Commit Graph

22 Commits

Author SHA1 Message Date
mkumbar
8cce8dea70 gpu: nvgpu: PMU NVRISCV BR failure HSI
- Add PMU NVRISCV BR failure HSI support.
- Created a falcon unit function to check for the
  BR competition status check and called from
  other units as needed.

Bug 3491596
Bug 3366818

Change-Id: I5c3c6a7e6aeaad68f77e6b24f21239e40d9a7f78
Signed-off-by: mkumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2686370
Reviewed-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit
2022-03-25 13:25:27 -07:00
Richard Zhao
851666b632 gpu: nvgpu: common/pmu: fix compile error of new compile flags
It's preparing to add bellow CFLAGS:
    -Werror -Wall -Wextra \
    -Wmissing-braces -Wpointer-arith -Wundef \
    -Wconversion -Wsign-conversion \
    -Wformat-security \
    -Wmissing-declarations -Wredundant-decls -Wimplicit-fallthrough

Jira GVSCI-11640

Signed-off-by: Richard Zhao <rizhao@nvidia.com>
Change-Id: Ide3ab484924bd5be976a9f335b55b136575ce428
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2555055
Reviewed-by: Shashank Singh <shashsingh@nvidia.com>
Reviewed-by: Aparna Das <aparnad@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
2022-01-13 12:36:02 -08:00
mkumbar
b92e8530fc gpu: nvgpu: ga10b: slcg and blcg update for PMU
Load register configuration for SLCG and BLCG for PMU.

Bug 3452217

Change-Id: Ib54077ee00d0b9247db8d792e5ed566fd4ca2efd
Signed-off-by: mkumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2641365
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: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-12-14 06:28:19 -08:00
Divya
c347b6e4ff gpu: nvgpu: print riscv pmu pc trace
- To print pmu RISCV PC trace, create a new flag
  which will be set to true after PMU is initialised.
- This flag is then used to used to print RISCV trace
  buffer when pmu halt occurrs.

JIRA NVGPU-7261

Change-Id: Ib3ad2f40efd1458d22b21e99ab151c11cfeb43be
Signed-off-by: Divya <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2624073
Reviewed-by: svcacv <svcacv@nvidia.com>
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>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-11-12 02:55:56 -08:00
Mahantesh Kumbar
7b29872bc4 gpu: nvgpu: swap the sequence of ACR & PERFMON
Swap the command sequence of ACR WPR init and PERFMON init sent
to PMU ucode upon init message, because perfmon init command read
is failing in PMU ucode when ACR WPR init command is processed
and accessed WPR info from system during un-rail-gate sequence.

And also flushing the FB-Q's for rail-gate and un-rail-gate sequence.

Bug 3400166

Change-Id: I23c38588d0ddc4e1621e83a72d5e232cf65371dc
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Signed-off-by: mkumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2617398
Tested-by: mobile promotions <svcmobile_promotions@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>
GVS: Gerrit_Virtual_Submit
2021-11-08 15:08:05 -08:00
Konsta Hölttä
f4ec400d5f gpu: nvgpu: simplify nvgpu_timeout_init
nvgpu_timeout_init() returns an error code only when the flags parameter
is invalid. There are very few possible values for flags, so extract the
two most common cases - cpu clock based and a retry based timeout - to
functions that cannot fail and thus return nothing. Adjust all callers
to use those, simplfying error handling quite a bit.

Change-Id: I985fe7fa988ebbae25601d15cf57fd48eda0c677
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2613833
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-10-26 13:47:32 -07:00
Divya Singhatwaria
b6ab227016 gpu: nvgpu: Enable pmu interrupt
- For secure RISCV boot, enable pmu interrupt
  during pmu_rtos_init
- As interrupts are enabled, PMU intr can be received
  before driver has changed the pmu firmware state. This
  can cause the RISCV boot to fail.
- To resolve this, first change the pmu firmware state
  from off to PMU_FW_STATE_STARTING and then wait
  for pmu priv lockdown release.

Change-Id: Ib2e8b033fec6320bf9ccff02696192a48172464b
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2586325
Reviewed-by: svc_kernel_abi <svc_kernel_abi@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: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
2021-09-07 16:00:05 -07:00
Vedashree Vidwans
e13ab1f9ea gpu: nvgpu: pmu: remove hw access from remove_pmu_support
GPU HW registers are locked before remove_pmu_support.
Remove functions accessing HW registers.

Bug 3357477

Change-Id: I34a1923bfdb3afacd462f2646e2821569573a81a
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2577627
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-08-17 09:45:42 -07:00
Sagar Kamble
f4571194b0 gpu: nvgpu: stop ELPG init thread during unload
ELPG initialization thread creation can fail when the process is killed.
That leads to driver resume failure.

That thread was stopped on suspend and re-created on resume. To avoid
the issue above, don't stop the ELPG thread in suspend and let the
first created thread handle the ELPG state transitions always.
And stop the ELPG thread during unload.

Also fix couple of instances of config flag as:

s/CONFIG_PMU_POWER_PG/CONFIG_NVGPU_POWER_PG

bug 3345977

Change-Id: I8952edf8d1664ed258f238e265002e716d1bf5c2
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2573763
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: Ashish Mhetre <amhetre@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
2021-08-11 01:55:46 -07:00
mkumbar
87984ea344 gpu: nvgpu: support nvriscv debug feature
Enable nvriscv debug buffer feature in NVGPU.
Debug buffer is a feature to print the debug log from ucode onto console
in real time.
Debug buffer feature uses the DMEM, queue and SWGEN1 interrupt to share
ucode debug data with NVGPU.
Ucode writes debug message to DMEM and updates offset in queue to trigger
interrupt to NVGPU.
NVGPU copies the debug message from DMEM to local buffer to process and
print onto console.

Debug buffer feature is added under falcon unit and required engine
can utilize the feature by providing required param through public
functions.

Currently GA10B NVRISCV NS/LS PMU ucode has support for this feature
and enabled support on NVGPU side by adding required changes, with this
feature enabled, it is now possible to see prints in real time.

JIRA NVGPU-6959

Change-Id: I9d46020470285b490b6bc876204f62698055b1ec
Signed-off-by: mkumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2548951
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: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-07-17 12:45:00 -07:00
mkumbar
fcf31d7063 gpu: nvgpu: ga10b: fix GSP/PMU priv error
- Fix GSP/PMU registers priv errors which are seen as part of boot sequence.
- Couple of GSP/PMU Falcon/NVRISCV registers are allowed to access
  upon NVRISCV bootrom completion but these registers were needed
  to configure on legacy chips to bootstrap/configure Falcon.
- Add is_falcon2_enabled or NVGPU_PMU_NEXT_CORE_ENABLED check
  to skip these registers.

JIRA NVGPU-7025

Change-Id: I087a477ade6736398dea113f89894a0ff73ae647
Signed-off-by: mkumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2553127
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-07-16 16:44:08 -07:00
Antony Clince Alex
c7d43f5292 gpu: nvgpu: remove usage of CONFIG_NVGPU_NEXT
The CONFIG_NVGPU_NEXT config is no longer required now that ga10b and
ga100 sources have been collapsed. However, the ga100, ga10b sources
are not safety certified, so mark them as NON_FUSA by replacing
CONFIG_NVGPU_NEXT with CONFIG_NVGPU_NON_FUSA.

Move CONFIG_NVGPU_MIG to Makefile.linux.config and enable MIG support
by default on standard build.

Jira NVGPU-4771

Change-Id: Idc5861fe71d9d510766cf242c6858e2faf97d7d0
Signed-off-by: Antony Clince Alex <aalex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2547092
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-06-27 05:02:47 -07:00
Ramesh Mylavarapu
a0b1b3f2be gpu: nvgpu: add priv lockdown release check for NVRISCV pmu
IRQ register access will cause priv errors if they
are accessed before priv lockdown is released.
This change adds a polling loop to check priv lockdown
before proceeding further while booting NVRISCV pmu.

Bug 200709761

Signed-off-by: Ramesh Mylavarapu <rmylavarapu@nvidia.com>
Change-Id: I44b8ce4c59b5a9f20901e5ce08610d17725da779
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2512351
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
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: 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
2021-05-04 18:04:25 -07:00
Sagar Kamble
ecfd675d9b gpu: nvgpu: free pmu variables allocated in early_init on error in rtos_init
On error in pmu_rtos_init, pmu state was freed partly. That lead to
invalid access on subsequent nvgpu poweron. Free all pmu state in
such case.

Bug 200575409

Change-Id: I11166b55dbe00a225e811425d21500c3143a354c
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2503577
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
2021-03-24 14:47:44 -07:00
mkumbar
65111b64f2 gpu: nvgpu: Add RISCV LS PMU support
-Add RISCV LS PMU support by adding RISCV LS PMU ucode to the blob.
-Modify the PMU RTOS sequence based on NEXT CORE enable flag.

JIRA NVGPU-6303

Change-Id: I4e2b989f9903b72a6327c931eb3c02f8cef2aa75
Signed-off-by: mkumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2447388
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2020-12-17 18:36:44 -06:00
Deepak Nibade
3b746dce0c gpu: nvgpu: use a falcon flag instead of enabled bit
common.gr unit right now makes use of a capability bit
NVGPU_PMU_FECS_BOOTSTRAP_DONE to ensure the recovery path hits a
different routine. This is actually needless and a common check
cannot be used for all GR instances anyways.

Delete this capability bit. Add and use a new flag
coldboot_bootstrap_done added under struct nvgpu_gr_falcon

Jira NVGPU-5648

Change-Id: I46faea6f07cf054f17a3215d4cbbe0fc8a6382ae
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2409533
Reviewed-by: automaticguardword <automaticguardword@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: Antony Clince Alex <aalex@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2020-12-15 14:13:28 -06:00
rmylavarapu
700bd83b41 gpu: nvgpu: Rename/clean boardobj unit
-Removed unwanded boardobj includes
-Renamed functions as struct as per usage

NVGPU-4484

Signed-off-by: rmylavarapu <rmylavarapu@nvidia.com>
Change-Id: I792a4b64075d5e87f911c1073717dbe7107227a1
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2335991
Reviewed-by: automaticguardword <automaticguardword@nvidia.com>
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
2020-12-15 14:13:28 -06:00
rmylavarapu
0115c26f1b gpu: nvgpu: Boardobj lite unit refactor
As boardobj unit is used only in PMU, the plan is to move
all the boardobj related functions/structures and Macros
to boardobj specific folders. This will remove unnecessary
usage of boardobj outside PMU.

NVGPU-4484

Change-Id: I9f0fda32e6affd1fce218eb0ac638a9dfc8b99c3
Signed-off-by: rmylavarapu <rmylavarapu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2335986
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: automaticguardword <automaticguardword@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: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
2020-12-15 14:13:28 -06:00
Thomas Fleury
28ccd63f69 gpu: nvgpu: enable CONFIG_NVGPU_LS_PMU for safety
Enable CONFIG_NVGPU_LS_PMU for dGPU safety build.
Add missing #ifdefs for CONFIG_NVGPU_POWER_PG and
CONFIG_NVGPU_CLK_ARB which are not defined for safety build.

Moved gm20b_mc_is_enabled to fusa code.

NVGPU_UNIT_PWR is only defined when CONFIG_NVGPU_HAL_NON_FUSA
is defined. Added #ifdefs to compile out gk20a_pmu functions
that are using it.

Jira NVGPU-4661

Change-Id: Ieb552f9374bad6f3dad777322f118931e0bc94ec
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2317085
Reviewed-by: automaticguardword <automaticguardword@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
2020-12-15 14:13:28 -06:00
Sagar Kamble
fba516ffae gpu: nvgpu: enable PMU ECC interrupt early
PMU IRQs were not enabled assuming entire functionality for LS PMU.
Debugging early init issues of PMU falcon ECC errors triggered
during nvgpu power-on will be cumbersome if interrupts are not
enabled early. FMEA analysis of the nvgpu init path also
requires this interrupt be enabled earlier.

Hence, Enable the PMU ECC IRQ early during nvgpu_finalize_poweron.
pmu_enable_irq is updated to enable interrupts differently for
safety and non-safety. PMU interrupts disabling is moved out
of nvgpu_pmu_destroy to nvgpu_prepare_poweroff. Prepared new
wrapper API nvgpu_pmu_enable_irq.

PMU ECC init and isr mutex init is moved to the beginning of
nvgpu_pmu_early_init as for safety, ls pmu code path is
disabled. Fixed the pmu_early_init dependent and mc
interrupt related unit tests.

Update the doxygen for changed functions.

JIRA NVGPU-4439

Change-Id: I1a1e792d2ad2cc7a926c8c1456d4d0d6d1f14d1a
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2251732
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2020-12-15 14:10:29 -06:00
Philip Elcan
b53ec4731e gpu: nvgpu: pmu: update init APIs
Remove the second parameter for the pmu_early_init() and pmu_init()
functions so they only require the gk20a object. The g->pmu was always
passed for this parameter. And this makes the API signature match the
other init functions in the driver.

JIRA NVGPU-3980

Change-Id: Iae9361a5f14bc5c1d02f4ddb6583f30b71b22d59
Signed-off-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2202968
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vinod Gopalakrishnakurup <vinodg@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2020-12-15 14:05:52 -06:00
Mahantesh Kumbar
5eeb751d58 gpu: nvgpu: Move PMU RTOS functions out from pmu.c
Moved PMU RTOS functions to new file from pmu.c to make clear
separation of PMU unit init & PMU RTOS init.

JIRA NVGPU-2457

Change-Id: I694bf561517b4b55f9396be8e132dc0da5cb29e6
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2199543
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2020-12-15 14:05:52 -06:00