Added support for GM20b GPCPLL frequency change in NA mode outside of
bypass. In this case the respective PLL DVFS detection settings are
updated in flight. The implemented algorithm relies on characterization
providing two frequency limits at the same voltage: max frequency on
the F/V curve (Fmax@V) in NA mode with characterized DVFS coefficient,
and safe frequency under the curve when DVFS coefficient is zero
(Fsafe@V, which is effectively the same as Fmax@V in legacy/non-DVFS
mode).
Transition between two Fmax@V points on the curve includes:
- Lowering frequency to Fsafe@V for the minimum V of the transition
end-points
- Setting DVFS coefficient to zero
- Changing DVFS calibration point to the new voltage
- Setting DVFS coefficient characterized for the new voltage
- Setting final target frequency
Note that voltage is changed by Tegra SoC DVFS before (when voltage
increases), or after (whet voltage decreases) the above procedure.
This commit kept NA mode disabled.
Bug 1555318
Change-Id: Ib5620aaa113dc1caa69ecd402d9c6f68e39c472c
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/501042
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Yu-Huan Hsu <yhsu@nvidia.com>
Added support for GM20b GPCPLL dynamic ramp in NA mode that requires
ramping of both integer NDIV and fractional SDM_DIN controls. If NA
mode is enabled, dynamic ramp is used only for transition to / from
disabled state. PLL frequency in NA mode is still changed under bypass
only.
This commit kept NA mode disabled.
Bug 1555318
Change-Id: I91f5722a485d1b66b6113aa9c35a2fe36c38ea80
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/500637
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Yu-Huan Hsu <yhsu@nvidia.com>
Added basic support for GM20b GPCPLL noise-aware(NA) mode. In this
mode PLL internal DVFS mechanism is engaged, and output frequency is
scaled with voltage automatically. The scaling coefficients in this
commit are preliminary, pending characterization.
If NA mode is enabled, any frequency change is done under PLL bypass,
with no dynamic ramp allowed.
This commit kept NA mode disabled.
Bug 1555318
Change-Id: I8d96a10006155635797331bae522fb048d3dc4a0
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/499488
GVS: Gerrit_Virtual_Submit
Reviewed-by: Yu-Huan Hsu <yhsu@nvidia.com>
Retrieve channel count from gm20b specific header instead of the
gk20a header. This increases channel count from 128 to 512.
Change-Id: I96d4887432852795f7f526e33f0d3d2458f3af0e
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/500623
- add and export API "gk20a_fifo_recover_tsg()" to
recover a TSG
- if TSG is running on any engine, then trigger MMU fault
on those engines
- otherwise, abort each channel in TSG
- modify channel specific API engines_on_ch() to generic
engines_on_id() which will take an ID and a flag to specify
whether ID is for channel or TSG and return engines running
on that ID
- modify channel specific API get_faulty_channel() to generic
get_faulty_id_type() which will take pointers to ID and type
of ID (either a regular channel or TSG)
- remove runlist update from recover_ch() since
no need to touch runlist during recovery
- set error notifier first and then only abort the channels
for TSG recovery path
- also, add necessary accessors to get engine
status type as TSG
Bug 1470692
Change-Id: I7137f611f80916b3d256d4b0dc6e5cf1e93eef6f
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/497873
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Restored changing GM20B GPCPLL post-divider in flight with the
following limitation: post divider transition is glitch-less only if
there is common "1" in binary representation of old and new settings.
Transitions that may create glitch are implemented in glitch-less steps
with minimum possible interim divider value (for example, 1 <=> 2
transition has interim value 3: 1 <=> 3 <=> 2).
Steps allowed for glitch-less transitions may not always have frequency
jump at/below VCO min/2 (in the example above 1st step jumps 2/3 of
VCOmin). Enabled external linear divider at 1:2 during such steps.
Used extra write of the same data when changing GM20b linear divider.
Bug 1552225
Change-Id: Ie8fba2fbe44afd34ca68f5f355bd302b7426a632
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/496319
(cherry picked from commit bdd21e0003032fe664bd20f163dbab9942fd1d1d)
Reviewed-on: http://git-master/r/499193
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Yu-Huan Hsu <yhsu@nvidia.com>
- Skipped PLL re-locking if only post-divider is changing under bypass
- Added 1us delay after switch to bypass clock source
- Changed wait for lock under bypass resolution from 2us to 1us
Change-Id: I259581c00c417752263ef3b2ea057200bb78ecbf
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/495473
(cherry picked from commit d90a19b8bf59c608a2a3a891b34ca714dfe990e9)
Reviewed-on: http://git-master/r/499192
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Yu-Huan Hsu <yhsu@nvidia.com>
Disable illegal compstat access interrupt. We access compstat backing
store to handle CDE swizzling.
Also change the magic number for evicted_cb to use a generated value.
Change-Id: I79b299abbffcb90497690ba4fc55d8517a3dbd87
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/496444
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Arto Merilainen <amerilainen@nvidia.com>
Reviewed-by: Jussi Rasanen <jrasanen@nvidia.com>
Tested-by: Jussi Rasanen <jrasanen@nvidia.com>
Add API gk20a_fifo_preempt_tsg() which takes ID of tsg
and preempts it
Bug 1514064
Bug 1470692
Change-Id: I1d52c1dd7a9aecc1314b0f223fe4eedecc033629
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/495583
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Alloc space writes the page size to a field that requires pgsz_idx.
That can cause corruption in internal kernel structures.
Clear_sparse treated a parameter as page size instead of index.
Bug 1549451
Change-Id: I73ce17b99aae6865056facce72d2ab9ca8b3f81d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/495692
Switch GM20b GPCPLL under bypass when changing post-divider setting
(for now, don't assume that post-divider is glitch-less).
Change-Id: I62b1285c035de0913207a86c41f37b7765da3893
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/495300
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Bo Yan <byan@nvidia.com>
Regenerate clock gating lists. Add new blocks, and takes them into
use. Also moves some clock gating settings to be applied at the
earliest possible moment right after reset.
Change-Id: I21888186c200f7a477c63bd3332e8ed578f63741
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/457698
Moved detection of idempotent GPCPLL operations from set_pll_freq()
function to its callers, e.g., explicitly check when enable operation
is called on already enabled PLL, instead of passing same frequency
to set_pll_freq() in such case. Similarly explicitly check when disable
operation is called on already disabled PLL.
Also moved check for GPU powered on from set_pll_freq() to callers,
and skip call to set interface if not.
Added last GPCPLL configuration structure updated after successful
completion of set_pll_freq() function.
Bug 1450787
Change-Id: I8c14b8cab2a8548e98c9b2d223c465c68fb87b61
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/488027
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Krishna Reddy <vdumpa@nvidia.com>
based on the config setting and fuse secure no non secure boot is done
Change-Id: I5937ba945c5a3a86f72e0f2a9078fcde01977137
Signed-off-by: Vijayakumar <vsubbu@nvidia.com>
Reviewed-on: http://git-master/r/487684
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Added direct read and write debugfs access to GM20b GPCPLL registers.
Change-Id: I203621906ee094991eecd5c18fd5b6c70b20a4c1
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/487314
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bo Yan <byan@nvidia.com>
Do not force GM20b GPCPLL post divider to 1:2 settings before switching
to bypass clock if PLL output frequency is increased as a result. Move
this step under bypass. However, this step is still needed in case when
PLL can be configured without switch to bypass.
Bug 1450787
Change-Id: Iab81b0e5a71f44f738a64e15b05df41fdbd61ebe
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/456505
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Tested-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: Krishna Reddy <vdumpa@nvidia.com>
The code took reference to gk20a by using gk20a_busy_noresume(). This
function takes pm runtime reference only to the GPU, however, the code
dropped the reference by calling gk20a_idle() which also drops the
reference to the platform dependencies (host1x).
This patch modifies gm20b_mm_mmu_vpr_info_fetch_wait() to drop only
the GPU reference.
Change-Id: Ied59381fa302452356768ed59e8ad9af18284e3d
Signed-off-by: Arto Merilainen <amerilainen@nvidia.com>
Reviewed-on: http://git-master/r/482721
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
The nvgpu driver now supports using the Tegra graphics virtualization
interfaces to support gk20a in a virtualized environment.
Bug 1509608
Change-Id: I6ede15ee7bf0b0ad8a13e8eb5f557c3516ead676
Signed-off-by: Aingara Paramakuru <aparamakuru@nvidia.com>
Reviewed-on: http://git-master/r/440122
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Current implementation is based on config GK20A_PHYS_PAGE_TABLES
to have APIs to create/free/map/unmap phys pages
Remove this config based implementaion and move the APIs so that
they are called at runtime based on tegra_platform_is_linsim()
In generic APIs, we first check if platform is linsim and if it
is then we forward the call to phys page specific APIs
Change-Id: I23eb6fa6a46b804441f18fc37e2390d938d62515
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/488843
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Disabled GPCPLL SYNC mode after GM20b is switched to bypass clock when
powering down GPU.
Bug 1450787
Change-Id: Ifaec2c562e51c0ae1328b7505faafd19607a77f2
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/456504
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bo Yan <byan@nvidia.com>
Gk20a unmaps the addresses binding to dummy page to clear sparse.
On Gm20b, we need to free the allocated page table entry for sparse
memory.
Bug 1538384
Change-Id: Ie2409ab016c29f42c5f7d97dd7287b093b47f9df
Signed-off-by: Kevin Huang <kevinh@nvidia.com>
Reviewed-on: http://git-master/r/448645
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Calculation for comptag backing store must use number of LTCs instead
of number of FBPs.
Change-Id: If0aa636e09a3d24459987e626fe53bb7c96f1b15
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/453809
Reviewed-by: Shridhar Rasal <srasal@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
"isr_enable_lock" was used to protect pmu's isr_enabled flag
and pmu enable/disable calls
Instead of this extra lock, we can reuse "isr_mutex" for this
purpose
Bug 200014542
Bug 200014887
Change-Id: Ifbb7d6108effc132266a20517820e470d52a7110
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/453348
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
1. Register tegra-throttle cooling device as a
platform driver.
2. Obtain all the platform data (throtlle table
info) for all instances of blanced-throtlled cdev
from device tree and register them.
Change-Id: Ie92685eea3eb5cb18068b195adc9ab5f83762399
Signed-off-by: Arun Kumar Swain <arswain@nvidia.com>
Reviewed-on: http://git-master/r/449104
Reviewed-by: Diwakar Tundlam <dtundlam@nvidia.com>
Tested-by: Diwakar Tundlam <dtundlam@nvidia.com>
Updated GM20B GPCPLL programming sequence to utilize new glitch-less
post divider:
- No longer bypass PLL for re-locking if it is already enabled, and
post divider as well as feedback divider are changing (input divider
change is still under bypass only).
- Use post divider instead of external linear divider to introduce
(VCO min/2) intermediated step when changing PLL frequency.
Bug 1450787
Signed-off-by: Alex Frid <afrid@nvidia.com>
Change-Id: I4fe60f8eb0d8e59002b641a6bfb29a53467dc8ce
