Originally the order for output was:
1. Dump platform deps (sync-points/host1x stuff)
2. Dump PBDMA status
3. Dump engine status
4. Dump channel status
The updated ordering is:
1. Dump channel status
2. Dump PBDMA status
3. Dump engine status
4. Dump platform deps (sync-points/host1x stuff)
The purpose of this is to put the useful information first and
relegate the less useful info to later in the dump. We naturally
scan downwards and treat stuff at the top as most important.
The end goal is to make the debug dump as useful in as little
time as possible. So instead of making an engineer dig through
a complex jumble of information to find the useful stuff the
hope is that the useful stuff is immediately available.
Change-Id: I9d2b755676b7e5dc2f8949f14dc36f3d337e2a3f
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1940514
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This function returns a boolean describing if a given log
mask is enabled for a given GPU. Previously this returned
and int but the bool type is far better suited for this.
Also implement this function in posix, as it may be useful
to have implemented there if any common code chooses to
use this function.
Change-Id: I7382e73df83282763df1bdbccbbb219c9f3e6f1b
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1938341
Reviewed-by: svc-misra-checker <svc-misra-checker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
MISRA rule 14.4 doesn't allow the usage of non-boolean variable as
boolean in the controlling expression of an if statement or an
iteration statement.
Fix violations where a non-boolean variable is used as a boolean in the
controlling expression of if and loop statements.
JIRA NVGPU-1022
Change-Id: I957f8ca1fa0eb00928c476960da1e6e420781c09
Signed-off-by: Amurthyreddy <amurthyreddy@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1941002
GVS: Gerrit_Virtual_Submit
Reviewed-by: Adeel Raza <araza@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add a fifo HAL for querying the doorbell token of a specific channel and
call it instead of doing the calculation directly. For Volta the token
is just the channel id plus the possible base number.
Bug 200145225
Change-Id: Ifbb150191575fdc72e413a14c799cab7e52d8c14
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1849639
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
MISRA Rule-15.6 requires that all if-else blocks and loop blocks
be enclosed in braces, including single statement blocks. Fix errors
due to single statement if-else and loop blocks without braces
by introducing the braces.
JIRA NVGPU-775
Change-Id: Ib70621d39735abae3fd2eb7ccf77f36125e2d7b7
Signed-off-by: Srirangan Madhavan <smadhavan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1928745
GVS: Gerrit_Virtual_Submit
Reviewed-by: Adeel Raza <araza@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
MISRA Rule-17.7 requires the return value of all functions to be used.
Fix is either to use the return value or change the function to return
void. This patch contains fix for all 17.7 violations instandard C functions
in OS/Posix interface.
JIRA NVGPU-1036
Change-Id: I2da417edc992f16de24cdff536c0538f1fde8b61
Signed-off-by: Nicolas Benech <nbenech@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1929901
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-misra-checker <svc-misra-checker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
MISRA Rule-17.7 requires the return value of all functions to be used.
Fix is either to use the return value or change the function to return
void. This patch contains fix for all 17.7 violations instandard C functions
in OS/Linux interface.
JIRA NVGPU-1036
Change-Id: I39b20f1d0e1a1da56d452f2c3d5ee049666cefe8
Signed-off-by: Nicolas Benech <nbenech@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1929900
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svc-misra-checker <svc-misra-checker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
MISRA rule 10.4 doesn't allow arithmetic conversions on operands of
different essential type category.
Fix violations where an arithmetic conversion is performed on boolean
and non-boolean types.
JIRA NVGPU-994
Change-Id: I2af9937678462b632bb6ec6178e10d02104fc3bc
Signed-off-by: Amurthyreddy <amurthyreddy@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1832337
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Import userd and gpfifo buffers from userspace if provided via
NVGPU_IOCTL_CHANNEL_ALLOC_GPFIFO_EX. Also supply the work submit token
(i.e., the hw channel id) to userspace.
To keep the buffers alive, store their dmabuf and attachment/sgt handles
in nvgpu_channel_linux. Our nvgpu_mem doesn't provide such data for
buffers that are mainly in kernel use. The buffers are freed via a new
API in the os_channel interface.
Fix a bug in gk20a_channel_free_usermode_buffers: also unmap the
usermode gpfifo buffer.
Bug 200145225
Change-Id: I8416af7085c91b044ac8ccd9faa38e2a6d0c3946
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1795821
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add a mmap callback on the control device node for mapping the usermode
register region to userspace. Each such mapping is removed when the GPU
railgates, and brought back again on unrailgate.
The mapping offset must be 0 and its size must be 4 KB.
Bug 200145225
Change-Id: Ie8d3758da745b958376292691d7d1d02a24e7815
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1795819
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Usermode submit needs to access the usermode region of registers from
userspace. Store the start address of register resource in struct
nvgpu_os_linux to be used in remap to userspace.
Bug 200145225
Change-Id: I3796b6bf67942af0cc16c86accb82a013032bfc8
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1811838
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This fixes a bug where the POSIX dma module was calling malloc for
dma allocation requests with the incorrect size.
Bug found as a result of JIRA NVGPU-1235
Change-Id: Ia2933dc90f12d84a7da91a50eb3eafd6eb0e129e
Signed-off-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1930799
GVS: Gerrit_Virtual_Submit
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
struct nvgpu_channel_sync is moved to a private header i.e.
channel_sync_priv.h present in common/sync/. All accesses to callback
functions inside the struct nvgpu_channel_sync in NVGPU driver is replaced by
the public channel_sync specific APIs.
Jira NVGPU-1093
Change-Id: I52d57b3d458993203a3ac6b160fb569effbe5a66
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1929783
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The nvlink needs to be de-initialized as part of system
shutdown or reboot. Add the shutdown callback of pci driver
and use it to trigger nvlink de-initialization.
Bug 200422323
Change-Id: Iec8193d9665bc77ddbf3680ea130dfa4c1b3b0ad
Signed-off-by: Nitin Kumbhar <nkumbhar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1928821
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Background:
In Hypervisor mode dGPU device is configured in pass through mode for
the Guest (QNX/Linux). GMMU programming is handled by the guest which
converts a mapped buffer's GVA into SGLes in IPA (Intermediate/Guest
Physical address) which is then translated into PA (Acutual Physical
address) and programs the GMMU PTEes with correct GVA to PA mapping.
Incase of the vgpu this work is delegated to the RM server which takes care
of the GMMU programming and IPA to PA conversion.
Problem:
The current GMMU mapping logic in the guest assumes that PA range is
continuous over a given IPA range. Hence, it doesn't account for holes being
present in the PA range. But this is not the case, a continous IPA range
can be mapped to dis-contiguous PA ranges. In this situation the mapping
logic sets up GMMU PTEes ignoring the holes in physical memory and
creates GVA => PA mapping which intrudes into the PA ranges which are
reserved. This results in memory being corrupted.
This change takes into account holes being present in a given PA range and
for a given IPA range it also identifies the discontiguous PA ranges and
sets up the PTE's appropriately.
Bug 200451447
Jira VQRM-5069
Change-Id: I354d984f6c44482e4576a173fce1e90ab52283ac
Signed-off-by: aalex <aalex@nvidia.com>
Signed-off-by: Antony Clince Alex <aalex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1850972
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Rename gk20a/dbg_gpu_gk20a.c to common/debugger.c and make it a
separate common unit
Also rename gk20a/dbg_gpu_gk20a.h to include/nvgpu/debugger.h
We had two different HALs for debugger - gops.debugger and
gops.dbg_session_ops
Combine them into one single HAL gops.debugger and remove
gops.dbg_session_ops
Rename all exported APIs from debugger.h to be in the form of
nvgpu_*()
Jira NVGPU-1013
Change-Id: I136dc7786e3b2065921eb03b99f16049212f3cd2
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1920075
Reviewed-by: Sachin Jadhav <sachinj@nvidia.com>
Tested-by: Sachin Jadhav <sachinj@nvidia.com>
For a long time now, the ALLOC_GPFIFO_EX channel IOCTL has done much
more than just gpfifo allocation, and its signature does not match
support that's needed soon. Add a new one called SETUP_BIND to hopefully
cover our future needs and deprecate ALLOC_GPFIFO_EX.
Change nvgpu internals to match this new naming as well.
Bug 200145225
Change-Id: I766f9283a064e140656f6004b2b766db70bd6cad
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1835186
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
When turning off CONFIG_DEBUG_FS, there are build errors:
drivers/gpu/nvgpu/os/linux/os_ops_gp106.o: In function `nvgpu_fecs_trace_init_debugfs':
os_ops_gp106.c:(.text+0x8): multiple definition of `nvgpu_fecs_trace_init_debugfs'
drivers/gpu/nvgpu/os/linux/os_ops_gp10b.o:os_ops_gp10b.c:(.text+0x0): first defined here
drivers/gpu/nvgpu/os/linux/os_ops_gv100.o: In function `gp106_therm_init_debugfs':
os_ops_gv100.c:(.text+0x0): multiple definition of `gp106_therm_init_debugfs'
drivers/gpu/nvgpu/os/linux/os_ops_gp106.o:os_ops_gp106.c:(.text+0x0): first defined here
drivers/gpu/nvgpu/os/linux/os_ops_tu104.o: In function `gv100_clk_init_debugfs':
os_ops_tu104.c:(.text+0x0): multiple definition of `gv100_clk_init_debugfs'
drivers/gpu/nvgpu/os/linux/os_ops_gv100.o:os_ops_gv100.c:(.text+0x10): first defined here
This is because those functions aren't marked as static.
So this patch just simply fixes the bug.
Bug 2284925
Change-Id: I1da39345c653dfb50c509adb0c822b4657646c56
Signed-off-by: Nicolin Chen <nicolinc@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1929355
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This adds a fault injection module to the POSIX build that can be added
to modules (for example k*alloc) then exercised in unit tests. The unit
tests can then verify that the driver units handle errors from the
modules.
JIRA NVGPU-1235
Change-Id: I9d9e443608e3d2026e165a62f7a7f011df96fd54
Signed-off-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1927455
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The chid member of the channel_gk20a struct was being used as a unsigned
value. By being declared as an int, it was causing MISRA 10.3 violations
for implicit assignment of different types.
JIRA NVGPU-647
Change-Id: I7477fad6f0c837cf7ede1dba803158b1dda717af
Signed-off-by: Philip Elcan <pelcan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1918470
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move the wait_cmd_buffer programming for channel_sync->wait_fd to
channel_sync.c. nvgpu_os_fence->program_waits
interface is now removed. channel_sync can directly retrieve
syncpt/semaphore from the interfaces of struct nvgpu_os_fence_syncpt
and struct nvgpu_os_fence_sema and use it for the wait programming.
Also, change int to u32 for some variables such as num_fences,
max_wait_size and wait_cmd_size.
Jira NVGPU-1093
Change-Id: I19c1b10d676caff49ce57861091f7f0ea65e7676
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1829719
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
channel_sync->wait_fd depends upon nvgpu_os_fence->program_waits which
invokes a channel_sync method and this leads to a circular dependency.
In order to resolve the above, constructed struct nvgpu_os_fence_sema and
struct nvgpu_os_fence_syncpt with interfaces that support conversion
between struct nvgpu_os_fence to above. Also, added the following
interfaces for retrieving syncpts and semaphore from the above structs
respectively.
void nvgpu_os_fence_sema_extract_nth_semaphore(...)
int nvgpu_os_fence_sema_get_num_semaphores(...)
void nvgpu_os_fence_syncpt_extract_nth_syncpt(...)
int nvgpu_os_fence_syncpt_get_num_syncpoints(...)
These enable channel_sync code to directly program the cmd_bufs
based on the syncpts and semaphore received using the above APIs
instead of the current state of doing the wait programming from
within nvgpu_os_fence's interfaces.
Jira NVGPU-1093
Change-Id: Ie411f0ba60bca38f66a0024f5dfca03ef0b836eb
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1827475
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
struct gk20a from gk20a.h needs defination of struct gk20a_ce_app
and ce2_gk20a.h needs defination of struct gk20a. This creates
a circular dependency.
Fix this by making gk20a_ce_app a pointer to skip knowing the
complete type details and using forward declarations for struct
gk20a_ce_app and struct gk20a.
The gk20a_ce_app pointer is alloc'ed in gk20a_init_ce_support()
and free'ed in gk20a_ce_destroy.
JIRA NVGPU-611
Change-Id: I4d62d5f2b2d1492db73bae69f90a1fe5586fba76
Signed-off-by: Nitin Kumbhar <nkumbhar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1917945
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add new separate unit common/perf/cyclestats_snapshot.c and add
corresponding header file include/nvgpu/cyclestats_snapshot.h
This unit is h/w independent and simply calls gops.perf.* HALs
exposed by perf unit to do the h/w configurations
Also remove gv11b/css_gr_gv11b.* files as h/w specific sequence
implemented in them is already moved to perf unit
Rename all cyclestats_snapshot HALs in the form nvgpu_css_*()
Jira NVGPU-1103
Change-Id: I303f6becb313ac918e06c495a5fe299947a1f0b1
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1916652
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add sysfs nodes to manage power of dGPU. Writing
pci dev name to poweroff/poweron sysfs node powers
off/on dGPU.
The format of pci dev name is DDDD:BB:DD.F i.e.
domain:bus:device.function
echo 0001:01:00.0 > /sys/bus/pci/drivers/nvgpu/poweroff
echo 0001:01:00.0 > /sys/bus/pci/drivers/nvgpu/poweron
The permissions of nodes are set such that only root
user can write to the sysfs node to control dGPU power
state.
JIRA NVGPU-1100
Change-Id: I904881cab58c5f553e94510a3a10000194238433
Signed-off-by: Nitin Kumbhar <nkumbhar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1749848
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The power rail of dGPU is managed with help of a set of
GPIOs. Using those GPIOs add an interface to power off and
power on dGPU.
Before dGPU is powered off, new work is blocked by setting
NVGPU_DRIVER_IS_DYING and current jobs are allowed to finish
by waiting for gpu to be idle.
The tegra PCIe controller driver provided APIs
tegra_pcie_attach_controller() and tegra_pcie_detach_controller()
are used to manage PCIe link shutdown, PCIe refclk management
and PCIe rescan.
JIRA NVGPU-1100
Change-Id: Ifae5b81535f40dceca5292a987d3daf6984f3210
Signed-off-by: Nitin Kumbhar <nkumbhar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1749847
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
