Use physical addresses in PDEs. All page table levels fit in 4k, so no
need for SMMU mapping.
Change-Id: Id9e418f35a79343f4a332a230e04abda5e0dd5d2
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/783748
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Fixed the following sparse warning by making the local function as static:
- symbol 'gp10b_pmu_load_multiple_falcons' was not declared.
Should it be static?
- symbol 'gp10b_load_falcon_ucode' was not declared.
Should it be static?
bug 200067946
Change-Id: I67d865aef6f57bf614db351929cd4bb1b6077c00
Signed-off-by: Vijayakumar <vsubbu@nvidia.com>
Reviewed-on: http://git-master/r/764646
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Amit Sharma (SW-TEGRA) <amisharma@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Implement support for privileged pages. Use them for kernel allocated buffers.
Change-Id: I24778c2b6063b6bc8a4bfd9d97fa6de01d49569a
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/761920
Use always physical addresses for page tables. In gp10b new format
each level fits in one page, so we do not need SMMU translation.
Change-Id: Ie46b2bce0f7a4e8d2904d74b1df616e389874141
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/758181
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Add regops whitelists for gp10b. The whitelist is generated, and is the
same for context switched and global registers.
Bug 1633363
Change-Id: I6d4d43d036d684c9f0d836a1a032f2c452604902
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/760935
gm20b clock registers do not exist in gp10b. Skip setting the clock
HAL to gm20b variants.
Change-Id: Ieaa9a04a8afbe772864d947d968e3e1c7f9968e9
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/760854
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Update sync point protection field only when we have a valid sync
point id, and the new id is different from old id.
Bug 1653328
Change-Id: Ie07e26f8abd7c8239ad562603b62fda00164cbc7
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/757102
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Alpha and beta sizes need to be clipped to a maximum value. For
alpha CB we were using beta size in clipping, and for both we were
not using number of TPCs to determine the max value.
Change-Id: I0c925464ba4c9f575e6e59dd5ba7759aa1cb6381
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/752667
Reviewed-by: Automatic_Commit_Validation_User
Some fields have different widths, so duplicate the code to program
global bundle CB.
Change-Id: Ib6af5abf3e90dfa1bcda2fbc6b97ad1031e6ab16
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/752635
When allocating betacb for a GfxP channel, add both alpha and beta
cb sizes together.
Change-Id: I8cef62f6272bfb3b5e9a3835a51590e5eb91dc92
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/752633
Reviewed-by: Automatic_Commit_Validation_User
Compbit backing store did not take into account number of GOBS
per comptagline per slice.
Bug 1604102
Change-Id: I42666e72ea54697b6fbc7318e65a6a09d867f5b6
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/754706
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sami Kiminki <skiminki@nvidia.com>
Kick channel off PBDMA before writing new sync point id to allowed
sync points.
Bug 1648297
Bug 1646477
Change-Id: I7c686d474c403fdd54bc64cff63b7d049feecb4d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/750981
Change for new VA space allocator is being reverted with
http://git-master/r/#/c/749291/ but only for Kernel3.18
In Kernel3.10, we support the new VA allocator
Since we support both the kernel versions as of now,
use a KERNEL_VERSION based mechanism to select
appropriate call
Define new macro NVGPU_USE_NEW_ALLOCATOR for Kernel3.10
where we want to use new allocator
Bug 200106514
Change-Id: I9af26d555278c40e03fe82b0912961a862c8bf55
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/751353
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
The produced wrappers for HW PM registers access which are required for
cyclestats support for snapshot buffers mapping.
See commit 589e7a9ffe2a5a70f8803a88fcf8429f553e2fba for tools:nvhost
generators update.
Bug 1573150
Bug 1517458
Change-Id: I9c9332a55f2282c0c626bc8ddbcfdce1289f778b
Signed-off-by: Leonid Moiseichuk <lmoiseichuk@nvidia.com>
Reviewed-on: http://git-master/r/747717
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
ZBC is safe to update and GPU is safe to rail gate when units are
in preempted or empty state. Idle may never be reached in case of
graphics preemption, so relax the ZBC update wait condition.
Bug 1640378
Change-Id: I40c59e9af22a7a30b777c6b9f87e69d130042e44
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/745655
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
We were dropping the part of address that span word bounary. The register
generator does not know how to real with multi-word fields, to edit things
in manually.
Bug 1646531
Change-Id: I3ef06d6dfcb0a499ed45456d165fe60c91492250
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/747468
This reverts commit 30e5947fa1f26ed6bb4f137fd76c8869e91b9829.
The original commit was actually fine.
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Change-Id: I0454415981d29ed0b877f7a21db6f54bc4c30470
Reviewed-on: http://git-master/r/743302
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
The comptag allocator is made in the chip-specific init code
for the comptags. Thus, a t18x change needs to be made to make
sure the new allocator code compiles and works on t18x.
Change-Id: I57a34f3c61ebd31f875caa577378e829812f2d4c
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/721171
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
VPR allocator needs to be used when allocating graphics context for
VPR channels. Define it for gp10b.
Bug 1625090
Change-Id: Ie2e3a865c310c34c629627891ac0b579f299983f
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/737846
Reviewed-by: Automatic_Commit_Validation_User
Used 128k comptag spacing, when 64k is the correct one.
Bug 1525976
Change-Id: Ie2f926929fa89cf715b86a57ffbf4dd1e4920473
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/737947
Enable new page table format for all platforms.
Bug 1525976
Change-Id: I9a3cfabdef7dc6ec33e18a8a4f32063c40f680fa
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/737364
If betacb size has been given via debugfs, use that instead of the
calculated number.
Bug 1628352
Change-Id: I8c68c27a2bfdd7f013776734ef846377a89b0033
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/733332
Fix sparse warnings of below type by making necessary
symbols static:
warning: symbol '<symbol>' was not declared. Should it be static?
Bug 200088648
Change-Id: I222bebd958e29b3a95d161f05a3052389200fc10
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/736663
GVS: Gerrit_Virtual_Submit
Reviewed-by: Amit Sharma (SW-TEGRA) <amisharma@nvidia.com>
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Implement the 5-level Pascal page table format. It is enabled
only for simulation.
Change-Id: I6767fac8b52fe0f6a2e2f86312de5fc93af6518e
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/682114
