In gp10b_round_clk_rate(), we right now return next
higher freq value than requested if requested value
matches a value in the table
Fix this by adding a right comparison
Bug 200194487
Change-Id: Ia99abfe4b247701d5ee1cda26b3ffcc18efba353
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1284302
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Remove EMC floor when GPU frequency is Fmin. At Fmin,
we most likely require a very low memory bandwidth.
At Fmin on load, actmon should sufficiently scale EMC
and hence not bottlenecking GPU.
Bug 1850297
Change-Id: I98b9dae648ea28910d534a9286ce2e9e91ea5fec
Signed-off-by: Cyril Raju <craju@nvidia.com>
Reviewed-on: http://git-master/r/1284572
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Move the gp10b HW headers to a new directory specially for them:
include/nvgpu/hw/gp10b
And change the code to include like so:
#include <nvgpu/hw/gp10b/hw_fb_gp10b.h>
This is part of the process to restructure the nvgpu driver.
Bug 1799159
Change-Id: Ic80ea5b7f5c280839e502e2178a345181f7a7ef9
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1280326
GVS: Gerrit_Virtual_Submit
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>
Fix small problems related to signed versus unsigned comparisons
throughout the driver. Bump up the warning level to prevent
such problems from occuring in future.
Change-Id: Ib7026728ef0e8c3c9e68956fc9794ec3a786a8a2
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1252069
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GPU frequencies can be set by powerhal when GPU is railgated,
and before this change that would cause EMC floors to remain
set until GPU is unrailgated.
After this change, EMC floors are not requested by the GPU
client when the GPU is railgated. It is ok to ignore the
requests, as the GPU client maxes the floor when powering
up.
Bug 1807560
Change-Id: I9a0d58b0288edbd03b2edf09580ecabd9b74f0c2
Signed-off-by: Juha Lainema <jlainema@nvidia.com>
Reviewed-on: http://git-master/r/1216233
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Ilan Aelion <iaelion@nvidia.com>
Reviewed-by: Cyril Raju <craju@nvidia.com>
Tested-by: Cyril Raju <craju@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Added control flag for nvgpu infra to allow kernel to create privileged
CE channels for page migration and clearing support between sysmem
and videmem.
JIRA DNVGPU-53
Change-Id: I1fc35eea60af3d1ea9a0b5582011f20d58958ccb
Signed-off-by: Lakshmanan M <lm@nvidia.com>
Reviewed-on: http://git-master/r/1173091
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
NvGPU is moving to use runtime PM only for its power
management
Remove pm_domain calls to register to nvhost
Jira DNVGPU-57
Change-Id: Idd01b680af0e8fd601801150fc663afa53b7ce6f
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1163217
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
This is done to mask a race issue seen where power refcount
is zero during ISR or bottom half.
Bug 200198908
Change-Id: I0a8ed774cd4fda9db65429b5aad03c5e001ff666
Signed-off-by: Bharat Nihalani <bnihalani@nvidia.com>
Reviewed-on: http://git-master/r/1162314
Reviewed-by: Juha Tukkinen <jtukkinen@nvidia.com>
Removed platform data parameter clockgate_delay, since it is not
really used for gpu clock gating any more.
Change-Id: I4c7148c70699cb5ed24f0b034ddc92bfb4b41887
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/1159594
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Set emc floor frequency as zero during rail-gate and set max emc
frequency as floor frequency during rail-ungate.
Bug 1770241
Change-Id: Ib6b6ea6c8b04518423126c3ca3600b4afac15180
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/1152848
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
For gp10b, set platform data for soc memory aperture type
as vidmem.
Bug 1749338
Change-Id: I7961734d3ebcca4af459c7c7d49bc31f0fc8ce5d
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/1129168
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Ken Adams <kadams@nvidia.com>
Use struct device instead of struct platform_device wherever
possible. This allows adding other bus types later.
Change-Id: I90623c020919ca8e2e5b31d53914c324d2dc6af9
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1120464
The ECC sysfs stat creation function is called on GR init. GR can get
initialized multiple times but we only need to create the ECC stats
once. Therefore, add a check to avoid creating duplicate stat sysfs
nodes.
Change-Id: Ifb338e57643f2f15492df137d2a7521e0c990cf2
Signed-off-by: Adeel Raza <araza@nvidia.com>
Reviewed-on: http://git-master/r/1021660
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Fixed the following sparse warning by making local symbol static:
- platform_gp10b_tegra.c:365: warning: symbol 'ecc_hash_table' was not declared.
Should it be static?
Bug 200088648
Change-Id: Iea1a682c3ee0609730366d44fab91849cd59c9ad
Signed-off-by: Amit Sharma <amisharma@nvidia.com>
Reviewed-on: http://git-master/r/1022410
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Tested-by: Sachin Nikam <snikam@nvidia.com>
This reverts commit 71b59d75fc49e2159830026bce387ef4d829faa8
since it causes suspend_sanity to fail on quill platform.
On system resume, we see the following error dump from GPU
gk20a 17000000.gp10b: gk20a_channel_timeout_handler: Job on channel 501 timed out
gk20a 17000000.gp10b: gk20a_fifo_set_ctx_mmu_error_ch: channel 501 generated a mmu fault
gk20a 17000000.gp10b: gk20a_set_error_notifier: error notifier set to 31 for ch 501
gk20a 17000000.gp10b: gk20a_channel_timeout_handler: Job on channel 509 timed out
Change-Id: I61bc3b0745fe136675ab79b13f54e9126602f51c
Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Reviewed-on: http://git-master/r/1017967
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
The powergating APIs only work if the BPMP is running. Skip
these calls if it's not available, instead of relying on
is_linsim, which doesn't work under all environments.
Change-Id: I34325847b2ebf33c5db2f31111c57d22ed28ef53
Signed-off-by: Matt Craighead <mcraighead@nvidia.com>
Reviewed-on: http://git-master/r/812415
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Implement platform specific reset_assert() and reset_deassert()
calls for gp10b
These APIs will in turn will use reset_control APIs to do
their work
Also, set force_reset_in_do_idle = true for gp10b, since
railgating is not supported yet
Bug 200137963
Change-Id: I2c0fe1273d3ecfd0c46704a44374712052ff51d6
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/797150
(cherry picked from commit 6ac04ca84cee8a4d3b089678c81534799880712d)
Reviewed-on: http://git-master/r/808240
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
VPR allocator needs to be used when allocating graphics context for
VPR channels. Define it for gp10b.
Bug 1625090
Change-Id: Ie2e3a865c310c34c629627891ac0b579f299983f
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/737846
Reviewed-by: Automatic_Commit_Validation_User
