[ICLLP][AVC Encode] AVC encode triggers GPU hang on ICLLP #267
Comments
|
Attaching the test stream. test.yuv.gz |
|
thanks for reporting, will check it ASAP |
|
Is this issue a regression? |
|
@walter-bai, we just upstream ICLLP. so , please help to check |
|
This is not a regression. ICLLP requires media-driver to shutdown subslices which miss VME, i.e. it needs to use i915 uAPI considered here: https://patchwork.freedesktop.org/series/42285/. Mind that this uAPI is not yet available in a stock kernel. On media-driver side the use of this uAPI may need to be activated via: However, I am not 100% sure that this will fully activate the use of uAPI (@zxye: can you, please, comment: what is needed to activate the use of required i915 uAPI?). Also, mind that media-driver may need to be updated to use correct ioctl number from the mentioned patch... |
|
The patch to shut down non-VME subslices for ICLLP is not merged to master branch yet. You can find it here: zxye@74a75f0 https://patchwork.freedesktop.org/series/42285/ is not the final version of the i915 series. Once the i915 uAPI is finalized, we need to update the ioctl number and the definition of struct drm_i915_gem_context_param, rework mos_set_context_param_sseu(), and then merge 74a75f0. |
|
@zxye : thank you. Yes, i915 series is not finalized. However, we may use it in the current state and try to finalize media-driver change as much as we can. Can you, please, create a PR to media-driver from the commit you have noted above? Please, add "Fixes #267" in the commit message to the patch to link the PR to the current issue. I know that we can't merge in this PR before i915 series, but creating it will be in full accordance with kernel drm maintainers guidelines. That's really important, please, do. I still don't quite understand one thing: does your patch require adding "Slice Count Set Support" config to /etc/igfx_user_feature.txt or on the ICLLP this uAPI will be used always after your change zxye/media-driver@74a75f0? |
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Aug 13, 2018
fengguang
pushed a commit
to 0day-ci/linux
that referenced
this issue
Aug 15, 2018
We want to allow userspace to reconfigure the subslice configuration for
its own use case. To do so, we expose a context parameter to allow
adjustment of the RPCS register stored within the context image (and
currently not accessible via LRI). If the context is adjusted before
first use, the adjustment is for "free"; otherwise if the context is
active we flush the context off the GPU (stalling all users) and forcing
the GPU to save the context to memory where we can modify it and so
ensure that the register is reloaded on next execution.
The overhead of managing additional EU subslices can be significant,
especially in multi-context workloads. Non-GPGPU contexts should
preferably disable the subslices it is not using, and others should
fine-tune the number to match their workload.
We expose complete control over the RPCS register, allowing
configuration of slice/subslice, via masks packed into a u64 for
simplicity. For example,
struct drm_i915_gem_context_param arg;
struct drm_i915_gem_context_param_sseu sseu = { .class = 0,
.instance = 0, };
memset(&arg, 0, sizeof(arg));
arg.ctx_id = ctx;
arg.param = I915_CONTEXT_PARAM_SSEU;
arg.value = (uintptr_t) &sseu;
if (drmIoctl(fd, DRM_IOCTL_I915_GEM_CONTEXT_GETPARAM, &arg) == 0) {
sseu.packed.subslice_mask = 0;
drmIoctl(fd, DRM_IOCTL_I915_GEM_CONTEXT_SETPARAM, &arg);
}
could be used to disable all subslices where supported.
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context (Chris)
v7: Synchronize the requests following a powergating setting change using a global
dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users (Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Aug 30, 2018
Updated slice shutdown uAPI to v12. Fixes intel#267.
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Sep 18, 2018
Updated slice shutdown uAPI to v12. Fixes intel#267.
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Sep 21, 2018
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Oct 10, 2018
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Oct 19, 2018
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Nov 2, 2018
fengguang
pushed a commit
to 0day-ci/linux
that referenced
this issue
Nov 14, 2018
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg =
{ .param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context (Chris)
v7: Synchronize the requests following a powergating setting change using a global
dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users (Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Nov 15, 2018
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Nov 19, 2018
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Nov 20, 2018
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Nov 23, 2018
cchunbo
pushed a commit
to cchunbo/media-driver
that referenced
this issue
Dec 5, 2018
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Dec 25, 2018
fengguang
pushed a commit
to 0day-ci/linux
that referenced
this issue
Jan 8, 2019
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg =
{ .param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context (Chris)
v7: Synchronize the requests following a powergating setting change using a global
dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users (Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
v21:
* Rebase for IS_GEN changes.
v22:
* Reserve uAPI for flags straight away. (Chris Wilson)
v23:
* Rebase for RUNTIME_INFO.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v21
fengguang
pushed a commit
to 0day-ci/linux
that referenced
this issue
Jan 8, 2019
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg =
{ .param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context (Chris)
v7: Synchronize the requests following a powergating setting change using a global
dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users (Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
v21:
* Rebase for IS_GEN changes.
v22:
* Reserve uAPI for flags straight away. (Chris Wilson)
v23:
* Rebase for RUNTIME_INFO.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v21
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
fengguang
pushed a commit
to 0day-ci/linux
that referenced
this issue
Jan 14, 2019
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg =
{ .param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context (Chris)
v7: Synchronize the requests following a powergating setting change using a global
dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users (Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
v21:
* Rebase for IS_GEN changes.
v22:
* Reserve uAPI for flags straight away. (Chris Wilson)
v23:
* Rebase for RUNTIME_INFO.
v24:
* Added some headline docs for the uapi usage. (Joonas/Chris)
v25:
* Renamed class/instance to engine_class/engine_instance to avoid clash
with C++ keyword. (Tony Ye)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tony Ye <tony.ye@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v21
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
fengguang
pushed a commit
to 0day-ci/linux
that referenced
this issue
Jan 16, 2019
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg =
{ .param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context (Chris)
v7: Synchronize the requests following a powergating setting change using a global
dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users (Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
v21:
* Rebase for IS_GEN changes.
v22:
* Reserve uAPI for flags straight away. (Chris Wilson)
v23:
* Rebase for RUNTIME_INFO.
v24:
* Added some headline docs for the uapi usage. (Joonas/Chris)
v25:
* Renamed class/instance to engine_class/engine_instance to avoid clash
with C++ keyword. (Tony Ye)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tony Ye <tony.ye@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v21
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Jan 22, 2019
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Jan 30, 2019
fengguang
pushed a commit
to 0day-ci/linux
that referenced
this issue
Jan 30, 2019
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg = {
.param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu()
(Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context
(Chris)
v7: Synchronize the requests following a powergating setting change using
a global dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users
(Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
v21:
* Rebase for IS_GEN changes.
v22:
* Reserve uAPI for flags straight away. (Chris Wilson)
v23:
* Rebase for RUNTIME_INFO.
v24:
* Added some headline docs for the uapi usage. (Joonas/Chris)
v25:
* Renamed class/instance to engine_class/engine_instance to avoid clash
with C++ keyword. (Tony Ye)
v26:
* Rebased for runtime pm api changes.
v27:
* Rebased for intel_context_init.
* Wrap commit msg to 75.
v28:
(Chris Wilson)
* Use i915_gem_ggtt.
* Use i915_request_await_dma_fence to show a better example.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tony Ye <tony.ye@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v21
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> # v27
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Jan 31, 2019
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Feb 13, 2019
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Feb 14, 2019
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Feb 19, 2019
zxye
added a commit
to zxye/media-driver
that referenced
this issue
Feb 25, 2019
fengguang
pushed a commit
to 0day-ci/linux
that referenced
this issue
Mar 7, 2019
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg = {
.param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu()
(Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context
(Chris)
v7: Synchronize the requests following a powergating setting change using
a global dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users
(Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
v21:
* Rebase for IS_GEN changes.
v22:
* Reserve uAPI for flags straight away. (Chris Wilson)
v23:
* Rebase for RUNTIME_INFO.
v24:
* Added some headline docs for the uapi usage. (Joonas/Chris)
v25:
* Renamed class/instance to engine_class/engine_instance to avoid clash
with C++ keyword. (Tony Ye)
v26:
* Rebased for runtime pm api changes.
v27:
* Rebased for intel_context_init.
* Wrap commit msg to 75.
v28:
(Chris Wilson)
* Use i915_gem_ggtt.
* Use i915_request_await_dma_fence to show a better example.
v29:
* i915_timeline_set_barrier can now fail. (Chris Wilson)
v30:
* Capture some acks.
v31:
* Drop the WARN_ON from use controllable paths. (Chris Wilson)
* Use overflows_type for all checks.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tony Ye <tony.ye@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v21
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> # v27
Acked-by: Timo Aaltonen <timo.aaltonen@canonical.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
djie1
pushed a commit
to djie1/media-driver
that referenced
this issue
May 20, 2019
jason77-wang
pushed a commit
to jason77-wang/oem
that referenced
this issue
Jun 18, 2019
BugLink: http://bugs.launchpad.net/bugs/1825940 We want to allow userspace to reconfigure the subslice configuration on a per context basis. This is required for the functional requirement of shutting down non-VME enabled sub-slices on Gen11 parts. To do so, we expose a context parameter to allow adjustment of the RPCS register stored within the context image (and currently not accessible via LRI). If the context is adjusted before first use or whilst idle, the adjustment is for "free"; otherwise if the context is active we queue a request to do so (using the kernel context), following all other activity by that context, which is also marked as barrier for all following submission against the same context. Since the overhead of device re-configuration during context switching can be significant, especially in multi-context workloads, we limit this new uAPI to only support the Gen11 VME use case. In this use case either the device is fully enabled, and exactly one slice and half of the subslices are enabled. Example usage: struct drm_i915_gem_context_param_sseu sseu = { }; struct drm_i915_gem_context_param arg = { .param = I915_CONTEXT_PARAM_SSEU, .ctx_id = gem_context_create(fd), .size = sizeof(sseu), .value = to_user_pointer(&sseu) }; /* Query device defaults. */ gem_context_get_param(fd, &arg); /* Set VME configuration on a 1x6x8 part. */ sseu.slice_mask = 0x1; sseu.subslice_mask = 0xe0; gem_context_set_param(fd, &arg); v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel) v3: Add ability to program this per engine (Chris) v4: Move most get_sseu() into i915_gem_context.c (Lionel) v5: Validate sseu configuration against the device's capabilities (Lionel) v6: Change context powergating settings through MI_SDM on kernel context (Chris) v7: Synchronize the requests following a powergating setting change using a global dependency (Chris) Iterate timelines through dev_priv.gt.active_rings (Tvrtko) Disable RPCS configuration setting for non capable users (Lionel/Tvrtko) v8: s/union intel_sseu/struct intel_sseu/ (Lionel) s/dev_priv/i915/ (Tvrtko) Change uapi class/instance fields to u16 (Tvrtko) Bump mask fields to 64bits (Lionel) Don't return EPERM when dynamic sseu is disabled (Tvrtko) v9: Import context image into kernel context's ppgtt only when reconfiguring powergated slice/subslices (Chris) Use aliasing ppgtt when needed (Michel) Tvrtko Ursulin: v10: * Update for upstream changes. * Request submit needs a RPM reference. * Reject on !FULL_PPGTT for simplicity. * Pull out get/set param to helpers for readability and less indent. * Use i915_request_await_dma_fence in add_global_barrier to skip waits on the same timeline and avoid GEM_BUG_ON. * No need to explicitly assign a NULL pointer to engine in legacy mode. * No need to move gen8_make_rpcs up. * Factored out global barrier as prep patch. * Allow to only CAP_SYS_ADMIN if !Gen11. v11: * Remove engine vfunc in favour of local helper. (Chris Wilson) * Stop retiring requests before updates since it is not needed (Chris Wilson) * Implement direct CPU update path for idle contexts. (Chris Wilson) * Left side dependency needs only be on the same context timeline. (Chris Wilson) * It is sufficient to order the timeline. (Chris Wilson) * Reject !RCS configuration attempts with -ENODEV for now. v12: * Rebase for make_rpcs. v13: * Centralize SSEU normalization to make_rpcs. * Type width checking (uAPI <-> implementation). * Gen11 restrictions uAPI checks. * Gen11 subslice count differences handling. Chris Wilson: * args->size handling fixes. * Update context image from GGTT. * Postpone context image update to pinning. * Use i915_gem_active_raw instead of last_request_on_engine. v14: * Add activity tracker on intel_context to fix the lifetime issues and simplify the code. (Chris Wilson) v15: * Fix context pin leak if no space in ring by simplifying the context pinning sequence. v16: * Rebase for context get/set param locking changes. * Just -ENODEV on !Gen11. (Joonas) v17: * Fix one Gen11 subslice enablement rule. * Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson) v18: * Update commit message. (Joonas) * Restrict uAPI to VME use case. (Joonas) v19: * Rebase. v20: * Rebase for ce->active_tracker. v21: * Rebase for IS_GEN changes. v22: * Reserve uAPI for flags straight away. (Chris Wilson) v23: * Rebase for RUNTIME_INFO. v24: * Added some headline docs for the uapi usage. (Joonas/Chris) v25: * Renamed class/instance to engine_class/engine_instance to avoid clash with C++ keyword. (Tony Ye) v26: * Rebased for runtime pm api changes. v27: * Rebased for intel_context_init. * Wrap commit msg to 75. v28: (Chris Wilson) * Use i915_gem_ggtt. * Use i915_request_await_dma_fence to show a better example. v29: * i915_timeline_set_barrier can now fail. (Chris Wilson) v30: * Capture some acks. v31: * Drop the WARN_ON from use controllable paths. (Chris Wilson) * Use overflows_type for all checks. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634 Issue: intel/media-driver#267 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Zhipeng Gong <zhipeng.gong@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Tony Ye <tony.ye@intel.com> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Acked-by: Timo Aaltonen <timo.aaltonen@canonical.com> Acked-by: Takashi Iwai <tiwai@suse.de> Acked-by: Stéphane Marchesin <marcheu@chromium.org> Link: https://patchwork.freedesktop.org/patch/msgid/20190205095032.22673-4-tvrtko.ursulin@linux.intel.com (backported from commit e46c2e99f60043bfdb63f18fe775db1f688906d9) Signed-off-by: Timo Aaltonen <timo.aaltonen@canonical.com>
jason77-wang
pushed a commit
to jason77-wang/oem
that referenced
this issue
Jun 24, 2019
BugLink: http://bugs.launchpad.net/bugs/1825940 We want to allow userspace to reconfigure the subslice configuration on a per context basis. This is required for the functional requirement of shutting down non-VME enabled sub-slices on Gen11 parts. To do so, we expose a context parameter to allow adjustment of the RPCS register stored within the context image (and currently not accessible via LRI). If the context is adjusted before first use or whilst idle, the adjustment is for "free"; otherwise if the context is active we queue a request to do so (using the kernel context), following all other activity by that context, which is also marked as barrier for all following submission against the same context. Since the overhead of device re-configuration during context switching can be significant, especially in multi-context workloads, we limit this new uAPI to only support the Gen11 VME use case. In this use case either the device is fully enabled, and exactly one slice and half of the subslices are enabled. Example usage: struct drm_i915_gem_context_param_sseu sseu = { }; struct drm_i915_gem_context_param arg = { .param = I915_CONTEXT_PARAM_SSEU, .ctx_id = gem_context_create(fd), .size = sizeof(sseu), .value = to_user_pointer(&sseu) }; /* Query device defaults. */ gem_context_get_param(fd, &arg); /* Set VME configuration on a 1x6x8 part. */ sseu.slice_mask = 0x1; sseu.subslice_mask = 0xe0; gem_context_set_param(fd, &arg); v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel) v3: Add ability to program this per engine (Chris) v4: Move most get_sseu() into i915_gem_context.c (Lionel) v5: Validate sseu configuration against the device's capabilities (Lionel) v6: Change context powergating settings through MI_SDM on kernel context (Chris) v7: Synchronize the requests following a powergating setting change using a global dependency (Chris) Iterate timelines through dev_priv.gt.active_rings (Tvrtko) Disable RPCS configuration setting for non capable users (Lionel/Tvrtko) v8: s/union intel_sseu/struct intel_sseu/ (Lionel) s/dev_priv/i915/ (Tvrtko) Change uapi class/instance fields to u16 (Tvrtko) Bump mask fields to 64bits (Lionel) Don't return EPERM when dynamic sseu is disabled (Tvrtko) v9: Import context image into kernel context's ppgtt only when reconfiguring powergated slice/subslices (Chris) Use aliasing ppgtt when needed (Michel) Tvrtko Ursulin: v10: * Update for upstream changes. * Request submit needs a RPM reference. * Reject on !FULL_PPGTT for simplicity. * Pull out get/set param to helpers for readability and less indent. * Use i915_request_await_dma_fence in add_global_barrier to skip waits on the same timeline and avoid GEM_BUG_ON. * No need to explicitly assign a NULL pointer to engine in legacy mode. * No need to move gen8_make_rpcs up. * Factored out global barrier as prep patch. * Allow to only CAP_SYS_ADMIN if !Gen11. v11: * Remove engine vfunc in favour of local helper. (Chris Wilson) * Stop retiring requests before updates since it is not needed (Chris Wilson) * Implement direct CPU update path for idle contexts. (Chris Wilson) * Left side dependency needs only be on the same context timeline. (Chris Wilson) * It is sufficient to order the timeline. (Chris Wilson) * Reject !RCS configuration attempts with -ENODEV for now. v12: * Rebase for make_rpcs. v13: * Centralize SSEU normalization to make_rpcs. * Type width checking (uAPI <-> implementation). * Gen11 restrictions uAPI checks. * Gen11 subslice count differences handling. Chris Wilson: * args->size handling fixes. * Update context image from GGTT. * Postpone context image update to pinning. * Use i915_gem_active_raw instead of last_request_on_engine. v14: * Add activity tracker on intel_context to fix the lifetime issues and simplify the code. (Chris Wilson) v15: * Fix context pin leak if no space in ring by simplifying the context pinning sequence. v16: * Rebase for context get/set param locking changes. * Just -ENODEV on !Gen11. (Joonas) v17: * Fix one Gen11 subslice enablement rule. * Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson) v18: * Update commit message. (Joonas) * Restrict uAPI to VME use case. (Joonas) v19: * Rebase. v20: * Rebase for ce->active_tracker. v21: * Rebase for IS_GEN changes. v22: * Reserve uAPI for flags straight away. (Chris Wilson) v23: * Rebase for RUNTIME_INFO. v24: * Added some headline docs for the uapi usage. (Joonas/Chris) v25: * Renamed class/instance to engine_class/engine_instance to avoid clash with C++ keyword. (Tony Ye) v26: * Rebased for runtime pm api changes. v27: * Rebased for intel_context_init. * Wrap commit msg to 75. v28: (Chris Wilson) * Use i915_gem_ggtt. * Use i915_request_await_dma_fence to show a better example. v29: * i915_timeline_set_barrier can now fail. (Chris Wilson) v30: * Capture some acks. v31: * Drop the WARN_ON from use controllable paths. (Chris Wilson) * Use overflows_type for all checks. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634 Issue: intel/media-driver#267 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Zhipeng Gong <zhipeng.gong@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Tony Ye <tony.ye@intel.com> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Acked-by: Timo Aaltonen <timo.aaltonen@canonical.com> Acked-by: Takashi Iwai <tiwai@suse.de> Acked-by: Stéphane Marchesin <marcheu@chromium.org> Link: https://patchwork.freedesktop.org/patch/msgid/20190205095032.22673-4-tvrtko.ursulin@linux.intel.com (backported from commit e46c2e99f60043bfdb63f18fe775db1f688906d9) Signed-off-by: Timo Aaltonen <timo.aaltonen@canonical.com>
jason77-wang
pushed a commit
to jason77-wang/oem
that referenced
this issue
Jul 5, 2019
BugLink: http://bugs.launchpad.net/bugs/1825940 We want to allow userspace to reconfigure the subslice configuration on a per context basis. This is required for the functional requirement of shutting down non-VME enabled sub-slices on Gen11 parts. To do so, we expose a context parameter to allow adjustment of the RPCS register stored within the context image (and currently not accessible via LRI). If the context is adjusted before first use or whilst idle, the adjustment is for "free"; otherwise if the context is active we queue a request to do so (using the kernel context), following all other activity by that context, which is also marked as barrier for all following submission against the same context. Since the overhead of device re-configuration during context switching can be significant, especially in multi-context workloads, we limit this new uAPI to only support the Gen11 VME use case. In this use case either the device is fully enabled, and exactly one slice and half of the subslices are enabled. Example usage: struct drm_i915_gem_context_param_sseu sseu = { }; struct drm_i915_gem_context_param arg = { .param = I915_CONTEXT_PARAM_SSEU, .ctx_id = gem_context_create(fd), .size = sizeof(sseu), .value = to_user_pointer(&sseu) }; /* Query device defaults. */ gem_context_get_param(fd, &arg); /* Set VME configuration on a 1x6x8 part. */ sseu.slice_mask = 0x1; sseu.subslice_mask = 0xe0; gem_context_set_param(fd, &arg); v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu() (Lionel) v3: Add ability to program this per engine (Chris) v4: Move most get_sseu() into i915_gem_context.c (Lionel) v5: Validate sseu configuration against the device's capabilities (Lionel) v6: Change context powergating settings through MI_SDM on kernel context (Chris) v7: Synchronize the requests following a powergating setting change using a global dependency (Chris) Iterate timelines through dev_priv.gt.active_rings (Tvrtko) Disable RPCS configuration setting for non capable users (Lionel/Tvrtko) v8: s/union intel_sseu/struct intel_sseu/ (Lionel) s/dev_priv/i915/ (Tvrtko) Change uapi class/instance fields to u16 (Tvrtko) Bump mask fields to 64bits (Lionel) Don't return EPERM when dynamic sseu is disabled (Tvrtko) v9: Import context image into kernel context's ppgtt only when reconfiguring powergated slice/subslices (Chris) Use aliasing ppgtt when needed (Michel) Tvrtko Ursulin: v10: * Update for upstream changes. * Request submit needs a RPM reference. * Reject on !FULL_PPGTT for simplicity. * Pull out get/set param to helpers for readability and less indent. * Use i915_request_await_dma_fence in add_global_barrier to skip waits on the same timeline and avoid GEM_BUG_ON. * No need to explicitly assign a NULL pointer to engine in legacy mode. * No need to move gen8_make_rpcs up. * Factored out global barrier as prep patch. * Allow to only CAP_SYS_ADMIN if !Gen11. v11: * Remove engine vfunc in favour of local helper. (Chris Wilson) * Stop retiring requests before updates since it is not needed (Chris Wilson) * Implement direct CPU update path for idle contexts. (Chris Wilson) * Left side dependency needs only be on the same context timeline. (Chris Wilson) * It is sufficient to order the timeline. (Chris Wilson) * Reject !RCS configuration attempts with -ENODEV for now. v12: * Rebase for make_rpcs. v13: * Centralize SSEU normalization to make_rpcs. * Type width checking (uAPI <-> implementation). * Gen11 restrictions uAPI checks. * Gen11 subslice count differences handling. Chris Wilson: * args->size handling fixes. * Update context image from GGTT. * Postpone context image update to pinning. * Use i915_gem_active_raw instead of last_request_on_engine. v14: * Add activity tracker on intel_context to fix the lifetime issues and simplify the code. (Chris Wilson) v15: * Fix context pin leak if no space in ring by simplifying the context pinning sequence. v16: * Rebase for context get/set param locking changes. * Just -ENODEV on !Gen11. (Joonas) v17: * Fix one Gen11 subslice enablement rule. * Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson) v18: * Update commit message. (Joonas) * Restrict uAPI to VME use case. (Joonas) v19: * Rebase. v20: * Rebase for ce->active_tracker. v21: * Rebase for IS_GEN changes. v22: * Reserve uAPI for flags straight away. (Chris Wilson) v23: * Rebase for RUNTIME_INFO. v24: * Added some headline docs for the uapi usage. (Joonas/Chris) v25: * Renamed class/instance to engine_class/engine_instance to avoid clash with C++ keyword. (Tony Ye) v26: * Rebased for runtime pm api changes. v27: * Rebased for intel_context_init. * Wrap commit msg to 75. v28: (Chris Wilson) * Use i915_gem_ggtt. * Use i915_request_await_dma_fence to show a better example. v29: * i915_timeline_set_barrier can now fail. (Chris Wilson) v30: * Capture some acks. v31: * Drop the WARN_ON from use controllable paths. (Chris Wilson) * Use overflows_type for all checks. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634 Issue: intel/media-driver#267 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Zhipeng Gong <zhipeng.gong@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Tony Ye <tony.ye@intel.com> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Acked-by: Timo Aaltonen <timo.aaltonen@canonical.com> Acked-by: Takashi Iwai <tiwai@suse.de> Acked-by: Stéphane Marchesin <marcheu@chromium.org> Link: https://patchwork.freedesktop.org/patch/msgid/20190205095032.22673-4-tvrtko.ursulin@linux.intel.com (backported from commit e46c2e99f60043bfdb63f18fe775db1f688906d9) Signed-off-by: Timo Aaltonen <timo.aaltonen@canonical.com>
|
I'm not familiar with this bug, but since it is closed here, should the text on the bottom of the README.md, where it is referenced as an open bug, be changed? |
amshafer
pushed a commit
to amshafer/kms-drm
that referenced
this issue
Apr 19, 2020
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg = {
.param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu()
(Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context
(Chris)
v7: Synchronize the requests following a powergating setting change using
a global dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users
(Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
v21:
* Rebase for IS_GEN changes.
v22:
* Reserve uAPI for flags straight away. (Chris Wilson)
v23:
* Rebase for RUNTIME_INFO.
v24:
* Added some headline docs for the uapi usage. (Joonas/Chris)
v25:
* Renamed class/instance to engine_class/engine_instance to avoid clash
with C++ keyword. (Tony Ye)
v26:
* Rebased for runtime pm api changes.
v27:
* Rebased for intel_context_init.
* Wrap commit msg to 75.
v28:
(Chris Wilson)
* Use i915_gem_ggtt.
* Use i915_request_await_dma_fence to show a better example.
v29:
* i915_timeline_set_barrier can now fail. (Chris Wilson)
v30:
* Capture some acks.
v31:
* Drop the WARN_ON from use controllable paths. (Chris Wilson)
* Use overflows_type for all checks.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tony Ye <tony.ye@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Acked-by: Timo Aaltonen <timo.aaltonen@canonical.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
Acked-by: Stéphane Marchesin <marcheu@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20190205095032.22673-4-tvrtko.ursulin@linux.intel.com
evadot
pushed a commit
to freebsd/drm-kmod
that referenced
this issue
May 7, 2020
We want to allow userspace to reconfigure the subslice configuration on a
per context basis.
This is required for the functional requirement of shutting down non-VME
enabled sub-slices on Gen11 parts.
To do so, we expose a context parameter to allow adjustment of the RPCS
register stored within the context image (and currently not accessible via
LRI).
If the context is adjusted before first use or whilst idle, the adjustment
is for "free"; otherwise if the context is active we queue a request to do
so (using the kernel context), following all other activity by that
context, which is also marked as barrier for all following submission
against the same context.
Since the overhead of device re-configuration during context switching can
be significant, especially in multi-context workloads, we limit this new
uAPI to only support the Gen11 VME use case. In this use case either the
device is fully enabled, and exactly one slice and half of the subslices
are enabled.
Example usage:
struct drm_i915_gem_context_param_sseu sseu = { };
struct drm_i915_gem_context_param arg = {
.param = I915_CONTEXT_PARAM_SSEU,
.ctx_id = gem_context_create(fd),
.size = sizeof(sseu),
.value = to_user_pointer(&sseu)
};
/* Query device defaults. */
gem_context_get_param(fd, &arg);
/* Set VME configuration on a 1x6x8 part. */
sseu.slice_mask = 0x1;
sseu.subslice_mask = 0xe0;
gem_context_set_param(fd, &arg);
v2: Fix offset of CTX_R_PWR_CLK_STATE in intel_lr_context_set_sseu()
(Lionel)
v3: Add ability to program this per engine (Chris)
v4: Move most get_sseu() into i915_gem_context.c (Lionel)
v5: Validate sseu configuration against the device's capabilities (Lionel)
v6: Change context powergating settings through MI_SDM on kernel context
(Chris)
v7: Synchronize the requests following a powergating setting change using
a global dependency (Chris)
Iterate timelines through dev_priv.gt.active_rings (Tvrtko)
Disable RPCS configuration setting for non capable users
(Lionel/Tvrtko)
v8: s/union intel_sseu/struct intel_sseu/ (Lionel)
s/dev_priv/i915/ (Tvrtko)
Change uapi class/instance fields to u16 (Tvrtko)
Bump mask fields to 64bits (Lionel)
Don't return EPERM when dynamic sseu is disabled (Tvrtko)
v9: Import context image into kernel context's ppgtt only when
reconfiguring powergated slice/subslices (Chris)
Use aliasing ppgtt when needed (Michel)
Tvrtko Ursulin:
v10:
* Update for upstream changes.
* Request submit needs a RPM reference.
* Reject on !FULL_PPGTT for simplicity.
* Pull out get/set param to helpers for readability and less indent.
* Use i915_request_await_dma_fence in add_global_barrier to skip waits
on the same timeline and avoid GEM_BUG_ON.
* No need to explicitly assign a NULL pointer to engine in legacy mode.
* No need to move gen8_make_rpcs up.
* Factored out global barrier as prep patch.
* Allow to only CAP_SYS_ADMIN if !Gen11.
v11:
* Remove engine vfunc in favour of local helper. (Chris Wilson)
* Stop retiring requests before updates since it is not needed
(Chris Wilson)
* Implement direct CPU update path for idle contexts. (Chris Wilson)
* Left side dependency needs only be on the same context timeline.
(Chris Wilson)
* It is sufficient to order the timeline. (Chris Wilson)
* Reject !RCS configuration attempts with -ENODEV for now.
v12:
* Rebase for make_rpcs.
v13:
* Centralize SSEU normalization to make_rpcs.
* Type width checking (uAPI <-> implementation).
* Gen11 restrictions uAPI checks.
* Gen11 subslice count differences handling.
Chris Wilson:
* args->size handling fixes.
* Update context image from GGTT.
* Postpone context image update to pinning.
* Use i915_gem_active_raw instead of last_request_on_engine.
v14:
* Add activity tracker on intel_context to fix the lifetime issues
and simplify the code. (Chris Wilson)
v15:
* Fix context pin leak if no space in ring by simplifying the
context pinning sequence.
v16:
* Rebase for context get/set param locking changes.
* Just -ENODEV on !Gen11. (Joonas)
v17:
* Fix one Gen11 subslice enablement rule.
* Handle error from i915_sw_fence_await_sw_fence_gfp. (Chris Wilson)
v18:
* Update commit message. (Joonas)
* Restrict uAPI to VME use case. (Joonas)
v19:
* Rebase.
v20:
* Rebase for ce->active_tracker.
v21:
* Rebase for IS_GEN changes.
v22:
* Reserve uAPI for flags straight away. (Chris Wilson)
v23:
* Rebase for RUNTIME_INFO.
v24:
* Added some headline docs for the uapi usage. (Joonas/Chris)
v25:
* Renamed class/instance to engine_class/engine_instance to avoid clash
with C++ keyword. (Tony Ye)
v26:
* Rebased for runtime pm api changes.
v27:
* Rebased for intel_context_init.
* Wrap commit msg to 75.
v28:
(Chris Wilson)
* Use i915_gem_ggtt.
* Use i915_request_await_dma_fence to show a better example.
v29:
* i915_timeline_set_barrier can now fail. (Chris Wilson)
v30:
* Capture some acks.
v31:
* Drop the WARN_ON from use controllable paths. (Chris Wilson)
* Use overflows_type for all checks.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100899
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107634
Issue: intel/media-driver#267
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Zhipeng Gong <zhipeng.gong@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tony Ye <tony.ye@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Acked-by: Timo Aaltonen <timo.aaltonen@canonical.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
Acked-by: Stéphane Marchesin <marcheu@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20190205095032.22673-4-tvrtko.ursulin@linux.intel.com
dyang23
pushed a commit
to dyang23/media-driver-intel-
that referenced
this issue
May 14, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
libva info: VA-API version 1.2.0
libva info: va_getDriverName() returns 0
libva info: User requested driver 'iHD'
libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so
libva info: Found init function __vaDriverInit_1_2
libva info: va_openDriver() returns 0
In dmesg:
......
[1138846.694198] i915 0000:00:02.0: Resetting chip for hang on rcs0
[1138854.694139] i915 0000:00:02.0: Resetting chip for hang on vcs0
The text was updated successfully, but these errors were encountered: