Bugzilla – Bug 113748
Lots of debug messages from SATA DVD Burner
Last modified: 2005-09-06 14:03:06 UTC
garbarek fills the syslog with: Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 Aug 29 09:24:58 linux kernel: ATA: abnormal status 0x58 on port 0xE407 this is with: Linux garbarek 2.6.13-rc7-4-smp #1 SMP Wed Aug 24 22:00:05 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux The machine has a SATA DVD Burner
What are you doing with the drive? SATA ATAPI still isn't 100% complete, as you know.
I'm not doing anything at all with it :-(. Btw. with the current kotd I get a kernel oops in ATA, I'll give the machine to Andreas G for looking into it.
Are you sure? Perhaps things like hald is polling the device. Can you shut it down, perhaps check with lsof if anyone has the device open.
Andreas G. has the machine now. Do you still need something from him?
I still need a response to my last question, comment #3.
Shutting down hal did stop the messages. They continue when I do something like ``mount /dev/sr0 /mnt'' though (and mount then takes ages until it aborts).
Hmm perhaps that ATAPI drive still does the irq completion on receiving the cdb. I'll see if I can buy a SATA ATAPI drive locally for testing purposes, I probably cannot do much without the hardware at hand.
I have found one that can be here by tomorrow morning, I just need approval to buy it now...
Happy hacking with the new drive!
The drive hasn't shown up yet, they screwed up the shipment. After yelling on the phone, someone will personally deliver it to me within an hour. So I hope it works out...
Drive just showed up, it refuses to work on a promise and via controller though. Great. I will try a piix/ahci controller next.
I've committed a lot of atapi pio libata fixes now, it should definitely improve things over what we have. Hannes will test on garbarek.
Looking at garbarek now, it seems to work as expected. No debug messages and mounting (and reading) the CD in the drive works as well. aj/hannes, can you confirm?
Hannes/aj has confirmed it works.
garbarek contains a custom build kernel. The fixes need to be reviewed and submitted to our kernel! Please remember to do so.
Andreas, the patches in question are already committed to HEAD. They are in mainline 2.6.13-git as well.