Using "shutdown -p now" to enter the S0 (shutdown) state or "zzz" to enter the S3 state doesn't work anymore. It used to work on 13.3-RELEASE, but I was also running an older BIOS version then, so the comparison might be a bit apples to oranges. Key parts of my hardware profile can be found in this gist: https://gist.github.com/ngie-eign/2e694de535c434fd3f29c9bc1d12db07
More info I forgot to include in the original report: - shutdown -p now/zzz results in a "power bounce" (it seems like devices or subsystems might be preventing the host from going to sleep?). - The VGA console is blank on "resume". -- Using "sysctl hw.acpi.reset_video=1" seems to deadlock the machine on resume. Basic power functions seem to come back, but the system doesn't fully power up (the CPU fan doesn't kick on again). - I have tried both enabling and disabling the APM/ERP (sic) functions in the BIOS and they don't seem to have an effect.
(In reply to Enji Cooper from comment #1) - The VGA console is blank on "resume". ... if I use zzz. Ah, I forgot -- the S0 problem was another issue. I'll do some more poking around the internet/file a different bug for that. In short, I was trying to get WoL to work a week ago and "shutdown -p now" was power bouncing the machine (rebooting). I need to see if it's a firmware bug or what have you. The S3 state issue is very much valid, though. I suppose I could try booting up a 13.3-RELEASE USB stick and see if using zzz works.
I just tried out the USB mini sticks on the following versions of FreeBSD: - 13.3-RELEASE: -- zzz works. -- No VGA console output. - 14.0-RELEASE: -- zzz works (!). -- No VGA console output. - 14.1-PRERELEASE 20240425-9857f824ec77-267512: -- zzz works. -- No VGA console output. I'm not entirely sure if VGA console after exiting S3 ever worked with this motherboard/CPU combo, to be frank, since most of the time I treat this host like a headless box; I'll save that issue for another bug. So, something changed between 14.0-RELEASE and 14.0-RELEASE-p6, OR one of the drivers/subsystems is failing to enter suspend properly (I am leaning more towards this possibility). I'll check to see if ada, graid, if_igb, or nvd with the USB stick image is causing problems when I get back sometime next week.
Apologies for taking so long. I'm not 100% sure, but the culprit might be the fact that I'm using graid on 2 SATA SSDs using Intel metadata (which, admittedly, is quite invalid on my AMD chipset motherboard). Looking at /var/log/messages for the timeframe when I did zzz, this is what I see: Jun 4 21:31:51 gilot kernel: pci0: set ACPI power state D0 on \_SB_.PCI0.GPP3 Jun 4 21:31:51 gilot kernel: pci0: set ACPI power state D0 on \_SB_.PCI0.GP17 Jun 4 21:31:51 gilot kernel: pci11: set ACPI power state D0 on \_SB_.PCI0.GP17.VGA_ Jun 4 21:31:51 gilot kernel: pci11: set ACPI power state D0 on \_SB_.PCI0.GP17.D018 Jun 4 21:31:51 gilot kernel: hdaa0: Pin sense: nid=3 sense=0x7fffffff (disconnected, ELD valid) Jun 4 21:31:51 gilot kernel: hdaa0: Pin sense: nid=5 sense=0x7fffffff (disconnected, ELD valid) Jun 4 21:31:51 gilot kernel: pci11: set ACPI power state D0 on \_SB_.PCI0.GP17.XHC0 Jun 4 21:31:51 gilot kernel: pci11: set ACPI power state D0 on \_SB_.PCI0.GP17.XHC1 Jun 4 21:31:51 gilot kernel: pci0: set ACPI power state D0 on \_SB_.PCI0.GP18 Jun 4 21:31:51 gilot kernel: pci12: set ACPI power state D0 on \_SB_.PCI0.GP18.SATA Jun 4 21:31:51 gilot kernel: ahcich6: AHCI reset... Jun 4 21:31:51 gilot kernel: ahcich6: SATA connect timeout time=10000us status=00000000 Jun 4 21:31:51 gilot kernel: ahcich6: AHCI reset: device not found Jun 4 21:31:51 gilot kernel: pci0: set ACPI power state D0 on \_SB_.PCI0.SBRG Jun 4 21:31:51 gilot kernel: ahcich1: AHCI reset: device ready after 100ms Jun 4 21:31:51 gilot kernel: ahcich5: AHCI reset: device ready after 100ms Jun 4 21:31:51 gilot kernel: SMP: passed TSC synchronization test Jun 4 21:31:51 gilot kernel: efirtc0: providing initial system time Jun 4 21:31:51 gilot kernel: (ada0:ahcich1:0:0:0): resume Jun 4 21:31:51 gilot kernel: (ada1:ahcich5:0:0:0): resume Jun 4 21:31:51 gilot acpi[6476]: resumed at 20240604 21:31:51 Jun 4 21:31:51 gilot ntpd[1310]: ntpd exiting on signal 15 (Terminated) Jun 4 21:31:51 gilot ntpd[6509]: ntpd 4.2.8p16-a (1): Starting Jun 4 21:31:51 gilot kernel: uhub2 on usbus3 Jun 4 21:31:51 gilot kernel: uhub2: <AMD XHCI root HUB, class 9/0, rev 3.00/1.00, addr 1> on usbus3 Jun 4 21:31:51 gilot kernel: uhub3 on usbus2 Jun 4 21:31:51 gilot kernel: uhub3: <AMD XHCI root HUB, class 9/0, rev 3.00/1.00, addr 1> on usbus2 Jun 4 21:31:51 gilot ntpd[6509]: Command line: /usr/sbin/ntpd -p /var/db/ntp/ntpd.pid -c /etc/ntp.conf -f /var/db/ntp/ntpd.drift Jun 4 21:31:51 gilot ntpd[6509]: ---------------------------------------------------- Jun 4 21:31:51 gilot ntpd[6509]: ntp-4 is maintained by Network Time Foundation, Jun 4 21:31:51 gilot ntpd[6509]: Inc. (NTF), a non-profit 501(c)(3) public-benefit Jun 4 21:31:51 gilot ntpd[6509]: corporation. Support and training for ntp-4 are Jun 4 21:31:51 gilot ntpd[6509]: available at https://www.nwtime.org/support Jun 4 21:31:51 gilot ntpd[6509]: ---------------------------------------------------- Jun 4 21:31:51 gilot ntpd[6510]: leapsecond file ('/var/db/ntpd.leap-seconds.list'): good hash signature Jun 4 21:31:51 gilot ntpd[6510]: leapsecond file ('/var/db/ntpd.leap-seconds.list'): loaded, expire=2024-12-28T00:00:00Z last=2017-01-01T00:00:00Z ofs=37 I'll see if I can reproduce this by activating the graid on a USB stick.
More information I didn't include in my prior message. - The SSDs aren't in SW RAID mode in the BIOS. - The only thing being mirrored is the EFI partition (which technically is not being mounted right now). I was just trying to avoid having the EFI partition get out of sync between the two SSDs. gilot% zpool status zroot pool: zroot state: ONLINE scan: scrub repaired 0B in 00:03:18 with 0 errors on Wed Apr 17 17:45:53 2024 config: NAME STATE READ WRITE CKSUM zroot ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 ada0p3 ONLINE 0 0 0 ada1p3 ONLINE 0 0 0 errors: No known data errors gilot% graid status Name Status Components raid/r0 OPTIMAL ada0p1 (ACTIVE (ACTIVE)) ada1p1 (ACTIVE (ACTIVE)) gilot% graid list Geom name: Intel-b68ba96e State: OPTIMAL Metadata: Intel Providers: 1. Name: raid/r0 Mediasize: 268431360 (256M) Sectorsize: 512 Mode: r0w0e0 Subdisks: ada0p1 (ACTIVE), ada1p1 (ACTIVE) Dirty: No State: OPTIMAL Strip: 131072 Components: 2 Transformation: RAID1 RAIDLevel: RAID1 Label: efi descr: Intel RAID1 volume Consumers: 1. Name: ada0p1 Mediasize: 268435456 (256M) Sectorsize: 512 Stripesize: 0 Stripeoffset: 20480 Mode: r1w1e1 ReadErrors: 0 Subdisks: r0(efi):0@0 State: ACTIVE (ACTIVE) 2. Name: ada1p1 Mediasize: 268435456 (256M) Sectorsize: 512 Stripesize: 0 Stripeoffset: 20480 Mode: r1w1e1 ReadErrors: 0 Subdisks: r0(efi):1@0 State: ACTIVE (ACTIVE)