Date: Wed, 8 Aug 2001 21:25:17 +0200 (MET DST) From: Per Andersson <pa@cdg.chalmers.se> To: Warner Losh <imp@harmony.village.org> Cc: <mobile@freebsd.org> Subject: Re: pccard patches problem Message-ID: <Pine.GSO.4.33.0108082107430.3579-100000@wilfer4.cdg.chalmers.se> In-Reply-To: <200108081627.f78GRh123212@harmony.village.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 8 Aug 2001, Warner Losh wrote: > In message <Pine.GSO.4.33.0108081714410.3429-500000@wilfer4.cdg.chalmers.se> Per Andersson writes: > : I have some trouble with your patches, I have applied them against -stable > : from august 6:th about 11:00 UTC. I'm running on a Thinkpad 570E and the > : behavior I see is quite strange. > : If I use slot 0 it "almost" works, the system doesn't "recognise" a card by > : itself when I insert it but if I do a "pccardc power 0 1" it finds it and > : assigns I/O adresse and a driver. Network cards doesn't seem to work at > : all and with modemcards I only get data from them when I send something, > : I guess all this is some interrupt problem. > : If I then do a "pccardc power 0 0" it detaches the card, but if I just > : remove it the system doesn't notice. > > OK. Given the date that you updated, the insert/remove part of this > is understandible. I didn't MFC the fixes for that on TI chipsets > until 14:47 UTC. It also sounds like the cards aren't getting their > interrupts for some reason. Not sure why. Maybe my patches will fix > it, or maybe not. It will definitely fix the card insert/remove > problems. > > Maybe you could update to Aug 7th or later -stable and try again. If > it still fails, can you send me > > pciconf -r pci0:2:0 0:0xff > pciconf -r pci0:2:1 0:0xff > pccardc rdreg > > so I can look at all the register settings. Maybe something isn't > getting setup properly that defaults to what we want on my laptop. I > notice you have a TI-1450 chipset. I have the TI-4451, which is very > close to yours except it also has the firewire interface in it too. > I have made an update, and applied the patches (this time there was some hunks that failed, but they seem to be in place already) again but it still fails, in the same way as far as I can see. Here is the output you asked for: bigblue# pciconf -r pci0:2:0 0:0xff 0xac1b104c 0x02100007 0x06070003 0x00824008 0x50000000 0x020000a0 0xb0040200 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x0760010b 0x01301014 0x00000001 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x0864d078 0x00000000 0x80010000 0x00001000 0x606002c0 0x00000000 0x00000000 0x00000000 0xfe110001 0x00c00000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 bigblue# pciconf -r pci0:2:1 0:0xff 0xac1b104c 0x02100007 0x06070003 0x00824008 0x50100000 0x020000a0 0xb0070500 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x0760020b 0x01301014 0x00000001 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x0864d078 0x00000000 0x80010000 0x00001000 0x606002c0 0x00000000 0x00000000 0x00000000 0xfe110001 0x00c00000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 bigblue# pccardc rdreg Registers for slot 0 00: 84 00 10 50 00 00 00 00 00 00 00 00 00 00 00 00 10: 00 00 00 00 00 00 c0 00 00 00 00 00 00 00 00 00 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Registers for slot 1 00: 84 00 10 50 00 00 00 00 00 00 00 00 00 00 00 00 10: 00 00 00 00 00 00 c0 00 00 00 00 00 00 00 00 00 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 bigblue# exit > Also, make sure that you are using the patched pccardd (but it looks > like you may be, but I'm just trying to cover all bases.). Yes I hope I have been able to use the right pccardd to the right kernel. > > : If I use slot 1, the system freezes and I have to power it off to recover, > : no output on the console. If I have a card in slot 1 when booting it freezes > : just after printing "pccard1: <PC Card bus (classic)> on pcic1" > > Again, sounds like an interrupt problem of some sort. Since it > freezes just after printing the pcic child attachment, my guess is > that it is enabling interrupts that are being routed somewhere else > and we're not fielding them. This surprises me a little since I'd > only expect that behavior on machines where pcic0 and pcic1 have > different irqs and it appears that your machine assigns 11 to both. I also noted an other thing, if I don't turn the laptop off before booting the old (unpatched) kernel it will not detect card insert/removal. /Per Andersson To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-mobile" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.GSO.4.33.0108082107430.3579-100000>