Date: Sun, 07 Jan 2001 02:36:54 -0800 From: Peter Wemm <peter@netplex.com.au> To: Jonathan Chen <jon@FreeBSD.org> Cc: Greg Lehey <grog@lemis.com>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/cardbus cardbus.c cardbus_cis.c cardbus_cis.h src/sys/dev/pccard card_if.m pccard.c pccardvar.h src/sys/dev/pccbb pccbb.c pccbbvar.h Message-ID: <200101071036.f07AasB01440@mobile.wemm.org> In-Reply-To: <20010106234646.B77540@spock.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Jonathan Chen wrote: > On Sun, Jan 07, 2001 at 10:54:25AM +1030, Greg Lehey wrote: > > On Saturday, 6 January 2001 at 10:04:56 -0800, Jonathan Chen wrote: > > > > > > * pccard begins to work > > > - pccard can now use higher memory space (and uses it by default). > > > > How high? Does this mean that, for normal PCMCIA cards, we're no > > longer bound to the 64 kB at 0xd0000? > > Correct. > > > I'm a little hazy about the status of CardBus/pccard support in > > -CURRENT. Is it still an either/or proposition? > > If you want stable pccard support, oldcard provides that, but not cardbus. > Under the latest newcard, cardbus appears to be working great. > Under the latest newcard, pccard support is very experimental, but should > mostly work with the pccbb bridge. There is a nasty bug in the pccard.c code that will break most of the better newbusified drivers that do proper resource allocation/deallocation. The only pccards that I'd expect to work at present are those that allocate their resources in their probe routine, and pass them to the attach via the persistant softc method. The newcard'ified pccard drivers tend to allocate their resources and free them at the end of probe, and then reallocate them if their attach method is called. > With newcard, cardbus and pccard can be used together. You can load the > cardbus module for cardbus support, pccard for pccard support, or both if > so desired. The pccbb bridge will do the right thing depending on what > card you inserted. One caveat.. the pccbb code does not seem to be disabling the old-style isa pcic interface at 0x3e0/3e1 on my machine. I've been looking through the ACPI AML code for my laptop and can see the inb/outb sequences that are used to turn this on and off, but we dont seem to do it in the driver. Hence, the isa pcic driver will "see" the compatability ports and get somewhat upset by thinking that you have four pccard slots. Fortunately, it fails to initialize the isa compatability ports once the pccbb code has initialized itself - the act of initializing it in cardbus mode seems to break the isa port access method just enough to stop it working. If the setup methods are pretty standard on the yenta interface chips, then perhaps we can take the 'disable 0x3e0' sequence and use that.... Alternatively, be sure to leave out 'device pcic' from your cardbus kernel config. Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200101071036.f07AasB01440>