The last major item preventing the core MM code in the nvgpu
driver from being platform agnostic is the usage of Linux
scattergather tables and scattergather lists. These data
structures are used throughout the mapping code to handle
discontiguous DMA allocations and also overloaded to represent
VIDMEM allocs.
The notion of a scatter gather table is crucial to a HW device
that can handle discontiguous DMA. The GPU has a MMU which
allows the GPU to do page gathering and present a virtually
contiguous buffer to the GPU HW. As a result it makes sense
for the GPU driver to use some sort of scatter gather concept
so maximize memory usage efficiency.
To that end this patch keeps the notion of a scatter gather
list but implements it in the nvgpu common code. It is based
heavily on the Linux SGL concept. It is a singly linked list
of blocks - each representing a chunk of memory. To map or
use a DMA allocation SW must iterate over each block in the
SGL.
This patch implements the most basic level of support for this
data structure. There are certainly easy optimizations that
could be done to speed up the current implementation. However,
this patches' goal is to simply divest the core MM code from
any last Linux'isms. Speed and efficiency come next.
Change-Id: Icf44641db22d87fa1d003debbd9f71b605258e42
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1530867
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
ctx header holds only gpu va for each address space.
All other information will be held in main
context. Ctx header will have gpu va for following
fields:
ctxsw_prog_main_image_context_buffer_ptr
ctxsw_prog_main_image_context_buffer_ptr_hi
ctxsw_prog_main_image_zcull_ptr
ctxsw_prog_main_image_zcull_ptr
ctxsw_prog_main_image_pm_ptr
ctxsw_prog_main_image_pm_ptr_hi
ctxsw_prog_main_image_full_preemption_ptr_hi
ctxsw_prog_main_image_full_preemption_ptr
ctxsw_prog_main_image_full_preemption_ptr_xxxx0
ctxsw_prog_main_image_full_preemption_ptr_xxxx0_v
ctxsw_prog_main_image_patch_adr_lo
ctxsw_prog_main_image_patch_adr_hi
Changes done as part of this CL:
- Read ctx_id from from main context header
- Golden context creation:
Use gold_mem for for golden context creation
and copy golden context from save gold local
memory to main context. No need to restore
golden context to context header.
- Write ctx_patch_count and smpc_ctxsw_mode in
main context header only.
- Update preemption mode in main context header and
preemption buffer va in context header.
- Updated image patch buffer va in context header.
Bug 1958308
Change-Id: Ic076aad8b1802f76f941d2d15cb9a8c07308e3e8
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1562680
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Initialize following counters in context header
for all legacy chips:
ctxsw_prog_main_image_num_save_ops
ctxsw_prog_main_image_num_restore_ops
This was already present in the code but move to a function
gk20a_gr_init_ctxsw_hdr_data, so that it can be re-used across
chips.
Additionally initialize following preemption related counters
for gp10b onwards in context header:
ctxsw_prog_main_image_num_wfi_save_ops
ctxsw_prog_main_image_num_cta_save_ops
ctxsw_prog_main_image_num_gfxp_save_ops
ctxsw_prog_main_image_num_cilp_save_ops
Bug 1958308
Change-Id: I0e45ec718a8f9ddb951b52c92137051b4f6a8c60
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1562654
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reorganize HAL initialization to remove inheritance and construct
the gpu_ops struct at compile time. This patch only covers the
gr 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: Ie37638f442fd68aca8a7ade5f297118447bdc91e
Signed-off-by: Sunny He <suhe@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1542989
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Add a pointer to struct gk20a to the FUSE APIs. This helps
QNX builds avoid any static data definitions.
Also this change plumbs struct gk20a in some of the Linux clk
code and fixes a few minor style nits.
Change-Id: I27dfb2c4e9a352f784d6cead150460d8e9e808d3
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1537611
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Richard Zhao <rizhao@nvidia.com>
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sourab Gupta <sourabg@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Replace privsecurity boolean flag in gpu_ops with entry in
common flag system.
The new common flag is NVGPU_SEC_PRIVSECURITY
Jira NVGPU-74
Change-Id: I4b258f5ffbe30a6344ffba0ece51c6f5d47ebec1
Signed-off-by: Sunny He <suhe@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1525713
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
In gr_gp10b_handle_sm_exception(), we do not check return value of
gr_gk20a_handle_sm_exception()
In case SM debug mode is disabled, gr_gk20a_handle_sm_exception()
returns an error. But due to uncaught error we return success from
gr_gp10b_handle_sm_exception() which results in not triggering the
recovery routine and fails the tests
Fix this by checking return value of gr_gk20a_handle_sm_exception()
We then parse ECC errors and then finally return error value back to
caller
Bug 200325081
Change-Id: I0c328a75494b9ca6a070f04c1f984e2653ffd90e
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1521563
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Tested-by: Ashutosh Jain <ashutoshj@nvidia.com>
The new SET_BES_CROP_DEBUG3 sw method is used to flip two fields
in the NV_PGRAPH_PRI_BES_CROP_DEBUG3 register. The sw method is
used by the user space driver to disable enough ROP optimizations
to maintain ZBC state of target tiles.
Bug 1942454
Change-Id: Id4e4d9d06c6c66080d06b6d4694546fe5cba8436
Signed-off-by: Lauri Peltonen <lpeltonen@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1516202
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This is required to support multiple SM and t19x
sm register address changes
JIRA GPUT19X-75
Change-Id: I844b5cf02a75ba397891a1100d917875e5a3e181
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1512217
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
This is required to support multiple SM and t19x
sm register address changes
JIRA GPUT19X-75
Change-Id: If8805bcc042c75ea70c1689306feb3c8bf011655
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1512216
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
This is required to support multiple SM and t19x
sm register address changes
JIRA GPUT19X-75
Change-Id: Icdae3b6ed67a3d3deeb17f29528184b2d7a70af5
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1512215
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
This is required to support multiple SM and t19x
sm register address changes
JIRA GPUT19X-75
Change-Id: Id104f611736535874cdaa5a2f768f692d799c2c5
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1512214
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
-Add sm input param for handle_sm_exception and
pre_process_sm_exception for gr ops/functions.
-Add functions to calculate gpc and tpc reg offsets.
-Add function to find SMs which raised SM exception.
JIRA GPUT19X-75
Change-Id: I257e7342ddabadb1556c9551c50a54d34b0f9d1e
Signed-off-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-on: https://git-master/r/1476108
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: David Martinez Nieto <dmartineznie@nvidia.com>
Remove <soc/tegra/fuse.h> includes and include
<nvgpu/fuse.h> header to remove direct dependency
on platform specific header
Use specific APIs like below to read/write fuses
nvgpu_tegra_fuse_write_bypass()
nvgpu_tegra_fuse_write_opt_gpu_tpc0_disable()
Remove old code which was compiled for kernel versions
less than 4.4 since we support only k4.4 and greater
versions now
Jira NVGPU-75
Change-Id: Iddd8e1a8da7effbce2aff217e8e25f7de04962d6
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1497518
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Since all debugfs code is Linux specific, remove
it from common code and move it to Linux module
Debugfs code is now divided into below
module specific files :
common/linux/debug.c
common/linux/debug_cde.c
common/linux/debug_ce.c
common/linux/debug_fifo.c
common/linux/debug_gr.c
common/linux/debug_mm.c
common/linux/debug_allocator.c
common/linux/debug_kmem.c
common/linux/debug_pmu.c
common/linux/debug_sched.c
Add corresponding header files for above modules too
And compile all of above files only if CONFIG_DEBUG_FS is set
Some more details of the changes made
- Move and rename gk20a/debug_gk20a.c to common/linux/debug.c
- Move and rename gk20a/debug_gk20a.h to include/nvgpu/debug.h
- Remove gm20b/debug_gm20b.c and gm20b/debug_gm20b.h and call
gk20a_init_debug_ops() directly from gm20b_init_hal()
- Update all debug APIs to receive struct gk20a as parameter
instead of receiving struct device pointer
- Update API gk20a_dmabuf_get_state() to receive struct gk20a
pointer instead of struct device
- Include <nvgpu/debug.h> explicitly in all files where debug
operations are used
- Remove "gk20a/platform_gk20a.h" include from HAL files
which no longer need this include
- Add new API gk20a_debug_deinit() to deinitialize debugfs
and call it from gk20a_remove()
- Move API gk20a_debug_dump_all_channel_status_ramfc() to
gk20a/fifo_gk20a.c
Jira NVGPU-62
Change-Id: I076975d3d7f669bdbe9212fa33d98529377feeb6
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1488902
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
nvgpu_log/info/warn/err() internally add a \n to the end of the message.
Hence, callers should not include a \n at the end of the message. Doing
so results in duplicate \n being printed, which ends up creating empty
log messages. Remove the duplicate \n from all err/warn messages.
Bug 1928311
Change-Id: I99362c5327f36146f28ba63d4e68181589735c39
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-on: http://git-master/r/1487232
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This change adds a new sysfs node to allow configuring CZF_BYPASS, to
enable platforms with low context-switching latency requirements.
/sys/devices/17000000.gp10b/czf_bypass
Values:
0 - always
1 - lateZ (default)
2 - single pass
3 - never
The specified value will apply only to newly allocated contexts.
Bug 1914014
Change-Id: Ibb9a8e86089acaadaa7260b00eedec5c80762d6f
Signed-off-by: Peter Boonstoppel <pboonstoppel@nvidia.com>
Reviewed-on: http://git-master/r/1478567
(cherry picked from commit 3bc022cb38)
Reviewed-on: http://git-master/r/1473820
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Begin moving (and renaming) the GMMU code into common/mm/gmmu.c. This
block of code will be responsible for handling the platform/OS
independent GMMU operations.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: Ide761bab75e5d84be3dcb977c4842ae4b3a7c1b3
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464083
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Added function pointers to check chip specific valid
gfx class and compute class. Also added function pointer
to update ctx header with preemption buffer pointers.
Bug 200292090
Change-Id: I8119ee082e2abb67186a8ac07088f8db7f410ba1
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/1293502
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Hide the Linux specific nvgpu_mem fields so that in subsequent patches
core code can instead of using struct sg_table it can use mem_desc.
Routines for accessing system specific fields will be added as needed.
This is the first step in a fairly major overhaul of the GMMU mapping
routines. There are numerous issues with the current design (or lack
there of): massively coupled code, system dependencies, disorganization,
etc.
JIRA NVGPU-12
JIRA NVGPU-30
Change-Id: I2e7d3ae3a07468cfc17c1c642d28ed1b0952474d
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1464076
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Use platform agnostic delay functions instead of Linux kernel APIs.
This allows removing dependency to Linux header linux/delay.h. At the
same time remove #include lines for other unused Linux headers.
JIRA NVGPU-16
Change-Id: I05df9d72edaf4bb061febe0cb40fc8a7cf9f51c7
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1460114
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
gk20a_err() and gk20a_warn() require a struct device pointer,
which is not portable across operating systems. The new nvgpu_err()
and nvgpu_warn() macros take struct gk20a pointer. Convert code
to use the more portable macros.
JIRA NVGPU-16
Change-Id: I8dc0ddf3b6ea38af6300c27558b60786c163da6d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1457344
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Rename the nvgpu DMA APIs from gk20a_gmmu_alloc* to nvgpu_dma_alloc*.
This better reflects the purpose of the APIs (to allocate DMA suitable
memory) and avoids confusion with GMMU related code.
JIRA NVGPU-12
Change-Id: I673d607db56dd6e44f02008dc7b5293209ef67bf
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1325548
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Make an nvgpu DMA API include file so that the intricacies of the
Linux DMA API can be hidden from the calling code.
Also document the nvgpu DMA API.
JIRA NVGPU-12
Change-Id: I7578e4c726ad46344b7921179d95861858e9a27e
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1323326
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Renaming was done with the following command:
$ find -type f | \
xargs sed -i 's/struct mem_desc/struct nvgpu_mem/g'
Also rename mem_desc.[ch] to nvgpu_mem.[ch].
JIRA NVGPU-12
Change-Id: I69395758c22a56aa01e3dffbcded70a729bf559a
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1325547
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
There were still a few remaining loops where we did not use
nvgpu_timeout and required Tegra specific functions for detecting if
timeout should be skipped. Replace all of them with nvgpu_timeout and
remove including chip-id.h where possible.
FE power mode timeout loop also used wrong delay value. It always
waited for the whole max timeout instead of looping with smaller
increments.
If SEC2 ACR boot fails to halt, we should not try to check ACR result
from mailbox. Add an early return for that case.
JIRA NVGPU-16
Change-Id: I9f0984250d7d01785755338e39822e6631dcaa5a
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1323227
Move clock APIs from gk20a_platform to gpu_ops. At the same time
allow use of internal get_rate/set_rate for querying both GPCCLK
and PWRCLK on iGPU.
At the same time we can replace calls to clk framework with the
new HAL and drop direct dependency to clk framework.
gp10b ops were replaced as a whole at HAL initialization. That
replaces anything set in platform probe stage, so reduce that to
touch only clock gating regs.
Change-Id: Iaf219b1f000d362dbf397d45832f52d25463b31c
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1300113
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Read fuse values from GPU's own fuse registers instead of Tegra fuse
registers whenever possible. This reduces the number of dependencies
to Linux fuse code.
Some fuses do not have a corresponding register in GPU, so they're
left as is.
Change-Id: Id9f2f4da897f3e20b20c300a67f705e3fa5ba35a
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1318278
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Remove restriction of setting compute preemption mode for channels
created with PASCAL_COMPUTE_A class only and allow it to be set
for PASCAL_A class too.
Also print compute preemption mode during channel closing.
Bug 200284575
Change-Id: I2de3b3acda128e91caa2ab0fd341915ce6e6520b
Signed-off-by: Sandeep Shinde <sashinde@nvidia.com>
Reviewed-on: http://git-master/r/1313286
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: Donghan Ryu <dryu@nvidia.com>
Instead of using Linux APIs for mutex and spinlocks
directly, use new APIs defined in <nvgpu/lock.h>
Replace Linux specific mutex/spinlock declaration,
init, lock, unlock APIs with new APIs
e.g
struct mutex is replaced by struct nvgpu_mutex and
mutex_lock() is replaced by nvgpu_mutex_acquire()
And also include <nvgpu/lock.h> instead of including
<linux/mutex.h> and <linux/spinlock.h>
Add explicit nvgpu/lock.h includes to below
files to fix complilation failures.
gk20a/platform_gk20a.h
include/nvgpu/allocator.h
Jira NVGPU-13
Change-Id: I81a05d21ecdbd90c2076a9f0aefd0e40b215bd33
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1293187
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This ioctl can be used on gp10b to set a flag in the context header
indicating this context should be run at elevated clock
frequency. FECS ctxsw ucode will read this flag as part of the context
switch and will request higher GPU clock frequencies from BPMP for the
duration of the context execution.
Bug 1819874
Change-Id: I84bf580923d95585095716d49cea24e58c9440ed
Signed-off-by: Peter Boonstoppel <pboonstoppel@nvidia.com>
Reviewed-on: http://git-master/r/1292746
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Move semaphore_gk20a.c drivers/gpu/nvgpu/common/ since the semaphore
code is common to all chips.
Move the semaphore_gk20a.h header file to drivers/gpu/nvgpu/include/nvgpu
and rename it to semaphore.h. Also update all places where the header
is inluced to use the new path.
This revealed an odd location for the enum gk20a_mem_rw_flag. This should
be in the mm headers. As a result many places that did not need anything
semaphore related had to include the semaphore header file. Fixing this
oddity allowed the semaphore include to be removed from many C files that
did not need it.
Bug 1799159
Change-Id: Ie017219acf34c4c481747323b9f3ac33e76e064c
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1284627
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
