I think the recent wdc/pciide change introduced a bug. I'm now getting a panic during boot. I'll get a full dmesg and trace in a bit, gotta find a ps2 keyboard and hook up a serial console. But, for now, I'll type it in:
pciide0: port 0: device present, speed: 1.5Gb/s panic: wdc_exec_command: polled command not done Stopped at Debugger+0x5: leave Debugger() at Debugger+0x5 panic() at panic+0xe4 wdc_exec_command() at wdc_exec_command+0x132 ata_get_params() at ataget_params+0x8f wdcattach() at wdcattach+0x145 sii3114_mapchan() at sii3114_mapchan+0x17a sii3114_chip_map() at sii3114_chip_map+0x270 config_attach() at config_attach+0x150 pci_probe_device() at pci_probe_device+0x3bb pci_enumerate_bus() at pci_enumerate_bus+0x3e9 end trace frame: 0xffffffff80e2dac0, count: 0 Tyan Tiger K8W main board. Western Digital Raptor WD360 drive (all others unplugged). Tested with my own build of the kernel and the two most recent snapshots. Problem exists across the board. I'll send a trace and ps (it crashes in config, so this will be the obvious list) as soon as I can wrangle up a ps2 keyboard since I lose USB during config. CVSROOT:/cvs Module name:src Changes by:deraadt< at >cvs.openbsd.org2011/04/17 22:16:14 Modified files: sys/dev/pci : pciide.c sys/dev/ata : ata_wdc.c wd.c sys/dev/ic : wdc.c wdcvar.h Log message: keep an eye out for wdc registers returning 0xff. if they return 0xff, this probably means the controller is dead -- as in, a hot plug eject event. in that case, mark the driver dead and try harder to return -1 in the interrupt handler. tested by many ok dlg, miod did not object Scot Bontrager cloudkucool...@gmail.com s...@indievisible.org
