Add support for cyclestats snapshots in the virtual case
Bug 1700143
JIRA EVLR-278
Change-Id: I376a8804d57324f43eb16452d857a3b7bb0ecc90
Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com>
Reviewed-on: http://git-master/r/1211547
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
We have completely different versions of probe for
nvgpu and pci device
Extract out common steps into nvgpu_probe() function
and separate it out in new file nvgpu_common.c
Divide task of nvgpu_probe() into further smaller
functions
Do platform specific things (like irq handling,
memresource management, power management) only in
individual probes and then call nvgpu_probe() to
complete the common initialization
Move all debugfs initialization to common gk20a_debug_init()
This also helps to bringup all debug nodes to pci device
Pass debugfs_symlink name as a parameter to gk20a_debug_init()
This allows us to set separate debugfs symlink for nvgpu
and pci device
In case of railgating, cde and ce debugfs, check if
platform supports them or not
Copy vidmem_is_vidmem from platform to mm structure
and set it to true for pci device
Return from gk20a_scale_init() if we don't have either of
governor or qos_notifier
Fix gk20a_alloc_debugfs_init() and gk20a_secure_page_alloc()
to receive device pointer instead of platform_device
Export gk20a_railgating_debugfs_init() so that we can call
it from gk20a_debug_init()
Jira DNVGPU-56
Jira DNVGPU-58
Change-Id: I3cc048082b0a1e57415a9fb8bfb9eec0f0a280cd
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1204207
(cherry picked from commit add6bb0a3d5bd98131bbe6f62d4358d4d722b0fe)
Reviewed-on: http://git-master/r/1204462
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Add a new debug message type: gpu_dbg_map_v. This is used for mapping
messages that are not specifically memory map operations.
Also cleanup the memory mapping debugging a bit since there was one
duplicate print and the memory map print was difficult to parse
visually. As a result the message has been modified to put the most
important information first in an easily readable format.
Bug 1732449
JIRA DNVGPU-12
Change-Id: Ib19c9371ee958009ab5a2d89b9610e699d070ee2
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1198593
(cherry picked from commit 51dba53b06ca171cdb13d1707f2d026b0ce29f07)
Reviewed-on: http://git-master/r/1147670
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Flushing timestamp record method can fail in case FECS is not
processing the main method queue. In particular, this occurs
in case of ctxsw timeout, where we process fifo sched interrupts
from the host, but FECS is still waiting for idle (grWFI).
In such scenario, this adds huge delay in fifo recovery
procedure (timeout on FECS method). Since flushing the last
(incomplete) record from FECS would only be useful in that case
(context switch ongoing), remove flush operation on engine
reset. Note that an explicit ENGINE_RESET event (with pid)
is inserted in user-facing ctxsw buffer on engine reset.
Bug 200228310
Change-Id: I885525f8f197f81266b50db161bb511867fc74f4
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1207305
(cherry picked from commit 44391b6204fd648949295f90481b0c424d9a5ddf)
Reviewed-on: http://git-master/r/1208414
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
We currently post bpt events (bpt.int and bpt.pause) even
before we process and clear the interrupts and this
could cause races with UMD
Fix this by posting bpt events only after we are done
processing the interrupts
Bug 200209410
Change-Id: Ic3ff7148189fccb796cb6175d6d22ac25a4097fb
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1184109
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Added preemption mode (WFI, GFXP, CTA and CILP) support for gp10x
family gr class (PASCAL_B and PASCAL_COMPUTE_B).
Bug 200221149
Change-Id: I859a4d2db518bca0ffeb0d85a6bb271f6b15db87
Signed-off-by: Lakshmanan M <lm@nvidia.com>
Reviewed-on: http://git-master/r/1193207
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Added interface to allow kernel to create privileged CE channels for
page migration and clearing support between sysmem and videmem.
JIRA DNVGPU-53
Change-Id: I3e18d18403809c9e64fa45d40b6c4e3844992506
Signed-off-by: Lakshmanan M <lm@nvidia.com>
Reviewed-on: http://git-master/r/1173085
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Added a dedicated device node to allow an
app manager to control TSG scheduling parameters:
- Get list of TSGs
- Get list of recent TSGs
- Get list of TSGs per pid
- Get TSG current scheduling parameters
- Set TSG timeslice
- Set TSG runlist interleave
Jira VFND-1586
Change-Id: I014c9d1534bce0eaea6c25ad114cf0cff317af79
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1160384
(cherry picked from commit 75ca739517cc7f7f76714b5f6a1a57c39b8cb38e)
Reviewed-on: http://git-master/r/1167021
Reviewed-by: Richard Zhao <rizhao@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
This patch calculates:
-Total time spent by GPU with rails gated.
-Total time spent by GPU with rails ungated.
-Total Railgating Cycles.
and dumps this information in debugfs file.
This feature requires CONFIG_DEBUG_FS set to true.
Bug 200195100
Change-Id: I1379f11237ce4900076947e18524caaa3304c7cb
Signed-off-by: Deepak Goyal <dgoyal@nvidia.com>
Reviewed-on: http://git-master/r/1178308
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
We currenlty initialize both runtime PM and pm_domains frameworks
and use pm_domain to control runtime power management of NvGPU
But since GPU has a separate rail, using pm_domain is not
strictly required
Hence remove pm_domain support and use runtime PM only for all
the power management
This also simplifies the code a lot
Initialization in gk20a_pm_init()
- if railgate_delay is set, set autosuspend delay of runtime PM
- try enabling runtime PM
- if runtime PM is now enabled, keep GPU railgated
- if runtime PM is not enabled, keep GPU unrailgated
- if can_railgate = false, disable runtime PM and keep
GPU unrailgated
Set gk20a_pm_ops with below callbacks for runtime PM
static const struct dev_pm_ops gk20a_pm_ops = {
.runtime_resume = gk20a_pm_runtime_resume,
.runtime_suspend = gk20a_pm_runtime_suspend,
.resume = gk20a_pm_resume,
.suspend = gk20a_pm_suspend,
}
Move gk20a_busy() to use runtime checks of pm_runtime_enabled()
instead of using compile time checks on CONFIG_PM
Clean up some pm_domain related code
Remove use of gk20a_pm_enable/disable_clk() since this
should be already done in platform specific unrailgate()/
railgate() APIs
Fix "railgate_delay" and "railgate_enable" sysfs to use
runtime PM calls
For VGPU, disable runtime PM during vgpu_pm_init()
With this, we will initialize vgpu with vgpu_pm_finalize_poweron()
upon first call to gk20a_busy()
Jira DNVGPU-57
Change-Id: I6013e33ae9bd28f35c25271af1239942a4fa0919
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1163216
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Propagate the buffer aperture flag in gk20a_locked_gmmu_map up so that
buffers represented as a mem_desc and present in vidmem can be mapped to
gpu.
JIRA DNVGPU-18
JIRA DNVGPU-76
Change-Id: I46cf87e27229123016727339b9349d5e2c835b3e
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1169308
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
add gk20a_aperture_mask() for memory target selection now that buffers
can actually be allocated from vidmem, and use it in all cases that have
a mem_desc available.
Jira DNVGPU-76
Change-Id: I4353cdc6e1e79488f0875581cfaf2a5cfb8c976a
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1169306
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Revamp the support the nvgpu driver has for semaphores.
The original problem with nvgpu's semaphore support is that it
required a SW based wait for every semaphore release. This was
because for every fence that gk20a_channel_semaphore_wait_fd()
waited on a new semaphore was created. This semaphore would then
get released by SW when the fence signaled. This meant that for
every release there was necessarily a sync_fence_wait_async() call
which could block. The latency of this SW wait was enough to cause
massive degredation in performance.
To fix this a fast path was implemented. When a fence is passed to
gk20a_channel_semaphore_wait_fd() that is backed by a GPU semaphore
a semaphore acquire is directly used to block the GPU. No longer is
a sync_fence_wait_async() performed nor is there an extra semaphore
created.
To implement this fast path the semaphore memory had to be shared
between channels. Previously since a new semaphore was created
every time through gk20a_channel_semaphore_wait_fd() what address
space a semaphore was mapped into was irrelevant. However, when
using the fast path a sempahore may be released on one address
space but acquired in another.
Sharing the semaphore memory was done by making a fixed GPU mapping
in all channels. This mapping points to the semaphore memory (the
so called semaphore sea). This global fixed mapping is read-only to
make sure no semaphores can be incremented (i.e released) by a
malicious channel. Each channel then gets a RW mapping of it's own
semaphore. This way a channel may only acquire other channel's
semaphores but may both acquire and release its own semaphore.
The gk20a fence code was updated to allow introspection of the GPU
backed fences. This allows detection of when the fast path can be
taken. If the fast path cannot be used (for example when a fence is
sync-pt backed) the original slow path is still present. This gets
used when the GPU needs to wait on an event from something which
only understands how to use sync-pts.
Bug 1732449
JIRA DNVGPU-12
Change-Id: Ic0fea74994da5819a771deac726bb0d47a33c2de
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1133792
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
This method is called when setting up gr
hardware. It is meant to adjust preemption
parameters.
Bug 1593548
Jira VFND-1894
Change-Id: I0f5aa3212bec3058a0493366bed6fe2a365c9542
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1162625
(cherry picked from commit c2e6d12570af28b3aae087401d7f670df40d40bd)
Reviewed-on: http://git-master/r/1166987
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Define specific QoS notifier for common clk framework
and protect it with CONFIG_COMMON_CLK
This new API will first get min/max requirements from
pm_qos and set min/max freq values in devfreq
A call to update_devfreq() will then ensure that
new estimated frequency is clipped appropriately
between min and max values
This also ensures that frequency is set along with
all the book-keeping
Add below platform specific notifier callback and use it
with pm_qos_add_notifier()
int (*qos_notify)()
If qos_notify is set, then only register the callback
We currently support only one qos_id which is treated
as notifier for min frequency
Remove dependency on qos_id, and use appropriate QoS
APIs like pm_qos_read_min/max_bound()
Store devfreq's min/max frequency in struct gk20a
for reference
Bug 1772462
Change-Id: I63d6d17451d19c9d376b67df7db775b38929287d
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1161161
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Add CE engine to vgpu engine list. CE engine is defined differently
for different GPUs, so we also add HAL for initializing the engine
info.
Bug 1780185
Change-Id: I5ae265551feac08d0c4d45402dd3277514e62b2d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1169720
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Aingara Paramakuru <aparamakuru@nvidia.com>
Tested-by: Aingara Paramakuru <aparamakuru@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Lakshmanan M <lm@nvidia.com>
-update get_netlist_name ops declaration to support
to load GPU FW based on GPU-ARCH
-"GAxxx" string used to get size for "gm204/" or
"gm206/" which will added to NETIMAGE path like
"gm204/NETC_img.bin"
Change-Id: I5bfa13df014533a885c4328d3c767e51c29f9255
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: http://git-master/r/1166783
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Move all places that read ptimer to use the callback.
It's for add vgpu implementation of read ptimer.
Bug 1395833
Change-Id: Ia339f2f08d75ca4969a443fffc9a61cff1d3d2b7
Signed-off-by: Richard Zhao <rizhao@nvidia.com>
Reviewed-on: http://git-master/r/1159587
(cherry picked from commit a01f804684f875c9cffc31eb2c1038f2f29ec66f)
Reviewed-on: http://git-master/r/1158449
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
When debug dump is called from an interrupt thread, we do not want
to call gk20a_busy() because it causes race in case rail gating is
being engaged at the same time. It has to be called from all debugfs
paths.
Bug 200198908
Bug 1770522
Change-Id: I7eda7d029b0a59cce0320ecc1b750dc2f4d7ccf0
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1163440
GVS: Gerrit_Virtual_Submit
Tested-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
This reverts commit 2219f38727. It leaves
GPU in on state for some tests that require powering down GPU.
Change-Id: I79d44fed729e98692021c57bbeff6a0ef2e8c983
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1161846
Read video memory size from hardware during initialization for devices
that support it.
JIRA DNVGPU-14
Change-Id: If190f2d89f7148520ee274ca674f972987c8056d
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1157215
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
We sometimes see race conditions where power refcount
is zero during ISR or bottom half.
If bottom half calls gk20a_busy(), it will lead to
boot up of GPU, but it is also possible that we are
already trying to poweroff GPU since power refcount
is zero
Fix this by taking a power refcount with gk20a_busy_noresume()
in ISR and then dropping this refcount at the end of
bottom half
Add new API gk20a_idle_nosuspend() to drop a refcount
without initiating suspend
Bug 200198908
Bug 1770522
Change-Id: Iec3d4dc8d468f49b71919d2bbc327da48b97bcab
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1160035
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
This CL covers the following modification,
1) Added multiple engine_info support
2) Added multiple runlist_info support
3) Initial changes for ASYNC CE support
4) Added ASYNC CE interrupt handling support
for gm206 GPU family
5) Added generic mechanism to identify the
CE engine pri_base address for gm206
(CE0, CE1 and CE2)
6) Removed hard coded engine_id logic and
made generic way
7) Code cleanup for readability
JIRA DNVGPU-26
Change-Id: I2c3846c40bcc8d10c2dfb225caa4105fc9123b65
Signed-off-by: Lakshmanan M <lm@nvidia.com>
Reviewed-on: http://git-master/r/1155963
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
-update HAL of ACR BL which can support
gm204/gm206 and DMATRFBASE method to global
JIRA DNVGPU-10
Change-Id: I56fc7ce040dadb6473f6f375ee6ce90783a046ad
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: http://git-master/r/1154954
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
To be able to scan, PCI devnodes need to be in a directory with read
permission. By default /dev is read protected by SELinux policy. Move
the devnodes to their own directory so that reading this one
directory can be allowed.
At the same time rename the nodes to start with string "card-".
JIRA DNVGPU-54
Change-Id: I0df4ced08afd1f3a468e983d07395ffcb8050365
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1152745
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
GVS: Gerrit_Virtual_Submit
Add support for chips gm204 and gm206. Adds also support for reading
VBIOS and booting devinit and pre-os images on PMU.
Change-Id: I4824b44245611e5379ace62793cc37158048f432
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1120467
GVS: Gerrit_Virtual_Submit
Reviewed-by: Ken Adams <kadams@nvidia.com>
Use an inline function instead of a macro to
"expand" all channel parameters.
Jira EVLR-244
Jira EVLR-318
Change-Id: I4e8c5ee6bc9da36564af171be809f50dd2dfd439
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: http://git-master/r/1150050
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Sequence to reset PMU is different for iGPU and dGPU. Specialize
and implement iGPU version.
Change-Id: I5b9ff2c018a736bc9e27b90d0942c52706b12a12
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1150540
Add support for hwpm broadcast registers (ltc and lts)
In gr_gk20a_find_priv_offset_in_buffer, replace "Unknown address type" error
with informational message: gr_gk20a_exec_ctx_ops calls
gk20a_get_ctx_buffer_offsets and if that fails,
calls gr_gk20a_get_pm_ctx_buffer_offsets; HWPM registers will fail the first
call, so an error or warning is overkill.
Bug 1648200
Change-Id: I197b82579e9894652add4ff254418f818981415a
Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com>
Reviewed-on: http://git-master/r/1131365
(cherry picked from commit 9f30a92c5d87f6dadd34cc37396a6b10e3a72751)
Reviewed-on: http://git-master/r/1133628
(cherry picked from commit 7eb7cfd998852ba7f7c4c40d3db286f66e83ab3a)
Reviewed-on: http://git-master/r/1127749
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
We were not using the engine_type field in device info, and the code
did not handle chained entries properly. The code assumed that first
entry is for graphics and second for CE, which is not always true.
Improve the code to go through all entries of device_info, and
preserve values across entries until we reach the last entry.
Only last entry triggers a write to fifo engine info.
There can also be multiple engines with same type, so accumulate
interrupts and reset ids from all of them.
As the code got fixed, now it reads the engine enum correctly from
hardware. We used to compare that against CE0, but we should compare
against CE2.
gk20a_fifo_reset_engine() uses wrong constants - it is passed a
internal numbering of engines, but it compares them against hardware
engine enum.
Change-Id: Ia59273921c602d2a090f7a5b1404afb0fca2532c
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1147746
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Program CWD TPC and SM registers correctly. The old code did not work
when there are more than 4 TPCs.
Refactor init_fs_mask to reduce code duplication.
Change-Id: Id93c1f8df24f1b7ee60314c3204e288b91951a88
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1143697
GVS: Gerrit_Virtual_Submit
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Setting max_ways_evict reserves some of L2 for CB. In gk20a CB is in
dedicated RAM, so we don't need to reserve space for it.
The code gets invoked only on gk20a.
Change-Id: Ib8efec8c5e90c135bd0c10bb1eaa3f797ec68698
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1144993
To support vidmem, pass g and mem_desc to the buffer memory accessor
functions. This allows the functions to select the memory access method
based on the buffer aperture instead of using the cpu pointer directly
(like until now). The selection and aperture support will be in another
patch; this patch only refactors these accessors, but keeps the
underlying functionality as-is.
gk20a_mem_{rd,wr}32() work as previously; add also gk20a_mem_{rd,wr}()
for byte-indexed accesses, gk20a_mem_{rd,wr}_n() for memcpy()-like
functionality, and gk20a_memset() for filling buffers with a constant.
The 8 and 16 bit accessor functions are removed.
vmap()/vunmap() pairs are abstracted to gk20a_mem_{begin,end}() to
support other types of mappings or conditions where mapping the buffer
is unnecessary or different.
Several function arguments that would access these buffers are also
changed to take a mem_desc instead of a plain cpu pointer. Some relevant
occasions are changed to use the accessor functions instead of cpu
pointers without them (e.g., memcpying to and from), but the majority of
direct accesses will be adjusted later, when the buffers are moved to
support vidmem.
JIRA DNVGPU-23
Change-Id: I3dd22e14290c4ab742d42e2dd327ebeb5cd3f25a
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1121143
Reviewed-by: Ken Adams <kadams@nvidia.com>
Tested-by: Ken Adams <kadams@nvidia.com>
Add below flag fields to gpu characteristics to indicate
supported and default preemption modes on platform for
graphics and compute
__u32 graphics_preemption_mode_flags;
__u32 compute_preemption_mode_flags;
__u32 default_graphics_preempt_mode;
__u32 default_compute_preempt_mode;
Add struct nvgpu_preemption_modes_rec to struct gr_gk20a
to store these values locally
Use platform specific get_preemption_mode_flags() to
get the flags and define gk20a/gm20b specific
get_preemption_mode_flags() API
Bug 1646259
Change-Id: I80193c0d988dc93bd96585f9aa631fd817f4dfa3
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1133595
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Add separate IOCTL NVGPU_IOCTL_CHANNEL_SET_PREEMPTION_MODE
to allow setting preemption modes from UMD
Define preemption modes in nvgpu.h and use them everywhere
Remove mode definitions from mm_gk20a.h
Also, we support setting only one preemption mode in a channel
But it is possible to have multiple preemption modes (one from
graphics and one from compute) set simultaneously
Hence, update struct gr_ctx_desc to include two separate
preemption modes (graphics_preempt_mode and compute_preempt_mode)
API NVGPU_IOCTL_CHANNEL_SET_PREEMPTION_MODE also supports
setting two separate preemption modes i.e. one for graphics
and one for compute
Make necessary changes in code to support two preemption
modes
Bug 1646259
Change-Id: Ia1dea19e609ba8cc0de2f39ab6c0c4cd6b0a752c
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1131805
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
gk20a_fifo_force_reset_ch() does not support vgpu now, so we need to
create a function pointer in gpu_ops and assign it differently for
vgpu and non-vgpu.
Bug 200184349
Change-Id: I5f8f4f731b4b970c4ff8de65531f25568e7691b6
Signed-off-by: Haley Teng <hteng@nvidia.com>
Reviewed-on: http://git-master/r/1130420
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
