When Host receives a page fault signal from a CE, Host will immediately
set _ENG_FAULTED bit in the NV_PCCSR_CHANNEL register for the channel
and will trigger a preempt of the TSG/channel.
A channel will only be scheduled when _ENABLED=1, _ENG_FAULTED=0
and _PBDMA_FAULTED=0 in pccsr_channel reg for the channel.
If a TSG has a faulted channel, Host will not schedule the entire TSG
agin until all _FAULTED bits from channels in the TSG are cleared by SW.
This function will be required for ce page fault handling.
JIRA GPUT19X-46
JIRA GPUT19X-12
Change-Id: Ib58dff7aa24aa144e970f11b5261877dec03f3e6
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1509776
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
When CE hits a page fault it needs to save out methods, it will
save it out to a per runqueue per TSG method buffer. The method buffers
(one per TSG runqueue) are allocated in BAR2 space during TSG creation
All channels in a TSG that are mapped to the same runqueue will point
to the same buffer.
S/w will insert channel's method buffer pointer in the channel's
instance block entries NV_RAMIN_ENG_METHOD_BUFFER_ADDR_LO and
NV_RAMIN_ENG_METHOD_BUFFER_ADDR_HI. Method buffer in memory will
be 32B aligned.
Eng method buffer allocated per tsg will be de-allocated during
tsg_release.
JIRA GPUT19X-46
Change-Id: Ib480ae5840d9815d24fe2eadc169ac3102854cd0
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1509747
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Implement get_num_pce ce ops to get number of physical
copy engines. This is required to calculate eng method
buffer size
JIRA GPUT19X-46
Change-Id: I5a37eb26ec11bc358700d1761cfdb6ca060e4287
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1511788
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Added handling for below ce interrupts
-INVALID_CONFIG interrupt will be generated if a floorswept
PCE is assigned to a valid LCE in the NV_CE_PCE2LCE_CONFIG
registers. This is a fatal error and the LCE will have to be
reset to get back to a working state.
-MTHD_BUFFER_FAULT interrupt will be triggered if any access to
a method buffer during context load or save encounters a fault.
This is a fatal interrupt and will require at least the LCE to be reset
before operations can start again, if not the entire GPU.
JIRA GPUT19X-12
JIRA GPUT19X-46
Change-Id: I2eeefc4e634f5bf53f20933c493c7594fe0ea755
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1510298
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Stop defining per-platform default big page size. It's defined via
HAL and inherited from gp10b.
JIRA NVGPU-38
Change-Id: If5eedd5d351d5504bdf87489d1aa091d430c43ba
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: https://git-master/r/1508069
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
We should check if BLCG/SLCG feature is enabled
before trying to enable/disable them in hardware.
Bug 200314250
Change-Id: I5431f97cc559444298b7bd4d53a9f4fc598fd268
Signed-off-by: Deepak Goyal <dgoyal@nvidia.com>
Reviewed-on: https://git-master/r/1509184
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Take care of t19x sm reg address changes and support multiple SM
JIRA GPUT19X-75
Change-Id: I675b76b90d08fe75331f0023f1fe722497d06373
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1477673
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Support multiple SM and take care of sm reg addr changes
JIRA GPUT19X-75
Change-Id: Id39e269034762c7a8347edaf1fff0b2efd7f153c
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1477705
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Support multiple SM and take care of SM hardware reg address changes
JIRA GPUT19X-75
Change-Id: I866011a85da06ca22bc10fda5ab59f84d0782902
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1477686
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Take care of t19x reg address changes to support multiple SM
JIRA GPUT19X-75
Change-Id: I92b97e60ac82c50a97fe44a85482437446479800
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1477694
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
- add fifo_gv11b to store usermode_regs
- consider channel_base and use usermode_regs when ring channel doorbell
It'll make kickoff code re-usable for vgpu.
Jira VFND-3796
Change-Id: Ia6974ccac137f201ad8763a7d372de81d5cca56b
Signed-off-by: Richard Zhao <rizhao@nvidia.com>
Reviewed-on: https://git-master/r/1510457
Reviewed-by: Aingara Paramakuru <aparamakuru@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
hw_chid is a relative id for vgpu. For native it's same as hw id.
Renaming it to chid to avoid confusing.
Jira VFND-3796
Change-Id: Ie94c1a15e9e45fc823d85790ce6a69da53a685bf
Signed-off-by: Richard Zhao <rizhao@nvidia.com>
Reviewed-on: https://git-master/r/1509531
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Depend on the main GK20A config for all subitems so that they get
grouped properly under the top level item in the menuconfig.
Change-Id: Ia44179cdebb1f5b24ea626b57ccea40bdd9b2bd8
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: https://git-master/r/1509336
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reorganize HAL initialization to remove inheritance and construct
the gpu_ops struct at compile time. This patch only covers the ltc
sub-module of the gpu_ops struct.
Perform HAL function assignments in hal_gxxxx.c through the
population of a chip-specific copy of gpu_ops.
Jira NVGPU-74
Change-Id: If8760efb7d8e94b63dc6f1fe9efec4ddf49c0b29
Signed-off-by: Sunny He <suhe@nvidia.com>
Reviewed-on: https://git-master/r/1507563
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Pbdma which encountered the ctxnotvalid interrupt will stall and
prevent the channel which was loaded at the time the interrupt fired
from being swapped out until the interrupt is cleared.
CTXNOTVALID pbdma interrupt indicates error conditions related
to the *_CTX_VALID fields for a channel. The following
conditions trigger the interrupt:
* CTX_VALID bit for the targeted engine is FALSE
* At channel start/resume, all preemptible eng have CTX_VALID FALSE but:
- CTX_RELOAD is set in CCSR_CHANNEL_STATUS,
- PBDMA_TARGET_SHOULD_SEND_HOST_TSG_EVENT is TRUE, or
- PBDMA_TARGET_NEEDS_HOST_TSG_EVENT is TRUE
JIRA GPUT19X-47
Change-Id: If65ce1fcdbaebd6b1d8313fdddf9e3e0fa51e885
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1329372
GVS: Gerrit_Virtual_Submit
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
In gv11b fbhub num_ltcs is read only, even though
register spec says it is rw. The number of ltcs
are populated by hw and no need for sw to set those
values.
GPUT19X-70
Change-Id: Ib9861894cacb70cf54b4958083e55d39a3a85e19
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/1497992
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
tegra/linux path was created to separate Tegra kernel specific
dependencies from common Linux specific dependencies. The split has
not really worked, so merge tegra/linux to common/linux.
JIRA NVGPU-38
Change-Id: I9efe078bfa5dfbef49408db9d8a3738dfda8bd1d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1505169
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Without non-cacheable, gpu filled subcontext data not visible
to cpu without additional l2 flush. Similarly, there will be issues
where cpu updates to subcontext header will not visible to gpu without
additional l2 flush.
Making subcontext header mapping non-cacheable fixes this issue.
Bug 1937331
Change-Id: I8e25b7cac165e7481eec7c9f1f93bc7992183c46
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/1505283
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
After enabling l2 write back in gv11b, for committing all
dirty data to sysmem correctly:
Added one fb_flush before l2_flush to commit dirty hshub data to l2/sysmem.
Added one more fb_flush after l2_flush, to commit any new dirty data on
hshub to sysmem.
This done by implementing gv11b specific l2_flush function.
Bug 1937331
Change-Id: Ie30edb12c98c4021783c88750bb4c4ca62e4a7ca
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/1503385
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
This patch:
- Adds a PMU command needed for enabling ELPG.
i.e. command to update sub-feature mask to enable ELPG.
- Adds a new version of PG-GR init param command function
which uses updated command interface.
JIRA GPUT19X-20.
Change-Id: If969c018e2e28264fdc9c897892eb28b021d12f2
Signed-off-by: Deepak Goyal <dgoyal@nvidia.com>
Reviewed-on: http://git-master/r/1504873
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Preempt type should be set to tsg and id should be set to tsgid
in fifo_preempt_r(). Preempt type channel and id set to channel
id does not initiate preemption.
Bug 200289427
Bug 200292090
Bug 200289491
Change-Id: I2ae96c0b9ca8a88a8405f42775744f0879994887
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: http://git-master/r/1497877
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
We added duplicate common/linux/nvhost_priv_t19x.h so that the definition of
struct nvgpu_nvhost_dev is available in nvgpu-t19x repo
But instead of duplicating the file, directly include original file with
path #include "common/linux/nvhost_priv.h
Jira NVGPU-29
Change-Id: I5d373227f0f6b2b4670d2fd3ad433a4655df8e4f
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1499167
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Remove use of linux specifix header files
<linux/nvhost.h> and <linux/nvhost_t194.h>
and use nvgpu specific header file
<nvgpu/nvhost_t19x.h> instead
This is needed to remove all Linux dependencies
from nvgpu driver
Replace all nvhost_*() calls by
nvgpu_nvhost_*() calls from new nvgpu library
Jira NVGPU-29
Change-Id: I32d59628ca5ab3ece80a10eb5aefa150b1da448b
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1494648
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Add new abstraction file common/linux/nvhost_t19x.c
for all nvhost APIs exported from linux/nvhost_t194.h
This file will be compiled only if config
CONFIG_TEGRA_GK20A_NVHOST is set
Export the new headers from file <nvgpu/nvhost_t19x.h>
Also add dummy private header file nvhost_priv_t19x.h
to store definition of private structure nvgpu_nvhost_dev
This file should be deleted when nvgpu-t19x repo
is merged into common nvhost repo
Jira NVGPU-29
Change-Id: I8c08c9242b08cc45f7c99cc400b3e1a720f9439c
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1493792
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
We pass (struct device_attribute *) to gp10b_ecc_stat_create()
and gr_gp10b_ecc_stat_create() and then assign a memory
allocation to this pointer
But since this pointer is local copy to function, static
pointer variables are never set in gr_gp10b_create_sysfs()
This also results in a resource leak since we never free
the storage assigned to local variable
Fix this by adding and passing correct parameter
(struct device_attribute **) so that the address of the
allocation is returned to the caller correctly
Bug 200291879
Coverity id : 2567934
Change-Id: I1b1d329265f4d32739abbbe3a4e419a2af62b874
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1495907
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
