I just got a new computer, an AlienWare Aurora R16. When I try to suspend the system, I am not sure what state it’s in because the case lighting doesn’t change but when I attempt to wake the system from what should be a suspend, it boots up. I suspect that it’s either powering off when attempting to suspend or it’s failing to resume and rebooting into a fresh boot. Could someone please help?
I’d try booting with the kernel option mem_sleep_default=s2idle
I had a Dell XPS that would power off and reboot cold after suspend if I used the default S3 sleep on Linux, but would suspend/wake fine with s2idle. Dell I think specifically advertised Windows Modern Sleep or something on that model.
2 Likes
Thanks for the help @Espionage724. No that did not work. I have noticed that when my system is supposedly in suspend, whether it’s s2idle or deep, if I press a keyboard button that the system comes back online but the case fans never stop while in “suspend”. My older Alienware Aurora R10 would change the lights to yellow and you could here the case fans stop but that’s not the case with this PC.
Also, before I was short pressing the power button to wake it and that’s when it booted instead of woke up.
Have a look in the Settings > Power > General > Suspend ; Power off; Nothing
On what is yours set?
Suspend.
Did you test “noting” ? To see if it really does nothing 
Just had a look what duckduckgo had to offer, and it looks bad:
Did you actualize the bios since you bought it?
For tesring purposes I am not using the button. I am choosing suspend in from the power menu and using keyboard or mouse to wake it. I noticed this, now that I have mem_sleep_default=s2idle set in my grub is that when I suspend the system, the light on graphics card turns off. I don’t know if that was always the case or if that has only been since I have made that change. The case lights not changing and the case fan constantly running is concerning. If the system is in suspend than why do we need a fan running. The lights would just be a nice indicator that it’s in suspend. As I had previously mentioned, my previous Alienware changed the case lights from blue to amber and the fans fell silent.
Thanks to @Espionage724 and @ilikelinux for the help so far but I feel like something is still missing or isn’t right. I am going to look into see if Windows users have the same issue. I am not a hardware engineer but I thought that ACPI was a standard and that interacting with it to cause a suspend / wake was standardized but I have never programmed interactions with it and I am really just assuming.
I remember a lot of ACPI-related quirks with Hackintosh, particularly also something about Suspend differences/patches depending on the CPU.
DSDT has OS-specific things; on my laptop Windows does something different with Intel HID. It exists as a visible device with driver on Windows, but with Windows OSI on Linux Intel HID doesn’t exist and I lose backlight/vol laptop keys. I think different Windows OSI versions also do something different with I2C and my touchpad but I didn’t notice a difference in Linux.
Dumping the DSDT and looking through it with a text editor can show what OSI’s do what; the devices it affects is kind-of easy but figuring out what it’s doing exactly is a little more complex.
ArchWiki has some more info; I remember having to dump and recompile with iasl too with macOS and there was a difference between ACPI’s official compiler and Intel’s.