params.h | 2 +-
drivers/gpu/drm/i915/intel_uc.c| 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/i915_params.h
b/drivers/gpu/drm/i915/i915_params.h
index 430f5f9d0ff4..3deae1e22974 100644
--- a/drivers/gpu/drm/i915/i915_p
---
drivers/gpu/drm/i915/i915_params.h | 2 +-
drivers/gpu/drm/i915/intel_uc.c| 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/i915_params.h
b/drivers/gpu/drm/i915/i915_params.h
index 430f5f9d0ff4..3deae1e22974 100644
--- a/drivers/gpu/drm/i915/i915_p
tasklet).
Fixes: 517aaffe0c1b ("drm/i915/execlists: Inhibit context save/restore for the
fake preempt context")
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Michel Thierry
Cc: Mika Kuoppala
Cc: Tvrtko Ursulin
---
drivers/
mption event. To do so we must always check inside
> execlists_dequeue() whether there is a priority inversion between the
> ports themselves as well as the head of the priority sorted queue, and we
> cannot just skip dequeuing if the queue is empty.
>
> Signed-off-by: Chris Wil
x27; is still much preferred for its clarity.
>
> Signed-off-by: Chris Wilson
> Cc: Joonas Lahtinen
> Cc: Mika Kuoppala
> Cc: Tvrtko Ursulin
> Cc: Michał Winiarski
> Cc: Michal Wajdeczko
Reviewed-by: Michał Winiarski
-Michał
> ---
> drivers/gpu/drm/i915/Makefile
On Tue, Feb 06, 2018 at 02:31:37PM +, Chris Wilson wrote:
> We only need to wake up the RPS worker once when initially enabling the
> client boost, it remains in effect then until the last client no longer
> requires the boost.
Reviewed-by: Michał Winiarski
-Michał
>
> Re
On Wed, Jan 31, 2018 at 11:44:38AM +0530, Sagar Arun Kamble wrote:
> On some systems like skl-gvtdvm, SSE4.1 movntdqa might not be available.
> movntdqa is needed for efficient capture of the logs from uncached GuC
> log buffer. GuC init was tied with this support and other setup needed
> for inter
On Tue, Jan 30, 2018 at 09:02:45AM +, Chris Wilson wrote:
> Quoting Daniele Ceraolo Spurio (2018-01-30 00:24:04)
> >
> >
> > > @@ -979,17 +979,19 @@ static int guc_clients_create(struct intel_guc *guc)
> > > }
> > > guc->execbuf_client = client;
> > >
> > > - client = guc_c
t try to unsubmit a completed request" in the subject?
(what does it mean to "finish the seqno"?)
Reviewed-by: Michał Winiarski
-Michał
>
> Signed-off-by: Chris Wilson
> ---
> drivers/gpu/drm/i915/i915_gem_request.c | 2 ++
> 1 file changed, 2 insertions(+)
References
It allows us to get test results while we're fixing the logging issues, so:
Reviewed-by: Michał Winiarski
-Michał
> Signed-off-by: Michal Wajdeczko
> Cc: Chris Wilson
> Cc: Jani Saarinen
> Cc: Tomi Sarvela
> Cc: Marta Lofstedt
> Cc: Michal Winiarski
> ---
&
kage. Assert that it doesn't move.
>
> Complete, but expensive checking is done by swfence, this assert is more
> about documenting the topologically ordered list.
>
> Signed-off-by: Chris Wilson
> Cc: Michał Winiarski
Reviewed-by: Michał Winiarski
-Michał
> ---
&
Kamble
> Fixes: 3176ff49bc3e ("drm/i915/guc: Move GuC workqueue allocations outside of
> the mutex")
> Cc: Michał Winiarski
> Cc: Chris Wilson
> ---
> drivers/gpu/drm/i915/i915_gem.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/gpu/d
20f480: fb fb fb fb fb fb fb fb fc fc fc fc fc fc
> fc fc
> [ 3083.158201] 8806bf20f500: fb fb fb fb fb fb fb fb fc fc fc fc fc fc
> fc fc
> [ 3083.158203]
> ==========
>
> Reported-by: Alexandru Chirvasitu
>
On Tue, Jan 02, 2018 at 07:25:00PM +, Chris Wilson wrote:
> We should never insert the invalid seqno into the wait tree, so assert
> we do not.
>
> Signed-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
-Michał
> ---
> drivers/gpu/drm/i915/intel_breadcrumbs.c | 2 ++
On Tue, Jan 02, 2018 at 03:12:25PM +, Chris Wilson wrote:
> Back up our comment that all signalers should have been signaled before
> we ourselves were retired with an assert to that effect.
>
> Signed-off-by: Chris Wilson
> Cc: Joonas Lahtinen
Reviewed-by: Michał Wini
>
> Signed-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
-Michał
> ---
> drivers/gpu/drm/i915/intel_lrc.c | 11 ---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_lrc.c
> b/drivers/gpu/drm/i915/intel_lrc.
On Tue, Jan 02, 2018 at 03:12:27PM +, Chris Wilson wrote:
> After staring at the list_for_each_safe macros for a bit, our current
> invocation of list_safe_reset_next in execlists_schedule() simply
> reduces to list_for_each.
>
> Signed-off-by: Chris Wilson
Reviewed-by: M
>
> Signed-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
-Michał
> ---
> drivers/gpu/drm/i915/i915_debugfs.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/i915_debugfs.c
> b/drivers/gpu/drm/i915/i915_debugfs.c
> index d8
d-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
-Michał
> ---
> drivers/gpu/drm/i915/intel_lrc.c | 33 +
> 1 file changed, 21 insertions(+), 12 deletions(-)
___
Intel-gfx mailing list
Intel-gfx@lis
On Tue, Jan 02, 2018 at 03:12:33PM +, Chris Wilson wrote:
> Currently, we record the elsp register offset inside init-hw but we only
> need to do it once during engine setup (after we know the mmio
> iomapping).
>
> Signed-off-by: Chris Wilson
Reviewed-by: Michał Wini
On Tue, Jan 02, 2018 at 03:12:34PM +, Chris Wilson wrote:
> Move the register settings for enabling execlists into its own function
> for clarity.
>
> Signed-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
-Michał
> ---
> drivers/gpu/drm/i91
processes/fds, it can only share GTT belonging
> to this fd.
> Testcase: igt/gem_ctx_shared
> Signed-off-by: Chris Wilson
> Cc: Joonas Lahtinen
> Cc: Tvrtko Ursulin
> Cc: Mika Kuoppala
> Cc: Michał Winiarski
> ---
> drivers/gpu/drm/i915/i915_gem_context.c | 62 +
We have the selftest that's checking doorbell create/destroy, so there's
no need to check all doorbells delaying the reset every time.
We do want to have that extra sanity check at module load/unload though.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
We can now move the clients allocation to submission_init path, rather
than keeping the condition inside submission_enable called on every
reset.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Michal Wajdeczko
Reviewed-by: Michel Thierry
---
drivers/gpu/drm/i915
caused by GuC action failures
on module unload.
Note that I've also removed one of the tests (bitmap out of sync), since
it doesn't make much sense anymore - bitmaps are now not expected to
change during the lifetime of a client.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas
-by: Chris Wilson
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Michal Wajdeczko
Reviewed-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
b/drivers/gpu/drm/i915/i915_gem_gtt.c
index a0
0 00 00 f6 80 98 00 00 00 01 0f
RIP: intel_guc_suspend+0x34/0x140 [i915] RSP: c9483df8
CR2: 00f9
---[ end trace 23a192a61d937a3e ]---
Fixes: b8e5eb960b28 ("drm/i915/guc: Allocate separate shared data object for
GuC communication")
Signed-off-by: Michał Winiarski
ble_guc on suspend/resume.
v2: Do not duplicate disable_communication / reset_guc_interrupts
v3: Add proper teardown after rebase
References: 04f7b24eccdf ("drm/i915/guc: Assert that we switch between known
ggtt->invalidate functions")
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
ris)
v3: Mention that we need the enable_guc=1 for lockdep splat (Chris)
Testcase: igt/debugfs_test/read_all_entries # with i915.enable_guc=1
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Michal Wajdeczko
Reviewed-by: Chris Wilson
---
driv
Generated using make header_install.
Generated from drm-intel-next-queued commit
53ff2641a817099e1c6d1aef409ba004c3a9f1ea
Signed-off-by: Michał Winiarski
---
include/drm/i915_drm.h | 215 ++---
1 file changed, 202 insertions(+), 13 deletions(-)
diff
Also:
Revert "drm/i915/GuC/GLK: Load GuC on GLK"
Now that we no longer have fallback to execlists submission available,
if the firmware is selected by the driver but is not available on the
system (like in this case - where the FW is not yet released), we're
unable to get a clean CI run.
This re
We have the selftest that's checking doorbell create/destroy, so there's
no need to check all doorbells delaying the reset every time.
We do want to have that extra sanity check at module load/unload though.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
We can now move the clients allocation to submission_init path, rather
than keeping the condition inside submission_enable called on every
reset.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Michal Wajdeczko
---
drivers/gpu/drm/i915/intel_guc_submission.c | 33
caused by GuC action failures
on module unload.
Note that I've also removed one of the tests (bitmap out of sync), since
it doesn't make much sense anymore - bitmaps are now not expected to
change during the lifetime of a client.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas
ble_guc on suspend/resume.
v2: Do not duplicate disable_communication / reset_guc_interrupts
References: 04f7b24eccdf ("drm/i915/guc: Assert that we switch between known
ggtt->invalidate functions")
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Michal Wa
1680 RSI: 7ffd6ffc3400 RDI: 7f98d39a2bc0
RBP: 7ffd6ffc33a0 R08: R09: 05a0
R10: 55e847c2a830 R11: 0002 R12: 0001
R13: 55e847c1d040 R14: 00007ffd6ffc3400 R15: 7f98d6752ba0
v2: Init preempt_work unconditiona
-by: Chris Wilson
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Michal Wajdeczko
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
b/drivers/gpu/drm/i915/i915_gem_gtt.c
index a0579b0c8581..c5f393870532 10
0 00 00 f6 80 98 00 00 00 01 0f
RIP: intel_guc_suspend+0x34/0x140 [i915] RSP: c9483df8
CR2: 00f9
---[ end trace 23a192a61d937a3e ]---
Fixes: b8e5eb960b28 ("drm/i915/guc: Allocate separate shared data object for
GuC communication")
Signed-off-by: Michał Winiarski
---
drivers/gpu/drm/i915/i915_params.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/i915_params.h
b/drivers/gpu/drm/i915/i915_params.h
index 792ce26d7449..9725c5ad8ac6 100644
--- a/drivers/gpu/drm/i915/i915_params.h
+++ b/drivers/gpu/drm/i915/i915_param
ble_guc on suspend/resume.
References: 04f7b24eccdf ("drm/i915/guc: Assert that we switch between known
ggtt->invalidate functions")
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Michal Wajdeczko
Cc: Sagar Arun Kamble
---
drivers/gpu/drm/i915/i915_drv.c
0 00 00 f6 80 98 00 00 00 01 0f
RIP: intel_guc_suspend+0x34/0x140 [i915] RSP: c9483df8
CR2: 00f9
---[ end trace 23a192a61d937a3e ]---
Fixes: b8e5eb960b28 ("drm/i915/guc: Allocate separate shared data object for
GuC communication")
Signed-off-by: Michał Winiarski
1680 RSI: 7ffd6ffc3400 RDI: 7f98d39a2bc0
RBP: 7ffd6ffc33a0 R08: R09: 05a0
R10: 55e847c2a830 R11: 0002 R12: 0001
R13: 55e847c1d040 R14: 7ffd6ffc3400 R15: 7f98d6752ba0
Testcase: igt/debugfs_tes
IP]
> +err_ggtt:
> +err_unlock:
So... Just unlock? :>
Does what it says on the tin (fixing WARN_ON galore on guc load failure):
Reviewed-by: Michał Winiarski
-Michał
> + intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
> + mutex_unlock(&dev_priv->drm.struct_m
ubmission
> flow of low priority contexts.
>
> -v2:
> - Rename subtest (Chris)
> - Use igt_hang_ctx to hang ring (Chris)
> - Add comment on execution order checks (Chris)
>
> Cc: Chris Wilson
> Cc: Michal Winiarski
> Signed-off-by: A
roup for the test (Chris)
>
> Cc: Chris Wilson
> Cc: Michal Winiarski
> Signed-off-by: Antonio Argenziano
Reviewed-by: Michał Winiarski
-Michał
> ---
> tests/gem_exec_schedule.c | 28 +++-
> 1 file changed, 27 insertions(+), 1 deletion(-)
_
we lost the skip over the first port if set for the
> non-preemption path. Restore it.
>
> Reported-by: Mika Kuoppala
> Fixes: c41937fd994a ("drm/i915/guc: Preemption! With GuC")
> Signed-off-by: Chris Wilson
> Cc: Michał Winiarski
> Cc: Mika Kuoppal
son
> Cc: Mika Kuoppala
> Cc: Michał Winiarski
Reviewed-by: Michał Winiarski
-Michał
> ---
> drivers/gpu/drm/i915/intel_guc_submission.c | 7 ++-
> 1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_guc_su
e still seeing the
long tail without exponential backoff if we manage to miss the fast timeout.
> References: 1758b90e38f5 ("drm/i915: Use a hybrid scheme for fast register
> waits")
> Signed-off-by: Chris Wilson
> Cc: Tvrtko Ursulin
> Cc: John Harrison
> Cc: Michał Wi
Gordon
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Michal Wajdeczko
---
drivers/gpu/drm/i915/i915_debugfs.c| 2 +
drivers/gpu/drm/i915/i915_guc_submission.c | 118
than
enable_guc_submission modparam, reorder preempt postprocessing (Chris)
v5: Make wq NULL after destroying it
v6: Swap struct guc_preempt_work members (Michał)
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Joonas Lahtinen
Cc: Oscar Mateo
Cc: Tvrtko Ursulin
Reviewe
: And move clients back from an array, to get rid of the enum (Michał)
v4: Use is_high_priority, move DRM_ERROR into __create_doorbell, move
GEM_BUG_ON inside guc_clients_create (Michał)
v5: Split the BUG_ON (Michał)
Signed-off-by: Dave Gordon
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
: And move clients back from an array, to get rid of the enum (Michał)
v4: Use is_high_priority, move DRM_ERROR into __create_doorbell, move
GEM_BUG_ON inside guc_clients_create (Michał)
Signed-off-by: Dave Gordon
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc
On Thu, Oct 26, 2017 at 12:04:11PM +0100, Chris Wilson wrote:
> Quoting Michał Winiarski (2017-10-26 11:52:31)
> > On Thu, Oct 26, 2017 at 08:31:27AM +0100, Chris Wilson wrote:
> > > For measuring the cost of preemption, inject a low priority spinner
> > > betwee
correctly).
Reviewed-by: Michał Winiarski
-Michał
>
> Signed-off-by: Chris Wilson
> CC: Michał Winiarski
> ---
> tests/gem_exec_latency.c | 88
>
> 1 file changed, 59 insertions(+), 29 deletions(-)
__
Argenziano
>
> Cc: Chris Wilson
> Cc: Michał Winiarski
> ---
> benchmarks/gem_exec_ctx.c | 6 ++---
> benchmarks/gem_exec_trace.c | 4 +--
> lib/i915/gem_context.c | 62
> +
> lib/i915/gem_context.h
than
enable_guc_submission modparam, reorder preempt postprocessing (Chris)
v5: Make wq NULL after destroying it
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Joonas Lahtinen
Cc: Oscar Mateo
Cc: Tvrtko Ursulin
Reviewed-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_dr
than
enable_guc_submission modparam, reorder preempt postprocessing (Chris)
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Joonas Lahtinen
Cc: Oscar Mateo
Cc: Tvrtko Ursulin
---
drivers/gpu/drm/i915/i915_drv.c| 3 +-
drivers/gpu/drm/i915/i915_ge
Also:
Revert "drm/i915/guc: Assert that we switch between known ggtt->invalidate
functions"
This reverts commit 04f7b24eccdfae680a36e9825fe0d61dcd5ed528.
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 8 ++--
drivers/gpu/drm/i915/i915_params.h | 4 ++--
2 files changed, 4 insertions(+), 8 deleti
ned-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Chris Wilson
Reviewed-by: Jeff McGee
---
drivers/gpu/drm/i915/intel_ringbuffer.h | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h
b/d
xcept we're introducing
HAS_LOGICAL_RING_PREEMPTION macro instad, converting other users that
were previously touching device info directly.
v2: s/intel_engine/execlists and pass execlists to unwind (Chris)
v3: use locked version for exporting, drop const qual (Chris)
Signed-off-by: Michał Win
We also want to support preemption with GuC submission backend.
In order to do that, we need to remember the priority, like we do on
execlists path.
v2: Remove completed prio == INT_MAX optimization
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Joonas Lahtinen
Reviewed
g GuC work item and
ringing the doorbell, and guc_wq_item_append - used by the function
above, not tied to the concept of gem request.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Chris Wilson
Reviewed-by: Jeff McGee
---
atch emit_pipe_control, s/render/rcs (Chris)
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Tvrtko Ursulin
Reviewed-by: Chris Wilson
---
drivers/gpu/drm/i915/intel_lrc.c| 28 +++-
drivers/gpu/drm/i915/intel_ringbuffer.h | 38 +++
: And move clients back from an array, to get rid of the enum (Michał)
Signed-off-by: Dave Gordon
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
---
drivers/gpu/drm/i915/i915_debugfs.c| 2 +
drivers
those new definitions.
v2: Drop unused INTEL_GUC_PREEMPT_OPTION_IMMEDIATE
Signed-off-by: Michał Winiarski
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Jeff McGee
---
drivers/gpu/drm/i915/intel_guc_fwif.h | 39 +++
1 file changed, 39 insertions(+)
di
Since it's a two-step process, we can have a cleaner error handling in
the caller if we do the allocations in a helper.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Daniele Ceraolo S
particular page?). Let's allocate a separate object instead.
v2: Drop kernel_context from GuC suspend/resume action handlers (Michel)
Suggested-by: Daniele Ceraolo Spurio
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Jeff McGee
Cc: Michal Wajdeczko
Apparently, this value is reserved and may be interpreted as changing
doorbell ownership. Even though we're not observing any side effects
now, let's skip over it to be consistent with the spec.
v2: Apply checkpatch (Sagar)
Suggested-by: Sagar Arun Kamble
Signed-off-by: Michał Win
No major changes from previous iteration.
Dropped the workaround for missing interrupt (which turned out to be
self-inflicted, now properly fixed by Chris), and applied the review comments.
Dave Gordon (1):
drm/i915/guc: Add a second client, to be used for preemption
Michał Winiarski (11
ll; be specific!
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103324
> Signed-off-by: Chris Wilson
> CC: Michał Winiarski
Reviewed-by: Michał Winiarski
-Michał
> ---
> tests/drv_hangman.c | 15 +++
> 1 file changed, 15 insertions(+)
>
>
On Tue, Oct 24, 2017 at 07:24:58PM +0100, Chris Wilson wrote:
> Since the headless subtest wants to compare execution latency of a
> headless mode vs a single head, it needs to be able to do a modeset and
> ergo it requires DRM_MASTER.
>
> Signed-off-by: Chris Wilson
Revi
() will do the job of exhausting the fdtable.
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103365
> Signed-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
But do we really need this test? Perhaps we could just remove it?
-Michał
> ---
> tests/gem_fd_exhaustion
ubsequent GPU hangs are not prominently recorded, as
> the kernel believes they are secondary hangs (and we only announce the
> first).
>
> Signed-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
-Michał
> ---
> lib/igt_gt.c | 31 ---
>
waiters.
> -Chris
With additional note that as a consequence, we're basically postponing the
irq_disable until idle_work_handler.
Reviewed-by: Michał Winiarski
-Michał
> ___
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> h
> Fixes: 1f181225f8ec ("drm/i915/execlists: Keep request->priority for its
> lifetime")
> Signed-off-by: Chris Wilson
> Cc: Michał Winiarski
> Cc: Joonas Lahtinen
Reviewed-by: Michał Winiarski
-Michał
> ---
> drivers/gpu/drm/i915/intel_lrc.c | 2 --
>
isper/*-priority
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: John Harrison
Cc: Joonas Lahtinen
---
drivers/gpu/drm/i915/i915_guc_submission.c | 36 +-
drivers/gpu/drm/i915/intel_guc.h | 1 +
2 files changed, 36 insertions(+), 1 deletio
Also:
Revert "drm/i915/guc: Assert that we switch between known ggtt->invalidate
functions"
This reverts commit 04f7b24eccdfae680a36e9825fe0d61dcd5ed528.
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 8 ++--
drivers/gpu/drm/i915/i915_params.h | 4 ++--
2 files changed, 4 insertions(+), 8 deleti
xcept we're introducing
HAS_LOGICAL_RING_PREEMPTION macro instad, converting other users that
were previously touching device info directly.
v2: s/intel_engine/execlists and pass execlists to unwind (Chris)
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Michal
We also want to support preemption with GuC submission backend.
In order to do that, we need to remember the priority, like we do on
execlists path.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Joonas Lahtinen
Reviewed-by: Chris Wilson
---
drivers/gpu/drm/i915
truct in intel_guc rather than engine (Michał)
Use ordered workqueue for inject_preempt worker to avoid GuC quirks.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Joonas Lahtinen
Cc: Oscar Mateo
Cc: Tvrtko Ursulin
---
drivers/gpu/drm/i915/i915_drv.c| 3 +-
dri
ned-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Chris Wilson
Reviewed-by: Jeff McGee
---
drivers/gpu/drm/i915/intel_ringbuffer.h | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h
b/d
atch emit_pipe_control, s/render/rcs (Chris)
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Tvrtko Ursulin
---
drivers/gpu/drm/i915/intel_lrc.c| 28 +++-
drivers/gpu/drm/i915/intel_ringbuffer.h | 38 +
2 fil
g GuC work item and
ringing the doorbell, and guc_wq_item_append - used by the function
above, not tied to the concept of gem request.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Chris Wilson
Reviewed-by: Jeff McGee
---
those new definitions.
Signed-off-by: Michał Winiarski
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Jeff McGee
---
drivers/gpu/drm/i915/intel_guc_fwif.h | 40 +++
1 file changed, 40 insertions(+)
diff --git a/drivers/gpu/drm/i915/intel_guc_fwif.h
.
Signed-off-by: Dave Gordon
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
---
drivers/gpu/drm/i915/i915_debugfs.c| 11 +--
drivers/gpu/drm/i915/i915_guc_submission.c | 107
submission mode.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Michal Wajdeczko
Cc: Michel Thierry
Cc: Mika Kuoppala
Cc: Oscar Mateo
Reviewed-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem.c| 10 +-
drivers/gpu/drm/i915/i915_guc_submission.c | 7 ---
drive
Since it's a two-step process, we can have a cleaner error handling in
the caller if we do the allocations in a helper.
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Jeff McGee
Cc: Michal Wajdeczko
Cc: Oscar Mateo
Reviewed-by: Daniele Ceraolo S
particular page?). Let's allocate a separate object instead.
v2: Drop kernel_context from GuC suspend/resume action handlers (Michel)
Suggested-by: Daniele Ceraolo Spurio
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc: Daniele Ceraolo Spurio
Cc: Jeff McGee
Cc: Michal Wajdeczko
Apparently, this value is reserved and may be interpreted as changing
doorbell ownership. Even though we're not observing any side effects
now, let's skip over it to be consistent with the spec.
Suggested-by: Sagar Arun Kamble
Signed-off-by: Michał Winiarski
Cc: Chris Wilson
Cc:
to be used for preemption
Michał Winiarski (13):
drm/i915/guc: Do not use 0 for GuC doorbell cookie
drm/i915/guc: Extract GuC stage desc pool creation into a helper
drm/i915/guc: Allocate separate shared data object for GuC
communication
drm/i915/guc: Initialize GuC before restarting engines
dfe556659f ("lib: Extract helpers for determining submission method")
Signed-off-by: Michał Winiarski
Cc: Arkadiusz Hiler
Cc: Chris Wilson
---
.../intel-gpu-tools/intel-gpu-tools-docs.xml | 1 +
lib/Makefile.sources | 2 +
lib/i
Another example of something that is used across different tests, and
should be moved to lib.
v2: Break the trend of expanding ioctl_wrappers
Signed-off-by: Michał Winiarski
Cc: Arkadiusz Hiler
Cc: Chris Wilson
Cc: Katarzyna Dec
Cc: Petri Latvala
Reviewed-by: Chris Wilson
Reviewed-by
We'd like to make ioctl_wrappers a bit thinner, and we plan to add new
helpers in the following patch. Let's move context related helpers before
adding more content.
Signed-off-by: Michał Winiarski
Cc: Arkadiusz Hiler
Cc: Chris Wilson
Cc: Katarzyna Dec
Cc: Petri Latvala
Reviewed
Couple of tests are using either determining submission method, or
pretty printing. Let's move those to helpers in lib.
v2: s/igt_show/gem_show
Signed-off-by: Michał Winiarski
Cc: Arkadiusz Hiler
Cc: Chris Wilson
Cc: Katarzyna Dec
Cc: Petri Latvala
Reviewed-by: Chris Wilson
Review
Couple of tests are using either determining scheduler capabilities or
pretty printing. Let's move those to helpers in lib. We can also keep
the value obtained from getparam static.
v2: Break the trend of expanding ioctl_wrappers
Signed-off-by: Michał Winiarski
Cc: Arkadiusz Hiler
Cc:
On Mon, Oct 16, 2017 at 08:12:46AM +, Chris Wilson wrote:
> For per-engine checks, check the engine actually exists (important for
> bsd aliasing).
>
> Signed-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
-Michał
> ---
> tests/gem_mocs_settings.c | 2 ++
&g
> ---
> tests/Makefile.sources | 1 +
> tests/core_suspend.c | 60
> ++
> 2 files changed, 61 insertions(+)
> create mode 100644 tests/core_suspend.c
Missing changes in meson.build
With that:
Review
201 - 300 of 621 matches
Mail list logo
