Description of problem: My hardware profile is ebf36068-3809-426c-ade0-d0c531b5b402, the laptop is Acer Aspire 5610 (got it with some strange linux "installed"). With F7t4's kernels (up to 2.6.21-1.3116.fc7 and 2.6.21-1.3123.fc7) after suspend to ram the machine does not wake up - the display stays black (Intel 945GM) and the disk led turns on forever. Version-Release number of selected component (if applicable): kernels 2.6.20-2925.5.fc7xen, 2.6.20-1.3104.fc7, 2.6.21-1.3116.fc7 and 2.6.21-1.3123.fc7). How reproducible: Every time. Actual results: The machine hangs after suspend to ram. Expected results: Should resume as it does with FC6. Additional info: To get hibernate (suspend to disk) working I had to create this file (/etc/pm/hooks/11iwlwifi): --- cut --- #!/bin/sh exit 0 case "$1" in hibernate|suspend) /sbin/lsmod | grep -q iwlwifi || exit 0 { touch /var/run/iwlwifi && /sbin/modprobe -r iwlwifi ; } \ 2>/dev/null ;; thaw|resume) [ -f /var/run/iwlwifi ] || exit 0 { /sbin/modprobe iwlwifi && rm -f /var/run/iwlwifi \ && /sbin/service NetworkManager restart ; } 2>/dev/null ;; *) ;; esac exit $? --- cut --- Once, just once, it did resume after suspend to ram in text console, but the keyboard acted "strange". When I pressed some special keys I saw the kernel complaining for unknown keys, but no normal key did anything. Using the mouse (copy a letter, paste it...) I succesfully rebooted it. Every other attempt after that just failed with the disk led constantly on (I tried with runlevel 1 too). Bug #238800 is maybe a related small problem with FC6 and this machine. I have no problems with the power led with F7t4.
With kernel 2.6.21-1.3141.fc7 I can suspend to disk/ram and resume without problems - tried 3 times in a row each, hurrah! iwlwifi got renamed to iwl3945 and still has some problems, but that's another story. I'll try not to remove the module while suspending and report the results.
Removing the module (iwl3945) before suspend helps (by hand), but sometimes hangs the machine. Sometimes inserting it also hangs the machine (after remove). New bug report (this one looks closed)?