In safety build, gops.gr.init.set_default_compute_regs() is invoked in
nvgpu_gr_obj_ctx_alloc() for all classes. Before enabling graphics
classes in safety this was executed only for compute class. But since
graphics classes are supported in safety now this call should be made
only for compute classes.
Add gops.gpu_class.is_valid_compute() check before calling this
function.
Bug 3482988
Change-Id: If3722be36e779195122f54925ad122871cf13317
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2667324
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Sagar Kamble <skamble@nvidia.com>
GVS: Gerrit_Virtual_Submit
In DRIVE 6.0, NvGPU is allowed to report only 32-bit metadata to
Safety_Services. So, there is no need to have distinct APIs for
reporting errors from units like GR, MM, FIFO to SDL unit. All
these error reporting APIs will be replaced with a single API. To
meet this objective, this patch does the following changes:
- Replaces nvgpu_report_*_err with nvgpu_report_err_to_sdl.
- Removes the reporting of error messages.
- Replaces nvgpu_log() with nvgpu_err(), for error reporting.
- Removes error reporting to Safety_Services from nvgpu_report_*_err.
However, nvgpu_report_*_err APIs and their related files are not
removed. During the creation of nvgpu-mon, they will be moved under
nvgpu-rm, in debug builds.
Note:
- There will be a follow-up patch to fix error IDs.
- As discussed in https://nvbugs/3491596 (comment #12), the high
level expectation is to report only errors.
JIRA NVGPU-7450
Change-Id: I428f2a9043086462754ac36a15edf6094985316f
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2662590
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
- implemented device info cmd to send device info to the gsp for
runlist submission. Currently GSP scheduler support only GR
engine '0' instance.
- implemented runlist submit cmd. GSP firmware will submit the
corresponding runlist by writing into submit registers. This
command is direct replacement of hw_submit ga10b hal for GR engine.
NVGPU-6790
Signed-off-by: Ramesh Mylavarapu <rmylavarapu@nvidia.com>
Change-Id: I5dc573a6ad698fe20b49a3466a8e50b94cae74df
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2608923
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
- NVGPU need to check for priv lockdown release before configuring
any priv registers. In current GSP bootstrap sequence has irq
configuration after GSP engine reset which is causing priv errors.
So irq configuration should be done after GSP firmware releases
priv lockdown.
- Removed clearing irq mask and dest registers before configuring
them as GSP firmware would have done partial irq configuration
before releasing the priv.
NVGPU-7342
Signed-off-by: Ramesh Mylavarapu <rmylavarapu@nvidia.com>
Change-Id: I4b6e83452c051654253e02bfb72330b3d6aec3fd
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2649826
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Separated gsp unit into three unit:
- GSP unit which holds the core functionality of GSP RISCV core,
bootstrap, interrupt, etc.
- GSP Scheduler to hold the cmd/msg management, IPC, etc.
- GSP Test to hold stress test ucode specific support.
NVGPU-7492
Signed-off-by: Ramesh Mylavarapu <rmylavarapu@nvidia.com>
Change-Id: I12340dc776d610502f28c8574843afc7481c0871
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2660619
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Initially, REMAP only worked with big pages but in some cases
only small pages are supported where REMAP functionality is
also needed.
This cleans up some page size assumptions. In particular, on a
remap request, the nvgpu_vm_area is found from the passed in VA,
but can only be done from virt_offset_in_pages if we're also
told the page size.
This now occurs from _PAGESIZE_ flags which are required by
both map and unmap operations.
Jira NVGPU-6804
Change-Id: I311980a1b5e0e5e1840bdc1123479350a5c9d469
Signed-off-by: Chris Johnson <cwj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2566087
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
A reference to the default scheduling domain is taken when a TSG is
opened. Although the explicit bind is designed to support only one bind,
the TSG is bound to the default one implicitly at that point. Release
the reference to avoid leaking it.
The domain might be null at that point if the default domain has been
removed; in that case there's just no domain to put back.
Change-Id: I7db43f7bbb2a8c86c391280eb7fa32431c8982da
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2663420
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
The nvgpu timeout API has an internal override for presilicon mode by
default: in presi simulation environments the timeouts never trigger.
This behaviour is intended in the original usecase of the timer unit
with hardware polling loops. In pure software logic though, the timer
must trigger after the specified timeout even in presi mode so add a new
init function to produce a timer for software logic. Use this new kind
of timer in channel and scheduling worker threads.
The channel worker currently times out for just the purpose of the
channel watchdog timer which has its own internal timer. Although that's
just software, the general expectation is that the watchdog does not
trigger in presilicon tests that run slower than usual. The internal
watchdog timer thus keeps the non-sw mode.
Bug 3521828
Change-Id: I48ae8522c7ce2346a930e766528d8b64195f81d8
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2662541
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
CBC contig allocation requires mempool node in DT and the
node can be used for contig allocations. The code duplication
can be avoided by unifying the code from vgpu.
Change-Id: I6eaa1d0c9db47b158602bf0ba68ce4e09cf487a7
Signed-off-by: Dinesh T <dt@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2650459
Reviewed-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-by: Richard Zhao <rizhao@nvidia.com>
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
GVS: Gerrit_Virtual_Submit
At present, for each resume cycle the driver sends the
"nvgpu_cbc_op_clear" command to L2 cache controller, this causes the
contents of the compression bit backing store to be cleared, and results
in corrupting the metadata for all the compressible surfaces already allocated.
Fix this by updating cbc.init function to be aware of resume state and
not clear the compression bit backing store, instead issue
"nvgpu_cbc_op_invalide" command, this should leave the backing store in a
consistent state across suspend/resume cycles.
The updated cbc.init HAL for gv11b is reusable acrosss multiple chips, hence
remove unnecessary chip specific cbc.init HALs.
Bug 3483688
Change-Id: I2de848a083436bc085ee98e438874214cb61261f
Signed-off-by: Antony Clince Alex <aalex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2660075
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
In DRIVE 6.0, NvGPU needs to support error reporting in QNX-Safety,
QNX-Standard, and Linux. To support error reporting in all these
platform variants, SDL unit will be moved from QNX to common code.
As part of this refactoring activity, this patch removes ops assignment
for report error. Also, it removes API calls that are used to take
time-stamp for stall interrupt thread. This time-stamp APIs will be
brought back later, if required to support periodic diagnostics.
JIRA NVGPU-7353
Change-Id: I38536019dc7165e6a97674863b37d009854af948
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2655958
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Shashank Singh <shashsingh@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Separate nvgpu_cg_blcg/slcg_fb_ltc_load_enable function
into nvgpu_cg_blcg/slcg_fb_load_enable and
nvgpu_cg_blcg/slcg_ltc_load_enable.
Program fb slcg/blcg prod values during fb init and
program ltc slcg/blcg prod values after acr boot to
have correct privilege for ltc cg programming.
Update unit tests to have sperate blcg/slcg hal for
fb and ltc programming.
Bug 3423549
Change-Id: Icdb45528abe1a3ab68a47f689310dee9a4fe9366
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2646039
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Allocator (bitmap, buddy, page) debugfs files are not cleaned up when
the allocators are destroyed. This leads to warning logs from nvgpu
like below:
[21073.493000] debugfs: File 'gk20a_as_17' in directory 'allocators' already present!
[21073.493026] debugfs: File 'gk20a_as_17-sys' in directory 'allocators' already present!
Remove the per-allocator debugfs node when destroying an allocator in
runtime.
While at this, add missing nvgpu_allocator locking to the function
nvgpu_bitmap_alloc_destroy. And create nop functions for the
functions nvgpu_init_alloc_debug and nvgpu_fini_alloc_debug
when CONFIG_DEBUG_FS is not defined to avoid adding the
CONFIG checks at multiple places.
Move gk20a_debug_deinit to the end of gk20a_free_cb called in nvgpu_put
as that tears down all debugfs entries. Allocator destroy happens as
part of nvgpu_put call and it can lead to invalid debugfs dentry
access if gk20a_debug_deinit is called before it.
Bug 3481097
Change-Id: I8a66bcf6ade7e5707f9207c78a54d12d7bd94c02
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2648012
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
nvs worker thread is created on each resume and deinitialized on every
suspend. nvgpu can be resumed when process is getting killed. Thread
creation can fail when the process is getting killed. That will lead
to driver resume failure.
To avoid the issue above, don't stop the nvs worker thread in suspend
and let the first created thread handle the nvs work always.
Deinitialize the nvs worker thread during nvgpu unload.
Also, log the error returned by nvgpu_thread_create in the function
nvgpu_worker_start.
bug 3480192
Change-Id: I8d5d9e7716a950b162cc3c2d9fcfde07c4edfcf6
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2646218
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
This patch performs the following improvements for VAB:
1) It avoids an infinite loop when collecting VAB information.
Previously, nvgpu incorrectly assumed that the valid bit would
be eventually set for the checker when polling. It may not be set
if a VAB-related fault has occurred.
2) It handles the VAB_ERROR mmu fault which may be caused for various
reasons: invalid vab buffer address, tracking in protected mode,
etc. The recovery sequence is to set the vab buffer size to 0 and
then to the original size. This clears the VAB_ERROR bit. After
reseting, the old register values are again set in the recovery
code sequence.
3) Use correct number of VAB buffers. There's only one VAB buffer on
ga10b, not two.
4) Simplify logic.
Bug 3374805
Bug 3465734
Bug 3473147
Change-Id: I716f460ef37cb848ddc56a64c6f83024c4bb9811
Signed-off-by: Martin Radev <mradev@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2621290
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
During gpu re-initialization(rail gate exit/sc7 exit), vab buffers
needs to programmed in hw by writing vab buffer address to
NV_PFB_PRI_MMU_VIDMEM_ACCESS_BIT_BUFFER_LO(HI)_ADDR and
setting vab entries to NV_PFB_PRI_MMU_VIDMEM_ACCESS_BIT_BUFFER_SIZE_VAL.
For this moved nvgpu nvgpu_fb_vab_init_hal from
nvgpu_init_mm_setup_sw to nvgpu_init_mm_support.
nvgpu_init_mm_setup_sw is skipping during re-init, because
sw_ready is set during first boot.
Bug 3468562
Change-Id: Iee2bd4bc5165397ea4f9cca0ba6744eaf361a342
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2643244
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Martin Radev <mradev@nvidia.com>
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
- Make the domain scheduler timeslice type nanoseconds to future proof
the interface
- Return -ENOSYS from ioctls if the nvs code is not initialized
- Return the number of domains also when user supplied array is present
- Use domain id instead of name for TSG binding
- Improve documentation in the uapi headers
- Verify that reserved fields are zeroed
- Extend some internal logging
- Release the sched mutex on alloc error
- Add file mode checks in the nvs ioctls. The create and remove ioctls
require writable file permissions, while the query does not; this
allows filesystem based access control on domain management on the
single dev node.
Jira NVGPU-6788
Change-Id: I668eb5972a0ed1073e84a4ae30e3069bf0b59e16
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2639017
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Move away from the prototype call in channel wdt worker and create a
separate worker thread for the domain scheduler. The details of runlist
domains are still encapsulated in the runlist code; the domain scheduler
controls when to switch domains. Switching happens based on domain
timeslices or when the current domain is deleted.
The worker thread is paused on railgate and spun back on poweron. The
scheduler data was also left dangling, so fix that by deinitializing all
nvs-related when gk20a_remove_support() is called. The runlist domains
already get freed as part of fifo removal.
Jira NVGPU-6427
Change-Id: I64f42498f8789448d9becdd209b7878ef0fdb124
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2632579
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
The flag pmu->pg->golden_image_initialized is set to
true during initial GPU context creation and is not
cleared while the GPU goes into pm_suspend (during railgate).
Hence, when the GPU resumes after un-railgate it retains
the previous value which can cause ELPG to kick in immediately.
Due to this, when ELPG and Railgating are enabled, IDLE_SNAP
is seen for read access of gr_gpc0_tpc0_sm_arch_r reg.
To resolve this, if golden image is ready set the
pmu->pg->golden_image_initialized to suspend state during railgate,
to delay the early enable of ELPG. Add a new
pmu_init_golden_img_state hal in the NVGPU_INIT_TABLE_ENTRY.
This will be called after all the GR access is done and GPU resumes
completely after un-railgate. This hal will then check if
golden_image_initialized flag is in suspend state, it will set it
to ready state and then re-enable ELPG.
Bug 3431798
Change-Id: I1fee83e66e09b6b78d385bbe60529d0724f79e79
Signed-off-by: Divya <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2639188
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Now that the main nvsched code exists in the nvgpu build, make it
control the runlist domains. As a new nvs domain is created, create the
relevant runlist data too. To support the default domain, create a
default nvs domain at boot.
The scheduling domain code owns the responsibility of domain lifetime,
and runlist domains exist to serve that logic although the RL domains
are directly used by channel and TSG logic. Add refcounting to the
scheduler uapi level to make sure that busy domains (that still have TSG
participants) do not get removed too early.
Adjust error injection sensitive unit tests to match the updated logic.
Jira NVGPU-6425
Jira NVGPU-6427
Change-Id: I1beec97c54c60ad334165b1c0acb5e827c24f2ac
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2632287
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add ioctls for creating, removing and querying scheduling domains and
interface with the "nvsched" entity that will be the core scheduler.
Include the scheduler in the Linux build.
The core scheduler code will ultimately hold data on and control what
gets scheduled, but this intermediate layer in nvgpu-rm needs a bit of
bookeeping to manage the userspace interface.
To keep changes isolated, this does not touch the internal runlist
domains yet. The core scheduler logic will eventually control the
runlist domains.
Jira NVGPU-6788
Change-Id: I7b4064edb6205acbac2d8c593dad019d517243ce
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2463625
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
As the docs of nvgpu_thread say, each thread (which the worker loop is)
should wake up and check also nvgpu_thread_should_stop() to manage
graceful and quick exit as requested. The loop does have that check
already, but the workqueue condition does not, so the cond wait might
end up waiting until its timeout hits.
It's not robust to trust the worker users to have a swift timeout for
exiting the thread, so read the should-stop flag in the wakeup condition
too.
Simplify the clk arb worker ops now that calling
nvgpu_worker_should_stop from there is no longer necessary. (Other
worker users did not have those, so they were technically buggy.)
Change-Id: I5409b8037564d4b6445a15cdbd4f1f3d616c4083
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2635808
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
- Separate out local golden context memory allocation from
nvgpu_gr_global_ctx_init_local_golden_image() into a new function
nvgpu_gr_global_ctx_alloc_local_golden_image().
- Add a new member local_golden_image_copy to struct
nvgpu_gr_obj_ctx_golden_image to store copy used for context
verification.
- Allocate local golden context memory from nvgpu_gr_obj_ctx_init()
which is called during poweron path.
- Remove memory allocation from nvgpu_gr_obj_ctx_save_golden_ctx().
- Disable test test_gr_obj_ctx_error_injection since it needs rework
to accomodate the new changes.
- Fix below tests to allocate local golden context memory :
test_gr_global_ctx_local_ctx_error_injection
test_gr_setup_alloc_obj_ctx
Bug 3307637
Change-Id: I2f760d524881fd328346838ea9ce0234358f8e51
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2633713
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
