CTS test dEQP-VK.api.object_management.max_concurrent.device_group
crashes with invalid userspace memory access.
Currently, nvgpu_submit_prepare_syncs() races with
nvgpu_channel_clean_up_jobs() and this race condition is exposed when
aggressive_sync_destroy_thresh is set to non-zero value.
nvgpu_submit_prepare_syncs() gets ref for c->sync to submit job and
releases channel sync_lock immediately. Meanwhile,
nvgpu_worker_poll_work() triggers nvgpu_channel_clean_up_jobs(), which
destroys ref'd c->sync pointer.
Channel sync is deleted by nvgpu_channel_clean_up_jobs() only if
aggressive_sync_destroy_thresh is non-zero.
So, nvgpu_channel_clean_up_jobs() and nvgpu_submit_prepare_syncs() will
race only in this scenario.
Hence, if aggressive_sync_destroy_thresh value is non-zero, this patch
protects channel's sync pointer by holding channel sync_lock
during complete execution of nvgpu_submit_prepare_syncs().
Bug 2613870
Change-Id: I030d8df7af10d4ed86f921b5cf60de2b1d60e5d3
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2181360
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Replaced ch->mmu_debug_mode_enabled with ch->mmu_debug_mode_refcnt.
If channel is enabled multiple times by userspace, then ref count is
updated accordingly. There is an expectation that enable/disable
calls are balanced for setting channel's mmu debug mode.
When unbinding the channel, decrease refcnt for the channel until it
reaches 0.
Also, removed tsg parameter from nvgpu_tsg_set_mmu_debug_mode as it
can be retrieved from ch.
Bug 2515097
Change-Id: If334e374a55bd14ae219edbfd3b1fce5ff25c226
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2184702
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This patch adds nvgpu API in linux and posix to query vpr resize.
The new API nvgpu_is_vpr_resize_enabled() is used in
nvgpu_submit_channel_gpfifo().
Previously, if non-deterministic channel has timeout disabled and
GPU cannot railgate on some platform, then channel doesn't power ref
count and results in video freeze. To resolve non-determinstic channel
job tracking needs to be enabled if vpr resize is supported or if GPU
can railgate.
Bug 200532122
Change-Id: Icfbff6253762b195b2f5955749343974b1a7a269
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2171093
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Fix misra errors in gr units
misra 14.3 rule - there shall be no dead code.
misra_c_2012_rule_14_3_violation: The condition
"graphics_preempt_mode != 0U" cannot be true.
misra_c_2012_rule_16_1_violation: The switch statement is not
well formed.
misra_c_2012_rule_10_8_violation: Cast from 32 bit width expression
"(regval >> 1U) & 1U" to a wider 64 bit type.
Jira NVGPU-3872
Change-Id: Ibb53d0756d464d2ae3279d1b841b3c91a16df9be
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2182562
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
For safety build, nvgpu driver should enter SW quiesce state
in case an uncorrectable error has occurred. In this state, any
activity on the GPU should be prevented, without powering off the GPU.
Also, a minimal set of operations should be used to enter SW quiesce
state.
Entering SW quiesce state does the following:
- set sw_quiesce_pending: when this flag is set, interrupt
handlers exit after masking interrupts. This should help mitigate
an interrupt storm.
- wake up thread to complete quiescing.
The thread performs the following:
- set NVGPU_DRIVER_IS_DYING to prevent allocation of new resources
- disable interrupts
- disable fifo scheduling
- preempt all runlists
- set error notifier for all active channels
Note: for channels with usermode submit enabled, userspace can
still ring doorbell, but this will not trigger any work on
engines since fifo scheduling is disabled.
Jira NVGPU-3493
Change-Id: I639a32da754d8833f54dcec1fa23135721d8d89a
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2172391
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Rule 10.1 Requires that operands shall not be of an inappropriate
essential type.
Rule 14.3 Requires that controlling expressions shall not be
invariant.
Rule 20.7 Requires that expressions resulting from the expansion
of macro parameters shall be enclosed in parentheses.
Fix violations of above rules in the macro defined in cond unit.
Jira NVGPU-3878
Change-Id: Icd5758ae12210e68333c4ab820a36dc48c0aa29e
Signed-off-by: ajesh <akv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2183647
GVS: Gerrit_Virtual_Submit
Reviewed-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The following violations are fixed in this patch
a) misra_c_2012_rule_2_1_violation: This code cannot be reached: "return
err;".
b) misra_c_2012_directive_4_7_violation: Calling function
"nvgpu_preempt_channel(g, ch)" which returns error information without
testing the error information.
c) misra_c_2012_rule_8_6_violation: "" is declared but never defined for
following functions
1) gm20b_dump_engine_status
2) gp10b_ramfc_setup
3) gp10b_ramfc_get_syncpt
4) gp10b_ramfc_set_syncpt
5) gk20a_fifo_intr_0_enable
6) gk20a_fifo_intr_0_isr
7) gk20a_fifo_handle_sched_error
8) gk20a_fifo_is_mmu_fault_pending
9) gk20a_fifo_intr_set_recover_mask
10) gk20a_fifo_intr_unset_recover_mask
11) gk20a_init_fifo_reset_enable_hw
12) gk20a_init_fifo_setup_hw
13) nvgpu_tsg_set_runlist_interleave
14) gm20b_dump_engine_status
15) gp10b_pbdma_channel_fatal_0_intr_descs
16) gp10b_pbdma_allowed_syncpoints_0_index_f
17) gp10b_pbdma_allowed_syncpoints_0_valid_f
18) gp10b_pbdma_allowed_syncpoints_0_index_v
19) gk20a_runlist_reschedule
The above functions declarations are now embedded within
CONFIG_NVGPU_HAL_NON_FUSA
d) The function nvgpu_channel_abort_clean_up has a UMD version and hence
its taken out of CONFIG_NVGPU_KERNEL_MODE_SUBMIT to avoid errors of
type c above.
Jira NVGPU-3881
Change-Id: I5f85c7070e1d2f0b18d14db07ce22a01c29f0e40
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2181032
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: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add doxygen documentation for gr/gr.h header.
Also add necessary plumbing to add common.gr unit documentation HTML
link.
Fix gr_init_reset_enable_hw() to return error code in case of error.
Compile nvgpu_gr_enable_ctxsw()/nvgpu_gr_disable_ctxsw() only if
debugger or recovery support is enabled.
Jira NVGPU-3911
Change-Id: Iaa942748a00aeffce9349cb0669d2cd5baca1e55
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2180312
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Nvgpu uses RPC to get target voltage for a freq, though this gets
the latest Voltage, there could be mismatch b/w data in nvgpu & PMU.
To make it consistent, use the local VF table for getting the voltage.
Also the slave ratio calculation is inaccurate due to quantization.
So instead of calculating, use the slave ratio from parsed vbios table.
Bug 200545403
Change-Id: Ibb064f2a0f5eba77166e2b3f9868da9e3fcc7193
Signed-off-by: Abdul Salam <absalam@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2183546
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
- Add doxygen documentation.
- Remove unused fields of nvgpu_tsg struct:
-- timeslice_timeout
-- timeslice_scale
- Remove unused functions:
-- nvgpu_tsg_set_runlist_interleave
- nvgpu_tsg_post_event_id is not supported in safety build.
This function is moved under CONFIG_NVGPU_CHANNEL_TSG_CONTROL
compiler flag.
- Below functions are moved under CONFIG_NVGPU_KERNEL_MODE_SUBMIT
nvgpu_tsg_ctxsw_timeout_debug_dump_state
nvgpu_tsg_set_ctxsw_timeout_accumulated_ms
- Rename
gk20a_is_channel_active -> nvgpu_tsg_is_channel_active
release_used_tsg -> nvgpu_tsg_release_used_tsg
- nvgpu_tsg_unbind_channel_common declared static
- Fix build issue when CONFIG_NVGPU_CHANNEL_TSG_CONTROL is disabled
Remove CONFIG_NVGPU_CHANNEL_TSG_CONTROL for
nvgpu_gr_setup_set_preemption_mode as it is needed in safety build.
By default compute preemption mode will be set to WFI. CUDA will
change it to CTA during context init time.
JIRA NVGPU-3595
Change-Id: I8ff6cabc8b892c691d951c37cdc0721e820a0297
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2151489
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
A call to exit the PMU state machine/kthread must
be prioritized over any other state change.
It was possible to set the state as PMU_STATE_EXIT,
signal the kthread and overwrite the state before
the kthread has had the chance to exit its loop.
This may lead to a "lost" signal, resulting in
indefinite wait during the destroy sequence.
Faulting sequence:
1. pmu_state = PMU_STATE_EXIT in nvgpu_pmu_destroy()
2. cond_signal()
3. pmu_state = PMU_STATE_LOADING_PG_BUF
4. PMU kthread wakes up
5. PMU kthread processes PMU_STATE_LOADING_PG_BUF
6. PMU kthread sleeps
7. nvgpu_pmu_destroy() waits indefinitely
This patch adds a sticky flag to indicate PMU_STATE_EXIT,
irrespective of any subsequent changes to pmu_state.
The PMU PG init kthread may wait on a call to
NVGPU_COND_WAIT_INTERRUPTIBLE, which requires a
corresponding call to nvgpu_cond_signal_interruptible()
as the core kernel code requires this task mask to
wake-up an interruptible task.
Bug 2658750
Bug 200532122
Change-Id: I61beae80673486f83bf60c703a8af88b066a1c36
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2181926
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Mahantesh Kumbar <mkumbar@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>
Change the directly called init functions to function pointers in the
HAL. This makes it more consistent. This also allows for writing more
comprehensive unit tests for nvgpu.common.init.
JIRA NVGPU-2239
Change-Id: I05d739a8f8a2e7d385322d93154206eb0bfddc10
Signed-off-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2173920
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
For getting the buffer size qnx issues a devctl to nvmap which can fail
as well. So, check the size that is returned by nvgpu_os_buf_get_size.
If 0 size is returned then return -EINVAL to the caller.
Jira NVGPU-3891
Change-Id: Id13e7612b044e9228d78469ab4e43961a6877ce8
Signed-off-by: Shashank Singh <shashsingh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2174458
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Rename reserved identifier __here to label_here to fix
around 180 DCL37-C violations - "The reserved identifier "__here",
which is reserved for use as identifiers with file scope in both
the ordinary and tag name spaces, is declared.
JIRA NVGPU-3908
Change-Id: Iecaaa03674800bec919d71dc5fd226d362fc6f56
Signed-off-by: Nitin Kumbhar <nkumbhar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2178457
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vinod Gopalakrishnakurup <vinodg@nvidia.com>
Reviewed-by: Adeel Raza <araza@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Advisory Rule 2.7 states that there should be no unused
parameters in functions.
This patch removes the unused 'post_event', 'fault_ch' and
'hww_global_esr' parameters from the following:
* gv11b_gr_intr_handle_l1_tag_exception()
* gv11b_gr_intr_handle_lrf_exception()
* gv11b_gr_intr_handle_cbu_exception()
* gv11b_gr_intr_handle_l1_data_exception()
* gv11b_gr_intr_handle_icache_exception()
These changes allowed the same parameters to be removed from the
the gr.intr.handle_tpc_sm_ecc_exception() interface.
Jira NVGPU-3178
Change-Id: I4d5dcbf2a5325e38782cdac67f9dd0b223fa1a18
Signed-off-by: Scott Long <scottl@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2171220
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
test_falcon_halt failed as nvgpu_timeout_expired returned -ETIMEDOUT when
time equal to timeout is reached and nvgpu_timeout_peek_expired returns
false when time is equal or less and true when time is greater than
timeout, leading to inconsistent return value.
Update nvgpu_timeout_expired_msg_cpu logic that is used by former.
JIRA NVGPU-3946
Change-Id: I365063cc12a584833c08ca710bb795c0e9d814cd
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2180233
Reviewed-by: Nicolas Benech <nbenech@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>
- This patch fixes disable fecs trace logic.
- If user does trace disable twice, enable_count will become
negative and when user tries to re-enable it, fecs trace
will not be enabled.
Bug 2672760
Change-Id: I895e48549970fa2faf06d5a531a423489f494cf4
Signed-off-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2171015
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Changes added to support "rmmod nvgpu" in dgpu simulation after gpu
poweron.
nvgpu_engine-wait_for_idle got stuck in busy mode for nvdec and nvec
engines in simulation as simulation doesnt support timeout.
These engines are not valid engines in nvgpu engine list.
Add nvgpu_engine_check_valid_id before checking engine status.
Simulation crash on accessing 0xb81604 top interrupt register.
Add func_priv_cpu_intr_top__size_1_v() function to get the supported
size than using default MAX_INTR_TOP_REGS.
nvlink is not supprted in dgpu simulation. Avoid warning for
-ENODEV return.
Avoid register read following gpu power off completion.
Bug 2498574
Change-Id: I9f9f1cf1ac4620242bda1d2cc0f29f51f81a6711
Signed-off-by: vinodg <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2179930
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>
