Hallo zusammen,
vorweg: mein S3-Suspend und auch der Resume daraus funktionieren einwandfrei. Aber es könnte natürlich immer noch ein bißchen schneller gehen 
Auf der Suche nach Einsparpotential habe ich erstmal den Floppy-Driver rausgeschmissen, der wartet nach meinem Dafürhalten ein 3s-Timeout ab, was schon mal 30% der insgesamt 10s Resume-Zeit bedeutet.
Dann ist mir aufgefallen, dass im Log zum Zeitpunkt des Resumes Meldungen über Suspend-Aktionen auftauchen. Es sieht so aus, als wenn einige Geräte beim Resume gerade erst schlafen gelegt werden:
Mar 18 04:55:34 dumbledore anacron[4895]: Anacron 2.3 started on 2011-03-18
Mar 18 04:55:34 dumbledore anacron[4895]: Normal exit (0 jobs run)
Mar 18 04:55:34 dumbledore kernel: [12082.288677] CPU0 attaching NULL sched-domain.
Mar 18 04:55:34 dumbledore kernel: [12082.288684] CPU1 attaching NULL sched-domain.
Mar 18 04:55:34 dumbledore kernel: [12082.324074] CPU0 attaching sched-domain:
Mar 18 04:55:34 dumbledore kernel: [12082.324078] domain 0: span 0-1 level MC
Mar 18 04:55:34 dumbledore kernel: [12082.324081] groups: 0 1
Mar 18 04:55:34 dumbledore kernel: [12082.324086] CPU1 attaching sched-domain:
Mar 18 04:55:34 dumbledore kernel: [12082.324087] domain 0: span 0-1 level MC
Mar 18 04:55:34 dumbledore kernel: [12082.324089] groups: 1 0
Mar 18 04:55:34 dumbledore init: logo-stop main process (4673) killed by TERM signal
Mar 18 07:23:45 dumbledore kernel: [12082.677308] PM: Syncing filesystems ... done.
Mar 18 07:23:45 dumbledore kernel: [12082.678840] PM: Preparing system for mem sleep
Mar 18 07:23:45 dumbledore kernel: [12082.678844] Freezing user space processes ... (elapsed 0.00 seconds) done.
Mar 18 07:23:45 dumbledore kernel: [12082.679159] Freezing remaining freezable tasks ... i2c_write error
Mar 18 07:23:45 dumbledore kernel: [12082.709515] (elapsed 0.03 seconds) done.
Mar 18 07:23:45 dumbledore kernel: [12082.709518] PM: Entering mem sleep
Mar 18 07:23:45 dumbledore kernel: [12082.709533] Suspending console(s) (use no_console_suspend to debug)
Mar 18 07:23:45 dumbledore kernel: [12082.751042] lirc_atiusb[3]: usb remote disconnected
Mar 18 07:23:45 dumbledore kernel: [12082.784060] sd 0:0:0:0: [sda] Synchronizing SCSI cache
Mar 18 07:23:45 dumbledore kernel: [12082.784118] sd 0:0:0:0: [sda] Stopping disk
Mar 18 07:23:45 dumbledore kernel: [12085.344338] PM: suspend of drv:sd dev:0:0:0:0 complete after 2560.278 msecs
Mar 18 07:23:45 dumbledore kernel: [12085.360919] ACPI handle has no context!
Mar 18 07:23:45 dumbledore kernel: [12085.360999] DDBridge 0000:03:00.0: PCI INT A disabled
Mar 18 07:23:45 dumbledore kernel: [12085.472026] HDA Intel 0000:02:00.1: PCI INT B disabled
Mar 18 07:23:45 dumbledore kernel: [12085.472057] ACPI handle has no context!
Mar 18 07:23:45 dumbledore kernel: [12085.488018] PM: suspend of drv:HDA Intel dev:0000:02:00.1 complete after 127.010 msecs
Mar 18 07:23:45 dumbledore kernel: [12085.509341] ath9k 0000:01:0a.0: PCI INT A disabled
Mar 18 07:23:45 dumbledore kernel: [12085.524017] ath9k 0000:01:0a.0: Refused to change power state, currently in D0
Mar 18 07:23:45 dumbledore kernel: [12085.524026] ivtv 0000:01:09.0: PCI INT A disabled
Mar 18 07:23:45 dumbledore kernel: [12085.524032] ieee1394: hpsb_bus_reset called while bus reset already in progress
Mar 18 07:23:45 dumbledore kernel: [12085.528007] ohci1394: fw-host0: Set PHY Reg timeout [0xffffffff/0x00004000/100]
Mar 18 07:23:45 dumbledore kernel: [12085.623332] hpet1: lost 5 rtc interrupts
Mar 18 07:23:45 dumbledore kernel: [12085.627318] ohci1394: fw-host0: Set PHY Reg timeout [0xffffffff/0x00004000/100]
Mar 18 07:23:45 dumbledore kernel: [12085.722704] hpet1: lost 6 rtc interrupts
Mar 18 07:23:45 dumbledore kernel: [12085.822061] ACPI handle has no context!
Mar 18 07:23:45 dumbledore kernel: [12085.836015] PM: suspend of drv:ohci1394 dev:0000:01:07.0 complete after 311.982 msecs
Mar 18 07:23:45 dumbledore kernel: [12085.948147] ahci 0000:00:09.0: PCI INT A disabled
Mar 18 07:23:45 dumbledore kernel: [12085.964014] PM: suspend of drv:ahci dev:0000:00:09.0 complete after 111.997 msecs
Mar 18 07:23:45 dumbledore kernel: [12086.172022] HDA Intel 0000:00:07.0: PCI INT A disabled
:
Ist das tatsächlich so, oder sind das Log-EInträge, die beim Suspend generiert werden, aber erst beim Wakeup (und mit dem Zeitstempel) ins Log geschrieben werden?
Danke,
Sebi