To avoid duplication of same code across multiple chips, export the
following functions through the corresponding headers for the
consumption of other GPU enabling functions:
- ga10b_gr_intr_report_tpc_sm_rams_ecc_err
- gv11b_gr_intr_report_l1_tag_uncorrected_err
- gv11b_gr_intr_report_l1_tag_corrected_err
- gv11b_gr_intr_report_icache_uncorrected_err
JIRA NVGPU-9075
Change-Id: I927285b6e638479ac52cd5d214711e490e5f151e
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2798371
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Due to changes in the host1x driver, dma_fence callbacks will be
executed in interrupt context instead of workqueue context as
previously. To allow for that, this patch effectively moves the
workqueue step into nvgpu so that the in-nvgpu fence callback gets
executed in workqueue context.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Change-Id: I7bfa294aa3b4bea9888921b79175a8fc218d8e3f
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2785968
Reviewed-by: Jonathan Hunter <jonathanh@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Added infrastructure for enabling parsing Control-Fifo's
ring buffers(i.e. send/receive).
Initialization of these buffers are handled as part of
nvgpu_nvs_buffer_alloc() call itself.
A follow-up change shall implement the methods defined here
as part of the existing NVS worker thread.
The changes adhered to the design laid out in the header
nvsched/include/nvs/nvs-control-interface.h.
Jira NVGPU-8619
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Change-Id: I2050e6fb681eba80e01cf547ada37a955e58315a
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2764518
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Use CONFIG_NVS_KMD_BACKEND to enclose all NVS KMD based scheduling
code.
Current configuration contains all the scheduling code managed within
CONFIG_NVS_PRESENT. Eventually, scheduling code shall only use GSP.
Hence, isolate KMD based scheduling code to a config
CONFIG_NVS_KMD_BACKEND. This shall make it easier to remove this code
later.
Jira NVGPU-8619
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Change-Id: I9dc668e0fa3e7706c111fda7a5e2415e1fc0dd03
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2769465
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
If NEXT bit remains set for a channel being unbound, it can lead to
MMU fault of type unbound inst block. When userspace is closing the
channel and NEXT bit is set, userspace retries.
When force killing the channel, nvgpu can retry few iterations to
ensure the channel is truly idle and unbound. If the channel is
really stuck then unbind will fail and TSG will be aborted.
Bug 3800844
Change-Id: I8fb024630ff2dd272245ae27116f3db6d6e0f788
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2787533
(cherry picked from commit 99e39f4b387743a93b05ba4b097c33b23fbbcf68)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2786479
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
When building NVGPU as an OOT module for upstream Linux kernels, the
NVGPU driver source is now copied into a common location with all the
other OOT modules. Therefore, we can now use the 'srctree.nvidia' path
for finding the necessary header files for Host1x and NVMAP. Update the
include search paths to use 'srctree.nvidia' when building NVGPU as an
OOT module.
Bug 3817518
Change-Id: I63066e4331c66a0f47ada83fde3e63402faaf38a
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2785910
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
The error injection code was enabled only when CONFIG_NVGPU_DGPU = n
so that the dGPUs do not attempt any error injection callback
function registration. But, this introduced dependency on DGPU
config when needs to be explicitly set to n for error injection to
be enabled.
Remove the dependency by moving the error injection callback
registration and deregistration to a HAL which is enabled only
on GA10b.
Bug 3819160
Change-Id: I4f4eb99189b1af3502d719536a91cc5e5d866bce
Signed-off-by: Tejal Kudav <tkudav@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2787202
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Changes
- Initialize virtual memory for gsp. This space is used for creating
queues for ctrl fifo. Also can be used to ro map sync-pt to this
instance where gsp firmware can poll the sync-pt with sync-pt id.
- Enabled gsp context interface and written the instance block pointer
to nxtctx register for the gsp firmware to access created virtual memory.
- Added required gsp registers for this feature.
NVGPU-8730
Bug 3770916
Change-Id: If538f615eca3f9b7840ffe2787826528b4808886
Signed-off-by: rmylavarapu <rmylavarapu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2764649
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
This patch fixes nvgpu_css_allocate_perfmon_ids which
leads to a buffer overflow if the allocation of perfmon
ids does not succeed.
If the allocation of perfmon ids cannot be satisfied,
bitmap_find... would return CSS_MAX_PERFMON_IDS and
nvgpu_bitmap_set would still be called with start after
the bitmap array. This results into a buffer overflow.
Bug 3814963
Change-Id: I4caff36cf0c920b4445e1841d16ba2b4c3d19aaa
Signed-off-by: Martin Radev <mradev@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2786747
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Prateek Sethi <prsethi@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
As part of the negative test case we replace the ACR binaries with
corrupted one(by editing the binary in hex editor). The expectaion
is that the process should log the error and exit properly but instead
the process crashed.
The root cause was because NVGPU driver was trying to pause the thread
using nvgpu_nvs_worker_pause but the but NVS isn't initialized at that
point. NVS is initialized after acr init.
Mitigated this failure by adding a checking condition in
nvgpu_nvs_worker_pause.
Bug 3670576
Change-Id: Ibfe66b253be034e7ca2c3ed298dc28d27e1d6de9
Signed-off-by: ht <ht@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2782937
Reviewed-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Prateek Sethi <prsethi@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Following changes are added to fix the issue.
1) Threads having higher priority e.g. RT may preempt
threads with sched-normal priority. As a consequence, higher priority
threads might not still see initialization of data in another thread
resulting in failures such as accessing a condition value before initialization.
Any initialization in the parent thread must be accompanied by a barrier
to make it visible in other thread. Added appropriate barriers to prevent
reordering of the initialization in the thread construction path.
2) There is a race condition between nvgpu_cond_signal() and
nvgpu_cond_destroy() in the asynchronous submit code and corresponding
worker thread's process_item callback for NVS. This may lead to
data corruption and resulting in the above errors as well. Fixed
that by adding a refcount based mechanism for ownership sharing
of the struct nvgpu_nvs_worker_item between the two threads.
Bug 3778235
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Change-Id: Ie9b9ba57bc1dcbb8780801be79863adc39690f72
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2771535
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Prateek Sethi <prsethi@nvidia.com>
Reviewed-by: Ketan Patil <ketanp@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
- Perfmon cmds are non-blocking calls and response
may/may-not come during railgate sequence for the
perfmon command sent as part of nvgpu_pmu_destroy call.
- if response is missed then payload allocated will not be
freed and allocation info will be present as part seq data
structure.
- This will be carried forward for multiple railgate/
rail-ungate sequence and that will cause the memleak
when new allocation request is made for same seq-id.
- Cleanup the sequence data struct as part of nvgpu_pmu_destroy
call by freeing the memory if cb_params is not NULL.
Bug 3747586
Bug 3722721
Change-Id: I1a0f192197769acec12993ae575277e38c9ca9ca
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2763054
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Tested-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
GR context associated with channel is updated in various driver paths.
Sequence to do the same is disable the TSG, preempt the TSG, update
the GR context or instance block and then enable the TSG.
These operations and runlist updates for channel have to be done under
TSG specific ctx_init_lock to avoid the race.
suspend_contexts and resume_contexts needs special handling which is
not covered in this patch.
Bug 3677982
Change-Id: I837257fe9d9ef3eb6f69f5d7e0707e0bb6d4ea72
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2720222
Reviewed-by: Scott Long <scottl@nvidia.com>
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Only certain combination of channels of GFX/Compute object classes can
be assigned to particular pbdma and/or VEID. CILP can be enabled only
in certain configs. Implement checks for the configurations verified
during alloc_obj_ctx and/or setting preemption mode.
Bug 3677982
Change-Id: Ie7026cbb240819c1727b3736ed34044d7138d3cd
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2719995
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Subcontext PDBs and valid mask in the instance blocks of the channels
in various subcontexts has to be updated when new subcontext is
created or a subcontext is removed.
Replayable fault state is cached in the channel structure. Replayable
fault state for subcontext is set based on first channel's bind
parameter. It was earlier programmed in function channel_setup_ramfc.
init_inst_block_core is updated to setup TSG level pdb map and mask.
Added new hal gv11b_channel_bind to enable the subcontext on channel
bind.
Bug 3677982
Change-Id: I58156c5b3ab6309b6a4b8e72b0e798d6a39c1bee
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2719994
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
With multiple subcontexts support added, global context buffers can be
committed as and when VEID0 channel is allocated in the TSG. Earlier,
it was committed on creating first channel in the TSG.
To allow committing the global buffers for VEID0 channels created after
ASYNC VEID channels, patch buffer data count has to be updated hence
change the patch write begin and end calls to update the data count.
Bug 3677982
Change-Id: I8c4e61105384168f157a6983d0f6a0167eb39aab
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2757239
Reviewed-by: Scott Long <scottl@nvidia.com>
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
This patch introduces following relationships among various nvgpu
objects to support multiple address spaces with subcontexts.
IOCTLs setting the relationships are shown in the braces.
nvgpu_tsg 1<---->n nvgpu_tsg_subctx (TSG_BIND_CHANNEL_EX)
nvgpu_tsg 1<---->n nvgpu_gr_ctx_mappings (ALLOC_OBJ_CTX)
nvgpu_tsg_subctx 1<---->1 nvgpu_gr_subctx (ALLOC_OBJ_CTX)
nvgpu_tsg_subctx 1<---->n nvgpu_channel (TSG_BIND_CHANNEL_EX)
nvgpu_gr_ctx_mappings 1<---->n nvgpu_gr_subctx (ALLOC_OBJ_CTX)
nvgpu_gr_ctx_mappings 1<---->1 vm_gk20a (ALLOC_OBJ_CTX)
On unbinding the channel, objects are deleted according
to dependencies.
Without subcontexts, gr_ctx buffers mappings are maintained in the
struct nvgpu_gr_ctx. For subcontexts, they are maintained in the
struct nvgpu_gr_subctx.
Preemption buffer with index NVGPU_GR_CTX_PREEMPT_CTXSW and PM
buffer with index NVGPU_GR_CTX_PM_CTX are to be mapped in all
subcontexts when they are programmed from respective ioctls.
Global GR context buffers are to be programmed only for VEID0.
Based on the channel object class the state is patched in
the patch buffer in every ALLOC_OBJ_CTX call unlike
setting it for only first channel like before.
PM and preemptions buffers programming is protected under TSG
ctx_init_lock.
tsg->vm is now removed. VM reference for gr_ctx buffers mappings
is managed through gr_ctx or gr_subctx mappings object.
For vGPU, gr_subctx and mappings objects are created to reference
VMs for the gr_ctx lifetime.
The functions nvgpu_tsg_subctx_alloc_gr_subctx and nvgpu_tsg_-
subctx_setup_subctx_header sets up the subcontext struct header
for native driver.
The function nvgpu_tsg_subctx_alloc_gr_subctx is called from
vgpu to manage the gr ctx mapping references.
free_subctx is now done when unbinding channel considering
references to the subcontext by other channels. It will unmap
the buffers in native driver case. It will just release the
VM reference in vgpu case.
Note that TEGRA_VGPU_CMD_FREE_CTX_HEADER ioctl is not called
by vgpu any longer as it would be taken care by native driver.
Bug 3677982
Change-Id: Ia439b251ff452a49f8514498832e24d04db86d2f
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2718760
Reviewed-by: Scott Long <scottl@nvidia.com>
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Below change added SPDX license identifier as svcacv complained.
However, these files have MIT license. SPDX need to be added to
GPL licensed sources.
commit 2b2beb7fb6 ("gpu: nvgpu: ga10b: restore the ptimer isr hal")
Remove the identifier from hal files.
Change-Id: Ic90c46721d06a43749e5a48f7077a837a96fb664
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2770085
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
With subcontexts support added, nvgpu has to allocate VEID0 channel
itself to initialize the golden context image. Allocate the channel
and init the golden context image at the beginning of alloc_obj_ctx
call for first user channel.
It can't be initialized at the end of probe as tpc pg settings need
to be updated before golden context image is initialized.
Bug 3677982
Change-Id: Ia82f6ad6e088c2bc1578a6bd32b7c7a707a17224
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2756289
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
The wait_pending HAL is now modified to simply
check the pending status of a given runlist.
The while loop is removed from this HAL.
A new function nvgpu_runlist_wait_pending_legacy() is
added that emulates the older wait_pending() HAL.
nvgpu_runlist_tick() is modified to accept a 64 bit
"preempt_grace_ns" value.
These changes prepare for upcoming control-fifo parser
changes.
Jira NVGPU-8619
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Change-Id: If3f288eb6f2181743c53b657219b3b30d56d26bc
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2766100
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
- Change enables CONFIG_NVS_PRESENT for safety build.
- Fixes misra vioations.
- Renames sched.h to nvs_sched.h to avoid the conflict with QNX system
sched.h file for the safety support.
- Disable test_channel_close, test_tsg_unbind_channel,
test_channel_enable_disable_tsg, test_gv11b_fifo_preempt_tsg,
test_tsg_unbind_channel_check_hw_state and test_rc_deinit unit tests.
Jira NVGPU-8619
Change-Id: I7c983de2f4910fcb23687ec23368a060ce89c918
Signed-off-by: prsethi <prsethi@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2763579
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
- With ELPG + RG enabled, gpu_module_reload test fails.
- This happens because the test tries to unload nvgpu.ko
module and then reload it. This all happens with RG enabled.
- During rmmod of nvgpu.ko module the code path taken is:
nvgpu_remove() -> nvgpu_quiesce() -> gk20a_pm_prepare_poweroff
-> nvgpu_prepare_poweroff -> pmu_destroy
- In this code path, NVGPU_DRIVER_IS_DYING flag is not set.
- Thus, in pmu_pg_task thread (which keeps on running in parallel),
commands are sent to the PMU and the driver keeps waiting for the
ACK in nvgpu_pmu_wait_fw_ack_status().
- Add nvgpu_start_gpu_idle() in nvgpu_remove() path, before calling
nvgpu_quiesce().
- This will set NVGPU_DRIVER_IS_DYING flag to true.
- nvgpu_can_busy() will return 0 when the driver is shutting down or
getting removed.
Bug 3676200
Change-Id: Ic24f58c210e4b477e5d560b053b70c16308e16f1
Signed-off-by: Divya <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2762310
Reviewed-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-by: Tejal Kudav <tkudav@nvidia.com>
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
(cherry picked from commit 8f1792565e71b822a6e9cc50af4b43c1b48518e0)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2765300
Tested-by: Mahantesh Kumbar <mkumbar@nvidia.com>
As part of the negative test case we replace the ACR binaries with
corrupted one(by editing the binary in hex editor). The expectaion
was the process should log the error and exit properly but instead
the process crashed.
We have found the root cause and it was because we were trying to
flush a memory which was not allocated.
To mitigate this issue we added a checking condition to check if the
memory was allocated before flushing.
Bug 3670576
Change-Id: I6b510388fb913695210c791e2253c7514bb7a0a9
Signed-off-by: ht <ht@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2762276
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
