On Fri, Jun 17, 2022 at 09:14:45AM +0100, Edd Barrett wrote: > Hi Mike, > > On Thu, Jun 16, 2022 at 09:19:50PM -0700, Mike Larkin wrote: > > From your original dmesg: > > > > > acpi0: wakeup devices PEG0(S4) PEGP(S4) PEG1(S4) PEGP(S4) PEG2(S4) > > > PEGP(S4) SIO1(S3) RP09(S4) PXSX(S4) RP10(S4) PXSX(S4) RP11(S4) PXSX(S4) > > > RP12(S4) PXSX(S4) RP13(S4) [...] > > > > Notice the [...] at the end, this is printed after 16 devices. What I'd > > suggest > > is this: > > > > 1. remove the code that truncates this list after 16, and note down all the > > wake > > devices. > > > > 2. If there are any in S3, try using ZZZ instead of zzz. If the machine > > does not instantly > > wake, it's possible it's because of one of those S3 devices doing the wake > > (since ZZZ > > uses S4). > > I'll try removing the truncation then. Bear with me. > > In the meantime, notice that the truncated list does include one S3 item > `SIO1(S3)`. I don't know if that's what we are looking for? > > FWIW, I have already tried `ZZZ` on this machine and it does succeed to > hibernate, but upon wake up, it hangs when decompressing the memory image. I > left it decompressing a ~50MB image for more than an hour and concluded it had > got stuck.
Oh, didn't read this closely enough the first time. If ZZZ doesn't instantly wake the machine, then it's one of the two S3 devices described in your next email. Since one is XHCI, I'd disable xhci in ukc and see if that helps. Or maybe the other *hci(4)s also. Now, why ZZZ fails to unpack is some other problem but the instant wake is not related to that. -ml > > > 3. If everything is S4, well, you're going to have to trace down those > > short names > > like PEGP, PXSX, etc, and disable one at a time until you find the one that > > is > > doing the wake. And it's possible it's none of these and is a fixed function > > button or something. > > One additional piece of info, which may be worthless. I tried a Debian live > USB > stick, to see if Linux was able to sleep this box. It was able to. > > I don't know if that rules out the idea of a fixed-function button? > > -- > Best Regards > Edd Barrett > > https://www.theunixzoo.co.uk
