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>
Debug boolean flag dump_ctxsw_stats_on_channel_close is right now stored
in gr_gk20a.ctx_vars struct
This flag logically is property of gr.ctx units since it indicates
whether each context should dump ctxsw stats on channel/context close
Move this flag to struct nvgpu_gr_ctx_desc and remove it from
gr_gk20a.ctx_vars
Expose below API from gr.ctx unit to check if flag is set
nvgpu_gr_ctx_desc_dump_ctxsw_stats_on_channel_close()
Move debugfs creation code to create corresponding debugfs to
gr_gk20a_debugfs_init() and change debugfs type from "u32" to "file"
Struct gr.gr_ctx_desc is created only during first poweron.
Return error if this struct is not available.
Remove unnecessary initialization of this variable from platform
specific probe functions
Jira NVGPU-3112
Change-Id: Id675e047237f82e9b8198a42082e99c95824578f
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2099399
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>
Debug boolean flags force_preemption_gfxp and force_preemption_cilp are
right now stored in gr_gk20a.ctx_vars struct
These flags logically are property of gr.ctx units since they indicate
whether each context should be forced to gfxp/cilp preemption mode by
default
Move these flags to struct nvgpu_gr_ctx_desc and remove them from
gr_gk20a.ctx_vars
Expose below APIs from gr.ctx unit to check if flags are set
nvgpu_gr_ctx_desc_force_preemption_gfxp()
nvgpu_gr_ctx_desc_force_preemption_cilp()
Move debugfs creation code to create corresponding debugfs to
gr_gk20a_debugfs_init() and change debugfs type from "u32" to "file"
Struct gr.gr_ctx_desc is created only during first poweron.
Return error if this struct is not available.
Remove unnecessary initialization of these variables from platform
specific probe functions
Jira NVGPU-3112
Change-Id: I8b2de27f0c71dd2ea5abcf94221c2e15c80073ea
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2099398
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>
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>
Renamed and moved from fifo_gk20a.c to common/fifo/engines.c
gk20a_fifo_should_defer_engine_reset -> nvgpu_engine_should_defer_reset
Renamed and moved from fifo_gk20a.c to common/fifo/channel.c
gk20a_fifo_deferred_reset -> nvgpu_channel_deferred_reset_engines
JIRA NVGPU-1314
Change-Id: Ifc32ff4dde398143b83c2c1b6fab896142574240
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2093910
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Rename the two native GPU GMMU map/unmap functions and update the
HAL initializations to reflect this:
gk20a_locked_gmmu_map -> nvgpu_gmmu_map_locked
gk20a_locked_gmmu_unmap -> nvgpu_gmmu_unmap_locked
This matches what other units do for handling vGPU "HAL" indirection.
Also move the function declarations to <nvgpu/gmmu.h> since these are
shared among all non-vGPU chips. But since these are still technically
HAL operations they should never be called directly. This is a bit of
an organixational issue that I have not thought through hwo to solve
yet.
Ideally they would go into a "hal/mm/gmmu/" include somewhere, but
that A) doesn't yet exist, and B) those are chip specific; these
functions are native specific. Ugh.
JIRA NVGPU-2042
Change-Id: Ibc614f2928630d12eafcec6ce73019628b44ad94
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2099692
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
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>
In gk20a_open_new_channel(), grab the ref_obtain_lock
before marking the channel as referenceable/gettable
to ensure that the ref_count and referenceable flag
are set in a consistent manner. This will ensure that
a thread getting the channel asynchronously while it's
not fully initialized yet (e.g.,
gk20a_channel_semaphore_wakeup()) will always see a
valid ref_count
Bug 200427711
Change-Id: I55ff43f979e7e0a00a49feb2d24be25cec698d95
Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2095559
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Unify qnx types unit with posix. Modify certain defines as part of
unification. Modify the fecs trace function mmap_user_buffer to
not have OS specific structure as parameter. Rename the function
pointer mmap_user_buffer as get_mmap_user_buffer_info to reflect the
functionality handled by the function.
Jira NVGPU-2146
Change-Id: I58ec9bf23a435c039d1b5f04194f56067b11aa28
Signed-off-by: ajesh <akv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2082619
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move the nvgpu_report_gr_exception call from gr_gk20a to
gr_intr.c as nvgpu_gr_intr_report_exception
Move local function gk20a_gr_get_channel_from_ctx to gr_intr.c
as nvgpu_gr_intr_get_channel_from_ctx
JIRA NVGPU-1891
Change-Id: I21521ad50989582d8f166a98a21ea3b1dcd3bbff
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2098229
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add a new MM HAL directory to contain all MM related HAL units.
As part of this change add cache unit to the MM HAL. This contains
several related fixes:
1. Move the cache code in gk20a/mm_gk20a.c and gv11b/mm_gv11b.c to
the new cache HAL. Update makefiles and header includes to take
this into account. Also rename gk20a_{read,write}l() to their
nvgpu_ variants.
2. Update the MM gops: move the cache related functions to the new
cache HAL and update all calls to this HAL to reflect the new
name.
3. Update some direct calls to gk20a MM cache ops to pass through
the HAL instead.
4. Update the unit tests for various MM related things to use the
new MM HAL locations.
This change accomplishes two architecture design goals. Firstly it
removes a multiple HW include from mm_gk20a.c (the flush HW header).
Secondly it moves code from the gk20a/ and gv11b/ directories into
more proper locations under hal/.
JIRA NVGPU-2042
Change-Id: I91e4bdca4341be4dbb46fabd72622b917769f4a6
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2095749
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move zcull size initialization to hal.gr.zcull unit.
This removes zcull dependency on falcon unit
Add new variable zcull_image_size to gr_gk20a.ctx_vars struct
Pass the size to nvgpu_gr_zcull_init()/vgpu_gr_init_gr_zcull() as
parameter to initialize zcull info
Jira NVGPU-3112
Change-Id: I54d966073dad658b4aad3a529f44c0478208b10c
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2098507
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>
Add new API nvgpu_gr_hwpm_map_get_size() in gr.hwpm_map unit to get
size of hwpm_map.
Use this API to get size and allocate each pm_ctx
Move nvgpu_gr_hwpm_map_init() call to gr.gr unit in gr_init_setup_sw()
instead of calling it from gr.falcon unit
Add nvgpu_gr_hwpm_map_init() to vGPU initialization to initialize
hwpm_map size on vGPU
Jira NVGPU-3112
Change-Id: Ifc669dcc9ecae273cea6978f5639f312cd451019
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2096160
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Moved the following HALs from fifo to usermode
- fifo.ring_channel_doorbell -> usermode.ring_doorbell
- fifo.doorbell_token -> usermode.doorbell_token
- fifo.usermode_base -> usermode.base
Created the following HAL
- usermode.setup_hw
Jira NVGPU-2978
Change-Id: I856ea24c126fa22d2f3fe860d4b14087c6d7330b
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2094813
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Added new function to add require sw initionaltions. before enabling
gr hw. Added nvgpu_netlist_init_ctx_vars and nvgpu_gr_falcon_init_support
as part of this function:
int nvgpu_gr_prepare_sw(struct gk20a *g)
Moved following structure defs from gr_gk20a.h to gr_falcon.h and
renamed appropriately:
gk20a_ctxsw_ucode_segment -> nvgpu_ctxsw_ucode_segment
gk20a_ctxsw_ucode_segments -> nvgpu_ctxsw_ucode_segments
Moved following struct to gr_falcon_priv.h:
gk20a_ctxsw_ucode_info -> nvgpu_ctxsw_ucode_info
Moved following data from struct gk20a to new structure in gr_falcon_priv.h
struct nvgpu_gr_falcon:
struct nvgpu_mutex ctxsw_disable_lock;
int ctxsw_disable_count;
struct gk20a_ctxsw_ucode_info ctxsw_ucode_info;
Also moved following data from gr_gk20.h to struct nvgpu_gr_falcon:
struct nvgpu_mutex fecs_mutex;
bool skip_ucode_init;
wait_ucode_status
GR_IS_UCODE related enums
eUcodeHandshakeInit enums
Now add a pointer to this new data structure from struct gr_gk20a to
access gr_falcon related data and modified code to reflect this
change:
struct nvgpu_gr_falcon *falcon;
Added following functions to access gr_falcon data:
struct nvgpu_mutex *nvgpu_gr_falcon_get_fecs_mutex(
struct nvgpu_gr_falcon *falcon);
struct nvgpu_ctxsw_ucode_segments *nvgpu_gr_falcon_get_fecs_ucode_segments(
struct nvgpu_gr_falcon *falcon);
struct nvgpu_ctxsw_ucode_segments *nvgpu_gr_falcon_get_gpccs_ucode_segments(
struct nvgpu_gr_falcon *falcon);
void *nvgpu_gr_falcon_get_surface_desc_cpu_va(
struct nvgpu_gr_falcon *falcon);
JIRA NVGPU-1881
Change-Id: I9100891989b0d6b57c49f2bf00ad839a72bc7c7e
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2091358
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
As part of PMU HAL separation, need to move non-HAL code to respective
UNIT & found still some more PG code left in these pmu_gk20a/gp106.c/h
files which needs to be moved PG UNIT.
JIRA NVGPU-2002
Change-Id: I583c3da35aff788f1bc0451af0b0dbdab0a62c00
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2089870
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add new hal g->ops.gr.init.detect_sm_arch() in hal.gr.init unit to get
SM arch information. Remove g->ops.gr.detect_sm_arch().
Move corresponding functions to hal.gr.init unit
Remove unused function declaration for gr_gv11b_init_sw_veid_bundle()
Jira NVGPU-2961
Change-Id: Idfd5ce19c06978dc31cbcec2cd01cb2912eb3cf9
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2097534
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add new hal g->ops.gr.init.get_patch_slots() in hal.gr.init unit to get
patch slot count. Remove g->ops.gr.get_patch_slots().
Move corresponding functions to hal.gr.init unit
This hal does not need to be set for vGPU since it is not called in
that case
Jira NVGPU-2961
Change-Id: Ide488ae93af53a755da95faa268563070bd24bea
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2097533
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move os_channel close after tsg unbind which internally sets syncpoint
to a safe value. Otherwise it causes syncpoint wait in syncpt waiter
thread to block indefinitely if signaler is killed.
Bug 200509048
Change-Id: Ifcb3c2efcabd94c0a4f7da3975db316926003cb5
Signed-off-by: Shashank Singh <shashsingh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2094476
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Many of the functions in common.gr.obj_ctx and common.gr.fs_state units
directly dereference struct gr_gk20a to obtain other structures
e.g. API nvgpu_gr_obj_ctx_set_ctxsw_preemption_mode() obtains pointer
to nvgpu_gr_config struct by direct access g->gr.config
Such accesses add dependency of these units on gr.h and hence create
circular dependency with common.gr.gr unit
Fix this by receiving all required structures in the function parameter
list itself
Jira NVGPU-1886
Change-Id: Iee973ae33fc7e1707b8f025ad61683f725dedb53
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2094995
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add new API nvgpu_gr_setup_free_subctx() in common.gr.ctx to free subctx
Call this via hal g->ops.gr.setup.free_subctx()
Subctx allocations happens through gr.setup api right now hence it makes
sense to provide subctx free api through same unit
Remove g->ops.channel.free_ctx_header() hal since we now have gr.setup
hal
Remove gv11b/subctx_gv11b.* files since they are no longer needed and
all the code in them has been moved to common units
Jira NVGPU-1886
Change-Id: I3d58fc3665ed9b6ffba830249a4cd30af7b857f4
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2094994
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move get_nonpes_aware_tpc hal to hal.gr.init . This hal is
implemented for gv11b.
Update sm_id_numbering hal to pass the gr_config struct pointer
as parameter to avoid dereferencing from gr inside hal.
JIRA NVGPU-2951
Change-Id: I1e06b634cc36741e116e41e581a18c7f5b373945
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2093835
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Created class unit under hal and moved all valid class check related
functionality to this unit. Moved all class defs from gr to a new header
include/nvgpu/class.h.
Moved following hals from gr to newly created class unit:
bool (*is_valid_class)(struct gk20a *g, u32 class_num); -->
bool (*is_valid)(u32 class_num);
bool (*is_valid_gfx_class)(struct gk20a *g, u32 class_num); -->
bool (*is_valid_gfx)(u32 class_num);
bool (*is_valid_compute_class)(struct gk20a *g, u32 class_num); -->
bool (*is_valid_compute)(u32 class_num);
JIRA NVGPU-3109
Change-Id: I01123e9b984613d4bddb2d8cf23d63410e212408
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2095542
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>
These two files contained the init_characteristics functions for
these two chips. move these to the hal_<chip>.c files for now.
This allows the gv11b.c and gp10b.c to be deleted and a bunch of
header includes removed.
The vGPU code required a slight update to the headers (moving the
<nvgpu/...> headers to before all the "common/..." and "hal/..."
includes since some of those headers make assumptions about what
is already present.
Also, delete the gv100/gv100.h header file and make the gv100
init_characteristics function static to be inline with the way the
gv11b and gp10b code now works. This further redecues the number
of sourece files and reduces needless complexity.
JIRA NVGPU-971
JIRA NVGPU-3074
Change-Id: Ic0ed722612ef9843e6219c7a00144ead12a73e78
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2090978
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Refactor the ECC code present under gp10b/, gv11b/, and tu104/. To
do this a few things were required:
Firstly, move the ecc related C and H files under the aforementioned
directories over to their new home under hal/gr/ecc/. Also copy over
the ECC HAL code present in gp10b/gp10b.c - not sure why this was
there but it, too, needed to be refactored. Also handle all the
updated header paths and Makefiles to take these movements into
account.
Secondly add a new HAL in GR (gr.ecc) to handle ECC HAL ops. There's
only two: detect and init. init() was copied over from gr and detect
was added so that the common ECC code can call it in a chip agnostic
way. This is required so that the ECC detect can be moved out of the
gp10b init characteristics function.
Lastly update the ECC init flow in the common ecc module to call
detect() right before init().
JIRA NVGPU-3074
Change-Id: Iba1464a904f9a91a45f30902beadf6f912607e40
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2090977
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
