Updated following hal functions for css gv11b and reused
them for gv100:
enable_snapshot
disable_snapshot
check_data_available
These changes are needed because of following reasons:
1. Register offsets for perf_pmasys_* are changed
for gv11b/gv100 from gk20a.
2. Updated memory type for perf_pmasys_mem_block_target
based on memory aperture used for hwpm inst_block.
Bug 200327596
Change-Id: I500d17670e2f389d8d0e77884374bcc3504a41f8
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1507546
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
SMID tables were generated according with the local tpc and the pagepool and cb
buffers from a different chip and did not take performance in consideration,
which made compute kernels hang with CTAs on the fly.
This change ensures we are using the right sizes and adds proper enumeration
of smids.
JIRA: NVGPUGV100-36
bug 2004378
Change-Id: Ic8f50c325d6d6720cca41d9740ae4f5f51e1100a
Signed-off-by: David Nieto <dmartineznie@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1581664
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Handle the possibility of failing gr init due to smid table initialization
failures
bug 2004378
Change-Id: I904b918a0ea31c32292edb3ab8ac3b1459c38a28
Signed-off-by: David Nieto <dmartineznie@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1581661
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Sched has been moved to be part of Linux implementation, and at the
same time sched_ctrl has been moved to be part of nvgpu_os_linux.
JIRA NVGPU-259
Change-Id: I4c1869628ad716bcd903ba99db926a8f8723828d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1580650
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Move sched parameter APIs to be Linux specific implementation. At
the same time the sched_ctrl fields were moved to nvgpu_os_linux.
JIRA NVGPU-259
Change-Id: I2397e2602e1c4783f2bebf3aec462634b7f86d4a
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1580649
GVS: Gerrit_Virtual_Submit
Use mc reset HAL function to reset perfmon unit.
This will take care of delays after writing to
mc_enable register.
Change-Id: I6118bdd2a9fc9192f7ffc44617a170b7982c26ba
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1581753
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
ecc_gk20a.h has no use for Linux IOCTL interface definitions. Remove
the stray #include.
JIRA NVGPU-259
Change-Id: I47a837ac86f1f83976d94cb9b30082fee827445a
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1580678
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Change required for equivalent change on nvgpu. This is required
since a few HALs were added that must be populated for all chips.
This patch adds those HAL definitions for gv11b, gv100, and the
vgpu.
JIRA NVGPU-30
JIRA NVGPU-138
Change-Id: I65374764350a5cacce8624b15d98947fada35a4a
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1579865
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Refactor the last nvgpu_vm functions from the mm_gk20a.c code. This
removes some usages of dma_buf from the mm_gk20a.c code, too, which
helps make mm_gk20a.c less Linux specific.
Also delete some header files that are no longer necessary in
gk20a/mm_gk20a.c which are Linux specific. The mm_gk20a.c code is now
quite close to being Linux free.
JIRA NVGPU-30
JIRA NVGPU-138
Change-Id: I72b370bd85a7b029768b0fb4827d6abba42007c3
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1566629
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Move most of the dma_buf usage present in the mm_gk20a.c code
out to Linux specific code and some commom/mm code. There's
two primary groups of code:
1. dma_buf priv field code (for holding comptag data)
2. Comptag usage that relies on dma_buf pointers
For (1) the dma_buf code was simply moved to common/linux/dmabuf.c
since most of this code is clearly Linux specific.
The comptag code was a bit more complicated since there is two
parts to the comptag code. Firstly there's the code that manages
the comptag memory. This is essentially a simple allocator. This
was moved to common/mm/comptags.c since it can be shared across
all chips. The second set of code is moved to
common/linux/comptags.c since it is the interface between dma_bufs
and the comptag memory.
Two other fixes were done as well:
- Add struct gk20a to the comptag allocator init so that
the proper nvgpu_vzalloc() function could be used.
- Add necessary includes to common/linux/vm_priv.h.
JIRA NVGPU-30
JIRA NVGPU-138
Change-Id: I96c57f2763e5ebe18a2f2ee4b33e0e1a2597848c
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1566628
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Updated following hal functions for gv11b and reused
them for gv100:
perfbuffer_enable
perfbuffer_disable
These changes are needed because of following reasons:
1. Register offsets for perf_pmasys_* are changed
for gv11b/gv100 from gk20a.
2. Updated memory type for perf_pmasys_mem_block_target
to sys_ncoh_f().
Bug 200327596
Change-Id: Ia672ac561917c8ed36caea9cc7e74b7fc7ce8188
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1571074
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Use abstract nvgpu_rwsem as TSG channel list lock instead of the Linux
specific rw_semaphore.
JIRA NVGPU-259
Change-Id: I5f6c918464315e3d140bea0c61a619c3712619c1
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1579934
GVS: Gerrit_Virtual_Submit
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
This change contains a generic cleanup of linux dependent parts of the
dbg_gpu_gk20a.* files. The following changes have been made
1) Moving methods into ioctl_dbg.* inside common/linux/
2) The structures dbg_session_gk20a and dbg_session_channel_data have
been split into two parts. struct device *dev is removed from
struct dbg_session_gk20a and instead packed into struct
dbg_session_gk20a_linux alongwith dbg_session_gk20a and is moved into
ioctl_dbg. dbg_session_gk20a is now rid of any linux dependencies and
remains in dbg_gpu_gk20a. Similarly, struct file is removed from
struct dbg_session_channel_data and is now packed into struct
dbg_session_channel_data_linux alongwith dbg_session_channel_data and
is moved into ioctl_dbg. struct dbg_session_channel_data is now rid of
linux dependencies and remains in dbg_gpu_gk20a.
3) A callback function is added in order to release the
dbg_session_channel_data.
JIRA NVGPU-205
Change-Id: I853da6dfbf9a96b7cd210beb77f2304445ff7ea6
Signed-off-by: Debarshi Dutta <ddutta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1575191
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The patch has the changes for compilation of
common nvgpu allocator for QNX.
This includes some cross-OS compilation changes
and removing some Linux'isms from the allocator.
Change-Id: Ib1ecceec77b497513a196597bff4441615577548
Signed-off-by: Sourab Gupta <sourabg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1540306
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
- Allocate memory for platform data for
each dgpu device detected on PCIe,
- Copy detected device static data to allocated
platform data
- Free allocated space in nvgpu_pci_remove()
Issue: Static platform data is overwritten When two
same SKU/device-id dGPU connected on single platform which
cause accessing wrong dGPU space,
Fix: Fixing issue by allocating space dynamically for each
dGPU device detected & copy data from detected dGPU static
data.
JIRA NVGPUGV100-18
Change-Id: Idf2d2d6d2016b831c21b9da6f8ee38b34304bd12
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1577913
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
linux/debugfs.h was included in gk20a.h because of the debugfs entry
bios_blob, which can be used for checking contents of VBIOS. That
has never been used, so instead of abstracting it, this patch removes
the feature altogether.
Two files were using debugfs but did not #include <linux/debugfs.h>.
They failed to build now that gk20a.h no longer #includes it, so
added explit #include.
JIRA NVGPU-259
Change-Id: Ie1ea9be1a8920441b1616f34e64e505e6e10e38c
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1570404
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
In gk20a.h, we use these multichar constants
'DONE', '0R32', '0W32'
But these constants fail compilation on some cross-os compilers
Hence remove them by creating a specific MULTICHAR_TAG()
All the User space components also form these constants in similar
fashion
Jira NVGPU-259
Change-Id: Ibe83617a8d15b657e450fdd59dbc171b3d822842
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1576933
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Forward declare struct gk20a_debug_output in gr_gp10b.h since we do not
explicitly include any header file for same
Jira NVGPU-259
Change-Id: I9a4048591363e2ddbe7e8b3c2e56ab1e4eae0b3a
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1576931
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
In include/nvgpu/pramin.h, we right include linux specific header
linux/types.h.
But since this is a common header, we should include nvgpu/types.h
which is generic and linux independent
Jira NVGPU-259
Change-Id: I0acff92b3de9431e615faa6de26306cccd5ca443
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1576930
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
We use struct stack_trace in struct channel_gk20a_ref_action
But since channel_gk20a_ref_action is needed only if
GK20A_CHANNEL_REFCOUNT_TRACKING is set, protect it with that config
Jira NVGPU-259
Change-Id: I6b2d6f470bf924bb1ddfd31ba9968b56c63c2372
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1576929
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Conditionally compile debug_kmem.c in the Makefile since the entire
file is not needed when CONFIG_NVGPU_TRACK_KMEM_USAGE is not enabled.
Removing as many conditional compilation flags from the C code itself
is highly desirable from a code complexity standard.
Change-Id: If1e87986ca1ee6d71485f1ab40f10c1645c1a628
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1576543
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Instead of calling the native HAL init function then adding
multiple layers of modification for VGPU, flatten out the sequence
so that all entry points are set statically and visible in a
single file.
JIRA ESRM-30
Change-Id: I8d277aaccb0e63b2d504e7aba32eb31ef82f4ec0
Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1574619
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Richard Zhao <rizhao@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Instead of calling the native HAL init function then adding
multiple layers of modification for VGPU, flatten out the sequence
so that all entry points are set statically and visible in a
single file.
JIRA ESRM-30
Change-Id: Ie424abb48bce5038874851d399baac5e4bb7d27c
Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1574616
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add abstraction of IO aperture accessors. Add new functions
gk20a_io_exists() and gk20a_io_valid_reg() to remove dependencies to
aperture fields from common code.
Implement Linux version of the abstraction by moving gk20a_readl()
and gk20a_writel() to new Linux specific io.c. Move the fields
defining IO aperture to nvgpu_os_linux.
Add t19x specific IO aperture initialization functions and add t19x
specific section to nvgpu_os_linux.
JIRA NVGPU-259
Change-Id: I09e79cda60d11a20d1099a9aaa6d2375236e94ce
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1569698
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reduce the usage of nvgpu_vidmem_get_page_alloc() and friends as much
as possible. This reduces the dependency of nvgpu on Linux SGLs. SGLs
still need to be used, however, since sharing buffers in userspace is
done by dma_buf FD. The best way to pass the vidmem buf through the
dma_buf is by SGL pointer.
JIRA NVGPU-30
JIRA NVGPU-138
Change-Id: Ide0e9e5a557f00aa63b063be085042101a5b34ee
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1540709
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Split the core vidmem allocation from the Linux component of vidmem
allocation. The core vidmem allocation allocates the nvgpu_mem struct
that defines the vidmem buffer in the core MM code. The Linux code
now allocates some Linux specific stuff (dma_buf, etc) and also
allocates the core vidmem buf.
JIRA NVGPU-30
JIRA NVGPU-138
Change-Id: I88e87e0abd5ec714610eacc6eac17e148bcee3ce
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1540708
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
For sync-point read map, create read only map per vm
and share with all channels that are using same vm.
Now restrict rw map to single syncpoint shim memory range.
JIRA GPUT19X-2
Change-Id: Ibd0b82d1cdb8861e1dbb073b27da1f9c9ab1d2ab
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1514339
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
For sync-point read map:
1. Added nvgpu_mem memory allocator in gk20a struct and
allocated memory for this in gk20a_finalize_poweron()
and freed this memory in gk20a_remove().
2. Added "u64 syncpt_ro_map_gpu_va" in vm_gk20a struct
for read map in vm.
Added nvgpu_quiesce() in nvgpu_remove() before freeing
syncpoint read map to ensure that nvgpu is idle.
JIRA GPUT19X-2
Change-Id: I7cbfec57f0992682dd833a1b0d92d694bcaf1eb3
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1514338
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Added new memory API _nvgpu_mem_create_from_phys for
creating nvgpu_mem from physical memory aperture.
With this new API, avoided usage of linux specific
"struct page" in general code and moved this code to
common linux code. This API internally uses
__nvgpu_mem_create_from_pages for creating nvgpu_mem
from physical pages.
JIRA GPUT19X-2
Change-Id: Iaf0193a7c33e71422e4ddabde01edf46f5a81794
Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1571073
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
