Deepak Nibade 00aeab6cca gpu: nvgpu: add gpc_mask to gr/config unit ...

We get gpc_mask by calling GR HAL g->ops.gr.get_gpc_mask()

But gpc_mask should be logically owned by gr/config unit
Hence add new gpc_mask field to nvgpu_gr_config

Initialize it in nvgpu_gr_config_init() by calling a new HAL
g->ops.gr.config.get_gpc_mask() if available
If HAL is not defined we just initialize it based on gpc_count

Expose new API nvgpu_gr_config_get_gpc_mask() to get gpc_mask
and use this API now

Remove gr_gm20b_get_gpc_mask() and HAL g->ops.gr.get_gpc_mask()

Update GV100 and TU104 chip HALs to remove old and add new HAL

Add gpc_mask to struct tegra_vgpu_constants_params to support this
on vGPU. Also get gpc_mask from vGPU private data in
vgpu_gr_init_gr_config()

Jira NVGPU-1879

Change-Id: Ibdc89ea51df944dc7085920509e3536a5721efc0
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2016084
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

2019-02-14 02:28:58 -08:00

..

nvgpu

gpu: nvgpu: add gpc_mask to gr/config unit

2019-02-14 02:28:58 -08:00

Makefile

nvgpu: use kernel overlay features

2017-10-29 11:00:55 -07:00