- Update uscript to check oc, max cpu and gpu freqs
- Update init to poke node to allow max scaling
Perf mode on perfmgr already allows full kernel scaling, so this
will just allow that to rise to max.
A better impl later may involve editing the json to have higher
game profiles etc, but for now this is more than fine.
Change-Id: I669fa523944dc0a99c2c9d6a6360593ee0a4bd8c
- Exercise greater control over hinting, power profiles, specific
behavior, etc.
- Better support in the future vs maintaining phs
- Totally nuke nvcpl to remove variables in display handling and
power profiles
Change-Id: If9076248fcacfdc8165bd814ad78786134997b97
* Makes sepolicy far easier.
* Use `exec_start` instead of `start` to ensure jc_setup exits before
`umount` is run.
Change-Id: I64d4f4eac991310db5ba9444044566b04bfbbedb
* We are aren't launched with Android at all to, but
our partitions were origianlly set up to match
foster, as it is our closest hardware counterpart.
* Those partition sizes are now to small to fit
almost any form of GMS.
* The fact that users have to exploit their console,
run a custom first stage bootloader (Hekate), and
applications bootloader (u-boot) that we control
means that instead of retrofitting dynamic partitions
we can just _launch with them_ (as we have yet to ship
this device).
* Treat this device as if it has always had dynamic
partitions and let Hekate set us up to reflect that.
* Do away with SAR in the process, as Hekate can turn
that off.
* Nuke MDA, we don't support hardware backed encryption
anyway, so it doesn't make any sense to house.
Reference: https://github.com/CTCaer/hekate/pull/922
Change-Id: I1e5d8c309d08559873ea3c93ac5b4b6e197b528e
Since this is built out of tree, it is not included in modules.dep,
thus modprobe does not work properly.
Change-Id: I3071e641040d777f82f6f42f136540e0498b33a1
Set up in polling mode--trigger and event mode will need testing. Will
also require testing for generalized calib profile for max range.
Change-Id: I8e58cac0288ba8b53f3451c239c15c0a906db8d4
* On odin boards, factory is used for mounting the hos-data
partition on the SD card which is FAT32.
* Just label in fstab instead.
Change-Id: I3477813906c5ce37df87bf9b170cc21d0fe1300b
This works together with a change to U-Boot to allow kernel and
userspace HDMI-CEC functionality to be disabled manually (or
automatically on vali). Now users who do not care about CEC do not
have to suffer from the CEC volume UI platform bug.
Change-Id: I34a89c78052193c38dc8217841a816615f135fb2
* Again, we do not plan to support vali display hacks for the
time being. If so, we will add env and do it that way.
* filter-out CEC for vali because common trees assume that unified
devices will all support CEC or none.
Change-Id: I1df0d895de870a3ebd76825fe278997e83435922
* We don't care about other nv devices.
Co-authored-by: Nolen Johnson <johnsonnolen@gmail.com>
Co-authored-by: Thomas Makin <halorocker89@gmail.com>
Co-authored-by: Billy Laws <blaws05@gmail.com>
Change-Id: Ib2f7711c87df2b25d8497eebdb9e71975566a3ab
* We don't need `update_js_touch_fw.sh` which is the only
thing housed in `init.loki_e_common.rc`.
Change-Id: I0025fc0fd7297af276964a73e54195cd30513be6
* This is a tree for all Nintendo Switch variants:
`odin`, `modin`, `vali`, and `frig`.
* `ro.hardware` will be NX on all variants with
`ro.hardware.sku` being set per variant by u-boot.
Changes are as follows:
* At the moment the packaged DTB is only for `odin`, more to follow.
* Note: We may make `modin`, `vail`, and `frig` device-tree-overlays
on the `odin` base DTB to simplify things.
* Use darcy's thermal config as it's the most lenient, and is a same SoC STB.
* Unlike `icosa_sr` previously did, we will use the BCM HAL for BT over
BTLinux, as A2DP is tested and verified, no need to deviate.
Changes that will require follow-up commits to build out:
* Update init to reflect nx's needs - will need more work to support devices
other than `odin`.
Removals are as follows:
* Remove all non-nx device support files.
Co-authored-by: Thomas Makin <halorocker89@gmail.com>
Co-authored-by: Billy Laws <blaws05@gmail.com>
Co-authored-by: Aaron Kling <webgeek1234@gmail.com>
Co-authored-by: CTCaer <ctcaer@gmail.com>
Co-authored-by: John Smith <nemui3353@gmail.com>
Co-authored-by: langerhans <langerhans@users.noreply.github.com>
Co-authored-by: Nolen Johnson <johnsonnolen@gmail.com>
Change-Id: I9bab4d94c6e94a9c6e2a7cf04d16c4b76c0b6ac5
The light hal is currently not picking up the lightbar since the
module finishes loading after the hal starts.
Change-Id: I7fd77d1e93a1966d98879aa98ca565247d7479c0
tmakin
Nolen Johnson
Aaron Kling
Billy Laws
