Commit Graph

5 Commits

Author SHA1 Message Date
dt
e1d6b8af8d gpu: nvgpu: ga10x: compute gnic_stride
GNIC register stride calculation is fixed by adding new hal to compute
the stride by getting the difference of gpc1 and gpc0 xbar_gnic strides
for ga10x GPUs.

Bug 200782045

Change-Id: Iaa84109bd9f1a974ef1af6fee136ca1fcc89bbb1
Signed-off-by: Dinesh T <dt@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2624848
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-12-01 08:40:36 -08:00
Antony Clince Alex
f9cac0c64d gpu: nvgpu: remove nvgpu_next files
Remove all nvgpu_next files and move the code into corresponding
nvgpu files.

Merge nvgpu-next-*.yaml into nvgpu-.yaml files.

Jira NVGPU-4771

Change-Id: I595311be3c7bbb4f6314811e68712ff01763801e
Signed-off-by: Antony Clince Alex <aalex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2547557
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2021-06-27 05:02:53 -07:00
Deepak Nibade
b23a114c63 gpu: nvgpu: ensure all perfmon writes are complete after reset
gr_gv100_reset_hwpm_pmm_registers() writes a bunch of registers in
sys/gpc/fbp chiplets to reset perfmons. To ensure all the writes have
completed it is necessary to readback each chiplet's PRI fence register.

Add and use new HAL g->ops.priv_ring.read_pri_fence() to achieve this.

Implement the HAL for gv11b in new source code file
hal/priv_ring/priv_ring_gv11b.c.

Bug 2510974
Jira NVGPU-5360

Change-Id: If4dd61cb4265422e8c2d16884790eb0fe7f2c103
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2453631
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2020-12-15 14:13:48 -06:00
Sagar Kamble
19c250e569 gpu: nvgpu: update hw headers with static inlines
There were following stale hw header files as their source script
changes were not present. These still had the static inline
functions.

Deleted below files as they are not used:
gk20a/hw_pri_ringstation_fbp_gk20a.h
gp106/hw_gc6_gp106.h

Regenerated with updated script changes:
gv11b/hw_usermode_gv11b.h

JIRA NVGPU-3733

Change-Id: I40b79b43b7f085c01858f3584fcf2c8928d62d13
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2152825
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>
2019-07-17 00:29:52 -07:00
Alex Waterman
6efe02f776 gpu: nvgpu: Add YAML arch to linux-nvgpu
Add the YAML arch to the linux-nvgpu repo so that the arch and source
code can be kept entirely in sync.

When the YAML was in a separate repo it was too difficult to keep things
in sync. Based on this work the next steps are:

  1. Gate compilation on no sources issues (missing or orphan source
     files.
  2. Perhaps use the arch to generate the Makefile.sources files.

JIRA NVGPU-3074
JIRA NVGPU-3075
Bug 2587555

Change-Id: Ic62be3e5e7ea1fb61ea50a2d611f905fa574988f
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2113154
Reviewed-by: Philip Elcan <pelcan@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>
2019-05-13 10:47:44 -07:00