Fix issues related with wrong storage type for
64 bit variables.
(1) Fixed width of HZ_TO_MHZ constant
(2) changed fence_wait timeout to store unsigned
long
bug 200299572
Change-Id: Ie8f2386b738f3aafce75fc2440947e36befac273
Signed-off-by: David Nieto <dmartineznie@nvidia.com>
Reviewed-on: http://git-master/r/1471611
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
We access ROM control register with xve_writel, but we also
add the base register address. This leads to adding the base
address twice, and the access goes to wrong register.
Bug 1846641
Change-Id: I46ef277aac661a08049935b08505120cad1a5e76
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1471505
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Write an error to UART on priv ring error. This uncovers any
accesses to missing registers or illegal accesses to registers.
Bug 1846641
Change-Id: Ic1e5ecadcd95777f2b3f7bd77accf98ddce97282
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1294683
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Use an nvgpu_mem created from a page in the ACR code instead of
a scatter gather table created by the Linux SGT APIs. The ACR code
needs to have the GPU map a physical page queried from an external
API (other than the regular DMA API).
Note that this code used to explicitly mark the SGT it makes as
bypassing the SMMU. However, that is now taken care of implicitly
by the __nvgpu_mem_create_from_pages() function.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: Ie40152a7611e985e1b97ac2ddc7e27664b71917c
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464082
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The gfp_t argument for dma_alloc_attrs() cannot take DMA_MEMORY_NOMAP -
it's interpreted as __GFP_RECLAIMABLE which has the same integer value.
Use GFP_KERNEL for the flag argument and set DMA_ATTR_NO_KERNEL_MAPPING
for dma attrs which the code was trying to do with the flag that is
meant for the coherent allocation API.
Bug 200299572
Change-Id: Ie4d988fbeeb954f6f7ccd4f9fb438968d76f0c6c
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1468315
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
nvgpu is being modularized for kernel-4.4 and newer kernels.
__alloc_fd() can't be used on modules as it is not exported.
So, use a wrapper tegra_fd_alloc() instead of __alloc_fd()
bug 200290850
Change-Id: I0e63307d429774440753698643ba7620c1e1f9f6
Signed-off-by: Sri Krishna chowdary <schowdary@nvidia.com>
Reviewed-on: http://git-master/r/1469313
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Use struct nvgpu_mem for DMA allocations (and the corresponding
nvgpu_dma_alloc_sys()) instead of custom rolled code. This migrates
away from using linux scatter gather tables directly. Instead this
is hidden in the nvgpu_mem struct. With this change the semaphore.c
code no longer has any direct Linux dependencies.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: I92167c98aac9b413ae87496744dcee051cd60207
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464081
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Add a function to create a nvgpu_mem from a list of arbitrary pages.
This API is useful for pages not necessarily allocated by the Linux
page allocator. It is useful for making nvgpu_mems that represent
carveouts or other things like that.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: Ibcb6432f077a6b0ecf9d183248e47a1f9ecb3ddd
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464080
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Replace a couple of boolean fields in nvgpu_mem with bits in a
bitmap introduced in an earlier patch.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: Iffc31bd629cab9a37e5a4fd13377eb9090353410
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464079
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Move and rename the functions that build sgtables for nvgpu_mems into
the Linux specific DMA code.
One place outside of the Linux code do include the Linux DMA header.
That will be fixed in a subsequent patch.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: Ie43c752b8f998f122af70f7c7eb727af0b0d98df
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464078
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add an API for creating a special sub-nvgpu_mem struct. This struct
comes with some fairly important caveats but is very useful for the
semaphore code.
Also, make sure that in nvgpu_mem_begin() and nvgpu_mem_end() no
additional mappings are made if not necessary. More importantly during
nvgpu_mem_end() it would be possible to vunmap() a CPU mapping of a
DMA allocation that does not expect this to happen.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: I579429da9ff7288488753a113bafc558e0f17a0f
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464077
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Get function declaration from fb_gm20b.h. This solves
Sparse warnings due to the missing of function prototype.
bug 200299572
Change-Id: Ie692f88e5bfc771bfcbb713adda3776521788f8c
Signed-off-by: Bo Yan <byan@nvidia.com>
Reviewed-on: http://git-master/r/1468858
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Put all debugfs dependencies inside #ifdef CONFIG_DEBUG_FS. This
includes some functions in allocators that were used only for
debugging.
Remove include of linux/debugfs.h on files that do not deal with
debugfs.
linux/debugfs.h implicitly included linux/fs.h, which we relied on.
Add explicit include of linux/fs.h for all files where this is the
case.
Change-Id: I16feffae6b0e3a2edf366075cdc01ade86be06f9
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1467897
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Hide the Linux specific nvgpu_mem fields so that in subsequent patches
core code can instead of using struct sg_table it can use mem_desc.
Routines for accessing system specific fields will be added as needed.
This is the first step in a fairly major overhaul of the GMMU mapping
routines. There are numerous issues with the current design (or lack
there of): massively coupled code, system dependencies, disorganization,
etc.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: I2e7d3ae3a07468cfc17c1c642d28ed1b0952474d
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464076
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
An extra semicolon was left in the nvgpu_log() macro. Remove this
semicolon.
Change-Id: I36d15d165b2f6ecf7e5f6c5cc477254fc9fc9252
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464075
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Move the Linux specific header files to include/nvgpu/linux/ to keep
the header files more organized. Having tons of Linux specific stuff
pollute the top level include seems like a bad idea.
Change-Id: I4913429c79d49f91c521021e5c0a613db453006a
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464074
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Separated parameters for GM20B GPC PLL revisions B1 (default), and C1
(new, not characterized, yet). For now, used C1 spec settings.
Skipped PL divider glitch WAR, since revision C1 is glitchless.
Bug 1851797
Bug 1867980
Change-Id: I12dc101d002a41230e9323c47af234f6bae8e034
Signed-off-by: Alex Frid <afrid@nvidia.com>
Reviewed-on: http://git-master/r/1461680
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
FECS trace used hrtimer for sleeping between frames. The resolution
does not need to be that high, and hrtimers are not portable, so
replace the delay with nvpgu_usleep_range().
Bug 200290850
Change-Id: I10fc56fcf6f6d407413384d557ae4bfdfa4239ef
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1465761
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
GVS: Gerrit_Virtual_Submit
perfbuf mapping fixes:
- Allocate VM specifically for perfbuf use: using the PMU's
results in mmu faults for larger buffers where 64k pages are used.
- Make 4GB boundary check work for large address spaces
- remove requirement to have allow_all flag set
- track perfbuf ownership and clean up appropriately
Bug 1880196
JIRA EVLR-1074
Change-Id: Ieee4eb17b64acf9b6ede37bf8e6a91892cda4a7e
Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com>
Reviewed-on: http://git-master/r/1460809
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Copy is_fmodel to struct gk20a at probe time, and access it from
gk20a instead of platform_gk20a.
JIRA NVGPU-16
Change-Id: Ib8d793ea2b02b62da3bfdbb6372d9927658b7ec6
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1463540
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
We have this recursion in headers as pointed
by Coverity
gk20a/gk20a.h" includes itself: gk20a.h -> clk.h -> clk_vin.h ->
boardobjgrp.h -> gk20a.h
Fix this by removing gk20a.h header from boardobjgrp.h
and declaring necessary structures
Also remove unused header pmu_gk20a.h from boardobjgrp.h
Coverity id : 2511592
Bug 200291879
Change-Id: I36b0bcc752d7c5d92ebe80bfc8f39d119526ae66
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1464637
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Add nvgpu_firmware data structure, and return it instead of Linux
struct firmare from nvgpu_request_firmware. Also add abstraction
for releasing firmware: nvgpu_release_firmware.
JIRA NVGPU-16
Change-Id: I6dae8262957c0d4506f710289e3a43a6c1729fc7
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1463538
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Remove linux list includes <linux/list.h>
and include <nvgpu/list.h> since we now use
nvgpu list APIs instead of linux APIs
Jira NVGPU-13
Change-Id: I59bd433a9bc5c15d4c40e6fe4b18cf44246ba3b2
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1462080
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Add new API nvgpu_list_last_entry() to get last
entry in the list
Jira NVGPU-13
Change-Id: I96d80996d4b82fdb4de4ea25a75d58fefaacd541
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1462078
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
The following ioctls:
- NVGPU_GPU_IOCTL_WAIT_FOR_PAUSE
- NVGPU_GPU_IOCTL_RESUME_FROM_PAUSE
- NVGPU_GPU_IOCTL_TRIGGER_SUSPEND
- NVGPU_GPU_IOCTL_CLEAR_SM_ERRORS
access hardware registers, so they should make sure that the gpu is
powered on first. Add gk20a_{busy,idle}() pairs for them.
Bug 1849661
Change-Id: I7deabf4a2c1c7d069a6134233f8e86df0a2722c8
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1461449
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
It is currently possible to set GPCCLK lower than the
minimum allowed frequency.
Clip target GPCCLK/MCLK according to valid min/max range
in arbiter. We could do this before submitting request to
arbiter, but then we would loose information on the
requested target frequency. Instead, we cache the clock
range in arbiter context, and check target frequency when
running arbiter.
Bug 200288036
Change-Id: I29f5176e6365a926d1041430c05a63f0c8447e2b
Reviewed-on: http://git-master/r/1460834
(cherry picked from commit eb626903e4fc046fe1f0eaee703c857e9a0f2b4d)
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1461715
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move Linux driver specific code to common/linux/module.c. This includes
module initialization, power management, and interrupt functions.
Move pci.c, pci.h and nvgpu_common.c under common/linux as they're
Linux only files.
JIRA NVGPU-16
Change-Id: If92b4dd78ebc0c2bbfa9fbca258e9552e4979b4b
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1456162
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Monitoring INA3221 is currently missing on some customer
designs. If power device is missing, skip programming
of power device and power policy in PMU. Also clear
related features in GPU characterictics (get voltage,
current and power features are not supported).
Bug 1903258
Change-Id: Ibd9d89e5833afb350c52b76527c54923055aee5e
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1459919
(cherry picked from commit e7e21c451843bddb8e8ac9e3db3b1d4b4a4c272c)
Reviewed-on: http://git-master/r/1463105
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Monitoring INA3221 is currently missing on some customer designs.
Use PCI vendor/device ids and VBIOS version to identify such device,
and set power_sensor_missing flag. This flag is subsequently used
to skip power device and power policy programming in PMU.
Bug 1903258
Change-Id: Ia6b5ce563b5cc3593e138635639025b830e4535e
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1460936
(cherry picked from commit 42f236c210931d28d2699858bb0663b91cf11046)
Reviewed-on: http://git-master/r/1463104
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
