[PARPORT] Re: New queueing code

From: Erik Inge Bolsų (knan@mo.himolde.no)
Date: Wed Mar 08 2000 - 10:42:14 EST

  • Next message: Eric Youngdale: "[PARPORT] Re: New queueing code"

    On Wed, 8 Mar 2000, Eric Youngdale wrote:
    > What kernel version did the problems start to occur? There have been
    >several rounds of changes to the ll_rw_blk code (not all of which I was
    >directly responsible for), and it would help a lot to narrow it down to a
    >specific kernel version. If the primary author of the change in question
    >was someone else, I can help to bring that person into the loop too.

    2.3.40: works
    2.3.45: broken

    I'll try compiling a few intermediate kernels to see if I can track it
    down. It broke in 2.3.41, 42, 43, 44 or 45.

    Later,
    Erik

    >----- Original Message -----
    >From: Erik Inge Bolsų <knan@mo.himolde.no>
    >To: Eric Youngdale <eric@andante.org>
    >Cc: Linux-parport mailing list <linux-parport@torque.net>
    >Sent: Wednesday, March 08, 2000 9:55 AM
    >Subject: Re: New queueing code
    >
    >
    >> On Wed, 8 Mar 2000, Eric Youngdale wrote:
    >> > Which source files are involved with ParIDE, and what are the
    >symptoms
    >> >of the breakage?
    >> >
    >> >-Eric
    >>
    >> Source files:
    >> drivers/block/paride/*, in particular pd.c, pf.c, pt.c, pcd.c and pg.c.
    >>
    >> Symptoms:
    >> >Did a little more experimenting with EPAT + PD in kernel 2.3.49pre1
    >> >today.
    >> >A log message I got, without taking the computer down with it this time:
    >> >
    >> >Mar 7 20:03:16 frankenstein kernel: pda: OUCH: b_reqnext != NULL
    >> >Mar 7 20:03:17 frankenstein last message repeated 46 times
    >> >
    >> >A PD bug? Or is that assertion failure the lower-level protocol's fault?
    >>
    >> (I was later assured this was a symptom of incompatible request queue
    >> changes.)
    >>
    >> >The next time I tried, it panicked, dumped a screenful of traceback (more
    >> >than 50 lines) that seemed, judging by the addresses, to be in an
    >> >infinite loop. Then the interrupt handler was killed so that I have no
    >> >log of the incident. (Sorry, no serial console to capture such things
    >> >handy)
    >> >
    >> >Sysrq-K in this state caused an immediate reboot.
    >>
    >> --
    >> Erik I. Bolsų <knan at mo.himolde.no> | <eriki at himolde.no>
    >> .. student of IT administration at Molde College, Norway.
    >>
    >

    --
    Erik I. Bolsų <knan at mo.himolde.no> | <eriki at himolde.no>
    .. student of IT administration at Molde College, Norway.
    

    -- To unsubscribe, send mail to: linux-parport-request@torque.net -- -- with the single word "unsubscribe" in the body of the message. --



    This archive was generated by hypermail 2b29 : Wed Mar 08 2000 - 10:46:29 EST