From: Patrick M. <mo...@os...> - 2002-03-29 21:52:08
|
On Fri, 29 Mar 2002, Pavel Machek wrote: > Hi! > > > (The first is not necessarily related to the other two, but the other two > > were created relative to the first, and it's otherwise innocuous). > > > > 1. Implements notion of 'system' bus, so system level devices can be added > > in a comon spot. This includes things like CPUs, PICs, timers, etc. > > Good thing, but naming is pretty inconsistent. > > You have device_register() but register_sys_device(). Noted. How about sys_device_register()? > > Testing welcome also, though I wouldn't expect one to get very far, since > > they're not actually used. ;) Which, brings up another question - what > > would be the proper place to call device_shutdown()? (I haven't looked > > very far into that part...) > > Tested, seems to work. > Pavel > PS: On toshiba 4030cdt, I can suspend once without no apparent ill > effects. On resume I get > > utmisc-0373 Ut_acquire_mutex : Invalid acquire order: Thread 5C owns > [ACPI_MTX_Hardware], wants [ACPI_MTX_Namespace]. > > followed by more warnings. Is there easy way to debug this? Maybe using one of the kernel debuggers? I haven't ever used one, but it sounds like you should have enough context where it would work.. -pat |