If the hangcheck timer expires, check if the fw's position in the
cmdstream has advanced (changed) since last timer expiration, and
allow it up to three additional "extensions" to it's alotted time.
The intention is to continue to catch "shader stuck in a loop" type
hangs quickly, but allow more time for things that are actually
making forward progress.
Because we need to sample the CP state twice to detect if there has
not been progress, this also cuts the the timer's duration in half.
v2: Fix typo (REG_A6XX_CP_CSQ_IB2_STAT), add comment
v3: Only halve hangcheck timer duration for generations which
support progress detection (hdanton); removed unused a5xx
progress (without knowing how to adjust for data buffered
in ROQ it is too likely to report a false negative)
v4: Comment updates to better describe the total hangcheck
duration when progress detection is applied
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Tested-by: Chia-I Wu <olvaffe@gmail.com> # dEQP-GLES2.functional.flush_finish.wait
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/511584/
Link: https://lore.kernel.org/r/20221114193049.1533391-3-robdclark@gmail.com
To avoid preventing the display from coming up before the rootfs is
mounted, without resorting to packing fw in the initrd, the GPU has
this limbo state where the device is probed, but we aren't ready to
start sending commands to it. This is particularly problematic for
a6xx, since the GMU (which requires fw to be loaded) is the one that
is controlling the power/clk/icc votes.
So defer enabling runpm until we are ready to call gpu->hw_init(),
as that is a point where we know we have all the needed fw and are
ready to start sending commands to the coproc's.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/489337/
Link: https://lore.kernel.org/r/20220613182036.2567963-1-robdclark@gmail.com
From testing on sc7180-trogdor devices, reading the GMU registers
needs the GMU clocks to be enabled. Those clocks get turned on in
a6xx_gmu_resume(). Confusingly enough, that function is called as a
result of the runtime_pm of the GPU "struct device", not the GMU
"struct device". Unfortunately the current a6xx_gpu_busy() grabs a
reference to the GMU's "struct device".
The fact that we were grabbing the wrong reference was easily seen to
cause crashes that happen if we change the GPU's pm_runtime usage to
not use autosuspend. It's also believed to cause some long tail GPU
crashes even with autosuspend.
We could look at changing it so that we do pm_runtime_get_if_in_use()
on the GPU's "struct device", but then we run into a different
problem. pm_runtime_get_if_in_use() will return 0 for the GPU's
"struct device" the whole time when we're in the "autosuspend
delay". That is, when we drop the last reference to the GPU but we're
waiting a period before actually suspending then we'll think the GPU
is off. One reason that's bad is that if the GPU didn't actually turn
off then the cycle counter doesn't lose state and that throws off all
of our calculations.
Let's change the code to keep track of the suspend state of
devfreq. msm_devfreq_suspend() is always called before we actually
suspend the GPU and msm_devfreq_resume() after we resume it. This
means we can use the suspended state to know if we're powered or not.
NOTE: one might wonder when exactly our status function is called when
devfreq is supposed to be disabled. The stack crawl I captured was:
msm_devfreq_get_dev_status
devfreq_simple_ondemand_func
devfreq_update_target
qos_notifier_call
qos_max_notifier_call
blocking_notifier_call_chain
pm_qos_update_target
freq_qos_apply
apply_constraint
__dev_pm_qos_update_request
dev_pm_qos_update_request
msm_devfreq_idle_work
Fixes: eadf79286a ("drm/msm: Check for powered down HW in the devfreq callbacks")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/489124/
Link: https://lore.kernel.org/r/20220610124639.v4.1.Ie846c5352bc307ee4248d7cab998ab3016b85d06@changeid
Signed-off-by: Rob Clark <robdclark@chromium.org>
Prior to the last commit, this could result in setting the GPU
written fence value back to an older value, if we had missed
updating completed_fence prior to suspend. This was mostly
harmless as the GPU would eventually overwrite it again with
the correct value. But we should just not do this. Instead
just leave a sanity check that the fence looks plausible (in
case the GPU scribbled on memory).
Reported-by: Steev Klimaszewski <steev@kali.org>
Fixes: 95d1deb02a ("drm/msm/gem: Add fenced vma unpin")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Tested-by: Steev Klimaszewski <steev@kali.org>
Patchwork: https://patchwork.freedesktop.org/patch/490138/
Link: https://lore.kernel.org/r/20220618161120.3451993-2-robdclark@gmail.com
Following commit 17e822f759 ("drm/msm: fix unbalanced
pm_runtime_enable in adreno_gpu_{init, cleanup}"), any call to
adreno_unbind() will disable runtime PM twice, as indicated by the call
trees below:
adreno_unbind()
-> pm_runtime_force_suspend()
-> pm_runtime_disable()
adreno_unbind()
-> gpu->funcs->destroy() [= aNxx_destroy()]
-> adreno_gpu_cleanup()
-> pm_runtime_disable()
Note that pm_runtime_force_suspend() is called right before
gpu->funcs->destroy() and both functions are called unconditionally.
With recent addition of the eDP AUX bus code, this problem manifests
itself when the eDP panel cannot be found yet and probing is deferred.
On the first probe attempt, we disable runtime PM twice as described
above. This then causes any later probe attempt to fail with
[drm:adreno_load_gpu [msm]] *ERROR* Couldn't power up the GPU: -13
preventing the driver from loading.
As there seem to be scenarios where the aNxx_destroy() functions are not
called from adreno_unbind(), simply removing pm_runtime_disable() from
inside adreno_unbind() does not seem to be the proper fix. This is what
commit 17e822f759 ("drm/msm: fix unbalanced pm_runtime_enable in
adreno_gpu_{init, cleanup}") intended to fix. Therefore, instead check
whether runtime PM is still enabled, and only disable it in that case.
Fixes: 17e822f759 ("drm/msm: fix unbalanced pm_runtime_enable in adreno_gpu_{init, cleanup}")
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Tested-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Link: https://lore.kernel.org/r/20220606211305.189585-1-luzmaximilian@gmail.com
Signed-off-by: Rob Clark <robdclark@chromium.org>
of_parse_phandle() returns a node pointer with refcount
incremented, we should use of_node_put() on it when not need anymore.
a6xx_gmu_init() passes the node to of_find_device_by_node()
and of_dma_configure(), of_find_device_by_node() will takes its
reference, of_dma_configure() doesn't need the node after usage.
Add missing of_node_put() to avoid refcount leak.
Fixes: 4b565ca5a2 ("drm/msm: Add A6XX device support")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Link: https://lore.kernel.org/r/20220512121955.56937-1-linmq006@gmail.com
Signed-off-by: Rob Clark <robdclark@chromium.org>
- Fourcc modifier for tiled but not compressed layouts
- Support for userspace allocated IOVA (GPU virtual address)
- Devfreq clamp_to_idle fix
- DPU: DSC (Display Stream Compression) support
- DPU: inline rotation support on SC7280
- DPU: update DP timings to follow vendor recommendations
- DP, DPU: add support for wide bus (on newer chipsets)
- DP: eDP support
- Merge DPU1 and MDP5 MDSS driver, make dpu/mdp device the master
component
- MDSS: optionally reset the IP block at the bootup to drop
bootloader state
- Properly register and unregister internal bridges in the DRM framework
- Complete DPU IRQ cleanup
- DP: conversion to use drm_bridge and drm_bridge_connector
- eDP: drop old eDP parts again
- DPU: writeback support
- Misc small fixes
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rob Clark <robdclark@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/CAF6AEGvJCr_1D8d0dgmyQC5HD4gmXeZw=bFV_CNCfceZbpMxRw@mail.gmail.com
The motivation at this point is mainly native userspace mesa driver in a
VM guest. The one remaining synchronous "hotpath" is buffer allocation,
because guest needs to wait to know the bo's iova before it can start
emitting cmdstream/state that references the new bo. By allocating the
iova in the guest userspace, we no longer need to wait for a response
from the host, but can just rely on the allocation request being
processed before the cmdstream submission. Allocation failures (OoM,
etc) would just be treated as context-lost (ie. GL_GUILTY_CONTEXT_RESET)
or subsequent allocations (or readpix, etc) can raise GL_OUT_OF_MEMORY.
v2: Fix inuse check
v3: Change mismatched iova case to -EBUSY
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Link: https://lore.kernel.org/r/20220411215849.297838-11-robdclark@gmail.com
Signed-off-by: Rob Clark <robdclark@chromium.org>
In the cause of using the GPU via virtgpu, the host side process is
really a sort of proxy, and not terribly interesting from the PoV of
crash/fault logging. Add a way to override these per process so that
we can see the guest process's name.
v2: Handle kmalloc failure, add comment to explain kstrdup returns
NULL if passed NULL [Dan Carpenter]
Signed-off-by: Rob Clark <robdclark@chromium.org>
Link: https://lore.kernel.org/r/20220317165144.222101-4-robdclark@gmail.com
Signed-off-by: Rob Clark <robdclark@chromium.org>
When building with CONFIG_PM=y and CONFIG_PM_SLEEP=n (such as ARCH=riscv
allmodconfig), the following warnings/errors occur:
drivers/gpu/drm/msm/adreno/adreno_device.c:679:12: error: 'adreno_system_resume' defined but not used [-Werror=unused-function]
679 | static int adreno_system_resume(struct device *dev)
| ^~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/msm/adreno/adreno_device.c:655:12: error: 'adreno_system_suspend' defined but not used [-Werror=unused-function]
655 | static int adreno_system_suspend(struct device *dev)
| ^~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
These functions are only used in SET_SYSTEM_SLEEP_PM_OPS(), which
evaluates to empty when CONFIG_PM_SLEEP is not set, making these
functions unused.
To resolve this, use the SYSTEM_SLEEP_PM_OPS() and RUNTIME_PM_OPS()
macros, which were introduced in commit 1a3c7bb088 ("PM: core: Add new
*_PM_OPS macros, deprecate old ones"). They are designed to avoid these
compiler warnings while still guarding their use on
CONFIG_PM{,_SLEEP}=y.
Fixes: 7e4167c9e0 ("drm/msm/gpu: Park scheduler threads for system suspend")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/r/20220411181249.2758344-1-nathan@kernel.org
Signed-off-by: Rob Clark <robdclark@chromium.org>
The mutex wasn't really protecting anything before. Before the previous
patch we could still be racing with the scheduler's kthread, as that is
not necessarily frozen yet. Now that we've parked the sched threads,
the only race is with jobs retiring, and that is harmless, ie.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Link: https://lore.kernel.org/r/20220310234611.424743-4-robdclark@gmail.com
In the system suspend path, we don't want to be racing with the
scheduler kthreads pushing additional queued up jobs to the hw
queue (ringbuffer). So park them first. While we are at it,
move the wait for active jobs to complete into the new system-
suspend path.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Link: https://lore.kernel.org/r/20220310234611.424743-3-robdclark@gmail.com
Add a SYSPROF param for system profiling tools like Mesa's pps-producer
(perfetto) to control behavior related to system-wide performance
counter collection. In particular, for profiling, one wants to ensure
that GPU context switches do not effect perfcounter state, and might
want to suppress suspend (which would cause counters to lose state).
v2: Swap the order in msm_file_private_set_sysprof() [sboyd] and
initialize the sysprof_active refcount to one (because the under/
overflow checking in refcount_t doesn't expect a 0->1 transition)
meaning that values greater than 1 means sysprof is active.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Link: https://lore.kernel.org/r/20220304005317.776110-4-robdclark@gmail.com
