On a DEC HiNote 433 notebook APM functions fine as long as i don't plug a 3Com 3C589C into the PCMCIA slot - if i do, and want to APM-resume after a APM-suspend, the display is switched on but nothing more happenes. (compiling with APM_DEBUG defined: without the card i get the suspend and the resume message, with the card i just get the suspend message but no resume message; compiling with APM_SLOWSTART=5 makes no difference if that matters) Fix: In if_zp.c got from the 2.1 tree: * From: if_ep.c,v 1.9 1994/01/25 10:46:29 deraadt Exp $ * $Id: if_zp.c,v 1.6.4.1 1995/08/19 23:27:14 davidg Exp $ delete the following statement on line 726 of function zp_suspend(): pcic_power_off(sc->slot); When compiled with APM support, the power switching is completely controlled by the APM BIOS. When the system is suspended by the apm driver, it is done with a parameter specifying to handle ALL devices and this obviously interferes with switching the power off for a PCMCICA slot in the device specific suspend hook routine. After the deletion of the above mentioned statement the system can be suspended and resumed in the above described configuration without errors or other side effects. hellmuth -- Hellmuth Michaelis hm@altona.hamburg.com Hamburg, Europe (A)bort, (R)etry, (I)nstall BSD ? How-To-Repeat: On a DEC HiNote 433, compile a kernel with APM support and the "zp" driver and place the system into suspend mode by either executing zzz or close the cover. The machine will freeze and never wake up again - the power has to be cycled.
Responsible Changed From-To: freebsd-bugs->nate Nate is doing APM/PCMCIA integration work.
State Changed From-To: open->analyzed Can you try using -current with the PC-CARD driver now? if_zp.c is soon to be gone now that we have generic PC-CARD support in -current. if_ep.c contains suspend hooks, so it may work better.
State Changed From-To: analyzed->closed The suggested patch was added to the if_zp driver a long time ago.