Only after we trigger the BUSRESET interrupt manually when booting, there should be no firewire interrupt. There must be wrong with anything but the interrupt handler 2009/3/28 JiSheng Zhang <jszhang3@xxxxxxxxx>: > There should be no interrupt related with firewire before the info > "firewire:Initiate bus reset" is printed. Such info is missing from > the log, so the interrupt isn't expected. IMO, it isn't related with > the interrupt handling. > > 2009/3/28 Stephan Assmus <superstippi@xxxxxx>: >> >> On 2009-03-28 at 08:53:29 [+0100], JiSheng Zhang <jszhang3@xxxxxxxxx> wrote: >>> The patch is tested a few times and attached, would someone help me apply >>> the patch and svn ci? Thanks in advance. >>> >>> PS: any ideas about ticket 3595? >> >> Unfortunately, I don't have any ideas to offer, but unless the interrupt >> handling of the FireWire stack can be made reliable, I would be in favor of >> not putting the FireWire bus_manager on the default Haiku image. The reason >> being that there is currently no DV media add-on, so it isn't really >> integrated. In case the FireWire bus_manager causes a hang at boot, it does >> more harm than it does good when it works, at least at the moment. Maybe >> some devs with kernel and driver experience have some ideas. >> >> Best regards, >> -Stephan >> >> >