In Drive 6.0, the error reporting is supported only for orin (ga10b)
in dev-main. For this purpose, this patch does the following:
- Removes the redundant reporting of following IDs from gv11b:
- GPU_HOST_PFIFO_SCHED_ERROR
- GPU_HOST_PFIFO_CTXSW_TIMEOUT_ERROR
- GPU_HOST_PBDMA_HCE_ERROR
- GPU_MMU_L1TLB_SA_DATA_ECC_UNCORRECTED
- GPU_MMU_L1TLB_FA_DATA_ECC_UNCORRECTED
- GPU_LTC_CACHE_DSTG_ECC_CORRECTED
- GPU_LTC_CACHE_TSTG_ECC_UNCORRECTED
- Migrates the reporting of following IDs from gv11b to ga10b:
- GPU_SM_L1_TAG_ECC_CORRECTED
- GPU_SM_L1_TAG_ECC_UNCORRECTED
- GPU_SM_CBU_ECC_UNCORRECTED
- GPU_SM_LRF_ECC_UNCORRECTED
- GPU_SM_L1_DATA_ECC_UNCORRECTED
- GPU_SM_ICACHE_L1_DATA_ECC_UNCORRECTED
- GPU_SM_ICACHE_L0_PREDECODE_ECC_UNCORRECTED
- GPU_SM_L1_TAG_MISS_FIFO_ECC_UNCORRECTED
- GPU_SM_L1_TAG_S2R_PIXPRF_ECC_UNCORRECTED
- Removes the unused ID that doesn't have any HSI related to it:
- GPU_HOST_PBDMA_PREEMPT_ERROR
In addition to the above, this patch does the following:
- Updates error IDs related to page fault error.
- Updates look-up table to remove unused error IDs.
JIRA NVGPU-8094
Bug 200729736
Change-Id: Ifea76d38ba609c894560e61ff5a6e406290f919e
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2685249
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Dinesh T <dt@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit
When an error is reported to Safety_Services, it will be cleared
at FSI and reported to SEH (System Error Handler). Since MISC_EC
interface provides only one register for error reporting, there
is a need to poll the status of previously reported error before
reporting the next error. For this purpose, this patch adds logic
to perform polling using epl_get_misc_ec_err_status(), in AV+L.
JIRA NVGPU-8094
Bug 200729736
Change-Id: Ia01a2fc42a7ce586b7965a82c90027a9a2dd252b
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2684141
Reviewed-by: Dinesh T <dt@nvidia.com>
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
GVS: Gerrit_Virtual_Submit
Patch udpates/fixes following issues.
- Updates nvgpu_dbg_gpu_get_mappings_entry.size to u64 to address >4G
limitations.
- Removes offset from original cpuva and unmaps only original mapped
address.
- Call nvgpu_vm_find_mapped_buf_range() in place of
nvgpu_vm_find_mapped_buf() to find the addresses which are not
page aligned.
- Update logic to parse the gpuva while trying to find gpu mappings so
that gpuva which are more than the mapped buffer base address can also
be considered.
Bug 200722275
Change-Id: If33d85db37a9f03a662984c212544a8b2ade471c
Signed-off-by: prsethi <prsethi@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2612129
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Dinesh T <dt@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit
While checking the GPU error status, userspace polls on the error
notifier 'status' and then verifies 'status' and 'info32'. nvgpu
sets 'info32' before 'status', so put a write barrier between
those two writes for the consistency between userspace and
kernel view of the error notifier state.
JIRA NVGPU-7538
Bug 200717195
Bug 3250920
Change-Id: I92ac0589283fee823f3366ac594d03b8f27f3590
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2680320
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
-Add check for ECC parity errors in IMEM, DMEM, EMEM, DCLS, REG
for ACR running in GSP engine.
The EXTIRQ3 external interrupt is set from ACR pointing towards host.
-Add function to check error type when ACR or Bootrom execution fails
and report accordingly to SDL with relevant error codes.
This is a part of HSI safety requirements.
Bug 3564039
Jira NVGPU-8108
Change-Id: I65407371f7a1d1ba50a10bdf443ef6b903eeaa36
Signed-off-by: mpoojary <mpoojary@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2678100
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Cast the return code of sprintf away when creating a scheduling domain
device name. While at it, use snprintf just in case.
CID 497476
CID 497479
Bug 3512545
Change-Id: I4e26c29b889de4b709d582ec3fdde28c50fca5b9
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2681274
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
a. LAUNCH_ERR
- Userspace error.
- Triggered due to faulty launch.
- Handle using recovery to reset CE engine and teardown the
faulty channel.
b. An INVALID_CONFIG -
- Triggered when LCE is mapped to floorswept PCE.
- On iGPU, we use the default PCE 2 LCE HW mapping.
The default mapping can be read from NV_CE_PCE2LCE_CONFIG
INIT value in CE refmanual.
- NvGPU driver configures the mapping on dGPUs (currently only on
Turing).
- So, this interrupt can only be triggered if there is
kernel or HW error
- Recovery ( which is killing the context + engine reset) will
not help resolve this error.
- Trigger Quiesce as part of handling.
c. A MTHD_BUFFER_FAULT -
- NvGPU driver allocates fault buffers for all TSGs or contexts,
maps them in BAR2 VA space and writes the VA into channel
instance block.
- Can be triggered only due to kernel bug
- Recovery will not help, need quiesce
d. FBUF_CRC_FAIL
- Triggered when the CRC entry read from the method fault buffer
does not match the computed CRC from the methods contained in
the buffer.
- This indicates memory corruption and is a fatal interrupt which
at least requires the LCE to be reset before operations can
start again, if not the entire GPU.
- Better to quiesce on memory corruption
CE Engine reset (via recovery) will not help.
e. FBUF_MAGIC_CHK_FAIL
- Triggered when the MAGIC_NUM entry read from the method fault
buf does not match NV_CE_MTHD_BUFFER_GLOBAL_HDR_MAGIC_NUM_VAL
- This indicates memory corruption and is a fatal interrupt
- Better to quiesce on memory corruption
f. STALLING_DEBUG
- Only triggered with SW write for debug purposes
- Debug interrupt, currently ignored
Move launch error handling from GP10b to GV11b HAL as -
1. LAUNCHERR_REPORT errcode METHOD_BUFFER_ACCESS_FAULT is not
defined on Pascal
2. We do not support GP10b on dev-main ToT
JIRA NVGPU-8102
Change-Id: Idc84119bc23b5e85f3479fe62cc8720e98b627a5
Signed-off-by: Tejal Kudav <tkudav@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2678893
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
g->fifo.runlists[] has size of g->fifo.max_runlists. During quiesce,
U32_MAX bitmask is passed to g->ops.runlist.write_state() HAL to
disable all the runlist. The Ga10b HAL implementation of
g->ops.runlist.write_state() references into runlists[] structure
for all the bits set in input runlist mask. For mask=U32_MAX,
there is NULL pointer dereference when runlist_id exceeds
g->fifo.max_runlists.
Add runlist_id boundary check before dereferencing the runlists[]
structure.
Update Gk20a HAL too with similar guard to make sure incorrect mask
doesn't get written to the register.
JIRA NVGPU-8102
Change-Id: Ic613aa38361b8b23d953c76d6924aba6bf6d5ea9
Signed-off-by: Tejal Kudav <tkudav@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2680847
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
GVS: Gerrit_Virtual_Submit
Add new HAL gops.gr.init.set_default_gfx_regs() to set graphics specific
PRI values for graphics contexts in function nvgpu_gr_obj_ctx_alloc().
Add new HAL gops.gr.init.capture_gfx_regs() to capture and save init
values for the PRIs. Add new struct nvgpu_gr_obj_ctx_gfx_regs to hold the
PRI init values.
Define HAL functions gv11b_gr_init_set_default_gfx_regs() and
gv11b_gr_init_capture_gfx_regs(). Set the HAL functions for
gv11b and ga10b.
Register accessors required to set PRIs are auto-generated.
Bug 3506078
Change-Id: I4c2843a274f3c924e402541e600e104ed0c9ed1c
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2671598
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Shashank Singh <shashsingh@nvidia.com>
Reviewed-by: Jonathan Mccaffrey <jmccaffrey@nvidia.com>
GVS: Gerrit_Virtual_Submit
- Below are warnings encountered when we use same function names in
multiple units
doxygenfunction: Unable to resolve multiple matches for function
“test_setup_env” with arguments () in doxygen xml output.
doxygenfunction: Unable to resolve multiple matches for function
“test_free_env” with arguments () in doxygen xml output.
- Fixing warnings by updating functions with unique names in multiple
units
JIRA NVGPU-7115
Change-Id: Iaa861040208e101c114f5c556096deb09d08b7fe
Signed-off-by: srajum <srajum@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2601798
(cherry picked from commit f57e408ba2fae4ff9b7c54a441e5cc3e75b0c87c)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2678347
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Shashank Singh <shashsingh@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
GVS: Gerrit_Virtual_Submit
- MISRA Directive 4.7
Calling function "nvgpu_tsg_unbind_channel(tsg, ch, true)" which returns
error information without testing the error information.
- MISRA Rule 10.3
Implicit conversion from essential type "unsigned 64-bit int" to different
or narrower essential type "unsigned 32-bit int"
- MISRA Rule 5.7
A tag name shall be a unique identifier
JIRA NVGPU-5955
Change-Id: I109e0c01848c76a0947848e91cc6bb17d4cf7d24
Signed-off-by: srajum <srajum@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2572776
(cherry picked from commit 073daafe8a11e86806be966711271be51d99c18e)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2678681
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
This patch does the following:
- Defines a new flag "CONFIG_NVGPU_ENABLE_MISC_EC" when the
build is not targeted for RM Server. Since iGPU is in pass-through
mode in both safety and standard build, EPL libraries will not be
included in RM Server. This is done with the help of the flag
"NV_BUILD_CONFIGURATION_IS_VM_SERVER".
- Updates error id that will be reported to Safety_Services. The
format of the error ID is:
- HW_unit_id: (4-bits: bit 0 to 3),
- Error_id: (5-bits: bit 4 to 8),
- Corrected/Uncorrected error: (1-bit: bit-9),
- Remaining 22-bits are unused.
- Defines macros that will be used to form error ID.
- Defines a macro for SW_ERR_CODE_0 register which is allocated for
NvGPU to report errors to Safety_Services via MISC_EC.
JIRA NVGPU-8094
Change-Id: I02f37db75ef3b82952ef5f196f4e065d6c5d1a3e
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2677373
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
This reverts commit a372ec9a38.
Earlier golden context image verification was failing on orin safety due
to mismatch. But on tot there is no mismatch obeserved (possibly due to
update of NET image from A to D). So, now golden context image
verification can be re-enabled for orin safety.
Bug 3482988
Change-Id: I2bda9be921987e6b6a3b933b3ff45b26cf3025ca
Signed-off-by: Shashank Singh <shashsingh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2678153
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
This patch does the following:
- Adds error IDs for GSP ACR and GSP SCHED.
- Updates error IDs for PMU.
- Removes reporting of DMEM ECC_CORRECTED since DMEM RAMs in PWR is
protected only with parity mechanism, (ref: T23x_UPROC_Safety_IAS)
- Removes reporting of IMEM ECC_CORRECTED since IMEM RAMs for PROC in
PWR is protected only with parity mechanism, (ref: T23x_UPROC_Safety_IAS)
JIRA NVGPU-8094
Change-Id: I127e78b1aa76b552758d1fff5bc7a01b5f8f3e54
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2677589
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
User space may erroneously provide extra flags for buffer
mappings which always get silently ignored without any
error. It would be better to return error to catch cases
of ABI mismatches.
This patch checks if any extra flags were provided and
returns error if this is the case.
Jira NVGPU-6640
Bug 3489827
Change-Id: Ib226f049f81bef48bf00656259ed97ba0a3eb47c
Signed-off-by: Martin Radev <mradev@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2676684
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
TEGRA_RAW mappings are only supposed to be used in special
circumstances by user space software, and the request cannot
be treated as a hint and be silently ignored if the feature
is not supported.
This patch updates the logic to return error if the feature is
not supported.
Jira NVGPU-6640
Bug 3489827
Change-Id: Ia2ce71df0202ab0c8676b815cf887cc7300aa07f
Signed-off-by: Martin Radev <mradev@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2676168
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Allow SC7 suspend/resume for platforms even if runtime pm
is disabled.
Currently, nvgpu can disable runtime pm by setting railgate_init
field to false for platform_{gk20a/gv11b/ga10b) files. This is done by
taking extra reference count in the PM Framework.
However, device suspend would still fail. Fix this by checking
for NVGPU_CAN_RAILGATE and removing the additional reference count
taken as mentioned above. Take the extra refcount back at
the end of the resume path.
Bug 3458643
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Change-Id: I413e09e2f9f380d78c0ce30196591e9c5b7544f3
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2668567
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
GVS: Gerrit_Virtual_Submit
Most SM RAMs are protected with parity (except L1 D-cache TAG mem
which is protected with SEC-DED ECC). The memory corruption errors
reported by these RAMs are therefore uncorrected errors only.
Remove the code to handle corrected errors from GR SM ECC.
The SM RAMS ECC errors currently report error to SDL using ID
GPU_SM_L1_TAG_ECC_(UN)CORRECTED. Update the error reporting to
use the newly created error IDs for Drive 6.0.
JIRA NVGPU-7987
Change-Id: Ic426d45f851d87aafaa7963b937535582cdafadf
Signed-off-by: Tejal Kudav <tkudav@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2674389
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
The current value of CTXSW_TIMEOUT (100ms) is too large and does not
meet the FTTI budget of 100ms. Update the value to 10 ms -
1. It seems well within FTTI - with some budget for recovery if
needed. The WCET for recovery is around 55ms.
2. It can be easily updated if needed later
Change-Id: If2ea3664c92d7426d1543d15614723e38b63aabd
Signed-off-by: Tejal Kudav <tkudav@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2672872
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
GVS: Gerrit_Virtual_Submit
Below CE interrupts do not have any users(usecases) on safety build;
disable them only on safety build.
1. BLOCKPIPE stall intr: Not used by GFX(VKSC) and CUDA on safety.
2. NONBLOCK_PIPE nonstall intr: Non-stall intrs are not supported
on safety build. Also, this one is not used by GFX(VKSC)
and CUDA.
3. STALLING_DEBUG intr: Added in Orin tree. It is only needed for
debugging. Disable on safety build as there is no current
usage in driver.
4. POISON_ERROR intr: Poison is a fault containment and not
supported on GA10b.
5. INVALID_CONFIG intr: Floor sweeping not supported on functional
safety SKU.
Bug 3548082
Change-Id: I8d97ccb38f138b2c04a780e1c255a64d28723405
Signed-off-by: Tejal Kudav <tkudav@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2671927
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
The PMA unit can only access GPU VAs within a 4GB window, hence both
the user allocated PMA buffer and the kernel allocated bytes available
buffer should lie in the same 4GB window. This is accomplished by
carving out and reserving a 4GB VA space in perbuf.vm and using fixed
GPU VAs to ensure that both buffers are bound within the same 4GB window.
In addition, update ALLOC_PMA_STREAM to use pma_buffer_offset,
pma_buffer_map_size fields correctly.
Bug 3503708
Change-Id: Ic5297a22c2db42b18ff5e676d565d3be3c1cd780
Signed-off-by: Antony Clince Alex <aalex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2671637
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@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
Falcon safety debug flag was previously disabled for safety debug
profile. This patch enables the flag support for safety debug.
copy_from_dmem function is required to copy the debug info from
dmem debug buffer whenever there's an error generated.
Hence, moved copy_from_dmem function to fusa file from non-fusa
and added ifdef condition to only enable when non-fusa or falcon debug
flag is set.
Also, some fixes for type conversion error in falcon_debug.c during
compilation.
Bug 3482988
Change-Id: Ic0ea32b3227b84d4ba0835e6e1aeb40f58ec7327
Signed-off-by: mpoojary <mpoojary@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2673900
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Allowlist and register ranges are declared global. Sparse throws warning
for them as:
- allowlist_ga100.c:351:5: warning: symbol
'ga100_cau_register_offset_allowlist' was not declared.
Should it be static?
- allowlist_ga100.c:389:47: warning: symbol
'ga100_hwpm_pma_trigger_register_ranges' was not declared.
Should it be static?
Make these arrays static global const.
Bug 3528472
Change-Id: I319f36c1579c630632b994295677c5831c1bff6b
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2676591
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
