I'm not opposed to the change per se. But I note a couple of problems: 1. There must be a bug. On a Fedora 28 system that does suspend to RAM in 20 minutes, the Power panel Automatic Suspend setting is set to Off. It makes no sense this computer suspends based on that setting. This is on a pre-mass rebuild system which hasn't yet been updated again so maybe it's a known bug. 2. I guarantee the change will expose suspend to ram bugs. I have rather common HP hardware (HP Spectre), and with an out of the box installation, it will suspend, wake, suspend, wake, suspend, wake, in an indefinite loop when connected to power and until battery death if not connected to power. To inhibit this, I have /etc/tmpfiles.d/suspendfix.conf which contains: w /proc/acpi/wakeup - - - - PWRB w /proc/acpi/wakeup - - - - XHC Exhaustive kernel regression testing with upstream developers yielded the claim this is a firmware bug, even though I found the exact commit where the bug was introduced in the kernel, and Windows 10 doesn't have this problem. Anyway, I have a seemingly stable work around, but I'm gonna bet other people will run into myriad suspend bugs. Sure they can just turn off suspend *if* they know to suspect this is the source of their problem. 3. Will suspend to ram be inhibited when running Fedora 28 in a VM? -- Chris Murphy _______________________________________________ desktop mailing list -- desktop@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to desktop-leave@xxxxxxxxxxxxxxxxxxxxxxx
