Re: Question regarding linux1394 and bug report
Brought to you by:
aeb,
bencollins
From: Adam G. <ad...@po...> - 2024-02-19 01:42:25
|
Hi Tasos, On Tue, Feb 13, 2024 at 12:18:01PM +0200, Tasos Sahanidis wrote: > Thank you for still looking into this. Apologies the late response once > more. Things got a bit out of hand... Unfortunately, I haven't had much time for FireWire recently, so my response is also late. Thank you for the additional logs. > I thought maybe it works if the system boots with the camera plugged in, > so I powered off the system, power cycled the camera, and booted > directly to 6.7. Except 6.7 crashed on boot without any logs. I did not > at the time have another machine with a serial port, so I was unable to > gather any logs, and I didn't have time to test out netconsole. A few > more reboots with the camera plugged in, and the system eventually > booted, but I got the following mess: > > [ 88.040334] firewire_ohci 0000:01:03.0: IRQ 00020000 busReset > [message repeated every 5~10us] > [ 88.062147] firewire_ohci 0000:01:03.0: IRQ 00020000 busReset > [ 88.062155] firewire_ohci 0000:01:03.0: IRQ 00020000 busReset > [ 88.062162] firewire_ohci 0000:01:03.0: IRQ 00020010 AR_req busReset > [ 88.062169] firewire_ohci 0000:01:03.0: IRQ 00020000 busReset Are you setting debug=-1 when loading the module? I believe OHCI_PARAM_DEBUG_BUSRESETS has been broken for years. It will freeze the system or spam tons of busReset log entries if enabled during module load, even without a camera connected. debug=7 should be safe. By the way, I will be submitting a patch relating to bus resets soon, but don't get excited. It won't help with the problem you're experiencing. -- Adam |