The simulator ring buffer DMA interface supports buffers of the following sizes:
4, 8, 12 and 16K. At present, it is configured to 4K and it happens to match
with the kernel PAGE_SIZE, which is used to wrap back the GET/PUT pointers once
4K is reached. However, this is not always true; for instance, take 64K pages.
Hence, replace PAGE_SIZE with SIM_BFR_SIZE.
Introduce macro NVGPU_CPU_PAGE_SIZE which aliases to PAGE_SIZE and replace
latter with former.
Bug 200658101
Jira NVGPU-6018
Change-Id: I83cc62b87291734015c51f3e5a98173549e065de
Signed-off-by: Antony Clince Alex <aalex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2420728
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
struct nvgpu_gr is right now initialized during probe and from OS
specific code. To support multiple instances of graphics engine,
nvgpu needs to initialize nvgpu_gr after number of engine instances
have been enumerated in poweron path.
Hence move nvgpu_gr_alloc() to poweron path and after gr manager has
been initialized.
Some of the members of nvgpu_gr are initialized in probe path and they
too are in OS specific code. Move them to common code in
nvgpu_gr_alloc()
Add field fecs_feature_override_ecc_val to struct gk20a to store the
override flag read from device tree. This flag is later copied to
nvgpu_gr in poweron path.
Update tpc_pg_mask_store() to check for g->gr being NULL before
accessing golden image pointer.
Update tpc_fs_mask_store() to return error if g->gr is not initialized.
This path needs nvgpu_gr struct initialized. Also fix the incorrect
NULL pointer check in tpc_fs_mask_store() which breaks the write path
to this sysfs.
Jira NVGPU-5648
Change-Id: Ifa2f66f3663dc2f7c8891cb03b25e997e148ab06
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2397259
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: Lakshmanan M <lm@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
In the upstream kernel ACCESS_ONCE is now deprecated with reason as
given in the following related commit:
commit 381f20fceba8e ("security: use READ_ONCE instead of deprecated
ACCESS_ONCE")
ACCESS_ONCE() does not work reliably on non-scalar types. For
example gcc 4.6 and 4.7 might remove the volatile tag for such
accesses during the SRA (scalar replacement of aggregates) step.
Replace usages of ACCESS_ONCE with READ_ONCE and WRITE_ONCE in nvgpu.
Bug 2834141
Change-Id: I9904c49e1a4d7b17ed2fe54360051d08595a2982
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2294096
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: Jonathan Hunter <jonathanh@nvidia.com>
Reviewed-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Writing same value to railgate_enable_store should be treated as nop
and made successfully. Doing so is not only an optimization for the
operation but also convention that users expect for "settings". This
change is primary for fixing a peculiar situation in the driver:
root@localhost:/sys/devices/17000000.gp10b# cat railgate_enable
0
root@localhost:/sys/devices/17000000.gp10b# echo 0 > railgate_enable
bash: echo: write error: Invalid argument
Attempt to disable railgating on a platform where railgating isn't
supported shouldn't be treated as 'invalid'. It's disabled after all.
Bug 200562094
Change-Id: I3c04934bdbaf337c33d7de9cac6d53c96b4dacae
Signed-off-by: Leon Yu <leoyu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2225476
(cherry picked from commit 10b3b5b1d5)
Reviewed-on: https://git-master.nvidia.com/r/2226185
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
IRQs can get triggered during nvgpu power-on due to MMU fault, invalid
PRIV ring or bus access etc. Handlers for those IRQs can't access the
full state related to the IRQ unless nvgpu is fully powered on.
In order to let the IRQ handlers know about the nvgpu power-on state
gk20a.power_on_state variable has to be protected through spinlock
to avoid the deadlock due to usage of earlier power_lock mutex.
Further the IRQs need to be disabled on local CPU while updating the
power state variable hence use spin_lock_irqsave and spin_unlock_-
irqrestore APIs for protecting the access.
JIRA NVGPU-1592
Change-Id: If5d1b5e2617ad90a68faa56ff47f62bb3f0b232b
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2203860
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
golden image size will be set when memory allocated.
See function:
- nvgpu_gr_obj_ctx_init
If golden image size is 0, gr_golden_image should be a NULL
pointer in most cases. So add NULL pointer checking in
tpc_pg_mask_store to avoid NULL pointer exception.
Bug 2403210
Change-Id: I14df5cd94d7a4418c3089c5f84b6eab93c485ba6
Signed-off-by: Sunny Li <sunnyl@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2161280
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Debarshi Dutta <ddutta@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
- In GV11B, read fuse_status_opt_tpc_gpc register
to read which TPCs are floorswept.
- The driver will also read sysfs node: tpc_pg_mask
- Based on these two values "can_tpc_powergate" will
be set to true or false and mask will be used to write to
fuse_ctrl_opt_tpc_gpc register to powergate the TPC.
- can_tpc_powergate = true indicates that the mask value
sent from userspace is valid and can be used to power gate
the desired TPC
- can_tpc_powergate = false indicates that the mask value
sent from userspace is not valid and cannot be used to
power gate the desired TPC.
Bug 200532639
Change-Id: Ib0806e4c96305a13b3574e8063ad8e16770aa7cd
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2170736
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Fix 8.2 violation for not specifying parameter name in prototype of
secure_alloc().
Fix 21.3 & 21.8 violations for using reserved names "free" and "exit."
Fix 8.6 and 21.2 violations for __gk20a_do_idle() and
__gk20a_do_unidle() by renaming the functions and wrapping them in a
missing #ifdef CONFIG_PM.
Fix 5.7 violation for reusing "class" as parameter name when already
defined as a struct.
JIRA NVGPU-3343
Change-Id: I976e95a32868fa0a657f4baf0845a32bd7aceb9e
Signed-off-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2117913
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Some functions are not accessing hardware directly
but are being called using HAL ops: For example
g->ops.pmu.pmu_elpg_statistics,
g->ops.pmu.pmu_pg_init_param,
g->ops.pmu.pmu_pg_supported_engines_list,
g->ops.pmu.pmu_pg_engines_feature_list,
g->ops.pmu.pmu_is_lpwr_feature_supported,
g->ops.pmu.pmu_lpwr_enable_pg,
g->ops.pmu.pmu_lpwr_disable_pg,
g->ops.pmu.pmu_pg_param_post_init,
g->ops.pmu.save_zbc
Change the function access by using sw ops, like:
Create new functions:
int nvgpu_pmu_elpg_statistics(struct gk20a *g, u32 pg_engine_id,
struct pmu_pg_stats_data *pg_stat_data);
void nvgpu_pmu_save_zbc(struct gk20a *g, u32 entries);
bool nvgpu_pmu_is_lpwr_feature_supported(struct gk20a *g,
u32 feature_id);
JIRA NVGPU-3209
Change-Id: I6db9b43c7c4a5054720a72487302b740b091044d
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2110963
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
GVS: Gerrit_Virtual_Submit
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>
Move remaining chip specific gr hal files to hal.gr.gr unit.
Remove unused headers include from hal files in hal.gr.gr unit
Update gr hal headers include location in the files currently
using these headers.
Jira NVGPU-3219
Change-Id: Ic632020a90ac4b8ac1e0359e979864b42f0ef2c0
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2105489
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Moved the following HALs from fifo to tsg
- set_timeslice
- default_timeslice_us
Renamed
- gk20a_tsg_set_timeslice -> nvgpu_tsg_set_timeslice
- min_timeslice_us -> tsg_timeslice_min_us
- max_timeslice_us -> tsg_timeslice_max_us
Scale timeslice to take into account PTIMER clock in
nvgpu_runlist_append_tsg.
Removed gk20a_channel_get_timescale_from_timeslice, and
instead moved timeout and scale computation into runlist HAL,
when building TSG entry:
- runlist.get_tsg_entry
Use ram_rl_entry_* accessors instead of hard coded values
for default and max timeslices.
Added #defines for min, max and default timeslices.
Jira NVGPU-3156
Change-Id: I447266c087c47c89cb6a4a7e4f30acf834b758f0
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2100052
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>
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>
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>
gfxp_wfi_timeout_count/unit fields were stored in gr_gk20a.ctx_vars
struct so that any user could configure them through sysfs nodes
But the sysfs nodes are legacy and not being actively used by anyone.
Hence delete the sysfs nodes to configure these fields.
Since the gfxp timeout unit/count can now be statically programmed,
make following changes
- remove g->ops.gr.init_gfxp_wfi_timeout_count() hal
- remove g->ops.gr.get_max_gfxp_wfi_timeout_count() hal
- update g->ops.gr.init.preemption_state() hals to configure the values
using macros instead of caller passing the values
- update g->ops.gr.init.gfxp_wfi_timeout() hals to configure the values
using macros instead of caller passing the values
Finally, we don't need to store gfxp_wfi_timeout_count/unit fields
anymore, hence delete them from gr_gk20a.ctx_vars
Jira NVGPU-3112
Change-Id: Idbe5ab3053228dd177aca253545aac36d38ca8ad
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2100219
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>
Removed unused struct from gr_gk20a.h
Change static allocation for struct gr_gk20a to dynamic type.
Change all the files that being affected by that change.
Call gr allocation from corresponding init_support functions, which
are part of the probe functions.
nvgpu_pci_init_support in pci.c
vgpu_init_support in vgpu_linux.c
gk20a_init_support in module.c
Call gr free before the gk20a free call in nvgpu_free_gk20a.
Rename struct gr_gk20a to struct nvgpu_gr
JIRA NVGPU-3132
Change-Id: Ief5e664521f141c7378c4044ed0df5f03ba06fca
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2095798
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Use API nvgpu_gr_obj_ctx_get/set_golden_image_size() exposed by
gr.obj_ctx unit to get/set size of golden image
Call nvgpu_gr_obj_ctx_init() from vgpu_gr_init_gr_setup_sw() to
initialize golden image size in gr.obj_ctx unit even on vGPU
Move g->ops.gr.falcon.init_ctx_state() call early in
vgpu_gr_init_gr_setup_sw() so that gr.ctx_vars struct is prepared
before fields in it accessed during rest of GR initialization
Jira NVGPU-3112
Change-Id: Ie827ad6f30cc3d931519a1f9a709861d26f8da26
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2096162
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
- Move the perfmon unit source code to common/pmu/perfmon/ folder
- Separate perfmon unit headers under include/nvgpu/pmu/pmu_perfmon.h
- Make a new structure: nvgpu_pmu_perfmon for perfmon unit
- This new struct combines all perfmon unit variables like
perfmon_query, perfmon_ready etc. into one
structure as a part of perfmon unit refactoring.
- Use pmu_perfmon struct to access all perfmon variables.
- Eg: pmu->pmu_perfmon->perfmon_query, pmu->pmu_perfmon->perfmon_ready
and so on.
JIRA NVGPU-1961
Change-Id: I57516c646bfb256004dd7b719e40fafd3c2a09b2
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2080555
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add a new unit common.gr.obj_ctx which allocates and initializes GR
context. This unit also takes care of creating global golden image
used to initialize every context.
Add private header obj_ctx_priv.h that defines struct
nvgpu_gr_obj_ctx_golden_image
Add public header obj_ctx.h that exposes functions supported by new unit
This unit now exposes below API to allocate and initialize context
nvgpu_gr_obj_ctx_alloc()
Remove below functions from gk20a/gr_gk20a.c and move them to new unit
with below renames
gr_gk20a_fecs_ctx_bind_channel() -> nvgpu_gr_obj_ctx_bind_channel()
gr_gk20a_fecs_ctx_image_save() -> nvgpu_gr_obj_ctx_image_save()
gk20a_init_sw_bundle() -> nvgpu_gr_obj_ctx_alloc_sw_bundle()
gr_gk20a_alloc_gr_ctx() -> nvgpu_gr_obj_ctx_gr_ctx_alloc()
gr_gk20a_init_golden_ctx_image() ->
nvgpu_gr_obj_ctx_alloc_golden_ctx_image()
Use new APIs in gk20a_alloc_obj_ctx() to allocate context
For now this unit includes <nvgpu/gr/gr.h> and some h/w headers.
But they will be removed in follow up patches
Jira NVGPU-1887
Change-Id: Ib95ec1c19c5b74810f85c2feed8fdd63889d3d22
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2087662
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: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Vinod Gopalakrishnakurup <vinodg@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
- Make a new structure: nvgpu_pmu_pg for PG unit
- This new struct combines all PG unit variables like
elpg_stat, elpg_refcnt, pg_init etc. into one
structure as a part of PG unit refactoring.
- Use pmu_pg struct to access all PG variables.
- Eg: &pmu->pmu_pg.elpg_mutex, &pmu->pmu_pg.pg_mutex
and so on.
NVGPU-1973
Change-Id: I9973072826f4089f6315827bce49fa30dbcbcdda
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2071306
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Moved cbc related code and data from gr to cbc unit.
Ltc and cbc related data is moved from gr header:
1. Ltc related data moved from gr_gk20a -> gk20a and it
will be moved eventually to ltc unit:
u32 slices_per_ltc;
u32 cacheline_size;
2. cbc data moved from gr_gk20a -> nvgpu_cbc
u32 compbit_backing_size;
u32 comptags_per_cacheline;
u32 gobs_per_comptagline_per_slice;
u32 max_comptag_lines;
struct gk20a_comptag_allocator comp_tags;
struct compbit_store_desc compbit_store;
3. Following config data moved gr_gk20a -> gk20a
u32 comptag_mem_deduct;
u32 max_comptag_mem;
These are part of initial config which should be available
during nvgpu_probe. So it can't be moved to nvgpu_cbc.
Modified code to use above updated data structures.
Removed cbc init sequence from gr and added in
common cbc unit. This sequence is getting called
from common nvgpu init code.
JIRA NVGPU-2896
JIRA NVGPU-2897
Change-Id: I1a1b1e73b75396d61de684f413ebc551a1202a57
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2033286
Reviewed-by: svc-misra-checker <svc-misra-checker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move GR HAL operation g->ops.gr.init_preemption_state() to hal.gr.init
unit as g->ops.gr.init.preemption_state()
Create hal.gr.init unit files for gp10b and gv11b and copy over
corresponding functions to new files
This API now takes gfxp_wfi_timeout_unit and gfxp_wfi_timeout_count as
parameter
Define gfxp_wfi_timeout_unit in struct gr_gk20a as a boolean flag named
gfxp_wfi_timeout_unit_usec
Remove GFXP_WFI_TIMEOUT_UNIT_SYSCLK/USEC macros
Jira NVGPU-2961
Change-Id: I4347b1e30c86c231e44cf274adccd8c70addcdab
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2072549
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add new power/clock gating functions that can be called by
other units.
New clock_gating functions will reside in cg.c under
common/power_features/cg unit.
New power gating functions will reside in pg.c under
common/power_features/pg unit.
Use nvgpu_pg_elpg_disable and nvgpu_pg_elpg_enable to disable/enable
elpg and also in gr_gk20a_elpg_protected macro to access gr registers.
Add cg_pg_lock to make elpg_enabled, elcg_enabled, blcg_enabled
and slcg_enabled thread safe.
JIRA NVGPU-2014
Change-Id: I00d124c2ee16242c9a3ef82e7620fbb7f1297aff
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2025493
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add new unit gr/config to initialize GR configuration like GPC/TPC
count, MAX count and mask
Create new structure nvgpu_gr_config that stores all the configuration
and that is owned by the new unit
Move below fields from struct gr_gk20a to nvgpu_gr_config in gr/config.h
Struct gr_gk20a now only holds the pointer to struct nvgpu_gr_config
u32 max_gpc_count;
u32 max_tpc_per_gpc_count;
u32 max_zcull_per_gpc_count;
u32 max_tpc_count;
u32 gpc_count;
u32 tpc_count;
u32 ppc_count;
u32 zcb_count;
u32 pe_count_per_gpc;
u32 *gpc_tpc_count;
u32 *gpc_ppc_count;
u32 *gpc_zcb_count;
u32 *pes_tpc_count[GK20A_GR_MAX_PES_PER_GPC];
u32 *gpc_tpc_mask;
u32 *pes_tpc_mask[GK20A_GR_MAX_PES_PER_GPC];
u32 *gpc_skip_mask;
u8 *map_tiles;
u32 map_tile_count;
u32 map_row_offset;
Remove gr->sys_count since it was already no longer used
common/gr/config/gr_config.c unit now exposes the APIs to initialize
the configuration and also to query the configuration values
nvgpu_gr_config_init() is called to initialize GR configuration from
gr_gk20a_init_gr_config() and gr_gk20a_init_map_tiles() is simply
renamed as nvgpu_gr_config_init_map_tiles()
Expose new API nvgpu_gr_config_deinit() to deinit the configuration
Expose nvgpu_gr_config_get_*() APIs to query above configuration
fields stored in nvgpu_gr_config structure
Update vgpu_gr_init_gr_config() to initialize the configuration
from gr->config structure
Chip specific HALs that access GR register for initialization
are implemented in common/gr/config/gr_config_gm20b.c
Set these HALs for all GPUs
Jira NVGPU-1879
Change-Id: Ided658b43124ea61b9f273b82b73fdde4ed3c8f0
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2012167
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
gk20a_remove_gr_support() is freeing the local_golden_image and
local_golden_image->context. But there are instances where
local_golden_image is not allocated since freeing an
unallocated golden context image accesses the contents of
local_golden_image causes a fault.
Check golden_image_initialized flag before freeing
local_golden_image->context.
Jira NVGPU-1648
Bug 2461665
Change-Id: I19235d2ec9d77ba4ef00257f43436448f5f70b25
Signed-off-by: Prateek sethi <prsethi@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1997665
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Rename __nvgpu_set_enabled() to nvgpu_set_enabled(). The original
double underscore was present to indicate that this function is a
function with potentially unintended side effects (enabling a feature
has wide ranging impact).
To not lose this documentation a comment was added to convey that this
function must be used with care.
JIRA NVGPU-1029
Change-Id: I8bfc6fa4c17743f9f8056cb6a7a0f66229ca2583
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1989434
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Local golden image is copy of global GR context buffer hence move its
ownership to global context unit
Add new structure nvgpu_gr_global_ctx_local_golden_image to hold all meta
data for local golden image and move it to struct gr_gk20a
Expose and use new APIs to initialize/deinitialize and load local golden image
Jira NVGPU-1625
Change-Id: Ieb68e52c205ca0ecd27f8bf4bb31922a01e7ae54
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1984952
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
MISRA Rule 21.15 prohibits use of memcpy() with incompatible ptrs
to qualified/unqualified types.
To circumvent this issue we've introduced a new MISRA-compliant
nvgpu_memcpy() function.
While linux os code does not need to be MISRA-compliant this
change switches over all memcpy() uses to nvgpu_memcpy()
with appropriate casts applied to maintain consistency within
the nvgpu source base.
JIRA NVGPU-849
Change-Id: I2c21a7845df5709dafa19508c121f8afa27cc4fc
Signed-off-by: Scott Long <scottl@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1950995
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Nitin Kumbhar <nkumbhar@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>
MISRA Rule-17.7 requires the return value of all functions to be used.
Fix is either to use the return value or change the function to return
void. This patch contains fix for all 17.7 violations instandard C functions
in OS/Linux interface.
JIRA NVGPU-1036
Change-Id: I39b20f1d0e1a1da56d452f2c3d5ee049666cefe8
Signed-off-by: Nicolas Benech <nbenech@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1929900
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svc-misra-checker <svc-misra-checker@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>
- TPC powergating should be done before
calling gk20a_enable_gr_hw.
gk20a_enable_gr_hw() issues a GR engine reset.
Without this fix, enabling 1 TPC from each PES
causes ctxsw timeout error while running GFX Benchmark.
- Adds valid tpc-pg mask for 1/2/3/4 active TPC configs.
TPC Config - TPC-MASK
4 TPC configuration - 0x0
3 TPC configuration - 0x1/0x2/0x4/0x8
2 TPC configuration - 0x5/0x9/0x6/0xa
- We should not write to gr_fe_tpc_pesmask_r()
as part of TPC-PG sequence. This register is for
debug purpose only.
Bug 200442360
Change-Id: I6fbe1ad8fbc836ace8cbaf00ec3d21a12c73e0bd
Signed-off-by: Deepak Goyal <dgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1809772
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add sysfs to indicate if GPU has ever been powered on.
This allows an application to avoid booting dGPU, when there is no
need for it. For instance the fan controller can check this value,
and refrain from booting the GPU, only to check the temperature.
This will save power, and also avoid locking some registers which
currently prevent from using VBIOS update tool.
Bug 200408936
Change-Id: I6344001e91e9eee8762ff8b227ca6698d8c6786f
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1802098
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Adding a comptag_mem_deduct in the platform_gk20a has certain problems:
1) It's not really convenient for platform users to configure it.
2) All products using the same GPU have to share the same configuration.
So this patch moves this comptag_mem_deduct from struct platform_gk20a
to struct gr_gk20a (per device). And it adds an sysfs node for products
or platform users to easily configure from user space.
Note: The comptag memory will not be allocated until the GPU driver goes
through the final poweron routine. So the user space has a small window
to configure this sysfs node.
Bug 2327574
Bug 2284925
Change-Id: Ie7d00b082704e422645c0ea254b59e22f9fc3b7f
Signed-off-by: Nicolin Chen <nicolinc@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1810334
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
In nvgpu repository, we have multiple accesses to methods in
pmu_gk20a.h which have register accesses. Instead of directly invoking
these methods, these are now called via HALs. Some common methods such
as pmu_wait_message_cond which donot have any register accesses
are moved to pmu_ipc.c and the method declarations are moved
to pmu.h. Also, changed gm20b_pmu_dbg to
nvgpu_dbg_pmu all across the code base. This would remove all
indirect dependencies via gk20a.h into pmu_gk20a.h. As a result
pmu_gk20a.h is now removed from gk20a.h
JIRA-597
Change-Id: Id54b2684ca39362fda7626238c3116cd49e92080
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1804283
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
In order to avoid the circular dependencies,
rearrange the static inline functions from
gk20a.h file.
Moved gk20a_gr_flush_channel_tlb function to
gr_gk20a.c and removed the #include gr_gk20a.h
from gk20a.h
Added a helper function utils.h to
move all generic static inline functions which
have no reference to gpu related structures.
ptimer related functions are moved to
ptimer.h
Implementations for as and pmu are moved to
corresponding files.
JIRA NVGPU-624
Change-Id: I4e956326e773ba037bf3a1696cc4c462085dbbe5
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1781941
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
- adds static tpc-powergating through sysfs.
- active tpc count will remain till the GPU/systems is not booted again.
- tpc_pg_mask can be written only after GPU probe finishes and
GPU boot is triggered.
Note:
To be able to use this feature, we need to change boot/init
scripts of the OS(used with nvgpu driver) to write to sysfs nodes before
posting discover image size query to FECS.
Bug 200406784
Change-Id: Id749c7a617422c625f77d0c1a9aada2eb960c4d0
Signed-off-by: Deepak Goyal <dgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1742422
Reviewed-by: svc-misra-checker <svc-misra-checker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
