fix: legolas NVMe suspend/resume crashes

Switch mem_sleep_default from deep (S3) to s2idle - the Phison E21
NVMe hangs on resume from the D3 transition S3 forces. Remove the
no-op d3cold_allowed udev rule (attribute doesn't exist) and the
broken nvme-resume-fix service (ExecStart redirect never ran). Add a
correct runtime-PM-off rule for the NVMe endpoint and its parent root port.

developmentlog.md

@@ -0,0 +1,18 @@
+# Development log
+
+## legolas: NVMe (Corsair MP600 ELITE / Phison E21) suspend/resume crashes
+
+The drive hung on resume from S3 (deep) suspend — boot logs ended at `PM: suspend entry (deep)` with no resume, followed by a hard reboot. The Phison E21 can't recover from the D3 transition that S3 forces.
+
+Investigation also found two existing workarounds were doing nothing:
+
+- The udev rule `ATTR{d3cold_allowed}="0"` was a no-op — that sysfs attribute doesn't exist on this Phison endpoint or its parent root port (`0000:00:1d.0`).
+- The `nvme-resume-fix` service was broken: `ExecStart` used `>` redirection without a shell, so `echo` just printed the string and never wrote to the `rescan` sysfs file. It never actually rescanned.
+
+Changes (`hosts/legolas/hardware-configuration.nix`):
+
+- Switched `mem_sleep_default` from `deep` (S3) to `s2idle` (modern standby). s2idle avoids the deep D3 path that hangs the drive.
+- Removed the no-op `d3cold_allowed` rule and the broken rescan service.
+- Added a correct runtime-PM-off udev rule on both the NVMe endpoint (`0000:6e:00.0`) and its parent root port (`0000:00:1d.0`), keeping the PCIe wakeup disable.
+
+Note: runtime PM was already off on the endpoint (`power/control=on`); that alone never fixed the crash because system suspend uses a separate code path. Verify after rebuild that the cmdline has `mem_sleep_default=s2idle` (the old running generation showed a stale duplicate `deep`).

hosts/legolas/hardware-configuration.nix

@@ -13,7 +13,10 @@
   boot.kernelPackages = pkgs.linuxPackages_latest;
 
   boot.kernelParams = [
-    "mem_sleep_default=deep"
+    # s2idle (modern standby) instead of deep (S3): the Phison E21 NVMe
+    # controller (Corsair MP600 ELITE) hangs on resume from the D3 transition
+    # that S3 forces. s2idle avoids that path.
+    "mem_sleep_default=s2idle"
     "nvme_core.default_ps_max_latency_us=0"
     "pcie_aspm=off"
   ];
@@ -67,22 +70,15 @@
     };
   };
 
-  # Disable PCIe wakeups and prevent NVMe D3cold (fixes resume from S3)
+  # Keep runtime PM off on the NVMe controller and its parent root port, and
+  # disable PCIe wakeup events. (The previous d3cold_allowed rule was a no-op:
+  # that sysfs attribute doesn't exist on this Phison endpoint or its root port.)
   services.udev.extraRules = ''
     ACTION=="add", SUBSYSTEM=="pci", DRIVER=="pcieport", ATTR{power/wakeup}="disabled"
-    ACTION=="add", SUBSYSTEM=="pci", KERNEL=="0000:6e:00.0", ATTR{d3cold_allowed}="0"
+    ACTION=="add", SUBSYSTEM=="pci", KERNEL=="0000:6e:00.0", ATTR{power/control}="on"
+    ACTION=="add", SUBSYSTEM=="pci", KERNEL=="0000:00:1d.0", ATTR{power/control}="on"
   '';
 
-  systemd.services.nvme-resume-fix = {
-    description = "NVMe resume workaround";
-    after = [ "suspend.target" ];
-    wantedBy = [ "suspend.target" ];
-    serviceConfig = {
-      Type = "oneshot";
-      ExecStart = "${pkgs.coreutils}/bin/echo 1 > /sys/class/nvme/nvme0/device/rescan";
-    };
-  };
-
   hardware.graphics = {
     enable = true;
     extraPackages = with pkgs; [ vpl-gpu-rt ];