- Added struct pmu_pg_stats_data to extract
data from multiple version of pmu pg statistics
- Added pmu_pg_stats_v2 interface to fetch
PG statistics data from PMU
- Added MSCG debugfs node to read mscg
statistics from PMU.
- Added pmu_elpg_statistics HAL support for
gp106 PG statistics read.
- Made changes to gp104/gp106
pmu_elpg_statistics HAL to support
for struct pmu_pg_stats_data
JIRA DNVGPU-165
Change-Id: I2b9e89c0fae90deb45006c4478170b9a97b56603
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: http://git-master/r/1252798
(cherry picked from commit 3c073b15fd991db8d65b3171b02c161294be40cd)
Reviewed-on: http://git-master/r/1271615
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Kickoff latencies when GPFIFO is in vidmem grow significantly as
function of number of GPFIFO entries. Move GPFIFO to sysmem to
improve kickoff latency.
Bug 1848369
Change-Id: Ie95d10df26b4f1370f7250a8fbf0f7ef0211df32
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
(cherry-picked from commit 897cb579a759bbe8455ce368413979e91eb0d475)
Reviewed-on: http://git-master/r/1281564
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move the GPU allocators to common/mm/ since the allocators are common
code across all GPUs. Also rename the allocator code to move away from
gk20a_ prefixed structs and functions.
This caused one issue with the nvgpu_alloc() and nvgpu_free() functions.
There was a function for allocating either with kmalloc() or vmalloc()
depending on the size of the allocation. Those have now been renamed to
nvgpu_kalloc() and nvgpu_kfree().
Bug 1799159
Change-Id: Iddda92c013612bcb209847084ec85b8953002fa5
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1274400
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Take interrupts as one kind of event message, and make it
easier to add new kind of events.
JIRA VFND-3291
Bug 200257899
Change-Id: I83482293230c0aa10b05caf61e249a042bf6653c
Signed-off-by: Richard Zhao <rizhao@nvidia.com>
Reviewed-on: http://git-master/r/1278396
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Aingara Paramakuru <aparamakuru@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
In gp10b_round_clk_rate(), we currently call
clk_round_rate() to round the clock rate for us
But since the frequency table is prepared
using the frequency values supported in h/w,
we can round the rate locally using the table
Bug 1827281
Change-Id: I85d034326539590352badceb4164aa5d89ee8842
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1280630
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
We right now get min and max frequencies, and then
interpolate rest of the frequencies.
With this approach, we do not select exact
frequencies as supported by h/w
Fix this so that we query all supported frequencies
using clk_round_rate() and then select every N'th
frequency to keep number of frequencies under limit
Use GP10B_FREQ_SELECT_STEP (currently set to 8)
to configure frequency selection step
Raise GP10B_MAX_SUPPORTED_FREQS to 200 since h/w
supported frequencies could be in that range
Bug 1827281
Change-Id: Id8678d7a0280a249e4affbba084ff2e33b6694e6
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1280629
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
In gk20a_busy() we check if dev is NULL and return -ENODEV if so. But
before that we've already dereferenced dev by passing it to
get_gk20a(). Defer call to get_gk20a() until after the NULL check.
Bug 200192125
Change-Id: I943a9e96d13ff8cb4333fe20a941c8e95d159a66
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1280349
GVS: Gerrit_Virtual_Submit
We retrieve perf table from VBIOS only if respective HAL op is
implemented. Later in code we unconditionally dereference a pointer
which can lead to NULL pointer access.
Fix two new cases by early aborting creation of devinit tables if the
perf VBIOS getter is missing.
Bug 200192125
Change-Id: I30f20f1829305ecc1117c6301e26ff1b869967eb
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1280347
GVS: Gerrit_Virtual_Submit
ctrlvolt.h #includes ctrlperf.h, which recursively #includes
ctrlvolt.h without needing anything from it. Remove the #include
to improve built time.
Bug 200192125
Change-Id: I8c917533127e6d42aae6ad1401860c1f5d110d81
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1280345
GVS: Gerrit_Virtual_Submit
gk20a.h includes pmu_gk20a.h, which #includes gpmuifboardobj.h,
which recursively #includes gk20a.h without needing anything from it.
Remove the #include to improve built time.
gpmuifboardobj.h also includes pmu_gk20a.h, when it actually needs a
definition from pmu_common.h. Change the #include to prevent another
recursive #include.
Bug 200192125
Change-Id: Idb8c13bceeae1a60a4efa466c07e78cb15ffa1fb
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1280348
GVS: Gerrit_Virtual_Submit
During construct of some VFE/CLK boardobjs, some data is
filled after a boardobj allocation is done.
Free up boardobj memory if an error is encountered in the data
filling.
Coverity ID 490171
Coverity ID 490172
Bug 200192125
Change-Id: I20621f7f9f9e379b8dced4905cd417c2ffa905b0
Signed-off-by: Vijayakumar <vsubbu@nvidia.com>
Reviewed-on: http://git-master/r/1280700
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Internally we use GPC2CLK in the arbiter, but we should expose
GPCCLK on kernel API and in user space. Added GPCCLK on the ioctl
API. Arbiter uses GPC2CLK to make queries, then converts to GPCCLK.
Jira DNVGPU-210
Change-Id: Id0b8134d0505c1f9bfd655a08e902bdcd03ebd96
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1280316
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
In lsfm_discover_ucode_images(), we currently collect
pmu ucode details with API pmu_ucode_details()
But if this API fails, we still continue loading
other firmwares
In case loading of pmu firmware fails due to some reason,
we should actually bail out immediately with error
Also, remove unnecessary chekc (ucode_img.lsf_desc != NULL)
since pmu_ucode_details() ensures that this cannot be
NULL
Bug 200265373
Change-Id: I68e7f9575e2a07b473ceacc528a5d172b58d6fb6
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1276555
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Add a platform flag and logic to disable ASPM on boot if a
platform does not want ASPM enabled.
Bug 200256272
Change-Id: I378997290377ef8ffa21bf8e7f3f59fa134b3d4d
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1263271
(cherry picked from commit 01e4700b74f4b4c6f0b9ffb40747653f2b63ea3c)
Reviewed-on: http://git-master/r/1274477
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Handle possible asynchronous GPU driver shutdown more gracefully.
This occurs when the GPU disappears from the PCI bus, for example,
if it overheats or detects an over current event.
Also add a preprocessor check to make sure that the
gk20a_channel_cancel_pending_sema_waits()
is always defined. In some builds CONFIG_SYNC is disabled but the
gk20a_remove_support() code does not check for this.
Bug 1816516
Bug 1807277
Change-Id: I932e312291c5c6a6ac5e13525ce8ca56a1be3652
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1250028
(cherry picked from commit 337810f8c478238a38d8553c1492622d5fa9aafa)
Reviewed-on: http://git-master/r/1274476
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Move the code that waits for deferred interrupts to nvgpu_common.c and
make it global. Also rename that function to use nvgpu_ as the function
prefix.
Bug 1816516
Bug 1807277
Change-Id: I42c4982ea853af5489051534219bfe8b253c2784
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1250027
(cherry picked from commit cb6fb03e20b08e5c3606ae8a5a9c237bfdf9e7da)
Reviewed-on: http://git-master/r/1274475
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Fix miscellaneous issues seen during driver shutdown.
o Make sure pointers are valid before accessing them.
o Busy the GPU during channel timeout.
o Cancel delayed work on channels.
o Avoid access to channels that may have been freed.
Bug 1816516
Bug 1807277
Change-Id: I62df40373fdfb1c4a011364e8c435176a08a7a96
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1250026
(cherry picked from commit 64a95fc96c8ef7c5af9c53c4bb3402626e0d2f60)
Reviewed-on: http://git-master/r/1274474
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
In many helper functions like gk20a_readl() the code assumed that
the GPU is present and registers and available. However, during
GPU shutdown this may not be the case.
In theory the driver should not be accessing GPU registers during
GPU shutdown (since shutdown is triggered by GPU registers being
unavailable) but these changes handle any missed cases where this
may happen.
This goes for GPU device access as well. Many parts of the code
assume that if the struct gk20a is valid, the the GPU dev must be
there are well. This isn't always the case, it seems.
Bug 1816516
Bug 1807277
Change-Id: Icaf6fd56ab7860724e77bda0f5e8d48f0da15642
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1250024
(cherry picked from commit e8c9997b2d7cd424d798ecfce1307e6193c0cf32)
Reviewed-on: http://git-master/r/1274473
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Enable FECS ctxsw tracing for gp106. Ensure that FECS records
have been written to memory before accessing the ring. Update
read index only once all records have been processed.
Jira EVLR-424
Change-Id: I1a21f841fcce1588397408906d77e2c3bf4a8c01
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1258243
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
nvgpu_clk_arb_init_arbiter() allocates master structure arb, and
sub-structures which is assigns as members of arb. At failure,
there's a single error label, and it tries to free the members of
arb without checking that arb was allocated.
Change-Id: Ifac7552b05dad17036835eb3ee624eae351ccb7d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1279226
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
nvgpu_clk_arb_install_fd() leaks dev if it cannot create an fd,
and it leaks both dev and fd if it fails to create a file.
Change-Id: I65c03401dba44d88be543d3aec2fa3fcd2fb0dcc
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1279225
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
clk_prog.h and clk_vin.h refer to boardobjgrp_e32 and
boardobjgrp_e255.h. Add explicit #includes for their definition.
Change-Id: Ib651b071f3c17bbc30ddaea0ef9bc5437ec9f2f4
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1279224
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
We retrieve perf table from VBIOS only if respective HAL op is
implemented. Later in code we unconditionally dereference the pointer
which can lead to NULL pointer access.
Fix by early aborting creation of devinit tables if the perf VBIOS
getter is missing.
Change-Id: If48aa6dac724056dd1feb2ef520e343736d4db85
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1279223
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
We use GPU reference clock as a divider. Check before division that
reference clock is not zero.
Change-Id: Ie453a78b422b2e740daeb7c12ce5b06faa52ba76
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1275743
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
In gk20a_submit_prepare_syncs(), after we have allocated wait_cmd
we check the results. If we failed to allocate wait_cmd, we still
jump to the error label that tries to free wait_cmd.
Create an own error labal for allocation before wait_cmd, and use
that if we fail to allocate wait_cmd.
Similarly create an error label for incr_cmd, and use that only once
incr_cmd has actually been allocated.
Change-Id: I1f8bc1d947c524038f5f237358a5e6b0dc2e6ac3
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1275742
GVS: Gerrit_Virtual_Submit
When allocating job channel_gk20a_job structure we assign the result
of allocation to *job_out, but we check the result of allocation
in job_out.
Change the check to check for result in *job_out.
Change-Id: Ia170cfa2dd5730665434b4c223c5a2f9502c744d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1275741
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
We first load VBIOS firmware from file, and then attempt to allocate
space into which we copy the data. If allocation fails, we do not
release the firmware.
Add a release_firmware() in the error path.
Change-Id: Iaa995e93bf8d5a23e08c5e332f70b203ac2e09db
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1275740
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
When we allocate boardobj object, we assign the allocated address
to *ppboardobj. The check for result of allocation checks looks
at address in ppboardobj.
Fix the check so that it actually checks results of kzalloc().
Change-Id: I90fda36afd1627efcf2fdea46bcd7ad1a7e3bb51
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1275738
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
voltage_get() defines det_out as u32. That variable is used to
calculate a 64-bit result, so cast det_out to u64.
Change-Id: I054ec299a4c7961cb38dd5ac0cbf0ac173b09efb
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1267692
GVS: Gerrit_Virtual_Submit
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
