uswsusp oddity

[Andreas Steinmetz]

I just tried uswsusp 0.2 and there is a strange problem (2.6.17.8, x86_64):

The swap device used for suspend must be the first swap device,
otherwise suspend fails.


Contents of /etc/suspend.conf:

snapshot device = /dev/snapshot
resume device = /dev/hda2
#image size = 350000000
suspend loglevel = 7
#compute checksum = y
#compress = y
#encrypt = y
early writeout = y
#splash = y


Output of /proc/swaps of the failing suspend:

Filename                Type            Size    Used    Priority
/dev/mapper/swap1       partition       1959920 0       2
/dev/mapper/swap2       partition       1959888 0       3
/dev/hda2               partition       1959920 0       1


Relevant portion of strace of suspend utility:

mlockall(MCL_CURRENT|MCL_FUTURE)        = 0
stat("/dev/hda2", {st_mode=S_IFBLK|0660, st_rdev=makedev(3, 2), ...}) = 0
open("/dev/hda2", O_RDWR)               = 3
stat("/dev/snapshot", {st_mode=S_IFCHR|0600, st_rdev=makedev(10, 231),
...}) = 0
open("/dev/snapshot", O_RDONLY)         = 4
ioctl(4, 0x4004330a, 0x302)             = -1 ENODEV (No such device)
write(2, "suspend: Could not use the resum"..., 57suspend: Could not use
the res
ume device (try swapon -a)


There seems to be something wrong with the SNAPSHOT_SET_SWAP_FILE ioctl.

If I do the following sequence suspend works:

swapoff -a
swapon -p 1 /dev/hda2
swapon -a
-- 
Andreas Steinmetz                       SPAMmers use robotrap@xxxxxxxx
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/