chore: remove system76-scheduler and fix resume

* system76-scheduler was unused. Or it did not really make much of a difference
  over the default scheduler.
* added `iommu=soft` to kernel param to make the ssd work fine after suspend.
  This might be related to the pcie_aspm policy we had set before. I need to test
  if removing the aspm policy has fixed this issue.

Signed-off-by: Chinmay D. Pai <chinmaydpai@gmail.com>
This commit is contained in:
Chinmay D. Pai 2024-09-23 11:48:19 +05:30
parent ff91127e85
commit 8dc7bf18d3
Signed by: thunderbottom
GPG Key ID: 75507BE256F40CED
2 changed files with 7 additions and 4 deletions

View File

@ -25,11 +25,8 @@
fileSystems = ["/"];
};
services.system76-scheduler.enable = true;
services.system76-scheduler.settings.cfsProfiles.enable = true;
# Power management, enable powertop and thermald.
powerManagement.powertop.enable = true;
powerManagement.cpuFreqGovernor = lib.mkDefault "schedutil";
services.thermald.enable = true;
# TODO: remove, temporary for mongoDB build

View File

@ -20,13 +20,19 @@
"iwlwifi"
"xe"
];
kernelPackages = lib.mkDefault pkgs.linuxPackages_testing;
kernelPackages = lib.mkForce pkgs.linuxPackages_testing;
kernelParams = [
# NixOS produces many wakeups per second, which is bad for battery life.
# This disables the timer tick on the last 7 cores.
"nohz_full=14-21"
"i915.force_probe=!7d55"
"xe.force_probe=7d55"
"initcall_blacklist=simpledrm_platform_driver_init"
# This solves an issue with resume after suspend where the SSD goes into
# a read-only state. We trust some random, obscure Arch wiki article over
# actually trying to figure out why it might be so.
# ref: https://wiki.archlinux.org/title/Solid_state_drive/NVMe#Controller_failure_due_to_broken_suspend_support
"iommu=soft"
# "resume_offset=2465529"
];
# resumeDevice = "/dev/disk/by-uuid/870fde90-a91a-4554-8b1c-d5702c789f4d";