It doesn't work on SMP systems without CONFIG_HOTPLUG_CPU enabled in the kernel, period.

After enabling it on my Lenovo T60, suspending to memory worked out of the box, even with ATI's evil^Wbinary drivers.  The kernel output leads me to believe that before suspending, it virtually removes the 2nd core and then proceeds to disable SMP.  When it wakes up, it initializes SMP, and brings the 2nd core back online.

Freezing cpus ...
CPU 1 is now offline
SMP alternatives: switching to UP code
CPU1 is down
Stopping tasks: ====================[...]=================|

And waking up ...

Thawing cpus ...
SMP alternatives: switching to SMP code
Booting processor 1/1 eip 3000
Initializing CPU#1
Calibrating delay using timer specific routine.. 3657.49 BogoMIPS (lpj=1828745)
CPU: After generic identify, caps: bfe9fbff 00100000 00000000 00000000 0000c1a9 00000000 00000000
CPU: After vendor identify, caps: bfe9fbff 00100000 00000000 00000000 0000c1a9 00000000 00000000
monitor/mwait feature present.
CPU: L1 I cache: 32K, L1 D cache: 32K
CPU: L2 cache: 2048K
CPU: Physical Processor ID: 0
CPU: Processor Core ID: 1
CPU: After all inits, caps: bfe9fbff 00100000 00000000 00000140 0000c1a9 00000000 00000000
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#1.
CPU1: Intel Genuine Intel(R) CPU           T2400  @ 1.83GHz stepping 08
APIC error on CPU1: 00(40)
CPU1 is up

Neat.  Well, whatever it's doing, suspending to memory is the first thing that worked "out of the box  on the T60.

Using:

• ati-drivers-8.26.18-r1
• gentoo-sources-2.6.17-r2
• xorg-x11-7.0-r1
• acpid-1.0.4-r3

New comments are currently disabled for this entry.