__intel_gt_reset() is really for resetting engines though
the name might suggest something else. So add a helper function
to remove confusions with no functional changes.
v2: Move intel_gt_reset_all_engines() next to
intel_gt_reset_engine() to make diff simple(John)
Cc: John Harrison <john.c.harrison@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240422201951.633-1-nirmoy.das@intel.com
Track every intel_gt_pm_get() until its corresponding release in
intel_gt_pm_put() by returning a cookie to the caller for acquire that
must be passed by on released. When there is an imbalance, we can see who
either tried to free a stale wakeref, or who forgot to free theirs.
v2: track recently added calls in gen8_ggtt_bind_get_ce and
destroyed_worker_func
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231030-ref_tracker_i915-v1-2-006fe6b96421@intel.com
Trace requires new-line at the end of message (in opposition to printk),
otherwise trace dump becomes messy.
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
Acked-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231115-eols-v1-1-d47a2f52b807@intel.com
Toggle binder context ready status when needed.
To issue gpu commands, the driver must be primed to receive
requests. Maintain binder-based GGTT update disablement until driver
probing completes. Moreover, implement a temporary disablement
of blitter prior to entering suspend, followed by re-enablement
post-resume. This is acceptable as those transition periods are
mostly single threaded.
v2: move changes to lower levels from i915_driver.c(Jani).
use new function for setting context ready status.
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Signed-off-by: Oak Zeng <oak.zeng@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230926083742.14740-7-nirmoy.das@intel.com
During resume, the steer semaphore on GT1 was observed to be held. The
hardware team has confirmed the safety of clearing steer semaphores
for all GTs during driver load/resume, as no lock acquisitions can occur
in this process by other agents.
v2: reset on resume not in intel_gt_init().
v3: do the reset on intel_gt_resume_early()
v4: do general sanitization for all GTs(Matt)
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230928130015.6758-3-nirmoy.das@intel.com
Currently a pointer to an intel_runtime_pm structure is stored in the
wake reference structures so the runtime data can be accessed. We can
save the entire device information (drm_i915_private) instead, since
we'll need to reference the new workqueue we'll add in subsequent
patches.
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ec0eb5149120d04f3d9870d7671ef10103e6fc29.1686231190.git.jani.nikula@intel.com
Driver Changes:
- Fix issue #6333: "list_add corruption" and full system lockup from
performance monitoring (Janusz)
- Give the punit time to settle before fatally failing (Aravind, Chris)
- Don't use stolen memory or BAR for ring buffers on LLC platforms (John)
- Add missing ecodes and correct timeline seqno on GuC error captures (John)
- Make sure DSM size has correct 1MiB granularity on Gen12+ (Nirmoy,
Lucas)
- Fix potential SSEU max_subslices array-index-out-of-bounds access on Gen11 (Andrea)
- Whitelist COMMON_SLICE_CHICKEN3 for UMD access on Gen12+ (Matt R.)
- Apply Wa_1408615072/Wa_1407596294 correctly on Gen11 (Matt R)
- Apply LNCF/LBCF workarounds correctly on XeHP SDV/PVC/DG2 (Matt R)
- Implement Wa_1606376872 for Xe_LP (Gustavo)
- Consider GSI offset when doing MCR lookups on Meteorlake+ (Matt R.)
- Add engine TLB invalidation for Meteorlake (Matt R.)
- Fix GSC Driver-FLR completion on Meteorlake (Alan)
- Fix GSC races on driver load/unload on Meteorlake+ (Daniele)
- Disable MC6 for MTL A step (Badal)
- Consolidate TLB invalidation flow (Tvrtko)
- Improve debug GuC/HuC debug messages (Michal Wa., John)
- Move fd_install after last use of fence (Rob)
- Initialize the obj flags for shmem objects (Aravind)
- Fix missing debug object activation (Nirmoy)
- Probe lmem before the stolen portion (Matt A)
- Improve clean up of GuC busyness stats worker (John)
- Fix missing return code checks in GuC submission init (John)
- Annotate two more workaround/tuning registers as MCR on PVC (Matt R)
- Fix GEN8_MISCCPCTL definition and remove unused INF_UNIT_LEVEL_CLKGATE (Lucas)
- Use sysfs_emit() and sysfs_emit_at() (Nirmoy)
- Make kobj_type structures constant (Thomas W.)
- make kobj attributes const on gt/ (Jani)
- Remove the unused virtualized start hack on buddy allocator (Matt A)
- Remove redundant check for DG1 (Lucas)
- Move DG2 tuning to the right function (Lucas)
- Rename dev_priv to i915 for private data naming consistency in gt/ (Andi)
- Remove unnecessary whitelisting of CS_CTX_TIMESTAMP on Xe_HP platforms (Matt R.)
-
- Escape wildcard in method names in kerneldoc (Bagas)
- Selftest improvements (Chris, Jonathan, Tvrtko, Anshuman, Tejas)
- Fix sparse warnings (Jani)
[airlied: fix unused variable in intel_workarounds]
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ZBMSb42yjjzczRhj@jlahtine-mobl.ger.corp.intel.com
The Wa_14017073508 require to send Media Busy/Idle mailbox while
accessing Media tile. As of now it is getting handled while __gt_unpark,
__gt_park. But there are various corner cases where forcewakes are taken
without __gt_unpark i.e. without sending Busy Mailbox especially during
register reads. Forcewakes are taken without busy mailbox leads to
GPU HANG. So bringing mailbox calls under forcewake calls are no feasible
option as forcewake calls are atomic and mailbox calls are blocking.
The issue already fixed in B step so disabling MC6 on A step and
reverting previous commit which handles Wa_14017073508
Fixes: 8f70f1ec58 ("drm/i915/mtl: Add Wa_14017073508 for SAMedia")
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230310061339.2495416-2-badal.nilawar@intel.com
When trying to analyse bug reports from CI, customers, etc. it can be
difficult to work out exactly what is happening on which GT in a
multi-GT system. So add GT oriented debug/error message wrappers. If
used instead of the drm_ equivalents, you get the same output but with
a GT# prefix on it.
v2: Go back to using lower case names (combined review feedback).
Convert intel_gt.c as a first step.
v3: Add gt_err_ratelimited() as well, undo one conversation that might
not have a GT pointer in some scenarios (review feedback from Michal W).
Split definitions into separate header (review feedback from Jani).
Convert all intel_gt*.c files.
v4: Re-order some macro definitions (Andi S), update (c) date (Tvrtko)
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230111200429.2139084-2-John.C.Harrison@Intel.com
Sync after v6.2-rc1 landed in drm-next.
We need to get some dependencies in place before we can merge
the fixes series from Gwan-gyeong and Chris.
References: https://lore.kernel.org/all/Y6x5JCDnh2rvh4lA@intel.com/
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Starting with MTL, there will be two GT-tiles, a render and media
tile. PXP as a service for supporting workloads with protected
contexts and protected buffers can be subscribed by process
workloads on any tile. However, depending on the platform,
only one of the tiles is used for control events pertaining to PXP
operation (such as creating the arbitration session and session
tear-down).
PXP as a global feature is accessible via batch buffer instructions
on any engine/tile and the coherency across tiles is handled implicitly
by the HW. In fact, for the foreseeable future, we are expecting this
single-control-tile for the PXP subsystem.
In MTL, it's the standalone media tile (not the root tile) because
it contains the VDBOX and KCR engine (among the assets PXP relies on
for those events).
Looking at the current code design, each tile is represented by the
intel_gt structure while the intel_pxp structure currently hangs off the
intel_gt structure.
Keeping the intel_pxp structure within the intel_gt structure makes some
internal functionalities more straight forward but adds code complexity to
code readability and maintainibility to many external-to-pxp subsystems
which may need to pick the correct intel_gt structure. An example of this
would be the intel_pxp_is_active or intel_pxp_is_enabled functionality
which should be viewed as a global level inquiry, not a per-gt inquiry.
That said, this series promotes the intel_pxp structure into the
drm_i915_private structure making it a top-level subsystem and the PXP
subsystem will select the control gt internally and keep a pointer to
it for internal reference.
This promotion comes with two noteworthy changes:
1. Exported pxp functions that are called by external subsystems
(such as intel_pxp_enabled/active) will have to check implicitly
if i915->pxp is valid as that structure will not be allocated
for HW that doesn't support PXP.
2. Since GT is now considered a soft-dependency of PXP we are
ensuring that GT init happens before PXP init and vice versa
for fini. This causes a minor ordering change whereby we previously
called intel_pxp_suspend after intel_uc_suspend but now is before
i915_gem_suspend_late but the change is required for correct
dependency flows. Additionally, this re-order change doesn't
have any impact because at that point in either case, the top level
entry to i915 won't observe any PXP events (since the GPU was
quiesced during suspend_prepare). Also, any PXP event doesn't
really matter when we disable the PXP HW (global GT irqs are
already off anyway, so even if there was a bug that generated
spurious events we wouldn't see it and we would just clean it
up on resume which is okay since the default fallback action
for PXP would be to keep the sessions off at this suspend stage).
Changes from prior revs:
v11: - Reformat a comment (Tvrtko).
v10: - Change the code flow for intel_pxp_init to make it more
cleaner and readible with better comments explaining the
difference between full-PXP-feature vs the partial-teelink
inits depending on the platform. Additionally, only do
the pxp allocation when we are certain the subsystem is
needed. (Tvrtko).
v9: - Cosmetic cleanups in supported/enabled/active. (Daniele).
- Add comments for intel_pxp_init and pxp_get_ctrl_gt that
explain the functional flow for when PXP is not supported
but the backend-assets are needed for HuC authentication
(Daniele and Tvrtko).
- Fix two remaining functions that are accessible outside
PXP that need to be checking pxp ptrs before using them:
intel_pxp_irq_handler and intel_pxp_huc_load_and_auth
(Tvrtko and Daniele).
- User helper macro in pxp-debugfs (Tvrtko).
v8: - Remove pxp_to_gt macro (Daniele).
- Fix a bug in pxp_get_ctrl_gt for the case of MTL and we don't
support GSC-FW on it. (Daniele).
- Leave i915->pxp as NULL if we dont support PXP and in line
with that, do additional validity check on i915->pxp for
intel_pxp_is_supported/enabled/active (Daniele).
- Remove unncessary include header from intel_gt_debugfs.c
and check drm_minor i915->drm.primary (Daniele).
- Other cosmetics / minor issues / more comments on suspend
flow order change (Daniele).
v7: - Drop i915_dev_to_pxp and in intel_pxp_init use 'i915->pxp'
through out instead of local variable newpxp. (Rodrigo)
- In the case intel_pxp_fini is called during driver unload but
after i915 loading failed without pxp being allocated, check
i915->pxp before referencing it. (Alan)
v6: - Remove HAS_PXP macro and replace it with intel_pxp_is_supported
because : [1] introduction of 'ctrl_gt' means we correct this
for MTL's upcoming series now. [2] Also, this has little impact
globally as its only used by PXP-internal callers at the moment.
- Change intel_pxp_init/fini to take in i915 as its input to avoid
ptr-to-ptr in init/fini calls.(Jani).
- Remove the backpointer from pxp->i915 since we can use
pxp->ctrl_gt->i915 if we need it. (Rodrigo).
v5: - Switch from series to single patch (Rodrigo).
- change function name from pxp_get_kcr_owner_gt to
pxp_get_ctrl_gt.
- Fix CI BAT failure by removing redundant call to intel_pxp_fini
from driver-remove.
- NOTE: remaining open still persists on using ptr-to-ptr
and back-ptr.
v4: - Instead of maintaining intel_pxp as an intel_gt structure member
and creating a number of convoluted helpers that takes in i915 as
input and redirects to the correct intel_gt or takes any intel_gt
and internally replaces with the correct intel_gt, promote it to
be a top-level i915 structure.
v3: - Rename gt level helper functions to "intel_pxp_is_enabled/
supported/ active_on_gt" (Daniele)
- Upgrade _gt_supports_pxp to replace what was intel_gtpxp_is
supported as the new intel_pxp_is_supported_on_gt to check for
PXP feature support vs the tee support for huc authentication.
Fix pxp-debugfs-registration to use only the former to decide
support. (Daniele)
- Couple minor optimizations.
v2: - Avoid introduction of new device info or gt variables and use
existing checks / macros to differentiate the correct GT->PXP
control ownership (Daniele Ceraolo Spurio)
- Don't reuse the updated global-checkers for per-GT callers (such
as other files within PXP) to avoid unnecessary GT-reparsing,
expose a replacement helper like the prior ones. (Daniele).
v1: - Add one more patch to the series for the intel_pxp suspend/resume
for similar refactoring
References: https://patchwork.freedesktop.org/patch/msgid/20221202011407.4068371-1-alan.previn.teres.alexis@intel.com
Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221208180542.998148-1-alan.previn.teres.alexis@intel.com
- gvt-next stuff mostly with refactor for the new MDEV interface.
i915 Changes:
- PSR fixes and improvements (Jouni)
- DP DSC fixes (Vinod, Jouni)
- More general display cleanups (Jani)
- More display collor management cleanup targetting degamma (Ville)
- remove circ_buf.h includes (Jiri)
- wait power off delay at driver remove to optimize probe (Jani)
- More audio cleanup targeting the ELD precompute readout (Ville)
- Enable DC power states on all eDP ports (Imre)
- RPL-P stepping info (Matt Atwood)
- MTL enabling patches (RK)
- Removal of DG2 force_probe (Matt)
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEEbSBwaO7dZQkcLOKj+mJfZA7rE8oFAmN3+28ACgkQ+mJfZA7r
E8rGZQf9HiTNW1yM1/YGq7ezD6/XN2sbCaH/iVNd0+FHxz4HHaTJaEVQ1xEBTdIb
poMQiWl64ig3t2LuDrRGwlsVmFYV3vK5byt758sMpLb/M3cWbbB2swsW4YTbORM6
qcatv9eYyuoiylwj6fVLFPMXpwNTyCdbngbLZt+tTcl1oxaZYp0PVXBkzlKFP02U
KL4p+Dv4OHSjP38beIzUaHz3IJJAik/ttsxa1JhhC8F9UUJs7kuo6GXMY1OzRNpc
jKj3+F0OtDRA2xaw7YpbLI8yt+pWzerqFsnZDAvvx8Js7SQLkyNtjkuHusp8OP77
x99MlyOfbSfBnkbdm4Rhm7IIPTiUnw==
=SHn9
-----END PGP SIGNATURE-----
Merge tag 'drm-intel-next-2022-11-18' of git://anongit.freedesktop.org/drm/drm-intel into drm-next
GVT Changes:
- gvt-next stuff mostly with refactor for the new MDEV interface.
i915 Changes:
- PSR fixes and improvements (Jouni)
- DP DSC fixes (Vinod, Jouni)
- More general display cleanups (Jani)
- More display collor management cleanup targetting degamma (Ville)
- remove circ_buf.h includes (Jiri)
- wait power off delay at driver remove to optimize probe (Jani)
- More audio cleanup targeting the ELD precompute readout (Ville)
- Enable DC power states on all eDP ports (Imre)
- RPL-P stepping info (Matt Atwood)
- MTL enabling patches (RK)
- Removal of DG2 force_probe (Matt)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/Y3f71obyEkImXoUF@intel.com
This workaround is added for Media tile of MTL A step. It is to help
pcode workaround which handles the hardware issue seen during package C2/C3
transitions due to RC6 entry/exit transitions on Media tile. As a part of
workaround pcode expect kmd to send mailbox message "media busy" when
components of Media tile are in use and "media idle" otherwise.
As per workaround description gucrc need to be disabled so enabled
host based RC for Media tile.
v2:
- Correct workaround id (Matt)
- Fix review comments (Rodrigo)
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Cc: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221103184559.2306481-1-badal.nilawar@intel.com
Turns out many of the files that need i915_reg.h get it implicitly via
{display/intel_de.h, gt/intel_context.h} -> i915_trace.h -> i915_irq.h
-> i915_reg.h. Since i915_trace.h doesn't actually need i915_irq.h,
makes sense to drop it, but that requires adding quite a few new
includes all over the place.
Prefer including i915_reg.h where needed instead of adding another
implicit include, because eventually we'll want to split up i915_reg.h
and only include the specific registers at each place.
Also some places actually needed i915_irq.h too.
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/6e78a2e0ac1bffaf5af3b5ccc21dff05e6518cef.1668008071.git.jani.nikula@intel.com
In order to get the GSC Support merged on drm-intel-gt-next
in a clean fashion we needed this ATS-M patch to avoid
conflict in i915_pci.c:
commit 412c942bdf ("drm/i915/ats-m: add ATS-M platform info")
--
Fixing a silent conflict on drivers/gpu/drm/i915/gt/intel_gt_gmch.c:
- if (!intel_vtd_active(i915))
+ if (!i915_vtd_active(i915))
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Initiating a reset when the command streamer is not idle or in the
middle of executing an MI_FORCE_WAKE can result in a hang. Multiple
command streamers can be part of a single reset domain, so resetting one
would mean resetting all command streamers in that domain.
To workaround this, before initiating a reset, ensure that all command
streamers within that reset domain are either IDLE or are not executing
a MI_FORCE_WAKE.
Enable GuC PRE_PARSER WA bit so that GuC follows the WA sequence when
initiating engine-resets.
For gt-resets, ensure that i915 applies the WA sequence.
Opens to address in future patches:
- The part of the WA to wait for pending forcewakes is also applicable
to execlists backend.
- The WA also needs to be applied for gen11
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220415224025.3693037-3-umesh.nerlige.ramappa@intel.com
On a multi-tile platform, each tile has its own registers + GGTT
space, and BAR 0 is extended to cover all of them.
Up to four GTs are supported in i915->gt[], with slot zero
shadowing the existing i915->gt0 to enable source compatibility
with legacy driver paths. A for_each_gt macro is added to iterate
over the GTs and will be used by upcoming patches that convert
various parts of the driver to be multi-gt aware.
Only the primary/root tile is initialized for now; the other
tiles will be detected and plugged in by future patches once the
necessary infrastructure is in place to handle them.
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@gmail.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220318233938.149744-4-andi.shyti@linux.intel.com
Remove the local yesno() implementation and adopt the str_yes_no() from
linux/string_helpers.h.
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220225234631.3725943-1-lucas.demarchi@intel.com
With GuC handling scheduling, i915 is not aware of the time that a
context is scheduled in and out of the engine. Since i915 pmu relies on
this info to provide engine busyness to the user, GuC shares this info
with i915 for all engines using shared memory. For each engine, this
info contains:
- total busyness: total time that the context was running (total)
- id: id of the running context (id)
- start timestamp: timestamp when the context started running (start)
At the time (now) of sampling the engine busyness, if the id is valid
(!= ~0), and start is non-zero, then the context is considered to be
active and the engine busyness is calculated using the below equation
engine busyness = total + (now - start)
All times are obtained from the gt clock base. For inactive contexts,
engine busyness is just equal to the total.
The start and total values provided by GuC are 32 bits and wrap around
in a few minutes. Since perf pmu provides busyness as 64 bit
monotonically increasing values, there is a need for this implementation
to account for overflows and extend the time to 64 bits before returning
busyness to the user. In order to do that, a worker runs periodically at
frequency = 1/8th the time it takes for the timestamp to wrap. As an
example, that would be once in 27 seconds for a gt clock frequency of
19.2 MHz.
Note:
There might be an over-accounting of busyness due to the fact that GuC
may be updating the total and start values while kmd is reading them.
(i.e kmd may read the updated total and the stale start). In such a
case, user may see higher busyness value followed by smaller ones which
would eventually catch up to the higher value.
v2: (Tvrtko)
- Include details in commit message
- Move intel engine busyness function into execlist code
- Use union inside engine->stats
- Use natural type for ping delay jiffies
- Drop active_work condition checks
- Use for_each_engine if iterating all engines
- Drop seq locking, use spinlock at GuC level to update engine stats
- Document worker specific details
v3: (Tvrtko/Umesh)
- Demarcate GuC and execlist stat objects with comments
- Document known over-accounting issue in commit
- Provide a consistent view of GuC state
- Add hooks to gt park/unpark for GuC busyness
- Stop/start worker in gt park/unpark path
- Drop inline
- Move spinlock and worker inits to GuC initialization
- Drop helpers that are called only once
v4: (Tvrtko/Matt/Umesh)
- Drop addressed opens from commit message
- Get runtime pm in ping, remove from the park path
- Use cancel_delayed_work_sync in disable_submission path
- Update stats during reset prepare
- Skip ping if reset in progress
- Explicitly name execlists and GuC stats objects
- Since disable_submission is called from many places, move resetting
stats to intel_guc_submission_reset_prepare
v5: (Tvrtko)
- Add a trylock helper that does not sleep and synchronize PMU event
callbacks and worker with gt reset
v6: (CI BAT failures)
- DUTs using execlist submission failed to boot since __gt_unpark is
called during i915 load. This ends up calling the GuC busyness unpark
hook and results in kick-starting an uninitialized worker. Let
park/unpark hooks check if GuC submission has been initialized.
- drop cant_sleep() from trylock helper since rcu_read_lock takes care
of that.
v7: (CI) Fix igt@i915_selftest@live@gt_engines
- For GuC mode of submission the engine busyness is derived from gt time
domain. Use gt time elapsed as reference in the selftest.
- Increase busyness calculation to 10ms duration to ensure batch runs
longer and falls within the busyness tolerances in selftest.
v8:
- Use ktime_get in selftest as before
- intel_reset_trylock_no_wait results in a lockdep splat that is not
trivial to fix since the PMU callback runs in irq context and the
reset paths are tightly knit into the driver. The test that uncovers
this is igt@perf_pmu@faulting-read. Drop intel_reset_trylock_no_wait,
instead use the reset_count to synchronize with gt reset during pmu
callback. For the ping, continue to use intel_reset_trylock since ping
is not run in irq context.
- GuC PM timestamp does not tick when GuC is idle. This can potentially
result in wrong busyness values when a context is active on the
engine, but GuC is idle. Use the RING TIMESTAMP as GPU timestamp to
process the GuC busyness stats. This works since both GuC timestamp and
RING timestamp are synced with the same clock.
- The busyness stats may get updated after the batch starts running.
This delay causes the busyness reported for 100us duration to fall
below 95% in the selftest. The only option at this time is to wait for
GuC busyness to change from idle to active before we sample busyness
over a 100us period.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211027004821.66097-2-umesh.nerlige.ramappa@intel.com
During the power event S3+ sleep/resume, hardware will lose all the
encryption keys for every hardware session, even though the
session state might still be marked as alive after resume. Therefore,
we should consider the session as dead on suspend and invalidate all the
objects. The session will be automatically restarted on the first
protected submission on resume.
v2: runtime suspend also invalidates the keys
v3: fix return codes, simplify rpm ops (Chris), use the new worker func
v4: invalidate the objects on suspend, don't re-create the arb sesson on
resume (delayed to first submission).
v5: move irq changes back to irq patch (Rodrigo)
v6: drop invalidation in runtime suspend (Rodrigo)
Signed-off-by: Huang, Sean Z <sean.z.huang@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210924191452.1539378-13-alan.previn.teres.alexis@intel.com
Just evict unpinned objects to system. For pinned LMEM objects,
make a backup system object and blit the contents to that.
Backup is performed in three steps,
1: Opportunistically evict evictable objects using the gpu blitter.
2: After gt idle, evict evictable objects using the gpu blitter. This will
be modified in an upcoming patch to backup pinned objects that are not used
by the blitter itself.
3: Backup remaining pinned objects using memcpy.
Also move uC suspend to after 2) to make sure we have a functional GuC
during 2) if using GuC submission.
v2:
- Major refactor to make sure gem_exec_suspend@hang-SX subtests work, and
suspend / resume works with a slightly modified GuC submission enabling
patch series.
v3:
- Fix a potential use-after-free (Matthew Auld)
- Use i915_gem_object_create_shmem() instead of
i915_gem_object_create_region (Matthew Auld)
- Minor simplifications (Matthew Auld)
- Fix up kerneldoc for i195_ttm_restore_region().
- Final lmem_suspend() call moved to i915_gem_backup_suspend from
i915_gem_suspend_late, since the latter gets called at driver unload
and we don't unnecessarily want to run it at that time.
v4:
- Interface change of ttm- & lmem suspend / resume functions to use
flags rather than bools. (Matthew Auld)
- Completely drop the i915_gem_backup_suspend change (Matthew Auld)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-5-thomas.hellstrom@linux.intel.com
With GuC submission on DG1, the execution of the requests times out
for the gem_exec_suspend igt test case after executing around 800-900
of 1000 submitted requests.
Given the time we allow elsewhere for fences to signal (in the order of
seconds), increase the timeout before we mark the gt wedged and proceed.
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Acked-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210922062527.865433-4-thomas.hellstrom@linux.intel.com
Reset implementation for new GuC interface. This is the legacy reset
implementation which is called when the i915 owns the engine hang check.
Future patches will offload the engine hang check to GuC but we will
continue to maintain this legacy path as a fallback and this code path
is also required if the GuC dies.
With the new GuC interface it is not possible to reset individual
engines - it is only possible to reset the GPU entirely. This patch
forces an entire chip reset if any engine hangs.
v2:
(Michal)
- Check for -EPIPE rather than -EIO (CT deadlock/corrupt check)
v3:
(John H)
- Split into a series of smaller patches
v4:
(John H)
- Fix typo
- Add braces around if statements in reset code
v5:
(Checkpatch)
- Fix warnings
Cc: John Harrison <john.c.harrison@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: John Harrison <john.c.harrison@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210727002348.97202-9-matthew.brost@intel.com
This essentially reverts
commit 84a1074920
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed Jan 24 11:36:08 2018 +0000
drm/i915: Shrink the GEM kmem_caches upon idling
mm/vmscan.c:do_shrink_slab() is a thing, if there's an issue with it
then we need to fix that there, not hand-roll our own slab shrinking
code in i915.
Also when this was added there was only one other caller of
kmem_cache_shrink (added 2005 to the acpi code). Now there's a 2nd one
outside of i915 code in a kunit test, which seems legit since that
wants to very carefully control what's in the kmem_cache. This out of
a total of over 500 calls to kmem_cache_create. This alone should have
been warning sign enough that we're doing something silly.
Noticed while reviewing a patch set from Jason to fix up some issues
in our i915_init() and i915_exit() module load/cleanup code. Now that
i915_globals.c isn't any different than normal init/exit functions, we
should convert them over to one unified table and remove
i915_globals.[hc] entirely.
v2: Improve commit message (Jason)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: David Airlie <airlied@linux.ie>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210721183229.4136488-1-daniel.vetter@ffwll.ch
Since we wake the GT up before executing a request, and go to sleep as
soon as it is retired, the GT wake time not only represents how long the
device is powered up, but also provides a summary, albeit an overestimate,
of the device runtime (i.e. the rc0 time to compare against rc6 time).
v2: s/busy/awake/
v3: software-gt-awake-time and I915_PMU_SOFTWARE_GT_AWAKE_TIME
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reported-by: kernel test robot <oliver.sang@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201215154456.13954-1-chris@chris-wilson.co.uk
Getting wedged device on driver init is pretty much unrecoverable.
Since we're running various scenarios that may potentially hit this in
CI (module reload / selftests / hotunplug), and if it happens, it means
that we can't trust any subsequent CI results, we should just apply the
taint to let the CI know that it should reboot (CI checks taint between
test runs).
v2: Comment that WEDGED_ON_INIT is non-recoverable, distinguish
WEDGED_ON_INIT from WEDGED_ON_FINI (Chris)
v3: Appease checkpatch, fixup search-replace logic expression mindbomb
in assert (Chris)
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Petri Latvala <petri.latvala@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20200706144107.204821-1-michal@hardline.pl
If we injected an error (such as pretending the GuC firmware was
broken), then suppress the error message as it is expected and our CI
complains if it sees any *ERROR*.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200603104657.25651-1-chris@chris-wilson.co.uk
We see that if the HW doesn't actually sleep, the HW may eat the poison
we set in its write-only HWSP during sanitize:
intel_gt_resume.part.8: 0000:00:02.0
__gt_unpark: 0000:00:02.0
gt_sanitize: 0000:00:02.0 force:yes
process_csb: 0000:00:02.0 vcs0: cs-irq head=5, tail=90
process_csb: 0000:00:02.0 vcs0: csb[0]: status=0x5a5a5a5a:0x5a5a5a5a
assert_pending_valid: Nothing pending for promotion!
The CS TAIL pointer should have been reset by reset_csb_pointers(), so
in this case it is likely that we have read back from the CPU cache and
so we must clflush our control over that page. In doing so, push the
sanitisation to the start of the GT sequence so that our poisoning is
assuredly before we start talking to the HW.
References: https://gitlab.freedesktop.org/drm/intel/-/issues/1794
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200427084000.10999-1-chris@chris-wilson.co.uk
For many configuration details within RC6 and RPS we are programming
intervals for the internal clocks. From gen11, these clocks are
configuration via the RPM_CONFIG and so for convenience, we would like
to convert to/from more natural units (ns).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Andi Shyti <andi.shyti@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200424162805.25920-2-chris@chris-wilson.co.uk
Before we resume, we reset the HW so we restart from a known good state.
However, as a part of the reset process, we drain our pending CS event
queue -- and if we are resuming that does not correspond to internal
state. On setup, we are scrubbing the CS pointers, but alas only on
setup.
Apply the sanitization not just to setup, but to all resumes.
Reported-by: Venkata Ramana Nayana <venkata.ramana.nayana@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Venkata Ramana Nayana <venkata.ramana.nayana@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200416114117.3460-1-chris@chris-wilson.co.uk
The workarounds are a common "feature" across gens and submission
mechanisms and we already call the other WA related functions from
common engine ones (<setup/cleanup>_common), so it makes sense to
do the same with WA application. Medium-term, This will help us
reduce the duplication once the GuC resume function is added, but short
term it will also allow us to use the workaround lists for pre-gen8
engine workarounds.
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20200131075716.2212299-2-chris@chris-wilson.co.uk
Before we idle, on parking, we switch to the kernel context such that we
have a scratch context loaded while the GPU idle, protecting any
precious user state. Be paranoid and assume that the idle state may have
been trashed, and reset the kernel_context image after idling.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Imre Deak <imre.deak@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200102131707.1463945-4-chris@chris-wilson.co.uk
Since intel_gt_resume() is always immediately proceeded by init_hw, pull
the call into intel_gt_resume, where we have the rpm and fw already
held.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Andi Shyti <andi.shyti@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191222144046.1674865-1-chris@chris-wilson.co.uk
Begin pulling the GT setup underneath a single GT umbrella; let intel_gt
take ownership of its engines! As hinted, the complication is the
lifetime of the probed engine versus the active lifetime of the GT
backends. We need to detect the engine layout early and keep it until
the end so that we can sanitize state on takeover and release.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Andi Shyti <andi.shyti@intel.com>
Acked-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191222120752.1368352-1-chris@chris-wilson.co.uk
For very light workloads that frequently park, acquiring the display
power well (required to prevent the dmc from trashing the system) takes
longer than the execution. A good example is the igt_coherency selftest,
which is slowed down by an order of magnitude in the worst case with
powerwell cycling. To prevent frequent cycling, while keeping our fast
soft-rc6, use a timer to delay release of the display powerwell.
Fixes: 311770173f ("drm/i915/gt: Schedule request retirement when timeline idles")
References: https://gitlab.freedesktop.org/drm/intel/issues/848
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191218093504.3477048-1-chris@chris-wilson.co.uk
New macros ENGINE_TRACE(), CE_TRACE(), RQ_TRACE() and
GT_TRACE() are introduce to tag device name and engine
name with contexts and requests tracing in i915.
Cc: Sudeep Dutt <sudeep.dutt@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Venkata Sandeep Dhanalakota <venkata.s.dhanalakota@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20191213155152.69182-2-venkata.s.dhanalakota@intel.com
Now that we rapidly park the GT when the GPU idles, we often find
ourselves idling faster than the RC6 promotion timer. Thus if we tell
the GPU to enter RC6 manually as we park, we can do so quicker (by
around 50ms, half an EI on average) and marginally increase our
powersaving across all execlists platforms.
v2: Now with a selftest to check we can enter RC6 manually
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Andi Shyti <andi.shyti@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Acked-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191127095657.3209854-1-chris@chris-wilson.co.uk
