Moved nvgpu_clk_pmupstate structure from gk20a to
nvgpu_pmu.The aim is to have single pmu structure inside
gk20a, that is "nvgpu_pmu" struct and all the global
structures of all units in PMU should be included in
"nvgpu_pmu" struct.
NVGPU-3220
Change-Id: I531aab568a692c55e640ca2c33aa7508b83a9593
Signed-off-by: rmylavarapu <rmylavarapu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2104129
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
MISRA rule 11.3 states that a cast shall not be performed between
a pointer to object type and a pointer to a different object type.
The following gr context routines convert a gr context buffer pointer
in the form of a u8 * to a u32 * before referencing the context buffer's
contents:
* gm20b_ctxsw_prog_check_main_image_header_magic
* gm20b_ctxsw_prog_check_local_header_magic
* gm20b_ctxsw_prog_get_num_gpcs
* gm20b_ctxsw_prog_get_num_tpcs
* gm20b_ctxsw_prog_get_extended_buffer_size_offset
* gm20b_ctxsw_prog_get_ppc_info
* gm20b_ctxsw_prog_get_local_priv_register_ctl_offset
Because the gr context buffer pointer starts out as a u32 * the
conversion to a u8 * isn't strictly necessary.
So this patch eliminates the conversion and the 11.3 rule violations
accordingly.
JIRA NVGPU-782
Change-Id: I0d24b539ad6ee8e56318287ce8640764285ed54d
Signed-off-by: Scott Long <scottl@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2102986
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
-Currently PG task is created for both iGPU & dGPU as part PMU init
sequence path, but task is not required for dGPU or can be skipped
if ELPG is not supported on iGPU, made changes to create PG task only
if supported else skip it, and made some functions to private as these
are required by PG UNIT only.
-PG instance is allocated & set to default properties as needed if
support is enabled else skip it.
-Made changes in dependent files as required to reflect above changes
JIRA NVGPU-1972
Change-Id: I4efb7f1814a9ad48770acea2173e66f0a4c8a9c1
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2094840
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Some variables/defines of PMU are not used currenlty, so removed.
Moved some struct from pmu header file to respective unit.
removed pmu_fw.c file, which is not required anymore, missed removing
as part of fw unit task.
JIRA NVGPU-1972
Change-Id: I6334f7b2f47b1965780ddeaa41edac69972e6038
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2094839
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
When flashing OBE ROM on PG189 with an internal chip SKU, it
ends up enumerating with unsupported device id. Since VBIOS
update will be done automatically in PDK, we want to detect
such boards, and throw a warning.
Use upper bits of driver_data to define some flags for pci device.
Added PCI_DEVICE_F_INTERNAL_CHIP_SKU for boards with internal
chip SKUs.
Allow enumeration of PG189 boards with internal chip SKUs
0x1eae and 0x1eaf, but throw a warning.
Bug 2569674
Change-Id: I0261a85139be5e58029f25fcb289f54dfd7c35b3
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2103831
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add a threaded test for nvgpu_cmpxchg API. This test implements an
atomic increment using cmpxchg. It uses the existing arithmetic
threaded framework to use the cmpxchg_inc() function to verify 100
parallel threads increment the atomic the correct number of times.
These are L1 tests since they have longer run times and are unlikely to
regress.
JIRA NVGPU-2251
Change-Id: I9c2b68052b3a1b6ef20adfa24e7d50746902f754
Signed-off-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2100748
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add a threaded test to verify xchg operation. Also validate the
non-atomic version will fail.
There are 100 threads spawned that each do exchanges on the same atomic
1000 times. Each thread starts with a unique value. When the test
completes, make sure the values are still unique.
These are L1 tests since they have longer run times and are unlikely to
regress.
JIRA NVGPU-2251
Change-Id: I4741df500d712e60d92e5221182be02dbeaa14ef
Signed-off-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2081067
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
cyclestats_snapshot data and lock is right now stored in struct nvgpu_gr
Use case itself is not specific to GR engine but in general it applies
to other units outside of GR too.
Hence it makes sense to move both data and lock to struct gk20a instead
of keeping them in struct nvgpu_gr
Update all cyclestats_snapshot code to refer data/lock from struct gk20a
Remove gr_priv.h header include from cyclestats_snapshot.c
Some of the functions were mistakenly declared in gr_gk20a.h.
Move them to cyclestats_snapshot.h and rename them to form nvgpu_css_*()
Jira NVGPU-1103
Change-Id: I3fb32fe96f0ca6613f4640c8bd227b9e0e02dca3
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2104848
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Separate out cyclestats handling code into separate unit
common.cyclestats
This unit now exposes new API nvgpu_cyclestats_exec() to perform
cyclestats operation. Call this API from common.gr.intr unit
Extract out all the private data structures from gk20a.h to
cyclestats_priv.h
Rename struct gk20a_cyclestate_buffer_elem to
nvgpu_cyclestate_buffer_elem
Jira NVGPU-1103
Change-Id: Id362675228fe23d03d6d277ff320bcc1066c3c64
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2104202
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@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>
Move chip specific recovery code for volta onwards
architecture to hal/rc/rc_gv11b.c
Rename
fifo.teardown_ch_tsg -> fifo.recover
gk20a_runlist_update_locked -> nvgpu_runlist_update_locked
Remove
Unused h/w headers from fifo_gv11b.c
Use local variable f instead of g->fifo
JIRA NVGPU-1314
Change-Id: Ia535bbe4780e7241fdd911a8f577c6b98cf0fe53
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2102897
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Currently, buddy, page and bitmap allocators have individual init()
functions. This patch creates common nvgpu_alloc_allocator_init()
function to trigger the individual functions based on allocator type
argument. This makes writing requirements for the allocators easier.
Jira NVGPU-991
Change-Id: If94e3496f46f036460ef9f1831852e6fc19d3a0b
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2097962
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
gr_gk20a.ctx_vars struct right now stores sizes for golden_image, zcull,
pm_ctxsw, and gfxp_preemption_buffer.
but these sizes should be really owned by respective units and should
be assigned to units as soon as they are queried from FECS
Add new structure to nvgpu_gr_falcon to hold sizes that will be queried
from FECS
struct nvgpu_gr_falcon_query_sizes {
u32 golden_image_size;
u32 pm_ctxsw_image_size;
u32 preempt_image_size;
u32 zcull_image_size;
};
gr.falcon unit now queries sizes from FECS and fills this structure.
gr.falcon unit also exposes below APIs to query above sizes
u32 nvgpu_gr_falcon_get_golden_image_size(struct nvgpu_gr_falcon *falcon);
u32 nvgpu_gr_falcon_get_pm_ctxsw_image_size(struct nvgpu_gr_falcon *falcon);
u32 nvgpu_gr_falcon_get_preempt_image_size(struct nvgpu_gr_falcon *falcon);
u32 nvgpu_gr_falcon_get_zcull_image_size(struct nvgpu_gr_falcon *falcon);
gr.gr unit now calls into gr.falcon unit to initailize sizes, and then
uses above exposed APIs to set sizes into respective units
vGPU will too fill up struct nvgpu_gr_falcon_query_sizes with all the sizes
and then above APIs will be used to set sizes into respective units
All of above means size variables in gr_gk20a.ctx_vars struct are no more
being referred. Delete them.
Jira NVGPU-3112
Change-Id: I8b8e64ee0840c3bdefabc8ee739e53a30791f2b3
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2103478
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
struct gr_gk20a defines boolean flag golden_image_initialized to
indicate if golden_image is initialized or not
common.gr.obj_ctx also added a flag of its own to check if golden_image
is ready
Add new API nvgpu_gr_obj_ctx_is_golden_image_ready() in
common.gr.obj_ctx unit to get status of golden_image
Use this new API everywhere to check if golden image is ready
Remove g->gr.ctx_vars.golden_image_initialized
Also remove ctx_mutex from struct gr_gk20a
Add new flag golden_image_initialized to struct nvgpu_pmu_pg and set it
when golden image is initialized. This is needed to avoid circular
dependency between GR and PMU
Jira NVGPU-3112
Change-Id: Id391294cede6424e15a9a9de29c40d013b509534
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2099400
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Only one h/w header is allowed per hal file. ramfc_*.c uses both
hw_ramfc_*.h and hw_pbdma_*.h. The pbdma dependencies are removed from
the HAL unit of ramfc by constructing new HAL functions for pbdma unit.
The HAL ops functions added are listed below.
get_gp_base
get_gp_base_hi
get_fc_formats
get_fc_pb_header
get_fc_subdevice
get_fc_target
get_ctrl_hce_priv_mode_yes
get_userd_aperture_mask
get_userd_addr
get_userd_hi_addr
get_fc_runlist_timeslice
get_config_auth_level_privileged
set_channel_info_veid
config_userd_writeback_enable
allowed_syncpoints_0_index_f
allowed_syncpoints_0_valid_f
allowed_syncpoints_0_index_v
These HAL ops uses the following new implementations.
gm20b_pbdma_get_gp_base
gm20b_pbdma_get_gp_base_hi
gm20b_pbdma_get_fc_formats
gm20b_pbdma_get_fc_pb_header
gm20b_pbdma_get_fc_subdevice
gm20b_pbdma_get_fc_target
gm20b_pbdma_get_ctrl_hce_priv_mode_yes
gm20b_pbdma_get_userd_aperture_mask
gm20b_pbdma_get_userd_addr
gm20b_pbdma_get_userd_hi_addr
gp10b_pbdma_get_fc_runlist_timeslice
gp10b_pbdma_get_config_auth_level_privileged
gp10b_pbdma_allowed_syncpoints_0_index_f
gp10b_pbdma_allowed_syncpoints_0_valid_f
gp10b_pbdma_allowed_syncpoints_0_index_v
gv11b_pbdma_get_fc_pb_header
gv11b_pbdma_get_fc_target
gv11b_pbdma_set_channel_info_veid
gv11b_pbdma_config_userd_writeback_enable
Jira NVGPU-3195
Change-Id: I849f16650046eca38c67b0d6e0e43cd2ab1ac224
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2102576
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Created PMU fw unit to hold PMU RTOS f/w specific ops, images,
flags & command arguments needed for PMU RTOS ucode support.
Moved PMU fw ops from gk20a.gpu_ops to pmu.fw.ops as these ops
are needed to support different version of PMU fw version for
different chips
JIRA NVGPU-1955
Change-Id: I51385d8c20524431f07cba3378676464663deb20
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2090769
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Currently, both clk_arb and channels use their own implementation of
a background worker. These implementations are almost identical and can
be extracted out into a single self-contained unit name nvgpu_worker.
Another advantage of using a single worker unit is to avoid duplication
of Unit Tests for this unit in other units.
channel and clk_arb units provide their own specific implementations
via an ops interface named nvgpu_worker_ops which is a part of the
nvgpu_worker struct.
The following high level APIs are exposed by the nvgpu_worker
nvgpu_worker_should_stop
nvgpu_worker_enqueue
nvgpu_worker_init
nvgpu_worker_deinit
The nvgpu_worker_ops containg the following function pointers
pre_process
wakeup_early_exit
wakeup_post_process
wakeup_process_item
wakeup_condition
wakeup_timeout
The specific code in channel and clk_arb is changed to use the above
implementations instead of their own separate implementations.
Jira NVGPU-3101
Change-Id: I14a0bba6a3d61a642b858dec70d5818d5a0472a4
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2090475
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move gr_gk20a_handle_fecs_error from gr_gk20a.c to
nvgpu_gr_intr_handle_fecs_error in common.gr.intr unit
Move gr_gp10b_handle_fecs_error and gr_gv11b_handle_fecs_error
to hal.gr.intr unit
JIRA NVGPU-3016
Change-Id: I5b7c48ebfd7b13f497980c4d0b64d718649154bd
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2103741
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move gr_gp10b_handle_sm_exception from gr_gp10b to
gp10b_gr_intr_handle_sm_exception in hal.gr.intr unit
Move gr_gk20a_handle_sm_exception from gr_gk20a to
nvgpu_gr_intr_handle_sm_exception in common.gr.intr
Move nvgpu_report_gr_sm_exception to common.gr.intr
JIRA NVGPU-3016
Change-Id: I545ddca052122f87685f35f515831841a246dab3
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2103736
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Make a hal/mm/gmmu sub-unit for the GMMU HAL code. Also move the
gk20a specific HAL code there. gp10b will happen in the next patch.
This change also updates all the GMMU related HAL usage, of which
there is quite a bit. Generally the only change is a .gmmu needs to
be inserted into the HAL path. Each HAL init was also updated.
JIRA NVGPU-2042
Change-Id: I6c46bdfddb8e021f56103d9457fb3e2a226f8947
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2099693
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
vgpu_finalize_poweron_common calls g->ops.fifo.channel_resume.
This call is not valid anymore as no such function pointer exists.
Correct this by replace g->ops.fifo.channel_resume() by
g->ops.channel.resume_all_serviceable_ch()
Change-Id: I09fdfce64d48ec6971396fb2ecabc43051e9bf10
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2102359
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
g->ops.gr.get_preemption_mode_flags() hal is used to fetch information
on supported preemption modes and default preemption mode
Temporary struct nvgpu_preemption_modes_rec is used for this purpose
and is defined in gk20a/gr_gk20a.h right now.
Split above hal into two separate hals and move them to hal.gr.init unit
g->ops.gr.init.get_supported__preemption_modes()
g->ops.gr.init.get_default_preemption_modes()
These hals now return respective flags in pointers passed in function
parameter list, so there is no need to use temporary structure anymore
Hence delete struct nvgpu_preemption_modes_rec
Implement gm20b/gp10b chip specific hals in hal.gr.init unit.
Delete g->ops.gr.get_preemption_mode_flags() hal
Jira NVGPU-3126
Change-Id: I84f507fcd8d122bb7f0ecf697e8b4f16c9339ce1
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2102455
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Description:
Present p-state unit handle both pstate boardobj and
initializing all the units. As part of restructuring,
the pstate unit is separated into two units:
1) Perf_pstate: This unit will handle pstate boardobjs.
2) Pmu_pstate: This unit will initialize all the units
which supoort performance states.
Changes:
1) Created pmu_pstate unit.
2) Pstate boardobjs are moved under perf_pstate which
is under perf unit.
NVGPU-1958
Change-Id: I2c428adfe6de4992c9eeda0d4356d30290f6e8a4
Signed-off-by: rmylavarapu <rmylavarapu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2096339
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Basic units like fifo, rc are having dependency on
gr_falcon. Avoided outside gr units dependency on gr_falcon
by moving following functions to gr:
int nvgpu_gr_falcon_disable_ctxsw(struct gk20a *g,
struct nvgpu_gr_falcon *falcon); ->
int nvgpu_gr_disable_ctxsw(struct gk20a *g);
int nvgpu_gr_falcon_enable_ctxsw(struct gk20a *g,
struct nvgpu_gr_falcon *falcon); ->
int nvgpu_gr_enable_ctxsw(struct gk20a *g);
int nvgpu_gr_falcon_halt_pipe(struct gk20a *g); ->
int nvgpu_gr_halt_pipe(struct gk20a *g);
HALs also moved accordingly and updated code to reflect this.
Also moved following data back to gr from gr_falcon:
struct nvgpu_mutex ctxsw_disable_mutex;
int ctxsw_disable_count;
JIRA NVGPU-3168
Change-Id: I2bdd4a646b6f87df4c835638fc83c061acf4051e
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2100009
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Instead of using GPC and TPC counts to allocate memory
to hold sm info, use nvgpu_gr_config_get_no_of_sm() to get
the actual number. This fixes memory issues (corruption and
segmentation fault) seem when nvgpu_gpu_ioctl_wait_for_pause
is used.
Bug 2559631
Change-Id: Idcf9983fbbec7ec7f53835c59164e04bc45cd041
Signed-off-by: Nitin Kumbhar <nkumbhar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2102557
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
