Hi, This is the problem I'm having as well - that of running the Xircom Realport REM56G on NEWCARD. The mapped attribute memory is bogusly all zeroes, and as a result the parsing code simply runs off the end of the mapped memory and thus faulting.
Testing for the condition avoids the fault but leaves the card unusable under -newcard. ====================================================================== --- pccard_cis.c Tue Oct 8 13:04:34 2002 +++ pccard_cis.c.new Tue Oct 8 13:04:01 2002 @@ -152,6 +152,11 @@ while (1) { /* get the tuple code */ + if (tuple.ptr >= PCCARD_CIS_SIZE) { + ret = 1; + goto done; + } + tuple.code = pccard_cis_read_1(&tuple, tuple.ptr); /* two special-case tuples */ ====================================================================== Regards -ah-zeep- On Mon, 7 Oct 2002, M. Warner Losh wrote: > Date: Mon, 07 Oct 2002 22:02:52 -0600 (MDT) > From: M. Warner Losh <[EMAIL PROTECTED]> > To: [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED] > Subject: Re: dc and PCMCIA still panic > > In message: <[EMAIL PROTECTED]> > Mikhail Teterin <[EMAIL PROTECTED]> writes: > : > Even though it doesn't make sense, can you turn on the debugging > : > information and run again? I use > : > > : > # Let's debug! > : > hw.cbb.debug=1 > : > hw.pccard.debug=1 > : > hw.pccard.cis_debug=1 > : > hw.cardbus.debug=1 > : > hw.cardbus.cis_debug=1 > : > : Actually, I lied... It is a Xircom RealPort Ethernet 10/100 + Modem 56 > : REM56G -- the xe card. > > Oh, that's different.... This is with my changes to pccard_cis.c? > I've never been able to get my IBM version of this card to work at > all... > > Do you get screen fulls of CIS parsing before the fall? That's what I > see both before and after that change. I think that something strange > is going on with a few cards that the code doesn't handle quite > right. I have maybe 4 of them at the moment. > > Warner > > > To Unsubscribe: send mail to [EMAIL PROTECTED] > with "unsubscribe freebsd-current" in the body of the message > To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message