nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-25 11:04:51 +03:00
Code Issues Packages Projects Releases Wiki Activity
6,767 Commits 57 Branches 84 Tags
fa49f99df8a5ece2416e1b4220d7faab6064a6e4
Commit Graph

6767 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Seema Khowala
fa49f99df8 gpu: nvgpu: Update doxygen format for pbdma_status.h 
Added @brief, @return and @retval where applicable.
Added [in/out] for function parameters.

JIRA NVGPU-3592

Change-Id: Ib870d7c0ba72c364fbada6b6ea3ce102d3eb9a7f
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2140912
Reviewed-by: Thomas Fleury <tfleury@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>
 2019-06-25 11:37:07 -07:00
ajesh
a1bbbd4cad gpu: nvgpu: fix CERT C violations in cond unit 
INT31-C requires that integer conversions do not result in lost or
misinterpreted data.
INT32-C requires that operations on signed integers do not result in
overflow.
Fix the above CERT C violations in cond unit.

Jira NVGPU-3607

Change-Id: I31da5b214ee3c39799066a785b0c19f45e48c3cd
Signed-off-by: ajesh <akv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2139212
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Nitin Kumbhar <nkumbhar@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-25 05:46:35 -07:00
Sagar Kamble
e60a381a39 gpu: nvgpu: compile out CE from safety build 
Now that VIDMEM support is compiled out from safety build, let us
compile out CE support as well.

JIRA NVGPU-3524
JIRA NVGPU-3646

Change-Id: If30e85f78c62917b99d26e5c502df3aada3e5841
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2136433
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-25 04:37:18 -07:00
Sagar Kamble
a16cc2dde3 gpu: nvgpu: compile out vidmem from safety build 
Safety build does not support vidmem. This patch compiles out vidmem
related changes - vidmem, dma alloc, cbc/acr/pmu alloc based on
vidmem and corresponding tests like pramin, page allocator &
gmmu_map_unmap_vidmem..
As vidmem is applicable only in case of DGPUs the code is compiled
out using CONFIG_NVGPU_DGPU.

JIRA NVGPU-3524

Change-Id: Ic623801112484ffc071195e828ab9f290f945d4d
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2132773
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-25 04:37:08 -07:00
Thomas Fleury
c2eb26436a gpu: nvgpu: Add doxygen documentation in runlist.h 
Removed the following unused fields from runlist context:
- total_entries
- stopped
- support_tsg

Renamed:
- nvgpu_fifo_runlist_set_state -> nvgpu_runlist_set_state

Removed RUNLIST_INVALID_ID which was redundant with
NVGPU_INVALID_RUNLIST_ID.

Jira NVGPU-3594

Change-Id: I23d1abdf87b73bc0138816dab6659249f2602b9f
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2139520
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@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>
 2019-06-24 17:36:29 -07:00
Seema Khowala
421e2d8187 gpu: nvgpu: Update doxygen format for engine_status.h 
Added @brief, @return and @retval where applicable.
Added [in/out] for function parameters.

JIRA NVGPU-3590

Change-Id: I6c2324b20033562d4cdc1790fb3d70fd09434b3d
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2140966
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
Tested-by: Thomas Fleury <tfleury@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>
 2019-06-24 16:45:49 -07:00
Thomas Fleury
06858c83d6 gpu: nvgpu: unit: stub userd.setup_sw 
libnvgpu-drv can be compiled with NVGPU_USERD, while unit
test is not. Make sure we use the stub in any case.

Jira NVGPU-3480

Change-Id: Id48d1bda8796dd600f726233978e4f3c4b158674
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2140790
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-24 12:56:21 -07:00
Sagar Kamble
5035564ffb gpu: nvgpu: disable compression in safety build 
With compression removal now handled in the nvrm driver let us compile
out the sources from nvgpu safety build.

JIRA NVGPU-3647

Change-Id: I04930f20bacd2c2a8a491d69d4a8ca8d39bad548
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2142189
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-24 04:26:33 -07:00
Sagar Kamble
fdcc156d8e gpu: nvgpu: fix the comments in Makefile.shared.configs 
List clearly the safety release/debug and regular build profiles while
defining the flags.

JIRA NVGPU-1949

Change-Id: I259e136e5941602fc263740f0d81384939d7ab11
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2140630
GVS: Gerrit_Virtual_Submit
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-24 04:26:19 -07:00
Seshendra Gadagottu
e364102f9a gpu: nvgpu: add graphics flag for gfxp related code 
Move GFXP related code under CONFIG_NVGPU_GRAPHICS flag.
Keep the NVGPU_PREEMPTION_MODE_GRAPHICS_WFI support.

JIRA NVGPU-3415

Change-Id: Ie690ac66df4b94eb113a5898d94a892fe0ce7b11
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2135427
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-24 02:46:03 -07:00
Vedashree Vidwans
57e31af3a6 gpu: nvgpu: disable logging for libnvgpu build 
This is a follow up patch for disabling logging for userspace libnvgpu
build. Logging is still force enabled for unit tests.

Jira NVGPU-869

Change-Id: Ieea84fad14e3caab33ae686d4f00fd04ec80850b
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2140311
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-23 23:15:37 -07:00
Aparna Das
21c7a00f18 gpu: nvgpu: vgpu: create hal vgpu unit 
File vgpu_fifo_gv11b.c contained syncpoint related implementation
specific to gv11b. Move the implementations to a new file in
hal directory for vgpu hal/vgpu/sync/syncpt_cmdbuf_gv11b_vgpu.c.
Also move function vgpu_gv11b_init_fifo_setup_hw() to a new
file in hal directory for vgpu hal/vgpu/fifo/fifo_gv11b_vgpu.c.

Add a new yaml file nvgpu-hal-vgpu.yaml that contains vgpu
specific hal files. Update arch yaml to reflect the above changes.

Jira GVSCI-994

Change-Id: Ie33614473d5fd3fcd624c70709b109c4e45725ef
Signed-off-by: Aparna Das <aparnad@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138390
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Nirav Patel <nipatel@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 18:45:09 -07:00
Aparna Das
8380e7bde6 gpu: nvgpu: vgpu: unify gpu tsg bind channel 
vgpu_gv11b_tsg_bind_channel() was specific to gv11b. Modify
function vgpu_tsg_bind_channel() to handle gv11b specific
case by checking if subctx is supported.
Delete gv11b specific file common/vgpu/gv11b/vgpu_tsg_gv11b.c
and update arch yaml file accordingly.

Jira GVSCI-994

Change-Id: I36c1f7392087573afa06cd3652a145aa92055f1c
Signed-off-by: Aparna Das <aparnad@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138389
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Nirav Patel <nipatel@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 18:45:00 -07:00
Aparna Das
7d3ae08847 gpu: nvgpu: vgpu: create unit fifo 
Move fifo related code to common/vgpu/fifo
and create new child units fifo, channel, tsg, preempt,
engines.

Also update arch YAML to include newly created files
related to fifo unit.

Jira GVSCI-994

Change-Id: I79897df4e729e0506702832ba62c1694c3f42280
Signed-off-by: Aparna Das <aparnad@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138388
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Nirav Patel <nipatel@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 18:44:51 -07:00
Aparna Das
35219753e2 gpu: nvgpu: vgpu: delete file vgpu.c 
File vgpu.c was renamed as part of change
https://git-master.nvidia.com/r/#/c/2081153/.
The file got erroneously added again. Delete
the file and update YAML arch.

Jira GVSCI-994

Change-Id: Iacbbedbc53ea12ea158c840ac0a5d96c0d366541
Signed-off-by: Aparna Das <aparnad@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138387
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Nirav Patel <nipatel@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 18:44:42 -07:00
Thomas Fleury
6d760dd075 gpu: nvgpu: Update doxygen format for pbdma.h 
Added @brief, @return and @retval where applicable.
Added [in/out] for function parameters.

Jira NVGPU-3591

Change-Id: Ibc555c631817a8acd28bf5d4343acbb3d6d0d669
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2139607
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 13:56:16 -07:00
Alex Waterman
893f28f775 gpu: nvgpu: MM SWUD 
Add a SWUD (SW Unit Design document) for MM as well as some glue in
the gk20a.h header for a nvgpu wide main-page.

JIRA NVGPU-3544

Change-Id: I7d6ee6ef94d7f0d2c10c0f4d8c860a20b33f7aae
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2129852
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 11:16:07 -07:00
Alex Waterman
2ba44f30c7 gpu: nvgpu: Add doxygen safety profile build 
Add a doxygen build to nvgpu. Use the following command from the
drivers/gpu/nvgpu directory:

  $ make -f Makefile.doxygen

This will autogenerate a list of files from the safety profile
architecture and then generate doxygen documentation from said
list of files.

Also add a .gitignore entry for the sources file that gets
generated during the build.

JIRA NVGPU-3544

Change-Id: I2415de631efcc8447c81ace1700ebac8b69a89df
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2131051
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 11:15:52 -07:00
ajesh
fa8e1904e2 gpu: nvgpu: fix CERTC violations in timers unit 
INT32-C requires that operations on signed integers do not result in
overflow.
Fix violations of INT32-C in timers unit.

Jira NVGPU-3605

Change-Id: I2d5ba6d1c2e3a84a5b3df1ecaf16cce13c1b2c97
Signed-off-by: ajesh <akv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2139971
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 10:26:50 -07:00
Vaibhav Kachore
318d6451e9 gpu: nvgpu: Improve accuracy of dGPU clk measurement 
- This patch replaces "nvgpu_current_time_ms" with "nvgpu_hr_timestamp_us".
- "nvgpu_hr_timestamp_us" gives timestamp in microseconds and has better
accuracy than "nvgpu_current_time_ms" (which gives timestamp in millisecond)

Bug 200503143

Change-Id: I6a10e8e1b3e8ff842aa23f58bf2ba9344af232a6
Signed-off-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2125959
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 10:25:48 -07:00
Thomas Fleury
80a5b864d3 gpu: nvgpu: unit: zero-init vm for as bound channels 
For some tests, we build ad-hoc structures to initialize ch->vm.
Make sure related vm and mm structures are zero-initialized to
avoid intermittent failures.

Jira NVGPU-3651

Change-Id: I20637e796a6f6b188f76c40aee918d2fa15490b1
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2140264
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Sagar Kamble <skamble@nvidia.com>
Tested-by: Sagar Kamble <skamble@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>
 2019-06-21 09:15:27 -07:00
Divya Singhatwaria
aab600a4f2 gpu: nvgpu: Fix CERT INT30-C violations in ACR 
CERT-C INT30 requires checking for wrapping when
doing arithmetic operations of unsigned value.

This fixes INT30 violations in acr_boostrap.c
and acr_sw_gv11b.c

JIRA NVGPU-3575

Change-Id: I9b73d9ca677b7e476ead4b67a257b37aeb3db6b3
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2139389
Reviewed-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 06:54:51 -07:00
ajesh
70d8a4555a gpu: nvgpu: fix CERTC violations in kmem unit 
INT30-C Requires that unsigned integer operations do not wrap.
INT31-C Requires that integer conversions do not result in lost or
misinterpreted data.
ERR33-C Requires to detect and handle standard library errors.
Fix the above CERT C violations in kmem unit.

Jira NVGPU-3603

Change-Id: I104551406e3cf6d2da22c5b6c6edc6694992ce62
Signed-off-by: ajesh <akv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2139529
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Nitin Kumbhar <nkumbhar@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 04:36:51 -07:00
Sagar Kadamati
d2444e85ed gpu: nvgpu: compile-out debug unit 
debug unit is not need to for safety build, so compile out it

JIRA NVGPU-3542

Change-Id: I60cc256a5659e72ae2e647ec4f1a810ba4aa959d
Signed-off-by: Sagar Kadamati <skadamati@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2133419
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 04:35:25 -07:00
Vinod G
eaae1afdae gpu: nvgpu: Fix CERT INT31-C error in gr.falcon unit 
Modify nvgpu_gr_checksum_u32 code from
(u32)(((u64)a + (u64)b) % (U32_MAX))
to nvgpu_safe_cast_u64_to_u32(((u64)a + (u64)b) & (U32_MAX)).
This function returns the last 32bits.

Jira NVGPU-3622

Change-Id: I8d348a624a5930b7ba813b0a74af270b49b03844
Signed-off-by: Vinod G <vinodg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2140254
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 01:07:42 -07:00
Seshendra Gadagottu
7ed189de44 gpu: nvgpu: fix CERT INT30-C in common.gr.falcon 
Fixed CERT C error by using nvgpu_gr_checksum_u32:
segments->boot_signature += bootimage[i]; -->
	segments->boot_signature = nvgpu_gr_checksum_u32(
				segments->boot_signature, bootimage[i]);

JIRA NVGPU-3622

Change-Id: I01de116ba2a8afacb8a93be7b88e356a48122c5a
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2132547
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-21 01:06:16 -07:00
rmylavarapu
2d66c2c928 gpu: nvgpu: Support sending VFE_EQU dependent masks to PMU 
Right now in PMU we have disabled usage of VFE cached
values as we have seen failures in 5.1.3.0 release.
This patch will send vfe_equ dependent masks to PMU which
will be used for initial invalidation of dependent VFE_EQU
cached values. As a result PMU will compute the dependent
EQUs and cache them.

NVGPU-3645

Change-Id: If99960481317a09df9499265650162245f404555
Signed-off-by: rmylavarapu <rmylavarapu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137867
Reviewed-by: Mahantesh Kumbar <mkumbar@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>
 2019-06-20 22:59:53 -07:00
rmylavarapu
0194d8640e gpu: nvgpu: Support sending VFE_VAR dependent masks to PMU 
Right now in PMU we have disabled usage of VFE cached values
as we have seen failures in 5.1.3.0 release. This patch
will send vfe_var dependent masks to PMU which will be
used for initial invalidation of dependent VFE_VAR cached
values. As a result PMU will compute the dependent VARs and
cache them.

NVGPU-3645

Change-Id: I814ae819d77270ac45b97cf2a32726992e6f4af3
Signed-off-by: rmylavarapu <rmylavarapu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137423
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Mahantesh Kumbar <mkumbar@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>
 2019-06-20 22:58:51 -07:00
Sagar Kamble
c85a7d7010 gpu: nvgpu: arch: update YAML for hal units 
therm unit is safe for dgpu hence add the gpu attribute. perf unit
is not valid anymore. remove it.

JIRA NVGPU-3660

Change-Id: Ib9d03baedbad3cc0c36fa4985361cf56c3f3b715
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138539
Reviewed-by: Automatic_Commit_Validation_User
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>
 2019-06-20 21:17:10 -07:00
Sagar Kamble
eb9bacc35e gpu: nvgpu: arch: add YAML gpu attribute to pramin unit 
pramin functionality will be used in the dgpu release with vidmem. It is
non-safe for igpu. Add the gpu field to the unit to specify that it
applies only to dgpu.

JIRA NVGPU-3660

Change-Id: Ice83cd335b97fc33ffcf34c87527f10ff262ee7f
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138538
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-20 21:17:01 -07:00
Sagar Kamble
3da1a06d68 gpu: nvgpu: arch: update YAML for pmu common units 
Created new unit pmu_pstate which is non-safe for igpu. Removed empty
units init and zbc. Updated gpu and safety attribute for perfmon, fw
super_surface and lsfm.

JIRA NVGPU-3660

Change-Id: Ia772e2d2779b90065549dd3713f89b152ef3d7d0
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138537
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-20 21:16:46 -07:00
Sagar Kamble
ffca99017b gpu: nvgpu: arch: update YAML safety attribute of mem queues 
SEC2 and its emem queues won't be used in future safety releases hence mark
emem queue unit as non-safe. With that, mem_queue can also be marked as non
-safe.

JIRA NVGPU-3660

Change-Id: I7fceb3eef6c05cf91a836e7df2fc060119592620
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138536
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-20 21:16:31 -07:00
Sagar Kamble
63c46857cd gpu: nvgpu: arch: mark page_allocator unit with dgpu attribute 
Since this unit is used by vidmem mark this unit as applicable to only
dgpu.

JIRA NVGPU-3660

Change-Id: I78e3f1fda579c4fdb3a80b95cd9209a8d68fa8d3
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2139336
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>
 2019-06-20 17:15:50 -07:00
Sagar Kamble
5d37a9e489 gpu: nvgpu: compile out sim changes from safety build 
As sim is non-safe unit compile it out. Also removed FMODEL related
nvgpu changes and unit tests from the safety build.

JIRA NVGPU-3527

Change-Id: I22c83e195a09f9150fb6f5a3afff91df2ea075b9
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2139455
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Raghuram Kothakota <rkothakota@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-20 16:05:33 -07:00
Thomas Fleury
ad92d2d8cf gpu: nvgpu: Add doxygen documentation in preempt.h 
Jira NVGPU-3593

Change-Id: Ia4491e6cd123a571b528eccb23727f747895ea86
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2133849
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: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-20 16:05:09 -07:00
Seeta Rama Raju
3ca084ae65 gpu: nvgpu: Add safe ops func for cast s32 to u64 
JIRA NVGPU-3482

Change-Id: Iff3dd13132057d3ed89c2955c31f3e14bf6cfeda
Signed-off-by: Seeta Rama Raju <srajum@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137505
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Nitin Kumbhar <nkumbhar@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-20 04:14:07 -07:00
Rajesh Devaraj
29ec6ad40f gpu: nvgpu: report fb_flush_timeout error 
This patch adds the support to report fb_flush_timeout error to 3LSS.
Specifically, it adds the following service-ID:
NVGUARD_SERVICE_IGPU_HOST_SWERR_PFIFO_FB_FLUSH_TIMEOUT_ERROR

JIRA NVGPU-3460
JIRA NVGPU-3461

Change-Id: Iddf978eedbc676197a19e47e72e08cd71c478a08
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2138051
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Raghuram Kothakota <rkothakota@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-19 22:51:20 -07:00
Divya Singhatwaria
c341c84039 gpu: nvgpu: Fix CERT INT30-C violations in ACR 
CERT-C INT30 requires checking for wrapping when
doing arithmetic operations of unsigned value.
This fixes INT30 violations in acr.c and acr_blob_construct_v1.c

JIRA NVGPU-3575

Change-Id: Ib44cc5675f0f3af7575b1f5340c78cabe0ce67af
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2136158
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-19 03:07:48 -07:00
Deepak Nibade
0755b25231 gpu: nvgpu: remove reset and enable/disable ctxsw hals 
Remove below hals since the corresponding functions are same on all
platforms and they are h/w independent
g->ops.gr.enable_ctxsw()
g->ops.gr.disable_ctxsw()
g->ops.gr.reset()

Call the functions directly at all places

Remove CONFIG_NVGPU_DEBUGGER from places where these functions are
called since they are not debugger dependent
This also helps to disable CONFIG_NVGPU_DEBUGGER and to keep recovery
sequence intact

Jira NVGPU-3506

Change-Id: Id2b208ca23dc4667e78edcd8ad242a8558e0ff64
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137255
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vinod Gopalakrishnakurup <vinodg@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-18 01:39:20 -07:00
Deepak Nibade
10fae67c21 gpu: nvgpu: add flag for debugger fields in struct gk20a 
Add CONFIG_NVGPU_DEBUGGER flag for debugger specific fields in struct
gk20a

Jira NVGPU-3506

Change-Id: Icfae87e16e0079a2c5f16714b8a8ced7c6572cd4
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137254
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-18 01:39:10 -07:00
Deepak Nibade
67350e2c9c gpu: nvgpu: add flags to debugger specific headers 
Add debugger/cyclestats/fecs_trace compile time flags to debugger
specific unit headers

Jira NVGPU-3506

Change-Id: Iedea5f274243a389dce91edecbc80c58753d4805
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137253
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-18 01:38:54 -07:00
Deepak Nibade
064f7a2f13 gpu: nvgpu: add debugger flag for netlist units 
Add CONFIG_NVGPU_DEBUGGER flag for debugger specific code in
common.netlist and common.sim.netlist units

Jira NVGPU-3506

Change-Id: I616eaea58e72ff104fef11140a0daa59afe7b5fb
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137252
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vinod Gopalakrishnakurup <vinodg@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-18 01:38:39 -07:00
Deepak Nibade
27a133aa4c gpu: nvgpu: add debugger flag for common.hal.ltc unit 
Add CONFIG_NVGPU_DEBUGGER flag for debugger specific code in
common.hal.ltc unit

Jira NVGPU-3506

Change-Id: I7a330cc60ea90e6b76bd1f783bcecd649032e279
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137251
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vinod Gopalakrishnakurup <vinodg@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-18 01:38:28 -07:00
Rajesh Devaraj
ab70c2e80f gpu: nvgpu: report class/method related errors 
This patch adds support to report class/method related errors to 3LSS.
Specifically, it adds the following service ID:
NVGUARD_SERVICE_IGPU_PGRAPH_SWERR_ILLEGAL_ERROR

JIRA NVGPU-3458
JIRA NVGPU-3461

Change-Id: I9b28ed3074f664254347e059ac699470f95610b3
Signed-off-by: Rajesh Devaraj <rdevaraj@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2136301
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Raghuram Kothakota <rkothakota@nvidia.com>
Reviewed-by: Ankur Kishore <ankkishore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-18 01:37:43 -07:00
Sagar Kamble
b7061a3263 gpu: nvgpu: compile out changes for dgpu falcons 
SW handling of dgpu falcons GSPLITE, NVDEC, SEC2, MINION needs to be
compiled out in the igpu safety build. Also compile out gp106 falcon
and nvdec sources.

JIRA NVGPU-3539

Change-Id: If4d21cec151b6c00f944457dc6cae4f457043b04
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2137226
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-17 23:16:00 -07:00
Divya Singhatwaria
203120deed gpu: nvgpu: Fix CERTC INT31-C violations in ACR 
CERT-C INT31 requires checking before casting from
"bool" to "unsigned int", "unsigned long" to "unsigned int"
otherwise it may result in lost or misinterpreted data.
This fixes INT31 violations in acr_blob_construct_v1.c

JIRA NVGPU-3575

Change-Id: I6da67a64758974f9fd04facac358a59782168470
Signed-off-by: Divya Singhatwaria <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2136255
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: Nitin Kumbhar <nkumbhar@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-17 22:15:35 -07:00
Nicolas Benech
6fb619aa30 gpu: nvgpu: unit: stage some nvgpu_channel tests 
The nvgpu_channel.setup_bind and enable_disable_tsg are causing
intermittent crashes in GVS, so stage them for now. Due to the
dependencies of the test functions, they just returns SUCCESS for now
(instead of disabling them entirely)

JIRA NVGPU-3640

Change-Id: I546b07b4633198f1a2c56e59e11eaa3355b1b285
Signed-off-by: Nicolas Benech <nbenech@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2136820
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
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>
 2019-06-17 11:39:37 -07:00
ajesh
b05a529219 gpu: nvgpu: Add safe ops for s64 
Add safe addition and multiplication functions for s64.

Jira NVGPU-3607

Change-Id: I8078679ee906dfcfcdab24ca221ec4e6b27e58db
Signed-off-by: ajesh <akv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2133656
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Nitin Kumbhar <nkumbhar@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>
 2019-06-17 04:06:25 -07:00
Vedashree Vidwans
6f37ac5de2 gpu: nvgpu: Disable logging for safety build 
This patch adds a conditional flag to filter out logging functions from
safety release build. Logging functions are replaced with stubs.

Jira NVGPU-869

Change-Id: If898b9ce8edb260727df28b407df83f0a92f61ad
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2109509
Reviewed-by: Alex Waterman <alexw@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>
 2019-06-17 00:16:03 -07:00
Adeel Raza
d8b2e22de9 gpu: nvgpu: unit: set fixed offset flag for VM tests 
Set the NVGPU_VM_MAP_FIXED_OFFSET flag when GPU VA is fixed.

JIRA NVGPU-3631

Change-Id: I8b0cc79e19367e2c729bc7d0744fd08d7c81e6c2
Signed-off-by: Adeel Raza <araza@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2135433
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
 2019-06-16 20:44:54 -07:00