-Add sm input param for handle_sm_exception and
pre_process_sm_exception for gr ops/functions.
-Add functions to calculate gpc and tpc reg offsets.
-Add function to find SMs which raised SM exception.
JIRA GPUT19X-75
Change-Id: I257e7342ddabadb1556c9551c50a54d34b0f9d1e
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1476108
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: David Martinez Nieto <dmartineznie@nvidia.com>
Add vgpu support for ModeE perfbuffers
- VM allocation is handled by the kernel, with final mapping
handled by the RM server
- Enabling/disabling the perfbuffer is handled by the RM server
Bug 1880196
JIRA EVLR-1074
Change-Id: Ifbeb5ede6b07e2e112b930c602c22b66a58ac920
Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com>
Reviewed-on: https://git-master/r/1506747
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reorder non-function pointer members of gpu_ops to be at the
very end of their respective sub-structs. This allows for
easier debug interpretation and slightly improves readability.
Jira NVGPU-107
Change-Id: Ife3279180306de70f7fad6760f616c6d69769b36
Signed-off-by: Sunny He <suhe@nvidia.com>
Reviewed-on: http://git-master/r/1506591
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
linux/version.h and linux/sched.h are not used by gk20a.h, so remove
them.
acr_gm20b.h and bus_gk20a.h are not needed by gk20a.h, so remove them.
pmu_gp106.c relies on implicit #include of acr_gm20b.h by gk20a.h, so
add that as an explicit #include.
Remove #include of iomap.h. platform_gk20a_tegra.c legacy rail gating
code still relies on access to that header, so add it as explicit
include.
JIRA NVGPU-38
Change-Id: I1cf57b9d3a7ee5e3cad298341107e317b4b8662f
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1505926
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
VBIOS memory settings have been updated for PG419, significantly
modifying MCLK switching sequences. This change adds support for
PG419 tables, while remaining backward compatible with PG418.
Bug 1921082
JIRA EVLR-1269
Change-Id: Ia8a1f8b3f482e348a46f0acb540af23287d9c11e
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1484110
(cherry picked from commit c2444ae89caf97da2702e8486cc8fb162b4f50b1)
Reviewed-on: http://git-master/r/1485300
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Pass struct gk20a pointer to gk20a_busy_noresume() and
gk20a_idle_nosuspend(). This reduces the number of dependencies to
Linux specific struct device.
JIRA NVGPU-38
Change-Id: I5e05be32e2376bc8be5402bb973c20e28c35a1c3
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1505177
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Delete a version of WRITE_ONCE() that gk20a.h defines only for
kernels older than 3.18. We don't support kernels below 4.4.
JIRA NVGPU-38
Change-Id: I0af50936523fde9929c21eea0547b91dac4a0081
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1505175
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
GVS: Gerrit_Virtual_Submit
The function tsg_gk20a_from_ch() is an operation on tsg_gk20a
structure, so move it to be part of tsg_gk20a.c and export
via tsg_gk20a.h.
JIRA NVGPU-38
Change-Id: I2afba3533ac829088a5edf8b16cf4e071b69b77a
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1505172
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
GVS: Gerrit_Virtual_Submit
In order to perform timestamps correlation for FECS
traces, we need to collect GPU / GPU timestamps
samples. In virtualization case, it is possible for
a guest to get GPU timestamps by using read_ptimer.
However, if the CPU timestamp is read on guest side,
and the GPU timestamp is read on vm-server side,
then it introduces some latency that will create an
artificial offset for GPU timestamps (~2 us in
average). For better CPU / GPU timestamps correlation,
Added a command to collect all timestamps on vm-server
side.
Bug 1900475
Change-Id: Idfdc6ae4c16c501dc5e00053a5b75932c55148d6
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1472447
(cherry picked from commit 56f56b5cd9)
Reviewed-on: http://git-master/r/1489183
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
To make do_idle work when nvgpu is built as a module, reverse the order
of call dependencies for do_idle. Don't provide visible
gk20a_do_{idle,unidle}() functions for the kernel but instead call the
kernel for registering and unregistering pointers to them when the
driver loads and unloads.
Refactor the internal __gk20a_do_{idle,unidle} functions to take a
struct gk20a * instead of struct device *, and use the callback api for
providing that g instead of retrieving the plat device from device tree.
Bug 200290850
Change-Id: Ibef8b069302e547b298069cbb97734f461a10cc3
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1493774
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
To support deterministic channels even with platforms where railgating
is supported, have each deterministic-marked channel hold a power
reference during their lifetime, and skip taking power refs for jobs in
submit path for those.
Previously, railgating blocked deterministic submits in general because
of gk20a_busy()/gk20a_idle() calls in submit path possibly taking time
and more significantly because the gpu may need turning on which takes a
nondeterministic and long amount of time.
As an exception, gk20a_do_idle() can still block deterministic submits
until gk20a_do_unidle() is called. Add a rwsem to guard this. VPR resize
needs do_idle, which conflicts with deterministic channels' requirement
to keep the GPU on. This is documented in the ioctl header now.
Make NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_NO_JOBTRACKING always
set in the gpu characteristics now that it's supported. The only thing
left now blocking NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_FULL is
the sync framework.
Make the channel debug dump show which channels are deterministic.
Bug 200291300
Jira NVGPU-70
Change-Id: I47b6f3a8517cd6e4255f6ca2855e3dd912e4f5f3
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1483038
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
- Moved PMU IPC related code to
drivers/gpu/nvgpu/common/pmu/pmu_ipc.c file,
-Below is the list which are moved
seq
mutex
queue
cmd/msg post & process
event handling
NVGPU-56
Change-Id: Ic380faa27de4e5574d5b22500125e86027fd4b5d
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: http://git-master/r/1478167
GVS: Gerrit_Virtual_Submit
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Remove use of linux specifix header files
<linux/nvhost.h> and <linux/nvhost_ioctl.h>
and use nvgpu specific header file <nvgpu/nvhost.h>
instead
This is needed to remove all Linux dependencies
from nvgpu driver
Replace all nvhost_*() calls by
nvgpu_nvhost_*() calls from new nvgpu library
Remove platform device pointer host1x_dev
from struct gk20a and add struct
nvgpu_nvhost_dev instead
Jira NVGPU-29
Change-Id: Ia7af70602cfc16f9ccc380752538c05a9cbb8a67
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1489726
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Add new abstraction file common/linux/nvhost.c for all
nvhost APIs and operations and export them from
header <nvgpu/nvhost.h>
This file will be compiled only if config
CONFIG_TEGRA_GK20A_NVHOST is set
Define struct nvgpu_nvhost_dev in a separate private
header nvhost_priv.h
Jira NVGPU-29
Change-Id: I17e1f7836d4854feadff0c339bc093e78ba7f3eb
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1489725
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Split handling of stalling interrupt to Linux specific chip
agnostic and OS independent chip specific parts.
Linux specific chip independent part contains handler for ISR
and passing the control to a bottom half worker. It uses the new MC
HALs intr_nonstall (query interrupt status), intr_nonstall_pause
(pause interrupts), intr_nonstall_resume (resume interrupts), and
is_intr1_pending (query per-engine interrupt bit).
MC HAL isr_nonstall is removed, because its work is now handled in
chip independent code.
JIRA NVGPU-26
Change-Id: I3e4c9905ef6eef7f1cc9f71b0278518ae663f87e
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1497048
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add MC HAL is_intr1_pending. At the same time introduce nvgpu_unit
that is passed as parameter to is_intr1_pending. The API is passed
contents of intr1 register and an engine number, and returns true
if there's an interrupt pending for the engine.
JIRA NVGPU-26
Change-Id: I8e6363dd78572f8e41dbab2b258036ed168b6f75
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1497870
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Unify the initialization routines for the vGPU and regular GPU paths.
This helps avoid any further code divergence. This also assumes that
the code running on the regular GPU essentially works for the vGPU.
The only addition is that the regular GPU path calls an API in the
vGPU code that sends the necessary RM server message.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: I37af1993fd8b50f666ae27524d382cce49cf28f7
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1480226
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move debugfs code from clk_gm20b.c to file in Linux module
common/linux/debug_clk.c
This file will be compiled only if CONFIG_DEBUG_FS is set
Define below new HAL APIs for various clock operations
which can be accessed from debug file
init_debugfs()
get_voltage()
get_gpcclk_clock_counter()
pll_reg_write()
get_pll_debug_data()
Export nvgpu_pl_to_div() and nvgpu_div_to_pl() so
that these can be accessed from debug_clk.c
Add new structure nvgpu_clk_pll_debug_data so that
all required register values for debugging can be
made available in debug_clk.c
Add new API gm20b_get_gpc_pll_parms() so that statically
defined variable can be accessed in debug_clk.c too
Remove global variable dvfs_safe_max_freq and add
it to struct clk_gk20a so that it can accessed
from both clk_gm20b.c and debug_clk.c
Jira NVGPU-62
Change-Id: I3ae70b40235e78141a686686930e1f178ad59453
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1488903
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
- Renamed "struct pmu_gk20a" to "struct nvgpu_pmu" then moved
to file "pmu.h" under folder "drivers/gpu/nvgpu/include/nvgpu/"
- Included header file "pmu.h" to dependent file &
removed "pmu_gk20a.h" include if its usage is not present.
- Replaced "struct pmu_gk20a" with "struct nvgpu_pmu" in dependent
source & header files.
JIRA NVGPU-56
Change-Id: Ia3c606616831027093d5c216959c6a40d7c2632e
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: http://git-master/r/1479209
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Since all debugfs code is Linux specific, remove
it from common code and move it to Linux module
Debugfs code is now divided into below
module specific files :
common/linux/debug.c
common/linux/debug_cde.c
common/linux/debug_ce.c
common/linux/debug_fifo.c
common/linux/debug_gr.c
common/linux/debug_mm.c
common/linux/debug_allocator.c
common/linux/debug_kmem.c
common/linux/debug_pmu.c
common/linux/debug_sched.c
Add corresponding header files for above modules too
And compile all of above files only if CONFIG_DEBUG_FS is set
Some more details of the changes made
- Move and rename gk20a/debug_gk20a.c to common/linux/debug.c
- Move and rename gk20a/debug_gk20a.h to include/nvgpu/debug.h
- Remove gm20b/debug_gm20b.c and gm20b/debug_gm20b.h and call
gk20a_init_debug_ops() directly from gm20b_init_hal()
- Update all debug APIs to receive struct gk20a as parameter
instead of receiving struct device pointer
- Update API gk20a_dmabuf_get_state() to receive struct gk20a
pointer instead of struct device
- Include <nvgpu/debug.h> explicitly in all files where debug
operations are used
- Remove "gk20a/platform_gk20a.h" include from HAL files
which no longer need this include
- Add new API gk20a_debug_deinit() to deinitialize debugfs
and call it from gk20a_remove()
- Move API gk20a_debug_dump_all_channel_status_ramfc() to
gk20a/fifo_gk20a.c
Jira NVGPU-62
Change-Id: I076975d3d7f669bdbe9212fa33d98529377feeb6
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1488902
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Begin removing all of the myriad flag variables in struct gk20a and
replace that with one API that checks for flags being enabled or
disabled. The API is as follows:
bool nvgpu_is_enabled(struct gk20a *g, int flag);
bool __nvgpu_set_enabled(struct gk20a *g, int flag, bool state);
These APIs allow many of the gk20a flags to be replaced by defines.
This makes flag usage consistent and saves a small amount of memory in
struct gk20a. Also it makes struct gk20a easier to read since there's
less clutter scattered through out.
JIRA NVGPU-84
Change-Id: I6525cecbe97c4e8379e5f53e29ef0b4dbd1a7fc2
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1488049
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This is needed to take care of gp10b h/w header
changes. gp10b changes as compared to legacy gpu chips
-fault_info_fault_type field width is changed
-fault_info_write field is removed
-fault_info_access_type field is added
-fault_info_engine_subid is removed
-fault_info_client_type is added
-fault_info_client field width has changed
JIRA GPUT19X-7
JIRA GPUT19X-12
Change-Id: Iebf28cc6c851830524049b67a71cd72fb4a28948
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: http://git-master/r/1487319
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Split handling of stalling interrupt to Linux specific chip
agnostic and OS independent chip specific parts.
Linux specific chip independent part contains handler for ISR
and passing the control to a bottom half. It uses the new MC HALs
intr_stall (query interrupt status), intr_pause (pause interrupts)
and intr_resume (resume interrupts).
MC HAL isr_stall now returns void and gets called in thread context
and thus remove isr_thread_stall and replace the implementation with
isr_stall.
JIRA NVGPU-26
Change-Id: I206f330f6fc4a1f4def47c5f986585ac4080216d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1480243
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Remove the VM init/de-init from the HAL and instead use a single
set of routines that init/de-init VMs. This prevents code divergence
between vGPUs and regular GPUs.
This patch also clears up the naming of the routines a little bit.
Since some VMs are used inplace and others are dynamically allocated
the APIs for freeing them were confusing. Also some free calls also
clean up an instance block (this is API abuse - but this is how it
currently exists).
The new API looks like this:
void __nvgpu_vm_remove(struct vm_gk20a *vm);
void nvgpu_vm_remove(struct vm_gk20a *vm);
void nvgpu_vm_remove_inst(struct vm_gk20a *vm,
struct nvgpu_mem *inst_block);
void nvgpu_vm_remove_vgpu(struct vm_gk20a *vm);
int nvgpu_init_vm(struct mm_gk20a *mm,
struct vm_gk20a *vm,
u32 big_page_size,
u64 low_hole,
u64 kernel_reserved,
u64 aperture_size,
bool big_pages,
bool userspace_managed,
char *name);
void nvgpu_deinit_vm(struct vm_gk20a *vm);
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: Ia4016384c54746bfbcaa4bdd0d29d03d5d7f7f1b
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1477747
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Make the watchdog notice also fine-grained changes within a single
pushbuffer - by tracking just the gpfifo get, the watchdog could wake
when the channel hasn't really been stuck but processing a relatively
large or slow pushbuf.
Jira NVGPU-72
Change-Id: I15374eea5d9abc9d3725a79d0b960503237e478c
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1485919
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
clk_gm20b.c has number of calls specific to linux and
tegra-soc environment
In order to unify the driver, move all of those calls
to tegra/linux specific file tegra/linux/clk.c
All the clk_*() and tegra_dvfs_*() calls are now
abstracted behind GPU's clock operations and shoule be
accessed using g->ops.clk.<API> format
Remove <linux/clk.h> and <soc/tegra/tegra-dvfs.h> from
clk_gm20b.c
Remove <linux/version.h> from clk_gm20b.c too since
we only support k4.4 and higher version only
Jira NVGPU-49
Change-Id: Ib26811e0423bbd3868b9a46e662b80a8ca088dc5
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1483092
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add the following ioctls
- NVGPU_CHANNEL_IOCTL_GET_TIMESLICE for channel timeslice in us
- NVGPU_TSG_IOCTL_GET_TIMESLICE for TSG timeslice in us
If timeslice has not been set explicitly, ioctl returns the
default timeslice that will be used when programming the runlist
entry.
Bug 1883271
Change-Id: Ib18fdd836323b1a2d4efceb1e27d07713bd6fca5
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1469040
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Remove gk20a_platform dependencies from gk20a.h. This makes
gk20a_platform a Linux platform specific data structure. Add #include
for platform_gk20a.h in the source files that still depend on Linux.
JIRA NVGPU-16
Change-Id: Ib098accd34a1f5066eb8680c387f9b178169f3f0
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1463547
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
