Supriya dbc46f0bf2 gpu: nvgpu: gm20b: WPR size 0, on railgate exit ...

Bug 200066741

ACR ucode has mechanism to skip WPR blob copy for second time,
in case WPR size is sent as 0 to acr ucode.
With above there is a saving of around 0.5 ms, but, in
conjunction with acr change to disable LS sig verification,
and scrubbing empty spaces in WPR sections to 0. This change
can reduce railgate exit latency by 4ms
ACR ucodes to be checked in main, as a different CL, and after
getting prod signs for ACR

Change-Id: I9d662027abf0b2615176d17433ff3ec3ae53d78a
Signed-off-by: Supriya <ssharatkumar@nvidia.com>
Reviewed-on: http://git-master/r/681892
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>

2015-04-04 18:06:41 -07:00

..

gpu/nvgpu

gpu: nvgpu: gm20b: WPR size 0, on railgate exit

2015-04-04 18:06:41 -07:00