From owner-freebsd-mobile Mon Oct 20 09:23:58 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA02917 for mobile-outgoing; Mon, 20 Oct 1997 09:23:58 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from fegmania.yahoo.com (sjx-ca115-42.ix.netcom.com [207.223.162.106]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA02910 for ; Mon, 20 Oct 1997 09:23:53 -0700 (PDT) (envelope-from bryan@fegmania.yahoo.com) Received: (from bryan@localhost) by fegmania.yahoo.com (8.8.7/8.8.7) id JAA26495; Mon, 20 Oct 1997 09:22:06 GMT Date: Mon, 20 Oct 1997 09:22:06 GMT Message-Id: <199710200922.JAA26495@fegmania.yahoo.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 From: bryan@yahoo-inc.com (bryan d. o'connor) To: freebsd-mobile@FreeBSD.ORG Subject: memory corruption? X-Mailer: VM 6.34 under 19.15p7 XEmacs Lucid X-Remote: break the ties that bind Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by hub.freebsd.org id JAA02913 Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk i'm running 2.2-Stable (i've been staying pretty up to date with it) on a Hitachi Visionbook Elite with 80M. for the most part, it works flawlessly. i spend the majority of my time running X, xemacs, netscape, ssh, and ppp. when i try to do a make world, though, it will fail at least 75% of the time, all in different places. it will complain about symbols not being found or variables not being defined... the problem is that the symbol or variable name is not correct.. for example -- gcc will complain: "300: foo_bar_wymbol not defined" and at line 300 it will instead be "foo_bar_symbol". or last night ld complained: "ld: Ùb: No such file or directory" i'm guessing that there some sort of memory corruption going on. it only seems to happen when i'm doing a make world. (granted, i haven't compiled much else on this system, though) i also sometimes get these messages about wd0: fegmania /kernel: wd0: interrupt timeout: fegmania /kernel: wd0: status 58 error 1 (but not necessarily all the times that "make world" dies) could i have a bad drive that's causing me swap problems? where do i begin to track down this problem? (also... i try to check my swap with "pstat -T" and it tells me pstat: sysctl: KERN_VNODE: No such file or directory did i miss some configuration needed for stable?) thanks.. ...bryan From owner-freebsd-mobile Mon Oct 20 11:53:20 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA12740 for mobile-outgoing; Mon, 20 Oct 1997 11:53:20 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from dominator.eecs.harvard.edu (dominator.eecs.harvard.edu [140.247.60.28]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id LAA12735 for ; Mon, 20 Oct 1997 11:53:15 -0700 (PDT) (envelope-from karp@eecs.harvard.edu) Received: (from karp@localhost) by dominator.eecs.harvard.edu (8.6.12/8.6.12) id OAA13219; Mon, 20 Oct 1997 14:53:06 -0400 Date: Mon, 20 Oct 1997 14:53:06 -0400 From: Brad Karp Message-Id: <199710201853.OAA13219@dominator.eecs.harvard.edu> To: bryan@yahoo-inc.com Subject: Re: memory corruption? Cc: freebsd-mobile@freebsd.org Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > i also sometimes get these messages about wd0: > fegmania /kernel: wd0: interrupt timeout: > fegmania /kernel: wd0: status 58 error 1 I posted a note to freebsd-mobile about similar messages a few weeks ago. I was getting these messages along with _long_ disk timeouts--like two minutes or more. And the problem only worsened. In the end, surmising a dying disk, I exchanged the laptop (an IBM 380). I've had the new one for over a week now, and it's not given me a single timeout message or long disk timeout. (Whereas the old one gave me both several times per hour.) So at least for me, the wd0 timeout messages _did_ indicate a bad disk. Seems to me there could be many reasons for the messages other than a bad disk, though. Just a data point... -Brad, karp@eecs.harvard.edu From owner-freebsd-mobile Mon Oct 20 13:53:03 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA19223 for mobile-outgoing; Mon, 20 Oct 1997 13:53:03 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from bex.boone.com (blink.boone.com [208.201.41.97]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id NAA19217 for ; Mon, 20 Oct 1997 13:52:55 -0700 (PDT) (envelope-from jce@bex.boone.com) Received: (from jce@localhost) by bex.boone.com (8.8.5/8.8.5) id PAA04121; Mon, 20 Oct 1997 15:52:55 -0500 (CDT) Message-ID: <19971020155254.42633@bex.boone.com> Date: Mon, 20 Oct 1997 15:52:54 -0500 From: Joe Ennis To: freebsd-mobile@freebsd.org Subject: ether pccard driver support. Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.74e Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Will the ed driver support the Megahertz XJ10BC? The XJ10BC is supposed to be using an SMC chip. Thanks,, -- Joe Ennis - jce@boone.com From owner-freebsd-mobile Mon Oct 20 14:23:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA20909 for mobile-outgoing; Mon, 20 Oct 1997 14:23:48 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from socrates.i-pi.com (socrates.i-pi.com [198.49.217.5]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA20904 for ; Mon, 20 Oct 1997 14:23:42 -0700 (PDT) (envelope-from ingham@i-pi.com) Received: (from ingham@localhost) by socrates.i-pi.com (8.8.5/8.8.5) id OAA07154; Mon, 20 Oct 1997 14:38:43 -0600 (MDT) Message-ID: <19971020143843.13906@socrates.i-pi.com> Date: Mon, 20 Oct 1997 14:38:43 -0600 From: Kenneth Ingham To: freebsd-mobile@FreeBSD.ORG Subject: Re: memory corruption? References: <199710201853.OAA13219@dominator.eecs.harvard.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.74 In-Reply-To: <199710201853.OAA13219@dominator.eecs.harvard.edu>; from Brad Karp on Mon, Oct 20, 1997 at 02:53:06PM -0400 Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > i also sometimes get these messages about wd0: > fegmania /kernel: wd0: interrupt timeout: > fegmania /kernel: wd0: status 58 error 1 I'll add my $0.02. I had similar messages which were cleared up by replacing the disk once, and the controller card another time (after I had replaced the disk and the messages continued). Neither of these were in a laptop, however. Kenneth From owner-freebsd-mobile Mon Oct 20 15:27:39 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA24670 for mobile-outgoing; Mon, 20 Oct 1997 15:27:39 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from DonaldBurr.dyn.ml.org (root@ppp12180.la.inreach.net [206.18.115.180]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA24655 for ; Mon, 20 Oct 1997 15:27:31 -0700 (PDT) (envelope-from dburr@POBoxes.com) Received: from DonaldBurr.dyn.ml.org (dburr@DonaldBurr.DonaldBurr.dyn.ml.org [192.160.60.1]) by DonaldBurr.dyn.ml.org (8.8.5/8.8.5) with SMTP id PAA03246; Mon, 20 Oct 1997 15:27:58 -0700 (PDT) Message-ID: X-Mailer: XFMail 1.1 [p0] on FreeBSD Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199710200922.JAA26495@fegmania.yahoo.com> Date: Mon, 20 Oct 1997 15:22:14 -0700 (PDT) Organization: Starfleet Command From: Donald Burr To: (bryan d. o'connor) Subject: RE: memory corruption? Cc: freebsd-mobile@FreeBSD.ORG Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk -----BEGIN PGP SIGNED MESSAGE----- My secret spy satellite informs me that on 20-Oct-97, bryan d. o'connor wrote: >when i try to do a make world, though, it will fail at least 75% >of the time, all in different places. it will complain about symbols >not being found or variables not being defined... the problem is >that the symbol or variable name is not correct.. for example -- >gcc will complain: > "300: foo_bar_wymbol not defined" >and at line 300 it will instead be "foo_bar_symbol". or last night >ld complained: > "ld: Ùb: No such file or directory" > >i'm guessing that there some sort of memory corruption going on. You got it in one. This sounds like some kind of memory problem. It could either be bad memory, or misconfigured memory and/or motherboard. Does it happen at random times (i.e. it doesn't always give the same error, on the same file and line?) This happened to me once, when overclocking the CPU (it couldn't take it, reducing the cpu back to published specs made it work fine), and once with a misconfigured memory bus (timings/interrupts/etc were set wrong). You might also want to check your GCC executables and all related pieces, to make sure they didn't get somehow clobbered. You wouldn't happen to use an AMD K6 CPU, would you? :) >it only seems to happen when i'm doing a make world. (granted, >i haven't compiled much else on this system, though) Compiling is generally a LOT more cpu-intensive than just running programs (except maybe for real math0intensive stuff like raytracing). 'make world' is even more demanding than compiling most (smaller) packages. I've seen systems where I could compile some stuff (elm, pine, etc. reliably, but make world bursts into flames. >i also sometimes get these messages about wd0: > fegmania /kernel: wd0: interrupt timeout: > fegmania /kernel: wd0: status 58 error 1 >(but not necessarily all the times that "make world" dies) >could i have a bad drive that's causing me swap problems? >where do i begin to track down this problem? Bad drive could be it, but given your other symptoms, I'd tend to think bad motherboard configuration. Do you run some other OS on the machine (e.g. Windows, Linux...?) Is it giving you problems? >(also... i try to check my swap with "pstat -T" and it tells me > pstat: sysctl: KERN_VNODE: No such file or directory >did i miss some configuration needed for stable?) Hmm, is your vnode device (pseudo-device vn in the config file) configured in? If it is, then I'd definitely say there's some weirdness going on... - --- Donald Burr - Ask me for my PGP key | PGP: Your WWW HomePage: http://DonaldBurr.base.org/ ICQ #1347455 | right to Address: P.O. Box 91212, Santa Barbara, CA 93190-1212 | 'Net privacy. Phone: (805) 957-9666 FAX: (800) 492-5954 | USE IT. -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBNEva7fjpixuAwagxAQElxAP6Aq5xpCGMrPxhEE3/oc2dnsYUIls7Vlnk 7WvebJKlxBNmTJDaupZCaXmf6EJ4POg541po7HEf9/j7s4UJZN32QlyhP+F4GDaF eb3mm29B52bDDLrs6iLft9n9ajtudLIGkIozxSTHxjT3WzTYrj3B44rxGCFhdWd5 SCC2ZVFXfvU= =2zKN -----END PGP SIGNATURE----- From owner-freebsd-mobile Mon Oct 20 16:19:56 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id QAA27960 for mobile-outgoing; Mon, 20 Oct 1997 16:19:56 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from coconut.itojun.org (root@coconut.itojun.org [210.160.95.97]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id QAA27943 for ; Mon, 20 Oct 1997 16:19:27 -0700 (PDT) (envelope-from itojun@itojun.org) From: itojun@itojun.org Received: from localhost (itojun@localhost.itojun.org [127.0.0.1]) by coconut.itojun.org (8.8.5/3.6Wbeta6) with ESMTP id IAA20503; Tue, 21 Oct 1997 08:18:59 +0900 (JST) To: Joe Ennis cc: freebsd-mobile@FreeBSD.ORG In-reply-to: jce's message of Mon, 20 Oct 1997 15:52:54 EST. <19971020155254.42633@bex.boone.com> X-Template-Reply-To: itojun@itojun.org X-Template-Return-Receipt-To: itojun@itojun.org X-PGP-Fingerprint: F8 24 B4 2C 8C 98 57 FD 90 5F B4 60 79 54 16 E2 Subject: Re: ether pccard driver support. Date: Tue, 21 Oct 1997 08:18:59 +0900 Message-ID: <20499.877389539@coconut.itojun.org> Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >Will the ed driver support the Megahertz XJ10BC? >The XJ10BC is supposed to be using an SMC chip. XJ10BC/BT has SMC91C92 chip on it, which is different from the chip supported by "ed" driver. XJ10BT(and possibly BC, I have no experience with BC) is is supported by "sn" driver that comes with PAO kit. http://www.jp.freebsd.org/PAO/ itojun From owner-freebsd-mobile Tue Oct 21 08:00:59 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id IAA17196 for mobile-outgoing; Tue, 21 Oct 1997 08:00:59 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from upsilon.cs.fsu.edu (upsilon.cs.fsu.edu [128.186.121.34]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA17190; Tue, 21 Oct 1997 08:00:57 -0700 (PDT) (envelope-from uh@NU.cs.fsu.edu) Received: (from uh@localhost) by upsilon.cs.fsu.edu (8.8.5/8.7.3) id LAA29317; Tue, 21 Oct 1997 11:00:53 -0400 (EDT) From: Gang-Ryung Uh Message-Id: <199710211500.LAA29317@upsilon.cs.fsu.edu> Subject: 3C589D PCMCIA card configuration To: questions@freebsd.org Date: Tue, 21 Oct 1997 11:00:52 -0400 (EDT) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, I am running 2.2.2 with PAO package on NEC versa laptop I would like to configure 3C589D PCMCIA card for the laptop. I looked at the config file and check IO address and interrupt queue seems correct. During the boot time: ------ ep0 not found at 0x300 <---- ??? .... PC-Card VLSI 82C146 (5 mem & 2 I/O windows) pcic: controller irq 3 Card inserted, slot 0 Slot 0, unfielded interrupt (0) ep0: utp/bnc[*UTP*] address 00:60:97:8d:d0:45 ----- After the boot, I typed "ifconfig -a" ---- lp0: .. tun0: .. sl0: .. lo0: .. ep0: flags=8802 mtu 1500 ether 00:60:..:..:..:.. ---- I set the "rc.conf" as follows, but ep0 is not up and running: ---- defalutrouter="128.xx.xx.1" .. router_enable="YES" pccard_ether="ep0" ifconfig_common="inet 128.xx.xx.99 netmask 255.255.255.0" ifconfig_ep0="$ifconfig_common" ---- Would you tell me what I did wrong? Thanks in advance. Regards, Gang-Ryung Uh (uh@cs.fsu.edu) From owner-freebsd-mobile Tue Oct 21 09:16:53 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA21429 for mobile-outgoing; Tue, 21 Oct 1997 09:16:53 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA21402 for ; Tue, 21 Oct 1997 09:16:50 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id KAA22801; Tue, 21 Oct 1997 10:16:43 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id KAA06176; Tue, 21 Oct 1997 10:16:42 -0600 (MDT) Date: Tue, 21 Oct 1997 10:16:42 -0600 (MDT) Message-Id: <199710211616.KAA06176@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Gang-Ryung Uh Cc: mobile@freebsd.org Subject: Re: 3C589D PCMCIA card configuration In-Reply-To: <199710211500.LAA29317@upsilon.cs.fsu.edu> References: <199710211500.LAA29317@upsilon.cs.fsu.edu> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [ Moved to -mobile ] > I am running 2.2.2 with PAO package on NEC versa laptop > I would like to configure 3C589D PCMCIA card for the laptop. It's recognized and ready to go, you just never had it setup. > I looked at the config file and check IO address and interrupt > queue seems correct. > > During the boot time: > > ------ > ep0 not found at 0x300 <---- ??? It didn't find a non-PCMCIA card. > .... > > PC-Card VLSI 82C146 (5 mem & 2 I/O windows) > pcic: controller irq 3 > Card inserted, slot 0 > Slot 0, unfielded interrupt (0) > ep0: utp/bnc[*UTP*] address 00:60:97:8d:d0:45 It's recognized, and the driver is allocated for it. > After the boot, I typed "ifconfig -a" > > ---- > lp0: .. > tun0: .. > sl0: .. > lo0: .. > ep0: flags=8802 mtu 1500 > ether 00:60:..:..:..:.. > ---- > > I set the "rc.conf" as follows, but ep0 is not up and running: > > ---- > defalutrouter="128.xx.xx.1" > .. > router_enable="YES" > > pccard_ether="ep0" > ifconfig_common="inet 128.xx.xx.99 netmask 255.255.255.0" > ifconfig_ep0="$ifconfig_common" You need to set the ifconfig_pccard line, since the ifconfig_ep0 line is used for 'desktop' units, and not PCMCIA card. Note, it still may not work right (it may be using the wrong network connector), so you may need to modify /etc/pccard.conf to switch between the different connectors, which is done by using the link flags. Different models require different settings. Nate From owner-freebsd-mobile Tue Oct 21 11:11:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA00200 for mobile-outgoing; Tue, 21 Oct 1997 11:11:43 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA00190; Tue, 21 Oct 1997 11:11:35 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id MAA23603; Tue, 21 Oct 1997 12:11:33 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id MAA06747; Tue, 21 Oct 1997 12:11:32 -0600 (MDT) Date: Tue, 21 Oct 1997 12:11:32 -0600 (MDT) Message-Id: <199710211811.MAA06747@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Nate Williams Cc: cvs-committers@freebsd.org, mobile@freebsd.org Subject: Re: cvs commit: src/sys/i386/isa aic6360.c In-Reply-To: <199710211757.KAA09096@freefall.freebsd.org> References: <199710211757.KAA09096@freefall.freebsd.org> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Nate Williams writes: > nate 1997/10/21 10:57:34 PDT > > Modified files: > sys/i386/isa aic6360.c > Log: > - PCCARD support to the much maligned SCSI driver. This gives up support > for a couple of external CD's (notably the Sony PRD-650). > > Note: In order to get my CD recognized, I had to configure the CD under > Win95, but it seems to work now even if I turn it off. > > Submitted by: PAO [minor mods by me] This doesn't work the first time I 'mount' the disk (if it boots with the disk in). However, if I open up the CD player, wait for the disk to stop, and then close the player, I can successfully mount the disk. It's not great, but it's better than nothing. (You might try booting with the disk gone to see if that helps...) Nate From owner-freebsd-mobile Tue Oct 21 12:23:57 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id MAA04385 for mobile-outgoing; Tue, 21 Oct 1997 12:23:57 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id MAA04379 for ; Tue, 21 Oct 1997 12:23:50 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id NAA24083; Tue, 21 Oct 1997 13:23:41 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id NAA07091; Tue, 21 Oct 1997 13:23:39 -0600 (MDT) Date: Tue, 21 Oct 1997 13:23:39 -0600 (MDT) Message-Id: <199710211923.NAA07091@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: mobile@freebsd.org CC: handy@sag.space.lockheed.com Subject: Patches from -current for -stable I'd like to commit after testing X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I'd like to bring in most of the recent changes to 2.2, not that 2.2.5 is out the door. (I didn't want to break things for the release, and didn't have time to test them enough for the release.) These changes provide very little additions to the default behavior (mostly fixes), but do add some additional functionality via sysctl's (machdep.pccard.pcic_resume_reset & machdep.pccard.apm_pccard_resume) if you enable them. Here are the changes in log message style --------------------- cut here --------------------- MFC: Bring in all the 'safe' changes from current - Speaker audio & low power mode on certain Cirrus Logic chips. - Update chip ID comments. - PCIC_RESUME_RESET & APM_PCCARD_RESUME sysctls. - make getb() and putb() member functions of struct pcic_slot. - Allocate the 'PCIC' interrupt from the last available (higher #) IRQ instead of the first available, like Win95 does. - Don't start at IRQ 0 in build-freelist, but instead start at IRQ 1. - Don't apply 5V when we want 3.3V. Kludge around the case where the PCIC refuses to use 5V because it knows better than us. - Make MOD_* macros almost consistent. --------------------- cut here --------------------- If I get get some yay/nay responses from this as to whether or not they work fine, I'll commit them. Unfortunately, I'm only able to test them in -current right now since my 2.2 disk is apparently dead. :( Note, I already committed changes to the aic6360 driver to both -current and -stable hat 'sort of' allows my external CD to work. They're not really robust, but it's better than nothing for now. Nate ----------- =================================================================== RCS file: /home/CVS/src/sys/pccard/i82365.h,v retrieving revision 1.5 diff -c -r1.5 i82365.h *** i82365.h 1996/06/14 11:01:56 1.5 --- i82365.h 1997/10/21 19:12:30 *************** *** 38,51 **** #define PCIC_VLSI 2 /* VLSI chip */ #define PCIC_PD672X 3 /* Cirrus logic 627x */ #define PCIC_PD6710 4 /* Cirrus logic 6710 */ ! #define PCIC_CL6729 5 ! #define PCIC_VG468 6 ! #define PCIC_VG469 7 #define PCIC_RF5C396 8 /* Ricoh RF5C396 */ #define PCIC_IBM_KING 9 /* IBM KING PCMCIA Controller */ ! #ifdef PC98 ! #define PCIC_PC98 10 ! #endif /* * Address of the controllers. Each controller can manage * two PCMCIA slots. Up to 8 slots are supported in total. --- 38,51 ---- #define PCIC_VLSI 2 /* VLSI chip */ #define PCIC_PD672X 3 /* Cirrus logic 627x */ #define PCIC_PD6710 4 /* Cirrus logic 6710 */ ! #define PCIC_CL6729 5 /* Cirrus logic 6729 */ ! #define PCIC_VG468 6 /* Vadem 468 */ ! #define PCIC_VG469 7 /* Vadem 469 */ #define PCIC_RF5C396 8 /* Ricoh RF5C396 */ #define PCIC_IBM_KING 9 /* IBM KING PCMCIA Controller */ ! #define PCIC_PC98 10 /* NEC PC98 PCMCIA Controller */ ! #define PCIC_TI1130 11 /* TI PCI1130 CardBus */ ! /* * Address of the controllers. Each controller can manage * two PCMCIA slots. Up to 8 slots are supported in total. *************** *** 57,66 **** * identify the port number, and the lower 6 bits * select one of the 64 possible data registers. */ ! #define PCIC_INDEX_0 0x3E0 /* index reg, chips 0 and 1 */ ! #define PCIC_DATA_0 0x3E1 /* data register, chips 0 and 1 */ ! #define PCIC_INDEX_1 0x3E2 /* index reg, chips 2 and 3 */ ! #define PCIC_DATA_1 0x3E3 /* data register, chips 2 and 3 */ /* * Register index addresses. */ --- 57,66 ---- * identify the port number, and the lower 6 bits * select one of the 64 possible data registers. */ ! #define PCIC_INDEX_0 0x3E0 /* index reg, chips 0 and 1 */ ! #define PCIC_DATA_0 (PCIC_INDEX_0 + 1) /* data reg, chips 0 and 1 */ ! #define PCIC_INDEX_1 (PCIC_INDEX_0 + 2) /* index reg, chips 2 and 3 */ ! #define PCIC_DATA_1 (PCIC_INDEX_1 + 1) /* data reg, chips 2 and 3 */ /* * Register index addresses. */ *************** *** 76,82 **** --- 76,84 ---- #define PCIC_IO1 0x0c /* I/O Address 1 */ #define PCIC_MEMBASE 0x10 /* Base of memory window registers */ #define PCIC_CDGC 0x16 /* Card Detect and General Control */ + #define PCIC_MISC1 0x16 /* PD672x: Misc control register 1 per slot */ #define PCIC_GLO_CTRL 0x1e /* Global Control Register */ + #define PCIC_MISC2 0x1e /* PD672x: Misc control register 2 per chip */ #define PCIC_TIME_SETUP0 0x3a #define PCIC_TIME_CMD0 0x3b *************** *** 197,208 **** --- 199,216 ---- #define PCIC_CDRES_EN 0x10 /* card detect resume enable */ #define PCIC_SW_CD_INT 0x20 /* s/w card detect interrupt */ + /* For Misc. Control Register 1 */ + #define PCIC_SPKR_EN 0x10 /* Cirrus PD672x: speaker enable */ + /* For Global Control register (PCIC_GLO_CTRL) */ #define PCIC_PWR_DOWN 0x01 /* power down */ #define PCIC_LVL_MODE 0x02 /* level mode interrupt enable */ #define PCIC_WB_CSCINT 0x04 /* explicit write-back csc intr */ #define PCIC_IRQ0_LEVEL 0x08 /* irq 14 pulse mode enable */ #define PCIC_IRQ1_LEVEL 0x10 + + /* For Misc. Control Register 2 */ + #define PCIC_LPDM_EN 0x02 /* Cirrus PD672x: low power dynamic mode */ /* * Mask of allowable interrupts. =================================================================== RCS file: /home/CVS/src/sys/pccard/pccard.c,v retrieving revision 1.27.2.1 diff -c -r1.27.2.1 pccard.c *** pccard.c 1997/03/04 00:27:24 1.27.2.1 --- pccard.c 1997/10/21 19:12:30 *************** *** 35,45 **** --- 35,47 ---- #include #include #include + #include #include #ifdef DEVFS #include #endif /*DEVFS*/ #include + /*#include */ #include #include *************** *** 56,61 **** --- 58,84 ---- #include #include + SYSCTL_NODE(_machdep, OID_AUTO, pccard, CTLFLAG_RW, 0, "pccard"); + + static int pcic_resume_reset = + #ifdef PCIC_RESUME_RESET + 1; + #else + 0; + #endif + + SYSCTL_INT(_machdep_pccard, OID_AUTO, pcic_resume_reset, CTLFLAG_RW, + &pcic_resume_reset, 0, ""); + + static int apm_pccard_resume = + #ifdef APM_PCCARD_RESUME + 1; + #else + 0; + #endif + + SYSCTL_INT(_machdep_pccard, OID_AUTO, apm_pccard_resume, CTLFLAG_RW, + &apm_pccard_resume, 0, ""); #define PCCARD_MEMSIZE (4*1024) *************** *** 332,338 **** struct pccard_dev *dp; for (dp = sp->devices; dp; dp = dp->next) ! (void) dp->drv->suspend(dp); if (!sp->suspend_power) sp->ctrl->disable(sp); return (0); --- 355,361 ---- struct pccard_dev *dp; for (dp = sp->devices; dp; dp = dp->next) ! (void)dp->drv->suspend(dp); if (!sp->suspend_power) sp->ctrl->disable(sp); return (0); *************** *** 342,359 **** slot_resume(void *arg) { struct slot *sp = arg; - struct pccard_dev *dp; ! #ifdef PCIC_RESUME_RESET ! sp->ctrl->resume(sp); ! #endif ! if (!sp->suspend_power) ! sp->ctrl->power(sp); ! if (sp->irq) ! sp->ctrl->mapirq(sp, sp->irq); ! for (dp = sp->devices; dp; dp = dp->next) ! (void) dp->drv->init(dp, 0); return (0); } #endif /* NAPM > 0 */ --- 365,396 ---- slot_resume(void *arg) { struct slot *sp = arg; ! if (pcic_resume_reset) ! sp->ctrl->resume(sp); ! if (apm_pccard_resume) { ! /* Fake card removal/insertion events */ ! if (sp->state == filled) { ! int s; ! s = splhigh(); ! disable_slot(sp); ! sp->state = empty; ! splx(s); ! sp->insert_seq = 1; ! timeout(inserted, (void *)sp, hz); ! selwakeup(&sp->selp); ! } ! } else { ! struct pccard_dev *dp; ! ! if (!sp->suspend_power) ! sp->ctrl->power(sp); ! if (sp->irq) ! sp->ctrl->mapirq(sp, sp->irq); ! for (dp = sp->devices; dp; dp = dp->next) ! (void)dp->drv->init(dp, 0); ! } return (0); } #endif /* NAPM > 0 */ *************** *** 439,445 **** int irq; unsigned int mask; ! for (irq = 1; irq < ICU_LEN; irq++) { mask = 1ul << irq; if (!(mask & imask)) continue; --- 476,482 ---- int irq; unsigned int mask; ! for (irq = ICU_LEN; irq > 0; irq--) { mask = 1ul << irq; if (!(mask & imask)) continue; *************** *** 614,622 **** */ sp->pwr.vcc = 50; sp->pwr.vpp = 0; ! untimeout(power_off_slot, (caddr_t)sp); ! if (sp->pwr_off_pending) sp->ctrl->disable(sp); sp->pwr_off_pending = 0; sp->ctrl->power(sp); printf("Card inserted, slot %d\n", sp->slot); --- 651,660 ---- */ sp->pwr.vcc = 50; sp->pwr.vpp = 0; ! if (sp->pwr_off_pending) { ! untimeout(power_off_slot, (caddr_t)sp); sp->ctrl->disable(sp); + } sp->pwr_off_pending = 0; sp->ctrl->power(sp); printf("Card inserted, slot %d\n", sp->slot); =================================================================== RCS file: /home/CVS/src/sys/pccard/pcic.c,v retrieving revision 1.23.2.1 diff -c -r1.23.2.1 pcic.c *** pcic.c 1997/03/04 00:27:29 1.23.2.1 --- pcic.c 1997/10/21 19:12:30 *************** *** 39,44 **** --- 39,45 ---- #include #include + #include #include #include *************** *** 81,86 **** --- 82,90 ---- char controller; /* Device type */ char revision; /* Device Revision */ struct slot *slotp; /* Back ptr to slot */ + u_char (*getb)(struct pcic_slot *sp, int reg); + void (*putb)(struct pcic_slot *sp, int reg, u_char val); + u_char *regs; /* Pointer to regs in mem */ } pcic_slots[PCIC_MAX_SLOTS]; static int pcic_irq; *************** *** 95,114 **** * Read a register from the PCIC. */ static inline unsigned char ! getb (struct pcic_slot *sp, int reg) { ! outb (sp->index, sp->offset + reg); ! return inb (sp->data); } /* * Write a register on the PCIC */ static inline void ! putb (struct pcic_slot *sp, int reg, unsigned char val) { ! outb (sp->index, sp->offset + reg); ! outb (sp->data, val); } /* --- 99,130 ---- * Read a register from the PCIC. */ static inline unsigned char ! getb1(struct pcic_slot *sp, int reg) { ! outb(sp->index, sp->offset + reg); ! return inb(sp->data); ! } ! ! static inline unsigned char ! getb2(struct pcic_slot *sp, int reg) ! { ! return (sp->regs[reg]); } /* * Write a register on the PCIC */ static inline void ! putb1(struct pcic_slot *sp, int reg, unsigned char val) { ! outb(sp->index, sp->offset + reg); ! outb(sp->data, val); ! } ! ! static inline void ! putb2(struct pcic_slot *sp, int reg, unsigned char val) ! { ! sp->regs[reg] = val; } /* *************** *** 117,123 **** static inline void clrb(struct pcic_slot *sp, int reg, unsigned char mask) { ! putb (sp, reg, getb (sp, reg) & ~mask); } /* --- 133,139 ---- static inline void clrb(struct pcic_slot *sp, int reg, unsigned char mask) { ! sp->putb(sp, reg, sp->getb(sp, reg) & ~mask); } /* *************** *** 126,142 **** static inline void setb(struct pcic_slot *sp, int reg, unsigned char mask) { ! putb (sp, reg, getb (sp, reg) | mask); } /* * Write a 16 bit value to 2 adjacent PCIC registers */ static inline void ! putw (struct pcic_slot *sp, int reg, unsigned short word) { ! putb (sp, reg, word & 0xFF); ! putb (sp, reg + 1, (word >> 8) & 0xff); } --- 142,158 ---- static inline void setb(struct pcic_slot *sp, int reg, unsigned char mask) { ! sp->putb(sp, reg, sp->getb(sp, reg) | mask); } /* * Write a 16 bit value to 2 adjacent PCIC registers */ static inline void ! putw(struct pcic_slot *sp, int reg, unsigned short word) { ! sp->putb(sp, reg, word & 0xFF); ! sp->putb(sp, reg + 1, (word >> 8) & 0xff); } *************** *** 213,219 **** int pcic_mod(struct lkm_table *lkmtp, int cmd, int ver) { ! DISPATCH(lkmtp,cmd,ver,pcic_handle,pcic_handle,lkm_nullcmd); } /* --- 229,236 ---- int pcic_mod(struct lkm_table *lkmtp, int cmd, int ver) { ! MOD_DISPATCH(pcic, lkmtp, cmd, ver, ! pcic_handle, pcic_handle, lkm_nullcmd); } /* *************** *** 226,236 **** int slot; struct pcic_slot *sp = pcic_slots; ! untimeout(pcictimeout,0); if (pcic_irq) { for (slot = 0; slot < PCIC_MAX_SLOTS; slot++, sp++) { if (sp->slotp) ! putb(sp, PCIC_STAT_INT, 0); } unregister_intr(pcic_irq, pcicintr); } --- 243,253 ---- int slot; struct pcic_slot *sp = pcic_slots; ! untimeout(pcictimeout, 0); if (pcic_irq) { for (slot = 0; slot < PCIC_MAX_SLOTS; slot++, sp++) { if (sp->slotp) ! sp->putb(sp, PCIC_STAT_INT, 0); } unregister_intr(pcic_irq, pcicintr); } *************** *** 275,281 **** freemask = 0; /* Walk through all of the IRQ's and find any that aren't allocated. */ ! for (irq = 0; irq < ICU_LEN; irq++) { /* * If the PCIC controller can't generate it, don't * bother checking to see if it it's free. --- 292,298 ---- freemask = 0; /* Walk through all of the IRQ's and find any that aren't allocated. */ ! for (irq = 1; irq < ICU_LEN; irq++) { /* * If the PCIC controller can't generate it, don't * bother checking to see if it it's free. *************** *** 356,364 **** * The values are all stored as the upper 12 bits of the * 24 bit address i.e everything is allocated as 4 Kb chunks. */ ! putw (sp, reg, sys_addr & 0xFFF); ! putw (sp, reg+2, (sys_addr + (mp->size >> 12) - 1) & 0xFFF); ! putw (sp, reg+4, ((mp->card >> 12) - sys_addr) & 0x3FFF); #if 0 printf("card offs = card_adr = 0x%x 0x%x, sys_addr = 0x%x\n", mp->card, ((mp->card >> 12) - sys_addr) & 0x3FFF, --- 373,381 ---- * The values are all stored as the upper 12 bits of the * 24 bit address i.e everything is allocated as 4 Kb chunks. */ ! putw(sp, reg, sys_addr & 0xFFF); ! putw(sp, reg+2, (sys_addr + (mp->size >> 12) - 1) & 0xFFF); ! putw(sp, reg+4, ((mp->card >> 12) - sys_addr) & 0x3FFF); #if 0 printf("card offs = card_adr = 0x%x 0x%x, sys_addr = 0x%x\n", mp->card, ((mp->card >> 12) - sys_addr) & 0x3FFF, *************** *** 385,408 **** printf("Map window to sys addr 0x%x for %d bytes, card 0x%x\n", mp->start, mp->size, mp->card); printf("regs are: 0x%02x%02x 0x%02x%02x 0x%02x%02x flags 0x%x\n", ! getb(sp, reg), getb(sp, reg+1), ! getb(sp, reg+2), getb(sp, reg+3), ! getb(sp, reg+4), getb(sp, reg+5), mp->flags); #endif /* * Enable the memory window. By experiment, we need a delay. */ ! setb (sp, PCIC_ADDRWINE, (1<start, mp->size, mp->card); printf("regs are: 0x%02x%02x 0x%02x%02x 0x%02x%02x flags 0x%x\n", ! sp->getb(sp, reg), sp->getb(sp, reg+1), ! sp->getb(sp, reg+2), sp->getb(sp, reg+3), ! sp->getb(sp, reg+4), sp->getb(sp, reg+5), mp->flags); #endif /* * Enable the memory window. By experiment, we need a delay. */ ! setb(sp, PCIC_ADDRWINE, (1<start, ip->size, win); #endif /* PCIC_DEBUG */ ! putw (sp, reg, ip->start); ! putw (sp, reg+2, ip->start+ip->size-1); x = 0; if (ip->flags & IODF_ZEROWS) x |= PCIC_IO_0WS; --- 491,498 ---- #ifdef PCIC_DEBUG printf("Map I/O 0x%x (size 0x%x) on Window %d\n", ip->start, ip->size, win); #endif /* PCIC_DEBUG */ ! putw(sp, reg, ip->start); ! putw(sp, reg+2, ip->start+ip->size-1); x = 0; if (ip->flags & IODF_ZEROWS) x |= PCIC_IO_0WS; *************** *** 490,513 **** * Flags for window 0 in lower nybble, and in upper nybble * for window 1. */ ! ioctlv = getb(sp, PCIC_IOCTL); DELAY(100); switch (win) { case 0: ! putb(sp, PCIC_IOCTL, x | (ioctlv & 0xf0)); break; case 1: ! putb(sp, PCIC_IOCTL, (x << 4) | (ioctlv & 0xf)); break; } DELAY(100); ! setb (sp, PCIC_ADDRWINE, mask); DELAY(100); } else { ! clrb (sp, PCIC_ADDRWINE, mask); DELAY(100); ! putw (sp, reg, 0); ! putw (sp, reg + 2, 0); } return(0); } --- 507,530 ---- * Flags for window 0 in lower nybble, and in upper nybble * for window 1. */ ! ioctlv = sp->getb(sp, PCIC_IOCTL); DELAY(100); switch (win) { case 0: ! sp->putb(sp, PCIC_IOCTL, x | (ioctlv & 0xf0)); break; case 1: ! sp->putb(sp, PCIC_IOCTL, (x << 4) | (ioctlv & 0xf)); break; } DELAY(100); ! setb(sp, PCIC_ADDRWINE, mask); DELAY(100); } else { ! clrb(sp, PCIC_ADDRWINE, mask); DELAY(100); ! putw(sp, reg, 0); ! putw(sp, reg + 2, 0); } return(0); } *************** *** 559,564 **** --- 576,583 ---- /* * Initialise the PCIC slot table. */ + sp->getb = getb1; + sp->putb = putb1; if (slot < 4) { sp->index = PCIC_INDEX_0; sp->data = PCIC_DATA_0; *************** *** 575,581 **** * ones would need to be probed at the new offset we set after * we assume it's broken. */ ! if (slot == 1 && maybe_vlsi && getb(sp, PCIC_ID_REV) != 0x84) { sp->index += 4; sp->data += 4; sp->offset = PCIC_SLOT_SIZE << 1; --- 594,600 ---- * ones would need to be probed at the new offset we set after * we assume it's broken. */ ! if (slot == 1 && maybe_vlsi && sp->getb(sp, PCIC_ID_REV) != 0x84) { sp->index += 4; sp->data += 4; sp->offset = PCIC_SLOT_SIZE << 1; *************** *** 585,591 **** * Intel PCMCIA controllers use 0x82 and 0x83 * IBM clone chips use 0x88 and 0x89, apparently */ ! c = getb (sp, PCIC_ID_REV); sp->revision = -1; switch(c) { /* --- 604,610 ---- * Intel PCMCIA controllers use 0x82 and 0x83 * IBM clone chips use 0x88 and 0x89, apparently */ ! c = sp->getb(sp, PCIC_ID_REV); sp->revision = -1; switch(c) { /* *************** *** 601,607 **** outb(sp->index, 0x0E); outb(sp->index, 0x37); setb(sp, 0x3A, 0x40); ! c = getb (sp, PCIC_ID_REV); if (c & 0x08) { sp->controller = ((sp->revision = c & 7) == 4) ? PCIC_VG469 : PCIC_VG468 ; --- 620,626 ---- outb(sp->index, 0x0E); outb(sp->index, 0x37); setb(sp, 0x3A, 0x40); ! c = sp->getb(sp, PCIC_ID_REV); if (c & 0x08) { sp->controller = ((sp->revision = c & 7) == 4) ? PCIC_VG469 : PCIC_VG468 ; *************** *** 611,617 **** /* * Check for RICOH RF5C396 PCMCIA Controller */ ! c = getb (sp, 0x3a); if (c == 0xb2) { sp->controller = PCIC_RF5C396; } --- 630,636 ---- /* * Check for RICOH RF5C396 PCMCIA Controller */ ! c = sp->getb(sp, 0x3a); if (c == 0xb2) { sp->controller = PCIC_RF5C396; } *************** *** 639,648 **** /* * Check for Cirrus logic chips. */ ! putb(sp, 0x1F, 0); ! c = getb(sp, 0x1F); if ((c & 0xC0) == 0xC0) { ! c = getb(sp, 0x1F); if ((c & 0xC0) == 0) { if (c & 0x20) sp->controller = PCIC_PD672X; --- 658,667 ---- /* * Check for Cirrus logic chips. */ ! sp->putb(sp, 0x1F, 0); ! c = sp->getb(sp, 0x1F); if ((c & 0xC0) == 0xC0) { ! c = sp->getb(sp, 0x1F); if ((c & 0xC0) == 0) { if (c & 0x20) sp->controller = PCIC_PD672X; *************** *** 688,694 **** * clear out the registers. */ for (i = 2; i < 0x40; i++) ! putb(sp, i, 0); #endif /* PCIC_NOCLRREGS */ /* * OK it seems we have a PCIC or lookalike. --- 707,713 ---- * clear out the registers. */ for (i = 2; i < 0x40; i++) ! sp->putb(sp, i, 0); #endif /* PCIC_NOCLRREGS */ /* * OK it seems we have a PCIC or lookalike. *************** *** 715,724 **** printf("pcic: controller irq %d\n", pcic_irq); } /* * Check for a card in this slot. */ ! setb (sp, PCIC_POWER, PCIC_PCPWRE| PCIC_DISRST); ! if ((getb (sp, PCIC_STATUS) & PCIC_CD) != PCIC_CD) { slotp->laststate = slotp->state = empty; } else { slotp->laststate = slotp->state = filled; --- 734,754 ---- printf("pcic: controller irq %d\n", pcic_irq); } /* + * Modem cards send the speaker audio (dialing noises) + * to the host's speaker. Cirrus Logic PCIC chips must + * enable this. There is also a Low Power Dynamic Mode bit + * that claims to reduce power consumption by 30%, so + * enable it and hope for the best. + */ + if (sp->controller == PCIC_PD672X) { + setb(sp, PCIC_MISC1, PCIC_SPKR_EN); + setb(sp, PCIC_MISC2, PCIC_LPDM_EN); + } + /* * Check for a card in this slot. */ ! setb(sp, PCIC_POWER, PCIC_PCPWRE| PCIC_DISRST); ! if ((sp->getb(sp, PCIC_STATUS) & PCIC_CD) != PCIC_CD) { slotp->laststate = slotp->state = empty; } else { slotp->laststate = slotp->state = filled; *************** *** 728,734 **** * Assign IRQ for slot changes */ if (pcic_irq > 0) ! putb(sp, PCIC_STAT_INT, (pcic_irq << 4) | 0xF); } #ifdef PC98 if (validslots == 0){ --- 758,764 ---- * Assign IRQ for slot changes */ if (pcic_irq > 0) ! sp->putb(sp, PCIC_STAT_INT, (pcic_irq << 4) | 0xF); } #ifdef PC98 if (validslots == 0){ *************** *** 775,780 **** --- 805,811 ---- static int pcic_ioctl(struct slot *slotp, int cmd, caddr_t data) { + struct pcic_slot *sp = slotp->cdata; switch(cmd) { default: *************** *** 784,793 **** */ case PIOCGREG: ((struct pcic_reg *)data)->value = ! getb(slotp->cdata, ((struct pcic_reg *)data)->reg); break; case PIOCSREG: ! putb(slotp->cdata, ((struct pcic_reg *)data)->reg, ((struct pcic_reg *)data)->value); break; } --- 815,824 ---- */ case PIOCGREG: ((struct pcic_reg *)data)->value = ! sp->getb(sp, ((struct pcic_reg *)data)->reg); break; case PIOCSREG: ! sp->putb(sp, ((struct pcic_reg *)data)->reg, ((struct pcic_reg *)data)->value); break; } *************** *** 864,870 **** reg |= PCIC_VCC_5V_KING; break; } ! reg |= PCIC_VCC_5V; if ((sp->controller == PCIC_VG468)|| (sp->controller == PCIC_VG469)) setb(sp, 0x2f, 0x03) ; --- 895,901 ---- reg |= PCIC_VCC_5V_KING; break; } ! reg |= PCIC_VCC_3V; if ((sp->controller == PCIC_VG468)|| (sp->controller == PCIC_VG469)) setb(sp, 0x2f, 0x03) ; *************** *** 886,897 **** } break; } ! putb (sp, PCIC_POWER, reg); DELAY(300*1000); if (slotp->pwr.vcc) { reg |= PCIC_OUTENA; ! putb (sp, PCIC_POWER, reg); ! DELAY (100*1000); } return(0); } --- 917,936 ---- } break; } ! sp->putb(sp, PCIC_POWER, reg); DELAY(300*1000); if (slotp->pwr.vcc) { reg |= PCIC_OUTENA; ! sp->putb(sp, PCIC_POWER, reg); ! DELAY(100*1000); ! } ! /* Some chips are smarter than us it seems, so if we weren't ! * allowed to use 5V, try 3.3 instead ! */ ! if (!(sp->getb(sp, PCIC_STATUS) & 0x40) && slotp->pwr.vcc == 50) { ! slotp->pwr.vcc = 33; ! slotp->pwr.vpp = 0; ! return (pcic_power(slotp)); } return(0); } *************** *** 936,942 **** if (irq == 0) clrb(sp, PCIC_INT_GEN, 0xF); else ! putb (sp, PCIC_INT_GEN, (getb (sp, PCIC_INT_GEN) & 0xF0) | irq); } /* --- 975,982 ---- if (irq == 0) clrb(sp, PCIC_INT_GEN, 0xF); else ! sp->putb(sp, PCIC_INT_GEN, ! (sp->getb(sp, PCIC_INT_GEN) & 0xF0) | irq); } /* *************** *** 965,993 **** case 0: /* Something funny happended on the way to the pub... */ return; case 1: /* Assert reset */ ! clrb (sp, PCIC_INT_GEN, PCIC_CARDRESET); slotp->insert_seq = 2; timeout(pcic_reset, (void*) slotp, hz/4); return; case 2: /* Deassert it again */ ! setb (sp, PCIC_INT_GEN, PCIC_CARDRESET|PCIC_IOCARD); slotp->insert_seq = 3; timeout(pcic_reset, (void*) slotp, hz/4); return; case 3: /* Wait if card needs more time */ ! if (!getb(sp, PCIC_STATUS) & PCIC_READY) { timeout(pcic_reset, (void*) slotp, hz/10); return; } } slotp->insert_seq = 0; if (sp->controller == PCIC_PD672X || sp->controller == PCIC_PD6710) { ! putb(sp, PCIC_TIME_SETUP0, 0x1); ! putb(sp, PCIC_TIME_CMD0, 0x6); ! putb(sp, PCIC_TIME_RECOV0, 0x0); ! putb(sp, PCIC_TIME_SETUP1, 1); ! putb(sp, PCIC_TIME_CMD1, 0xf); ! putb(sp, PCIC_TIME_RECOV1, 0); } selwakeup(&slotp->selp); } --- 1005,1033 ---- case 0: /* Something funny happended on the way to the pub... */ return; case 1: /* Assert reset */ ! clrb(sp, PCIC_INT_GEN, PCIC_CARDRESET); slotp->insert_seq = 2; timeout(pcic_reset, (void*) slotp, hz/4); return; case 2: /* Deassert it again */ ! setb(sp, PCIC_INT_GEN, PCIC_CARDRESET|PCIC_IOCARD); slotp->insert_seq = 3; timeout(pcic_reset, (void*) slotp, hz/4); return; case 3: /* Wait if card needs more time */ ! if (!sp->getb(sp, PCIC_STATUS) & PCIC_READY) { timeout(pcic_reset, (void*) slotp, hz/10); return; } } slotp->insert_seq = 0; if (sp->controller == PCIC_PD672X || sp->controller == PCIC_PD6710) { ! sp->putb(sp, PCIC_TIME_SETUP0, 0x1); ! sp->putb(sp, PCIC_TIME_CMD0, 0x6); ! sp->putb(sp, PCIC_TIME_RECOV0, 0x0); ! sp->putb(sp, PCIC_TIME_SETUP1, 1); ! sp->putb(sp, PCIC_TIME_CMD1, 0xf); ! sp->putb(sp, PCIC_TIME_RECOV1, 0); } selwakeup(&slotp->selp); } *************** *** 1005,1012 **** return; } #endif ! putb(sp, PCIC_INT_GEN, 0); ! putb(sp, PCIC_POWER, 0); } /* --- 1045,1052 ---- return; } #endif ! sp->putb(sp, PCIC_INT_GEN, 0); ! sp->putb(sp, PCIC_POWER, 0); } /* *************** *** 1053,1061 **** #endif /* PC98 */ s = splhigh(); for (slot = 0; slot < PCIC_MAX_SLOTS; slot++, sp++) ! if (sp->slotp && (chg = getb(sp, PCIC_STAT_CHG)) != 0) if (chg & PCIC_CDTCH) { ! if ((getb(sp, PCIC_STATUS) & PCIC_CD) == PCIC_CD) { pccard_event(sp->slotp, card_inserted); --- 1093,1101 ---- #endif /* PC98 */ s = splhigh(); for (slot = 0; slot < PCIC_MAX_SLOTS; slot++, sp++) ! if (sp->slotp && (chg = sp->getb(sp, PCIC_STAT_CHG)) != 0) if (chg & PCIC_CDTCH) { ! if ((sp->getb(sp, PCIC_STATUS) & PCIC_CD) == PCIC_CD) { pccard_event(sp->slotp, card_inserted); *************** *** 1073,1078 **** static void pcic_resume(struct slot *slotp) { if (pcic_irq > 0) ! putb(slotp->cdata, PCIC_STAT_INT, (pcic_irq << 4) | 0xF); } --- 1113,1123 ---- static void pcic_resume(struct slot *slotp) { + struct pcic_slot *sp = slotp->cdata; if (pcic_irq > 0) ! sp->putb(sp, PCIC_STAT_INT, (pcic_irq << 4) | 0xF); ! if (sp->controller == PCIC_PD672X) { ! setb(sp, PCIC_MISC1, PCIC_SPKR_EN); ! setb(sp, PCIC_MISC2, PCIC_LPDM_EN); ! } } Index: i82365.h Index: pccard.c Index: pcic.c From owner-freebsd-mobile Tue Oct 21 13:09:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA07012 for mobile-outgoing; Tue, 21 Oct 1997 13:09:50 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from church.cse.ogi.edu (cse.ogi.edu [129.95.20.2]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id NAA07002 for ; Tue, 21 Oct 1997 13:09:42 -0700 (PDT) (envelope-from jrb@church.cse.ogi.edu) Received: from church.cse.ogi.edu (localhost [127.0.0.1]) by church.cse.ogi.edu (8.8.6/8.8.6) with ESMTP id NAA17839 for freebsd-mobile@freebsd.org; Tue, 21 Oct 1997 13:09:40 -0700 (PDT) Message-Id: <199710212009.NAA17839@church.cse.ogi.edu> To: freebsd-mobile@freebsd.org Subject: query re pccard/ethernet cards Date: Tue, 21 Oct 1997 13:09:40 -0700 From: Jim Binkley Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Which ethernet pccards could I go out and buy that would both support multicast and promiscuous mode? Jim Binkley jrb@cse.ogi.edu From owner-freebsd-mobile Tue Oct 21 19:18:17 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id TAA26823 for mobile-outgoing; Tue, 21 Oct 1997 19:18:17 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA26818 for ; Tue, 21 Oct 1997 19:18:13 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.gsoft.com.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id LAA00615; Wed, 22 Oct 1997 11:44:45 +0930 (CST) Message-Id: <199710220214.LAA00615@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: mobile@FreeBSD.ORG, handy@sag.space.lockheed.com Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Tue, 21 Oct 1997 13:23:39 CST." <199710211923.NAA07091@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 22 Oct 1997 11:44:44 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I'd like to bring in most of the recent changes to 2.2, not that 2.2.5 > is out the door. (I didn't want to break things for the release, and > didn't have time to test them enough for the release.) Can you hold off a couple of days? Something you've changed recently has broken PCCARD support on my Toshiba, but I haven't been able to track it down yet. (One killer is that the code now allocates IRQ 11 by default for the pcic, but IRQ 11 is used by "something" else that's not probed. There needs to be a mechanism for explicitly specifying an IRQ for the pcic.) mike From owner-freebsd-mobile Tue Oct 21 21:19:03 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA03539 for mobile-outgoing; Tue, 21 Oct 1997 21:19:03 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA03532 for ; Tue, 21 Oct 1997 21:18:57 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id WAA27233; Tue, 21 Oct 1997 22:18:52 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id WAA08580; Tue, 21 Oct 1997 22:18:50 -0600 (MDT) Date: Tue, 21 Oct 1997 22:18:50 -0600 (MDT) Message-Id: <199710220418.WAA08580@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org, handy@sag.space.lockheed.com Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710220214.LAA00615@word.smith.net.au> References: <199710211923.NAA07091@rocky.mt.sri.com> <199710220214.LAA00615@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > I'd like to bring in most of the recent changes to 2.2, not that 2.2.5 > > is out the door. (I didn't want to break things for the release, and > > didn't have time to test them enough for the release.) > > Can you hold off a couple of days? Something you've changed recently > has broken PCCARD support on my Toshiba, but I haven't been able to > track it down yet. > > (One killer is that the code now allocates IRQ 11 by default for the > pcic, but IRQ 11 is used by "something" else that's not probed. I changed the code to use the 'highest' un-allocated IRQ, because that's what Win95 does and many machines used built-in IRQ's for IRQ 3. > There > needs to be a mechanism for explicitly specifying an IRQ for the pcic.) Yes, there does. But, because of the current 'ISA' leanings of the configuration, there isn't an easy way. In the meantime, you could try backing out the change I made to see if things start working again. Nate ------ cut-n-pasto reverse patch follows ----------- RCS file: /home/CVS/src/sys/pccard/pccard.c,v retrieving revision 1.35 retrieving revision 1.36 diff -u -r1.35 -r1.36 --- pccard.c 1997/09/21 22:01:53 1.35 +++ pccard.c 1997/10/06 02:56:15 1.36 @@ -451,7 +451,7 @@ int irq; unsigned int mask; - for (irq = 1; irq < ICU_LEN; irq++) { + for (irq = ICU_LEN; irq > 0; irq--) { mask = 1ul << irq; if (!(mask & imask)) continue; Index: pccard.c From owner-freebsd-mobile Tue Oct 21 21:23:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA03736 for mobile-outgoing; Tue, 21 Oct 1997 21:23:37 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA03731 for ; Tue, 21 Oct 1997 21:23:34 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id WAA27280; Tue, 21 Oct 1997 22:23:22 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id WAA08641; Tue, 21 Oct 1997 22:23:21 -0600 (MDT) Date: Tue, 21 Oct 1997 22:23:21 -0600 (MDT) Message-Id: <199710220423.WAA08641@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Nate Williams Cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710220418.WAA08580@rocky.mt.sri.com> References: <199710211923.NAA07091@rocky.mt.sri.com> <199710220214.LAA00615@word.smith.net.au> <199710220418.WAA08580@rocky.mt.sri.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Nate Williams writes: > > > I'd like to bring in most of the recent changes to 2.2, not that 2.2.5 > > > is out the door. (I didn't want to break things for the release, and > > > didn't have time to test them enough for the release.) > > > > Can you hold off a couple of days? Something you've changed recently > > has broken PCCARD support on my Toshiba, but I haven't been able to > > track it down yet. > > > > (One killer is that the code now allocates IRQ 11 by default for the > > pcic, but IRQ 11 is used by "something" else that's not probed. A quick followup. The only other change I've made recently are the modem support for certain Cirrus chipsets, the rest need a sysctl to enable them or are 'style' changes and/or comment updates. It's gotta be the 'grab the highest IRQ' change. Do you know what's using IRQ 11, so that you can enable the driver for it? Nate From owner-freebsd-mobile Tue Oct 21 21:32:53 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA04216 for mobile-outgoing; Tue, 21 Oct 1997 21:32:53 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA04210 for ; Tue, 21 Oct 1997 21:32:48 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.gsoft.com.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id NAA01198; Wed, 22 Oct 1997 13:58:55 +0930 (CST) Message-Id: <199710220428.NAA01198@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: Mike Smith , mobile@freebsd.org, handy@sag.space.lockheed.com Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Tue, 21 Oct 1997 22:18:50 CST." <199710220418.WAA08580@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 22 Oct 1997 13:58:55 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > (One killer is that the code now allocates IRQ 11 by default for the > > pcic, but IRQ 11 is used by "something" else that's not probed. > > I changed the code to use the 'highest' un-allocated IRQ, because that's > what Win95 does and many machines used built-in IRQ's for IRQ 3. Understood entirely; all I meant was that you moved the problem from your machine to mine. 8) > > There > > needs to be a mechanism for explicitly specifying an IRQ for the pcic.) > > Yes, there does. But, because of the current 'ISA' leanings of the > configuration, there isn't an easy way. I hacked an option into kern_intr that allows you to specify a mask of interrupts that can never be allowed on the system; this seemed to work pretty well. > In the meantime, you could try backing out the change I made to see if > things start working again. I did that; it didn't help. I need to spend some more time chasing kernel builds; I just don't have that right now. (Assignments, exams, new product lagging on a solid deadline...) mike From owner-freebsd-mobile Tue Oct 21 21:37:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA04442 for mobile-outgoing; Tue, 21 Oct 1997 21:37:12 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA04432 for ; Tue, 21 Oct 1997 21:37:09 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id WAA27374; Tue, 21 Oct 1997 22:37:07 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id WAA08864; Tue, 21 Oct 1997 22:37:05 -0600 (MDT) Date: Tue, 21 Oct 1997 22:37:05 -0600 (MDT) Message-Id: <199710220437.WAA08864@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710220428.NAA01198@word.smith.net.au> References: <199710220418.WAA08580@rocky.mt.sri.com> <199710220428.NAA01198@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > (One killer is that the code now allocates IRQ 11 by default for the > > > pcic, but IRQ 11 is used by "something" else that's not probed. > > > > I changed the code to use the 'highest' un-allocated IRQ, because that's > > what Win95 does and many machines used built-in IRQ's for IRQ 3. > > Understood entirely; all I meant was that you moved the problem from > your machine to mine. 8) Actually, I never had a problem, but many users complained to me about this. And, it made sense because Win95 'does it this way', and they are the reference implementation. > > > There > > > needs to be a mechanism for explicitly specifying an IRQ for the pcic.) > > > > Yes, there does. But, because of the current 'ISA' leanings of the > > configuration, there isn't an easy way. > > I hacked an option into kern_intr that allows you to specify a mask of > interrupts that can never be allowed on the system; this seemed to > work pretty well. It's an option, but it still doesn't allow you to specify the interrupt. > > In the meantime, you could try backing out the change I made to see if > > things start working again. > > I did that; it didn't help. I need to spend some more time chasing > kernel builds; I just don't have that right now. (Assignments, exams, > new product lagging on a solid deadline...) *sigh* Does your laptop have a Cirrus Logic controller? Nate From owner-freebsd-mobile Tue Oct 21 21:38:07 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA04474 for mobile-outgoing; Tue, 21 Oct 1997 21:38:07 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA04467 for ; Tue, 21 Oct 1997 21:37:51 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.gsoft.com.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id OAA01232; Wed, 22 Oct 1997 14:03:04 +0930 (CST) Message-Id: <199710220433.OAA01232@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Tue, 21 Oct 1997 22:23:21 CST." <199710220423.WAA08641@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 22 Oct 1997 14:03:02 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > Can you hold off a couple of days? Something you've changed recently > > > has broken PCCARD support on my Toshiba, but I haven't been able to > > > track it down yet. > > > > > > (One killer is that the code now allocates IRQ 11 by default for the > > > pcic, but IRQ 11 is used by "something" else that's not probed. > > A quick followup. The only other change I've made recently are the > modem support for certain Cirrus chipsets, the rest need a sysctl to > enable them or are 'style' changes and/or comment updates. There's the "low power" bit as well as the modem bit. I *think* the modem bit change fixed the intermittent modem sound output I was getting (sometimes it would, sometimes it wouldn't). > It's gotta be the 'grab the highest IRQ' change. Do you know what's > using IRQ 11, so that you can enable the driver for it? It's the "PCI IRQ", and now you know as much as I do. It might be the USB controller, for which we don't have a driver. I just know that IRQ 11 can't be assigned to a PCCARD and have it work. FWIW, the kernel that works was built 5th Oct from sources current to that day. I haven't said anything because I wanted to come back with real answers rather than unsupportable complaints. mike From owner-freebsd-mobile Tue Oct 21 22:00:58 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA05682 for mobile-outgoing; Tue, 21 Oct 1997 22:00:58 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA05672 for ; Tue, 21 Oct 1997 22:00:54 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.gsoft.com.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id OAA01311; Wed, 22 Oct 1997 14:27:35 +0930 (CST) Message-Id: <199710220457.OAA01311@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Tue, 21 Oct 1997 22:37:05 CST." <199710220437.WAA08864@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 22 Oct 1997 14:27:33 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > > (One killer is that the code now allocates IRQ 11 by default for the > > > > pcic, but IRQ 11 is used by "something" else that's not probed. > > > > > > I changed the code to use the 'highest' un-allocated IRQ, because that's > > > what Win95 does and many machines used built-in IRQ's for IRQ 3. > > > > Understood entirely; all I meant was that you moved the problem from > > your machine to mine. 8) > > Actually, I never had a problem, but many users complained to me about > this. And, it made sense because Win95 'does it this way', and they are > the reference implementation. Understood. There's probably a quirk handler somwhere in there for this unit, or it's possible that the ESCD data lists it as not available. Still haven't got around to talking to that. 8( > > I hacked an option into kern_intr that allows you to specify a mask of > > interrupts that can never be allowed on the system; this seemed to > > work pretty well. > > It's an option, but it still doesn't allow you to specify the interrupt. Yes; my feelings were that given that it's trying to automatically assign one it makes more sense to correct the information it's using rather than to heavy-hand it into a choice for which it has no justification. > > I did that; it didn't help. I need to spend some more time chasing > > kernel builds; I just don't have that right now. (Assignments, exams, > > new product lagging on a solid deadline...) > > *sigh* Does your laptop have a Cirrus Logic controller? No. 8( Toshiba mostly do their own stuff. The pcic is, I *think*, the part marked TC183GT6JF. I seem to recall Tatsumi saying that he had tried to get data on this one and failed. The controller is detected as an i82365, FWIW. mike From owner-freebsd-mobile Tue Oct 21 22:18:22 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA06373 for mobile-outgoing; Tue, 21 Oct 1997 22:18:22 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from sag.space.lockheed.com (sag.space.lockheed.com [192.68.162.134]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id WAA06368 for ; Tue, 21 Oct 1997 22:18:20 -0700 (PDT) (envelope-from handy@sag.space.lockheed.com) Received: from localhost by sag.space.lockheed.com; (5.65v3.2/1.1.8.2/21Nov95-0423PM) id AA30406; Tue, 21 Oct 1997 22:18:17 -0700 Date: Tue, 21 Oct 1997 22:18:17 -0700 (PDT) From: "Brian N. Handy" To: Nate Williams Cc: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710211923.NAA07091@rocky.mt.sri.com> Message-Id: X-Files: The truth is out there Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Well, I tried the NateBSD patches and ... I can hear the modem! However, on a TP560, suspending with a card in still hangs the system hard. (I set the two new sysctl's to 1, for lack of knowing any better; I've never sysctl'd before. Irregardless, my results mirror those of John Polstra exactly. I think he's also packing a 560.) SoooOOOoo...I'm not adversely affected by the changes. I think the modem speaker at least should go in! Aside from that, Nate and Mike can fight the rest of it over out in the parking lot. :-) (Oh, and what I'm talking about here is -STABLE with the Nate patches of this morning.) Brian From owner-freebsd-mobile Tue Oct 21 22:53:46 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA07760 for mobile-outgoing; Tue, 21 Oct 1997 22:53:46 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from sag.space.lockheed.com (sag.space.lockheed.com [192.68.162.134]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id WAA07755 for ; Tue, 21 Oct 1997 22:53:43 -0700 (PDT) (envelope-from handy@sag.space.lockheed.com) Received: from localhost by sag.space.lockheed.com; (5.65v3.2/1.1.8.2/21Nov95-0423PM) id AA03039; Tue, 21 Oct 1997 22:53:11 -0700 Date: Tue, 21 Oct 1997 22:53:11 -0700 (PDT) From: "Brian N. Handy" Reply-To: "Brian N. Handy" To: Nate Williams Cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710220418.WAA08580@rocky.mt.sri.com> Message-Id: X-Files: The truth is out there Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk OK, I jumped to conclusions a little too fast... My box won't suspend now. When I pull the pccards, invoke 'zzz', and sit back... the lights blink for a second, the laptop appears to suspend, the screen goes black and the sounds go away. Then, a second later, it starts back up again! The error logs say: Oct 21 22:49:22 kriek /kernel: sio2: unload,gone Oct 21 22:49:22 kriek /kernel: Return IRQ=10 Oct 21 22:49:22 kriek /kernel: Card removed, slot 0 Oct 21 22:49:22 kriek /kernel: stray irq 10 [Now my trusty assistant types 'zzz' at the console] Oct 21 22:49:32 kriek /kernel: sio2: suspending Oct 21 22:49:32 kriek /kernel: Entire system suspend failure: errcode = 96 Oct 21 22:49:32 kriek /kernel: calcru: negative time: -804770 usec Oct 21 22:49:55 kriek /kernel: resumed from suspended mode (slept 00:00:23) Oct 21 22:50:00 kriek /kernel: Card inserted, slot 0 Oct 21 22:50:11 kriek /kernel: sio2: type 16550A (I included the first part as proof I pulled out the modem first. :-) It puzzles me that it also burps up this sio2 error, with the card out. But it does. This with with the sysctls both set and unset. (But not one or the other.) Sooo...something seems amiss. Brian From owner-freebsd-mobile Tue Oct 21 23:00:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA08015 for mobile-outgoing; Tue, 21 Oct 1997 23:00:05 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from u2.farm.idt.net (root@u2.farm.idt.net [169.132.8.11]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA08006 for ; Tue, 21 Oct 1997 23:00:02 -0700 (PDT) (envelope-from garycorc@idt.net) Received: from idt.net (ppp-11.ts-1.mlb.idt.net [169.132.71.11]) by u2.farm.idt.net (8.8.5/8.8.5) with ESMTP id BAA28413 for ; Wed, 22 Oct 1997 01:59:54 -0400 (EDT) Message-ID: <344D9623.90B4E255@idt.net> Date: Wed, 22 Oct 1997 01:58:59 -0400 From: "Gary T. Corcoran" X-Mailer: Mozilla 4.02b7 [en] (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: freebsd-mobile@freebsd.org Subject: Does 3C589D using ep0 need PAO? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Good day, all. I've just bought my first notebook PC and I'm trying to get my 3Com 3C589D ethernet card going. After searching through my year's worth of -mobile messages, I've made some progress (used DOS to read card settings, setup pccard.conf, pccard_ifconfig in rc.conf, compiled custom kernel) but I'm to the point where I'm getting the "driver allocation failed for 3Com Corporation" message. This is a new Toshiba Tecra with Cardbus compatible slots, but they are set in the BIOS to "automatic" mode which seems to allow them to operate in Non-Cardbus mode also (set to Cardbus mode as an experiment, even Win95 didn't see the card) I'm presuming that the fact that pccardc dumpcis does give me info, plus the fact that I get a beep and "Card removed" / "Card inserted" messages (but NOT the strings in pccard.conf) means that the PC-card slots themselves are "working". Plus, the LED on the dongle lights up, which I took as a good sign. For now, I just have one simple question: If I want to use the "ep0" driver with pccardd, do I, or do I _not_ need to get the PAO patches for the 3C589D? My -mobile archives seemed to show different opinions on this question! I'm running FreeBSD 2.2.5 (BETA from Saturday). Okay, how about one more question: is the "driver allocation failed" message indicative of a simple known problem? Thanks, Gary From owner-freebsd-mobile Tue Oct 21 23:57:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA11158 for mobile-outgoing; Tue, 21 Oct 1997 23:57:43 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA11151 for ; Tue, 21 Oct 1997 23:57:35 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.gsoft.com.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id QAA01641; Wed, 22 Oct 1997 16:24:12 +0930 (CST) Message-Id: <199710220654.QAA01641@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: "Gary T. Corcoran" cc: freebsd-mobile@FreeBSD.ORG Subject: Re: Does 3C589D using ep0 need PAO? In-reply-to: Your message of "Wed, 22 Oct 1997 01:58:59 -0400." <344D9623.90B4E255@idt.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 22 Oct 1997 16:24:12 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > For now, I just have one simple question: > If I want to use the "ep0" driver with pccardd, do I, or do > I _not_ need to get the PAO patches for the 3C589D? You explicitly DO NOT require the PAO patches. > Okay, how about one more question: is the "driver allocation > failed" message indicative of a simple known problem? No. It's "something is wrong". You need to check the configuration entry for the device as listed in /etc/pccard.conf and ensure that the resources that it claims are marked as available at the top of /etc/pccard.conf. These are good places to start. mike From owner-freebsd-mobile Wed Oct 22 00:03:16 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA11528 for mobile-outgoing; Wed, 22 Oct 1997 00:03:16 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from u2.farm.idt.net (root@u2.farm.idt.net [169.132.8.11]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA11523 for ; Wed, 22 Oct 1997 00:03:13 -0700 (PDT) (envelope-from garycorc@idt.net) Received: from idt.net (ppp-11.ts-1.mlb.idt.net [169.132.71.11]) by u2.farm.idt.net (8.8.5/8.8.5) with ESMTP id DAA09333 for ; Wed, 22 Oct 1997 03:03:08 -0400 (EDT) Message-ID: <344DA4F4.3E902429@idt.net> Date: Wed, 22 Oct 1997 03:02:12 -0400 From: "Gary T. Corcoran" X-Mailer: Mozilla 4.02b7 [en] (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: freebsd-mobile@FreeBSD.ORG Subject: Re: Does 3C589D using ep0 need PAO? References: <344D9623.90B4E255@idt.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Gary T. Corcoran wrote: > I've just bought my first notebook PC and I'm trying to get my > 3Com 3C589D ethernet card going. <...> > For now, I just have one simple question: > If I want to use the "ep0" driver with pccardd, do I, or do > I _not_ need to get the PAO patches for the 3C589D? > My -mobile archives seemed to show different opinions on > this question! I'm running FreeBSD 2.2.5 (BETA from Saturday). > > Okay, how about one more question: is the "driver allocation > failed" message indicative of a simple known problem? > Sorry folks. I accidentally made an IRQ conflict when I compiled my custom kernel. Removing the IRQ conflict made the above problem go away (sound familiar, anybody? ;-) (I knew I should've gone to bed earlier...) After changing the 'link' flags, I get a green light, and I can ping my other PC now! Now if only the XFree86 folks can get me going with X - the video chip in my machine is too new and is not (directly) supported... At least we now know that you _don't_ need the PAO patches with 2.2.5 on the new Toshiba's... Good night! Gary From owner-freebsd-mobile Wed Oct 22 00:33:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA13442 for mobile-outgoing; Wed, 22 Oct 1997 00:33:48 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA13422 for ; Wed, 22 Oct 1997 00:33:23 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.gsoft.com.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id QAA01863; Wed, 22 Oct 1997 16:59:45 +0930 (CST) Message-Id: <199710220729.QAA01863@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: "Gary T. Corcoran" cc: freebsd-mobile@FreeBSD.ORG Subject: Re: Does 3C589D using ep0 need PAO? In-reply-to: Your message of "Wed, 22 Oct 1997 03:02:12 -0400." <344DA4F4.3E902429@idt.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 22 Oct 1997 16:59:45 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Now if only the XFree86 folks can get me going with X - the video > chip in my machine is too new and is not (directly) supported... Which chip is this? mike From owner-freebsd-mobile Wed Oct 22 03:52:58 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id DAA22218 for mobile-outgoing; Wed, 22 Oct 1997 03:52:58 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from mail.kornet.nm.kr (mail.kornet.nm.kr [168.126.63.5]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id DAA22211 for ; Wed, 22 Oct 1997 03:52:55 -0700 (PDT) (envelope-from hwlee@netlab.korea.ac.kr) Received: from kuccgx.korea.ac.kr (kuccgx.korea.ac.kr [163.152.1.1]) by mail.kornet.nm.kr (8.6.12h2/8.6.9) with SMTP id TAA14791 for ; Wed, 22 Oct 1997 19:52:30 +0900 Received: from netlab.korea.ac.kr by kuccgx.korea.ac.kr (AIX 3.2/UCB 5.64/4.03) id AA10461; Wed, 22 Oct 1997 19:44:19 +0900 Received: from hwlee.korea.ac.kr ([163.152.40.60]) by netlab.korea.ac.kr.korea.ac.kr (5.x/SMI-SVR4) id AA14246; Wed, 22 Oct 1997 19:53:38 +0900 Message-Id: <9710221053.AA14246@netlab.korea.ac.kr.korea.ac.kr> From: "Hyung-Woo Lee" To: Date: Wed, 22 Oct 1997 19:54:41 +0900 X-Msmail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1155 Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk unsubscribe From owner-freebsd-mobile Wed Oct 22 08:09:33 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id IAA04181 for mobile-outgoing; Wed, 22 Oct 1997 08:09:33 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from u1.farm.idt.net (root@u1.farm.idt.net [169.132.8.10]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA04170 for ; Wed, 22 Oct 1997 08:09:31 -0700 (PDT) (envelope-from garycorc@idt.net) Received: from idt.net (ppp-11.ts-1.mlb.idt.net [169.132.71.11]) by u1.farm.idt.net (8.8.5/8.8.5) with ESMTP id LAA27176; Wed, 22 Oct 1997 11:07:50 -0400 (EDT) Message-ID: <344E168C.9832F0C@idt.net> Date: Wed, 22 Oct 1997 11:06:52 -0400 From: "Gary T. Corcoran" X-Mailer: Mozilla 4.02b7 [en] (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: Mike Smith CC: freebsd-mobile@FreeBSD.ORG Subject: Re: Does 3C589D using ep0 need PAO? References: <199710220729.QAA01863@word.smith.net.au> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Mike Smith wrote: > > > Now if only the XFree86 folks can get me going with X - the video > > chip in my machine is too new and is not (directly) supported... > > Which chip is this? (BTW, thanks for your earlier reply, Mike) The video chip it uses is the S3 Virge/MX. While XFree86 version 3.3.1 supports the S3 Virge in a few flavors, it doesn't recognize this new MX flavor. Gary From owner-freebsd-mobile Wed Oct 22 09:11:01 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA07731 for mobile-outgoing; Wed, 22 Oct 1997 09:11:01 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA07717 for ; Wed, 22 Oct 1997 09:10:57 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id KAA01354; Wed, 22 Oct 1997 10:10:43 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id KAA10346; Wed, 22 Oct 1997 10:10:41 -0600 (MDT) Date: Wed, 22 Oct 1997 10:10:41 -0600 (MDT) Message-Id: <199710221610.KAA10346@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: "Brian N. Handy" Cc: Nate Williams , Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: References: <199710220418.WAA08580@rocky.mt.sri.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > OK, I jumped to conclusions a little too fast... > > My box won't suspend now. When I pull the pccards, invoke 'zzz', and sit > back... the lights blink for a second, the laptop appears to suspend, the > screen goes black and the sounds go away. Then, a second later, it starts > back up again! The error logs say: > > Oct 21 22:49:22 kriek /kernel: sio2: unload,gone > Oct 21 22:49:22 kriek /kernel: Return IRQ=10 > Oct 21 22:49:22 kriek /kernel: Card removed, slot 0 > Oct 21 22:49:22 kriek /kernel: stray irq 10 > > [Now my trusty assistant types 'zzz' at the console] > > Oct 21 22:49:32 kriek /kernel: sio2: suspending This is weird here, since sio2 is already gone. But, I'm seeing this on -current as well, so I'm not sure what's up. I thought it was due to something weird on my box, but it appears that drivers aren't fully unloaded for some reason although they should be. > Oct 21 22:49:32 kriek /kernel: Entire system suspend failure: errcode = 96 > Oct 21 22:49:32 kriek /kernel: calcru: negative time: -804770 usec > Oct 21 22:49:55 kriek /kernel: resumed from suspended mode (slept 00:00:23) > Oct 21 22:50:00 kriek /kernel: Card inserted, slot 0 > Oct 21 22:50:11 kriek /kernel: sio2: type 16550A > > (I included the first part as proof I pulled out the modem first. :-) Hmm, that's no good. > It puzzles me that it also burps up this sio2 error, with the card out. That's very confusing. I'll look into it on my box to see if I can track anything down. > But it does. This with with the sysctls both set and unset. (But not one > or the other.) Sooo...something seems amiss. Can you yank *all* the patches and see if they are making the difference? It's possible the 'low-power' changes may have an adverse affect, since with both sysctl's unset, the only changes are the modem speaker and the power. Nate From owner-freebsd-mobile Wed Oct 22 09:13:47 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA07972 for mobile-outgoing; Wed, 22 Oct 1997 09:13:47 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from sag.space.lockheed.com (sag.space.lockheed.com [192.68.162.134]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id JAA07965 for ; Wed, 22 Oct 1997 09:13:44 -0700 (PDT) (envelope-from handy@sag.space.lockheed.com) Received: from localhost by sag.space.lockheed.com; (5.65v3.2/1.1.8.2/21Nov95-0423PM) id AA06366; Wed, 22 Oct 1997 09:13:43 -0700 Date: Wed, 22 Oct 1997 09:13:43 -0700 (PDT) From: "Brian N. Handy" To: Nate Williams Cc: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710221610.KAA10346@rocky.mt.sri.com> Message-Id: X-Files: The truth is out there Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Can you yank *all* the patches and see if they are making the >difference? It's possible the 'low-power' changes may have an adverse >affect, since with both sysctl's unset, the only changes are the modem >speaker and the power. I was about to go back to -STABLE as of this morning and give that a try, or should I step back to before yesterday's upgrade? Brian From owner-freebsd-mobile Wed Oct 22 09:20:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA08672 for mobile-outgoing; Wed, 22 Oct 1997 09:20:06 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA08605 for ; Wed, 22 Oct 1997 09:19:58 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id KAA01416; Wed, 22 Oct 1997 10:19:50 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id KAA10415; Wed, 22 Oct 1997 10:19:49 -0600 (MDT) Date: Wed, 22 Oct 1997 10:19:49 -0600 (MDT) Message-Id: <199710221619.KAA10415@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: "Gary T. Corcoran" Cc: freebsd-mobile@freebsd.org Subject: Re: Does 3C589D using ep0 need PAO? In-Reply-To: <344D9623.90B4E255@idt.net> References: <344D9623.90B4E255@idt.net> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I've just bought my first notebook PC and I'm trying to get my > 3Com 3C589D ethernet card going. After searching through my > year's worth of -mobile messages, I've made some progress (used DOS > to read card settings, setup pccard.conf, pccard_ifconfig in rc.conf, > compiled custom kernel) but I'm to the point where I'm getting the > "driver allocation failed for 3Com Corporation" message. If you use the if_zp driver, you don't need the PCCARD code. But, it appears that you want it. Can you send your kernel config, /etc/rc.conf, and /etc/pccard.conf so we can look? In case they help, the relevant snippets from my configuration are posted here. kernel config: -------------- # PCCARD (PCMCIA) support controller crd0 device pcic0 at crd? device pcic1 at crd? # Not used with pccard support #device zp0 at isa? port 0x300 net irq 10 iomem 0xde000 vector zpintr # PCCARD version of the 3Com driver device ep0 at isa? port 0x300 net irq 10 vector epintr /etc/rc.conf: ------------- apm_enable="YES" # Set to YES if you want APM enabled. pccard_enable="YES" # Set to YES if you want to configure PCCARD devices. pccard_mem="DEFAULT" # If pccard_enable=YES, this is card memory address. pccard_ifconfig="inet 206.127.76.107 netmask 0xffffffe0" /etc/pccard.conf: ----------------- # Generally available IO ports io 0x240-0x360 # Generally available IRQs (Built-in sound-card owners remove 5) irq 10 11 13 15 # Available memory slots memory 0xd4000 96k # 3Com Etherlink III 3C589B, 3C589C card "3Com Corporation" "3C589" config 0x1 "ep0" 11 insert echo 3Com Etherlink III inserted insert /etc/pccard_ether ep0 link0 -link1 # insert /etc/pccard_ether ep0 -link0 link1 remove echo 3Com Etherlink III removed remove /sbin/ifconfig ep0 delete > I'm presuming that the fact that pccardc dumpcis does give me > info, plus the fact that I get a beep and "Card removed" / > "Card inserted" messages (but NOT the strings in pccard.conf) > means that the PC-card slots themselves are "working". Yep. > For now, I just have one simple question: > If I want to use the "ep0" driver with pccardd, do I, or do > I _not_ need to get the PAO patches for the 3C589D? No, you do *NOT* need the PAO patches. > Okay, how about one more question: is the "driver allocation > failed" message indicative of a simple known problem? Yes, if the above snippets don't help get you on the right track, send those files and we can debug it remotely. Nate From owner-freebsd-mobile Wed Oct 22 09:26:20 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA09117 for mobile-outgoing; Wed, 22 Oct 1997 09:26:20 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA09109 for ; Wed, 22 Oct 1997 09:26:15 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id KAA01459; Wed, 22 Oct 1997 10:26:12 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id KAA10426; Wed, 22 Oct 1997 10:26:11 -0600 (MDT) Date: Wed, 22 Oct 1997 10:26:11 -0600 (MDT) Message-Id: <199710221626.KAA10426@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: "Brian N. Handy" Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: References: <199710221610.KAA10346@rocky.mt.sri.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >Can you yank *all* the patches and see if they are making the > >difference? It's possible the 'low-power' changes may have an adverse > >affect, since with both sysctl's unset, the only changes are the modem > >speaker and the power. > > I was about to go back to -STABLE as of this morning and give that a try, > or should I step back to before yesterday's upgrade? Either or. The patches I sent out were against -stable as of yesterday evening, so re-cvsup'ing would be the same as backing out the changes. (But, keep them around in case since I may have you do some experimentation with them if you don't mind.) Nate From owner-freebsd-mobile Wed Oct 22 10:54:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id KAA15043 for mobile-outgoing; Wed, 22 Oct 1997 10:54:37 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id KAA15031 for ; Wed, 22 Oct 1997 10:54:27 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id DAA01746; Thu, 23 Oct 1997 03:21:03 +0930 (CST) Message-Id: <199710221751.DAA01746@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: "Gary T. Corcoran" cc: Mike Smith , freebsd-mobile@FreeBSD.ORG Subject: Re: Does 3C589D using ep0 need PAO? In-reply-to: Your message of "Wed, 22 Oct 1997 11:06:52 -0400." <344E168C.9832F0C@idt.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 23 Oct 1997 03:21:03 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > The video chip it uses is the S3 Virge/MX. > While XFree86 version 3.3.1 supports the S3 Virge in a few flavors, > it doesn't recognize this new MX flavor. Hmm. How about the Xi server? (http://www.xig.com) mike From owner-freebsd-mobile Wed Oct 22 14:22:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA29017 for mobile-outgoing; Wed, 22 Oct 1997 14:22:54 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA29008 for ; Wed, 22 Oct 1997 14:22:50 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id PAA03286; Wed, 22 Oct 1997 15:22:39 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id PAA11535; Wed, 22 Oct 1997 15:22:37 -0600 (MDT) Date: Wed, 22 Oct 1997 15:22:37 -0600 (MDT) Message-Id: <199710222122.PAA11535@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Nate Williams Cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710220423.WAA08641@rocky.mt.sri.com> References: <199710211923.NAA07091@rocky.mt.sri.com> <199710220214.LAA00615@word.smith.net.au> <199710220418.WAA08580@rocky.mt.sri.com> <199710220423.WAA08641@rocky.mt.sri.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > > I'd like to bring in most of the recent changes to 2.2, not that 2.2.5 > > > > is out the door. (I didn't want to break things for the release, and > > > > didn't have time to test them enough for the release.) > > > > > > Can you hold off a couple of days? Something you've changed recently > > > has broken PCCARD support on my Toshiba, but I haven't been able to > > > track it down yet. Can you try out this patch. It was made after the 2nd, but it causes Brian Handy's box to blow chunks. It's related to Justin's timeout changes, and changed the flow of the code, so I'm just reverting to the original behavior. (Why it changes things I'm not sure, since I don't see why it should, but it seems to be doing something bad.) Nate ------- =================================================================== RCS file: /home/CVS/src/sys/pccard/pccard.c,v retrieving revision 1.38 diff -c -r1.38 pccard.c *** pccard.c 1997/10/06 05:46:03 1.38 --- pccard.c 1997/10/22 21:19:34 *************** *** 652,661 **** */ sp->pwr.vcc = 50; sp->pwr.vpp = 0; ! if (sp->pwr_off_pending) { ! untimeout(power_off_slot, (caddr_t)sp, sp->poff_ch); sp->ctrl->disable(sp); - } sp->pwr_off_pending = 0; sp->ctrl->power(sp); printf("Card inserted, slot %d\n", sp->slot); --- 652,660 ---- */ sp->pwr.vcc = 50; sp->pwr.vpp = 0; ! untimeout(power_off_slot, (caddr_t)sp, sp->poff_ch); ! if (sp->pwr_off_pending) sp->ctrl->disable(sp); sp->pwr_off_pending = 0; sp->ctrl->power(sp); printf("Card inserted, slot %d\n", sp->slot); Index: pccard.c From owner-freebsd-mobile Wed Oct 22 14:34:21 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA29954 for mobile-outgoing; Wed, 22 Oct 1997 14:34:21 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA29940 for ; Wed, 22 Oct 1997 14:34:17 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id PAA03358; Wed, 22 Oct 1997 15:32:28 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id PAA11576; Wed, 22 Oct 1997 15:32:26 -0600 (MDT) Date: Wed, 22 Oct 1997 15:32:26 -0600 (MDT) Message-Id: <199710222132.PAA11576@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Nate Williams Cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710222122.PAA11535@rocky.mt.sri.com> References: <199710211923.NAA07091@rocky.mt.sri.com> <199710220214.LAA00615@word.smith.net.au> <199710220418.WAA08580@rocky.mt.sri.com> <199710220423.WAA08641@rocky.mt.sri.com> <199710222122.PAA11535@rocky.mt.sri.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I wrote: I'd like to bring in most of the recent changes to 2.2, not that 2.2.5 is out the door. (I didn't want to break things for the release, and didn't have time to test them enough for the release.) Mike: Can you hold off a couple of days? Something you've changed recently has broken PCCARD support on my Toshiba, but I haven't been able to track it down yet. My response: Can you try out this patch. [ Note, This patch is against -current. If you're running -current, can you apply this patch and see if it makes any difference? If it helps (and solves Mike's problems), then I'll back this change out of -current and continue on the with 2.2 merge. ] Nate ps. This is the same patch I mailed out, but I'm re-sending just in case folks deleted it. =================================================================== RCS file: /home/CVS/src/sys/pccard/pccard.c,v retrieving revision 1.38 diff -c -r1.38 pccard.c *** pccard.c 1997/10/06 05:46:03 1.38 --- pccard.c 1997/10/22 21:19:34 *************** *** 652,661 **** */ sp->pwr.vcc = 50; sp->pwr.vpp = 0; ! if (sp->pwr_off_pending) { ! untimeout(power_off_slot, (caddr_t)sp, sp->poff_ch); sp->ctrl->disable(sp); - } sp->pwr_off_pending = 0; sp->ctrl->power(sp); printf("Card inserted, slot %d\n", sp->slot); --- 652,660 ---- */ sp->pwr.vcc = 50; sp->pwr.vpp = 0; ! untimeout(power_off_slot, (caddr_t)sp, sp->poff_ch); ! if (sp->pwr_off_pending) sp->ctrl->disable(sp); sp->pwr_off_pending = 0; sp->ctrl->power(sp); printf("Card inserted, slot %d\n", sp->slot); Index: pccard.c From owner-freebsd-mobile Wed Oct 22 22:36:21 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA00169 for mobile-outgoing; Wed, 22 Oct 1997 22:36:21 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA00162 for ; Wed, 22 Oct 1997 22:36:16 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id PAA00243; Thu, 23 Oct 1997 15:02:52 +0930 (CST) Message-Id: <199710230532.PAA00243@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Wed, 22 Oct 1997 15:32:26 CST." <199710222132.PAA11576@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 23 Oct 1997 15:02:50 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I wrote: > I'd like to bring in most of the recent changes to 2.2, not that 2.2.5 > is out the door. (I didn't want to break things for the release, and > didn't have time to test them enough for the release.) > > Mike: > Can you hold off a couple of days? Something you've changed recently > has broken PCCARD support on my Toshiba, but I haven't been able to > track it down yet. > > My response: > > Can you try out this patch. Sure. Doesn't help though. 8( Allocating IRQ 11 to the pcic actually seems harmless; we still get card insertion/removal interrupts, so it must be working. Unfortunately I just get the dreaded "driver allocation failed for ..." messages, where I didn't before. The kernel that works predates your hiding of the "interrupt configuration" messages, if that's any help. I'm onto this this afternoon, and will report ASAP. mike From owner-freebsd-mobile Wed Oct 22 22:56:39 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA00947 for mobile-outgoing; Wed, 22 Oct 1997 22:56:39 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA00942 for ; Wed, 22 Oct 1997 22:56:37 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id XAA06429; Wed, 22 Oct 1997 23:56:34 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id XAA13744; Wed, 22 Oct 1997 23:56:33 -0600 (MDT) Date: Wed, 22 Oct 1997 23:56:33 -0600 (MDT) Message-Id: <199710230556.XAA13744@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710230532.PAA00243@word.smith.net.au> References: <199710222132.PAA11576@rocky.mt.sri.com> <199710230532.PAA00243@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > My response: > > > > Can you try out this patch. > > Sure. Doesn't help though. 8( Allocating IRQ 11 to the pcic actually > seems harmless; we still get card insertion/removal interrupts, so it > must be working. OK, so that's not it. (Although it was on Brian Handy's TP560). > Unfortunately I just get the dreaded "driver allocation failed for ..." > messages, where I didn't before. Umm, if you're using IRQ 11, then I suspect your 3c589 card can't use it for itself. Do you have your configuration hard-coded to use IRQ 11? (I do, since I can't get it to work anywhere else for some silly reason.) > The kernel that works predates your hiding of the "interrupt > configuration" messages, if that's any help. I'm not sure I follow. Also note that I've made other changes recently which shouldn't make things any worse, but should make suspend/resume better in the default case. (More coming, I just did a code review and want to try out some more 'new' changes on my laptop to see if they make things better/worse.) Nate From owner-freebsd-mobile Wed Oct 22 23:13:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA01787 for mobile-outgoing; Wed, 22 Oct 1997 23:13:06 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from u1.farm.idt.net (root@u1.farm.idt.net [169.132.8.10]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA01781 for ; Wed, 22 Oct 1997 23:13:03 -0700 (PDT) (envelope-from garycorc@idt.net) Received: from idt.net (ppp-54.ts-1.mlb.idt.net [169.132.71.54]) by u1.farm.idt.net (8.8.5/8.8.5) with ESMTP id CAA13126 for ; Thu, 23 Oct 1997 02:13:01 -0400 (EDT) Message-ID: <344EEAB4.752D66BB@idt.net> Date: Thu, 23 Oct 1997 02:12:04 -0400 From: "Gary T. Corcoran" X-Mailer: Mozilla 4.02b7 [en] (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: mobile@FreeBSD.ORG Subject: Re: Patches from -current for -stable I'd like to commit after testing References: <199710230532.PAA00243@word.smith.net.au> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Mike Smith wrote: > > > Can you hold off a couple of days? Something you've changed recently > > has broken PCCARD support on my Toshiba, but I haven't been able to > > track it down yet. > > > > My response: > > > > Can you try out this patch. > > Sure. Doesn't help though. 8( Allocating IRQ 11 to the pcic actually > seems harmless; we still get card insertion/removal interrupts, so it > must be working. > > Unfortunately I just get the dreaded "driver allocation failed for ..." > messages, where I didn't before. > When I had my PCCARD woes yesterday due to a mis-allocated IRQ in my kernel config, I too seemed to get the insertion/removal interrupts (in 2.2.5), but got the "driver allocation failed for..." message in that setup. Once I corrected the situation of having another device on the same IRQ as the ethernet pc-card, then things started working. Of course you're running different code than I am, but was just wondering if that message might be due to the same symptom of an IRQ conflict? I don't know which Toshiba notebook you have Mike, but if it's anything like mine, IRQ 11 is dedicated as the "PCI" interrupt, and has about 3-4 things hung on it - Infrared ports, USB ports, and, hey(?) Win95 assigns the "Toshiba ToPIC97 CardBus Controller"s to IRQ 11 also (the 3C589D gets to use IRQ 10, the only free IRQ, in both W95 and FreeBSD). Don't know if any of this will help the two of you resolve how to automatically assign things, but thought I'd pass it along just in case... Gary From owner-freebsd-mobile Wed Oct 22 23:48:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA03571 for mobile-outgoing; Wed, 22 Oct 1997 23:48:11 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA03565 for ; Wed, 22 Oct 1997 23:48:06 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id QAA00413; Thu, 23 Oct 1997 16:14:45 +0930 (CST) Message-Id: <199710230644.QAA00413@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Wed, 22 Oct 1997 23:56:33 CST." <199710230556.XAA13744@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 23 Oct 1997 16:14:44 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Sure. Doesn't help though. 8( Allocating IRQ 11 to the pcic actually > > seems harmless; we still get card insertion/removal interrupts, so it > > must be working. > > OK, so that's not it. (Although it was on Brian Handy's TP560). Hmm. Maybe his *does* have something uncivilised on the high interrupt? It'd be nice to have a generic probe-time test-for-an-interrupt function, along the lines of what the sio(4) driver does to test the interrupt function of the 8250. > > Unfortunately I just get the dreaded "driver allocation failed for ..." > > messages, where I didn't before. > > Umm, if you're using IRQ 11, then I suspect your 3c589 card can't use it > for itself. Do you have your configuration hard-coded to use IRQ 11? > (I do, since I can't get it to work anywhere else for some silly > reason.) Sorry, I will clarify. There are two interrupts that I make available for PCCARDs, 9 and 10. These are declared in /etc/pccard.conf, and are not used by anything else in the system. Older kernels allocate IRQ 3 to the pcic, and the newer top-down code allocates IRQ 11. Both of these seem to work, insofar as card insertion/removal events are signalled as expected. > > The kernel that works predates your hiding of the "interrupt > > configuration" messages, if that's any help. > > I'm not sure I follow. I was just trying to give you a "feel" for the vintage of working kernel. I'm just cvsupping the 2.2.5 tag monster; once that's finished I'll be building a batch of kernels to see where things fell down. > Also note that I've made other changes recently > which shouldn't make things any worse, but should make suspend/resume > better in the default case. (More coming, I just did a code review and > want to try out some more 'new' changes on my laptop to see if they make > things better/worse.) Qool. If there's anything I can help with testing, let me know. mike From owner-freebsd-mobile Wed Oct 22 23:55:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA03839 for mobile-outgoing; Wed, 22 Oct 1997 23:55:37 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from u3.farm.idt.net (root@u3.farm.idt.net [169.132.8.12]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA03834 for ; Wed, 22 Oct 1997 23:55:35 -0700 (PDT) (envelope-from garycorc@idt.net) Received: from idt.net (ppp-54.ts-1.mlb.idt.net [169.132.71.54]) by u3.farm.idt.net (8.8.5/8.8.5) with ESMTP id CAA18549 for ; Thu, 23 Oct 1997 02:55:29 -0400 (EDT) Message-ID: <344EF4A8.A1F57A44@idt.net> Date: Thu, 23 Oct 1997 02:54:32 -0400 From: "Gary T. Corcoran" X-Mailer: Mozilla 4.02b7 [en] (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: freebsd-mobile@FreeBSD.ORG Subject: Re: Does 3C589D using ep0 need PAO? References: <199710221751.DAA01746@word.smith.net.au> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Mike Smith wrote: > > > > > The video chip it uses is the S3 Virge/MX. > > While XFree86 version 3.3.1 supports the S3 Virge in a few flavors, > > it doesn't recognize this new MX flavor. > > Hmm. How about the Xi server? (http://www.xig.com) Looked around there tonight, but it's also too new for them. Someone from XFree86 gave me something to try, but I haven't gotten anything but a mess on my screen so far (almost scary - looks like the screen is 'melting' like when an old celluloid film (movie) gets burned through due to the film getting stuck on a single frame - round 'burn pattern' and all :-) I'm going to try again tomorrow though... Gary From owner-freebsd-mobile Wed Oct 22 23:56:21 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA03888 for mobile-outgoing; Wed, 22 Oct 1997 23:56:21 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA03882 for ; Wed, 22 Oct 1997 23:56:18 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id AAA06742; Thu, 23 Oct 1997 00:56:16 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id AAA13857; Thu, 23 Oct 1997 00:56:15 -0600 (MDT) Date: Thu, 23 Oct 1997 00:56:15 -0600 (MDT) Message-Id: <199710230656.AAA13857@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710230644.QAA00413@word.smith.net.au> References: <199710230556.XAA13744@rocky.mt.sri.com> <199710230644.QAA00413@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > Sure. Doesn't help though. 8( Allocating IRQ 11 to the pcic actually > > > seems harmless; we still get card insertion/removal interrupts, so it > > > must be working. > > > > OK, so that's not it. (Although it was on Brian Handy's TP560). > > Hmm. Maybe his *does* have something uncivilised on the high > interrupt? Given the # of TP560's in the ranks of the developers, I figured breaking what appears to be the most popular laptops seemed a bad things to do. > It'd be nice to have a generic probe-time > test-for-an-interrupt function, along the lines of what the sio(4) > driver does to test the interrupt function of the 8250. Yep. Sounds like a job for 'Laptop-Mike'. The hooks are are there in pcic.c:build_freelist. :) > > > Unfortunately I just get the dreaded "driver allocation failed for ..." > > > messages, where I didn't before. > > > > Umm, if you're using IRQ 11, then I suspect your 3c589 card can't use it > > for itself. Do you have your configuration hard-coded to use IRQ 11? > > (I do, since I can't get it to work anywhere else for some silly > > reason.) > > Sorry, I will clarify. There are two interrupts that I make available > for PCCARDs, 9 and 10. These are declared in /etc/pccard.conf, and are > not used by anything else in the system. So, cards that previously worked with these two interrupts no longer work? *dang* > Older kernels allocate IRQ 3 to the pcic, and the newer top-down code > allocates IRQ 11. Both of these seem to work, insofar as card > insertion/removal events are signalled as expected. Good (I think??). > > > The kernel that works predates your hiding of the "interrupt > > > configuration" messages, if that's any help. > > > > I'm not sure I follow. > > I was just trying to give you a "feel" for the vintage of working > kernel. I'm just cvsupping the 2.2.5 tag monster; once that's finished > I'll be building a batch of kernels to see where things fell down. I don't remember my "hiding the 'interrupt configuration' messages" changes. Or at least that's not how I think of them, so I'm not sure what changes I made you're talking about. > > Also note that I've made other changes recently > > which shouldn't make things any worse, but should make suspend/resume > > better in the default case. (More coming, I just did a code review and > > want to try out some more 'new' changes on my laptop to see if they make > > things better/worse.) > > Qool. If there's anything I can help with testing, let me know. Do you have a 3.0 box, or is it now running 2.2? The changes will initially be made to 3.0, since it's already (apparently) broken there. (Although, I'm hoping that it may have been fixed by backing out the change....) If you want them for 2.2, I can make them there as well, although I can't test them on my laptop. :( Finally, any help you can give to figuring out why the current code is not working on your laptop, or better yet a pointer to the code that breaks it would be greatly appreciated. Nate From owner-freebsd-mobile Thu Oct 23 00:11:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA04748 for mobile-outgoing; Thu, 23 Oct 1997 00:11:50 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA04740 for ; Thu, 23 Oct 1997 00:11:45 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id QAA00530; Thu, 23 Oct 1997 16:38:28 +0930 (CST) Message-Id: <199710230708.QAA00530@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Thu, 23 Oct 1997 00:56:15 CST." <199710230656.AAA13857@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 23 Oct 1997 16:38:27 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Given the # of TP560's in the ranks of the developers, I figured > breaking what appears to be the most popular laptops seemed a bad things > to do. Definitely. (I'll pause to swear vigorously about documentation and the difficulties in getting the (&^%$*&^%$* stuff.) > > It'd be nice to have a generic probe-time > > test-for-an-interrupt function, along the lines of what the sio(4) > > driver does to test the interrupt function of the 8250. > > Yep. Sounds like a job for 'Laptop-Mike'. The hooks are are there in > pcic.c:build_freelist. :) Actually, I'd like something like: interrupt_capture(NULL); poke_hardware(); interrupt_capture(&interrupts); which would come back with a bitmask of interrupts that were generated between the two capture calls. Unfortunately, I'm not sure I know how to get this right, especially in the mixed interrupts disabled/enabled environment. One of these days I'll work out how to enthuse Bruce about something, and it'll just, er, pop out. 8) > > Sorry, I will clarify. There are two interrupts that I make available > > for PCCARDs, 9 and 10. These are declared in /etc/pccard.conf, and are > > not used by anything else in the system. > > So, cards that previously worked with these two interrupts no longer > work? *dang* That's correct. Boot old kernel, cards work. Boot new kernel, "driver allocation failed for...". We *really*must* improve that error message. I've been loth to get really dirty with the pccard* utilities because the PAO people have so much layered on top of them, but IMHO they really have to be redone from scratch. 8( > > Older kernels allocate IRQ 3 to the pcic, and the newer top-down code > > allocates IRQ 11. Both of these seem to work, insofar as card > > insertion/removal events are signalled as expected. > > Good (I think??). Yup. > I don't remember my "hiding the 'interrupt configuration' messages" > changes. Or at least that's not how I think of them, so I'm not sure > what changes I made you're talking about. The kernel that works says: device combination doesn't support shared irq4 intr_connect(irq4) failed, result=-1 device combination doesn't support shared irq5 intr_connect(irq5) failed, result=-1 device combination doesn't support shared irq7 intr_connect(irq7) failed, result=-1 device combination doesn't support shared irq12 intr_connect(irq12) failed, result=-1 device combination doesn't support shared irq14 intr_connect(irq14) failed, result=-1 device combination doesn't support shared irq15 intr_connect(irq15) failed, result=-1 PC-Card Intel 82365 (5 mem & 2 I/O windows) pcic: controller irq 3 Card inserted, slot 0 Card inserted, slot 1 sio1: type 16550A ep0: utp/bnc[*UTP*] address 00:60:97:88:e7:93 The one that doesn't says: pcic: controller irq 11 Card inserted, slot 0 Card inserted, slot 1 driver allocation failed for Megahertz driver allocation failed for 3Com Corporation > Do you have a 3.0 box, or is it now running 2.2? The changes will > initially be made to 3.0, since it's already (apparently) broken there. Real hackers run -current on their laptop, remember? > Finally, any help you can give to figuring out why the current code is > not working on your laptop, or better yet a pointer to the code that > breaks it would be greatly appreciated. As soon as this ^&%$&^% 2.2.5 tag is laid down I'll be started. Someone forgot to move his cvsup cron job when he shifted his repo. 8( mike From owner-freebsd-mobile Thu Oct 23 00:17:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA05073 for mobile-outgoing; Thu, 23 Oct 1997 00:17:15 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA05068 for ; Thu, 23 Oct 1997 00:17:10 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id QAA00559; Thu, 23 Oct 1997 16:43:53 +0930 (CST) Message-Id: <199710230713.QAA00559@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: "Gary T. Corcoran" cc: mobile@FreeBSD.ORG Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Thu, 23 Oct 1997 02:12:04 -0400." <344EEAB4.752D66BB@idt.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 23 Oct 1997 16:43:52 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > > Sure. Doesn't help though. 8( Allocating IRQ 11 to the pcic actually > > seems harmless; we still get card insertion/removal interrupts, so it > > must be working. ... > When I had my PCCARD woes yesterday due to a mis-allocated IRQ in my kernel > config, I too seemed to get the insertion/removal interrupts (in 2.2.5), > but got the "driver allocation failed for..." message in that setup. > > Once I corrected the situation of having another device on the same IRQ > as the ethernet pc-card, then things started working. Of course you're > running different code than I am, but was just wondering if that message > might be due to the same symptom of an IRQ conflict? Could be. However I pulled it back to IRQ 3 and it still failed, so in this case I don't think it was. > I don't know which Toshiba notebook you have Mike, but if it's anything > like mine, IRQ 11 is dedicated as the "PCI" interrupt, and has about 3-4 > things hung on it - Infrared ports, USB ports, and, hey(?) Win95 assigns > the "Toshiba ToPIC97 CardBus Controller"s to IRQ 11 also > (the 3C589D gets to use IRQ 10, the only free IRQ, in both W95 and FreeBSD). Yeah, same here. This is a 220CDS; sounds like there's a lot in common across their range. 8) > Don't know if any of this will help the two of you resolve how to > automatically assign things, but thought I'd pass it along just in case... It's kinda odd; I don't actually see much PCI action in this guy at all: Probing for devices on PCI bus 0: chip0: rev 0x23 on pci0.0.0 vga0: rev 0xc2 on pci0.4.0 Now, I may have disabled some stuff using Tsetup (why don't they have a BIOS setup like everyone else?), but I'd have expected to see more stuff than that. Certainly nothing claiming any interrupts. I'll have to see what W95 thinks; more data later. mike From owner-freebsd-mobile Thu Oct 23 06:48:30 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id GAA26712 for mobile-outgoing; Thu, 23 Oct 1997 06:48:30 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id GAA26699 for ; Thu, 23 Oct 1997 06:48:21 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id XAA00290 for ; Thu, 23 Oct 1997 23:14:55 +0930 (CST) Message-Id: <199710231344.XAA00290@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Wed, 22 Oct 1997 10:10:41 CST." <199710221610.KAA10346@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 23 Oct 1997 23:14:54 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Well, it's been an interesting day, reliving the booboos in -current over the last few weeks as I wandered around trying to work out what was happening with my system. I can only conclude that I must have been the victim of pilot error; certainly -current as it stands now is working as well as it ever has, no problems detecting cards at all. Apologies for the panic, and back to your regular service. Nate, hit me with them post-fishin' changes. mike From owner-freebsd-mobile Thu Oct 23 08:15:27 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id IAA03045 for mobile-outgoing; Thu, 23 Oct 1997 08:15:27 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from thorazine.neuron.net (thorazine.neuron.net [208.132.136.190]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA03028 for ; Thu, 23 Oct 1997 08:15:16 -0700 (PDT) (envelope-from amir@thorazine.neuron.net) Received: (from amir@localhost) by thorazine.neuron.net (8.8.7/8.8.6) id LAA10886; Thu, 23 Oct 1997 11:20:08 -0400 (EDT) Message-ID: <19971023111953.63608@neuron.net> Date: Thu, 23 Oct 1997 11:19:53 -0400 From: "Amir Y. Rosenblatt" To: mobile@FreeBSD.ORG Subject: new xig X server out Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81 Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Xig (formerly X-Inside) just put out v4.1 of their Accelerated-X server. I installed the demo (along with updated drivers for the cyber9385 chipset that were just uploaded yesterday) on my TP 765D -- works like a charm. I just ordered the full package -- should have it tommorow. I used to use their desktop X server when I hadmy FreeBSD server at home (the XFree86 drivers for my Number 9 Imagine 128 were pretty poor when I first got the machine rolling). I've always had awful luck with the XFree86 config files so this was totally worth the money to me. Just thought I'd mention it. -Amir -- / \ I'm the *big* yak-daddy | Amir Y. Rosenblatt /<@>\ | amir@neuron.net / \ FNORD | http://www.neuron.net/~amir _/_______\____________________________________|____________________________ From owner-freebsd-mobile Thu Oct 23 08:55:44 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id IAA05864 for mobile-outgoing; Thu, 23 Oct 1997 08:55:44 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA05857 for ; Thu, 23 Oct 1997 08:55:42 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id JAA09939; Thu, 23 Oct 1997 09:55:39 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id JAA15370; Thu, 23 Oct 1997 09:55:38 -0600 (MDT) Date: Thu, 23 Oct 1997 09:55:38 -0600 (MDT) Message-Id: <199710231555.JAA15370@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710230708.QAA00530@word.smith.net.au> References: <199710230656.AAA13857@rocky.mt.sri.com> <199710230708.QAA00530@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > We *really*must* improve that error message. I've been loth to get > really dirty with the pccard* utilities because the PAO people have so > much layered on top of them, but IMHO they really have to be redone > from scratch. 8( Ahh, it makes my heart glad to hear this. The current code is too much of a mess to be much help. > > I don't remember my "hiding the 'interrupt configuration' messages" > > changes. Or at least that's not how I think of them, so I'm not sure > > what changes I made you're talking about. > > The kernel that works says: > > device combination doesn't support shared irq4 > intr_connect(irq4) failed, result=-1 > device combination doesn't support shared irq5 > intr_connect(irq5) failed, result=-1 > device combination doesn't support shared irq7 > intr_connect(irq7) failed, result=-1 > device combination doesn't support shared irq12 > intr_connect(irq12) failed, result=-1 > device combination doesn't support shared irq14 > intr_connect(irq14) failed, result=-1 > device combination doesn't support shared irq15 > intr_connect(irq15) failed, result=-1 > PC-Card Intel 82365 (5 mem & 2 I/O windows) > pcic: controller irq 3 > Card inserted, slot 0 > Card inserted, slot 1 > sio1: type 16550A > ep0: utp/bnc[*UTP*] address 00:60:97:88:e7:93 > > The one that doesn't says: > > pcic: controller irq 11 > Card inserted, slot 0 > Card inserted, slot 1 > driver allocation failed for Megahertz > driver allocation failed for 3Com Corporation Ahh, I remember now. The commit was merely cosmetic, so it really couldn't affect anything (unless somehow by slowing things down from printing the timing was somehow negatively affected.) > > Do you have a 3.0 box, or is it now running 2.2? The changes will > > initially be made to 3.0, since it's already (apparently) broken there. > > Real hackers run -current on their laptop, remember? *grin* You mentioned 2.2 in a previous email. Nate From owner-freebsd-mobile Thu Oct 23 09:06:31 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA06454 for mobile-outgoing; Thu, 23 Oct 1997 09:06:31 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA06448 for ; Thu, 23 Oct 1997 09:06:26 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id BAA00993 for ; Fri, 24 Oct 1997 01:33:15 +0930 (CST) Message-Id: <199710231603.BAA00993@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Thu, 23 Oct 1997 09:55:38 CST." <199710231555.JAA15370@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Oct 1997 01:33:15 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > We *really*must* improve that error message. I've been loth to get > > really dirty with the pccard* utilities because the PAO people have so > > much layered on top of them, but IMHO they really have to be redone > > from scratch. 8( > > Ahh, it makes my heart glad to hear this. The current code is too much > of a mess to be much help. Argh. The problem is that the whole current mechanism is messy. At the other extreme, an elegant and logical solution would be far too heavyweight for the practicalities of the situation. > > > Do you have a 3.0 box, or is it now running 2.2? The changes will > > > initially be made to 3.0, since it's already (apparently) broken there. > > > > Real hackers run -current on their laptop, remember? > > *grin* You mentioned 2.2 in a previous email. I did? I think you have me and Brian confused. I wouldn't let 2.2 near this thing. mike From owner-freebsd-mobile Thu Oct 23 09:12:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA06942 for mobile-outgoing; Thu, 23 Oct 1997 09:12:13 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA06918 for ; Thu, 23 Oct 1997 09:12:05 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id KAA10047; Thu, 23 Oct 1997 10:11:58 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id KAA15468; Thu, 23 Oct 1997 10:11:57 -0600 (MDT) Date: Thu, 23 Oct 1997 10:11:57 -0600 (MDT) Message-Id: <199710231611.KAA15468@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710231344.XAA00290@word.smith.net.au> References: <199710221610.KAA10346@rocky.mt.sri.com> <199710231344.XAA00290@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Well, it's been an interesting day, reliving the booboos in -current > over the last few weeks as I wandered around trying to work out what > was happening with my system. > > I can only conclude that I must have been the victim of pilot error; > certainly -current as it stands now is working as well as it ever has, > no problems detecting cards at all. Really? That's *GREAT* to hear.:) > Nate, hit me with them post-fishin' changes. Will do when I get them wrapped up. Nate From owner-freebsd-mobile Thu Oct 23 13:28:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id NAA25315 for mobile-outgoing; Thu, 23 Oct 1997 13:28:04 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from fatman.i-pi.com (fatman.i-pi.com [198.49.217.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id NAA25282 for ; Thu, 23 Oct 1997 13:27:58 -0700 (PDT) (envelope-from ingham@i-pi.com) Received: (from ingham@localhost) by fatman.i-pi.com (8.8.5/8.8.5) id OAA13712; Thu, 23 Oct 1997 14:27:55 -0600 (MDT) Message-ID: <19971023142754.64160@fatman.i-pi.com> Date: Thu, 23 Oct 1997 14:27:55 -0600 From: Kenneth Ingham To: mobile@freebsd.org Subject: 2.2.5-RELEASE and PC Cards Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Mailer: Mutt 0.67 Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Sigh. I'm fairly sure that I have seen the answer to this go by recently, but I can't find it searching the mailing list archives at www.freebsd.org. I just upgraded my laptop which used to run 2.2.2+PAO to 2.2.5-RELEASE. The upgrade went OK except for the PCCArds no longer work. I built a new kernel with the ep, crd, and pcic drivers and booted it. Relevant parts of the boot messages (from dmesg) are: FreeBSD 2.2.5-RELEASE #0: Wed Oct 22 21:04:11 MDT 1997 ingham@socrates.i-pi.com:/usr/src/sys/compile/SOCRATES CPU: i486 DX4 (486-class CPU) Origin = "GenuineIntel" Id = 0x480 Stepping=0 Features=0x3 pccard driver sio added sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 not found at 0x2f8 sio2 not found at 0x3e8 sio3 not found at 0x2e8 pccard driver ep added ep0 not found at 0x300 apm0 on isa apm: found APM BIOS version 1.1 PC-Card Vadem 468 (5 mem & 2 I/O windows) pcic: controller irq 3 Card inserted, slot 1 Card inserted, slot 0 I've verified with pccardc dumpcis that the manufacturer and version match. This is not the PAO pccard.conf file that I was using (but it didn't work either). In fact, here's my pccard.conf file (I removed everything except the info for the cards I am trying to use (hmmm, the date in this is old, but I didn't see a newer one in the upgrade files): # Sample PCCARD configuration file # # Removing all IRQ conflicts from this file can't be done because of some # IRQ-selfish PC-cards. So if you want to use some of these cards in # your machine, you will be forced to modify their IRQ parameters from # the following list. # # IRQ == 0 means "allocate free IRQ from IRQ pool" # IRQ == 16 means "do not use IRQ (e.g. PIO mode)" # # $Id: pccard.conf.sample,v 1.4 1996/06/19 01:28:07 nate Exp $ # Generally available IO ports io 0x240-0x360 # Generally available IRQs (Built-in sound-card owners remove 5) irq 3 5 10 11 13 15 # Available memory slots memory 0xd4000 96k # 3Com Etherlink III 3C589B, 3C589C card "3Com Corporation" "3C589" config 0x1 "ep0" ? insert echo 3Com Etherlink III inserted insert /etc/pccard_ether ep0 -link0 link1 # insert /etc/pccard_ether ep0 link0 -link1 remove echo 3Com Etherlink III removed remove /sbin/ifconfig ep0 delete # US Robotics Sportster PCMCIA V.34 # US Robotics COURIER PCMCIA V.34 card "USRobotics" "PCMCIA 28800 Data/Fax" config 0x3 "sio2" 3 insert echo USRobotics PCMCIA Modem inserted remove echo USRobotics PCMCIA Modem removed And, I used the undocumented pccardc dumpcis on the cards and got the following. USR Sportster: code Manufacturer ID ignored Code 129 not found Code 129 not found code Unknown ignored Configuration data for card in slot 0 Tuple #1, code = 0x1 (Common memory descriptor), length = 3 000: 00 01 ff Common memory device information: Device number 1, type No device, WPS = OFF Speed = No speed, Memory block size = 2Kb, 1 units Tuple #2, code = 0x15 (Version 1 info), length = 41 000: 04 01 55 53 52 6f 62 6f 74 69 63 73 00 50 43 4d 010: 43 49 41 20 32 38 38 30 30 20 44 61 74 61 2f 46 020: 61 78 00 56 36 2e 31 00 ff Version = 4.1, Manuf = [USRobotics],card vers = [PCMCIA 28800 Data/Fax] Addit. info = [V6.1],[ÿ] Tuple #3, code = 0x0 (Null tuple), length = 9 000: 15 01 30 37 34 39 2d 00 ff Tuple #4, code = 0x21 (Functional ID), length = 3 000: 02 00 ff Serial port/modem Tuple #5, code = 0x22 (Functional EXT), length = 4 000: 00 02 0f 7c Serial interface extension: 16550 UART, Parity - Space,Mark,Odd,Even, Tuple #6, code = 0x22 (Functional EXT), length = 9 000: 01 1f 09 00 08 00 00 06 00 Modem interface capabilities: Tuple #7, code = 0x22 (Functional EXT), length = 13 000: 02 06 00 3f 0c 03 03 07 07 00 00 b5 20 Data modem services available: Tuple #8, code = 0x22 (Functional EXT), length = 8 000: 13 06 00 0f 00 02 b5 20 Tuple #9, code = 0x22 (Functional EXT), length = 8 000: 23 06 00 0f 00 02 b5 20 Tuple #10, code = 0x0 (Null tuple), length = 22 000: 55 53 52 01 04 01 f0 01 08 02 01 00 03 04 30 33 010: 31 37 04 01 14 ff Tuple #11, code = 0x1a (Configuration map), length = 5 000: 01 04 e6 01 07 Reg len = 2, config register addr = 0x1e6, last config = 0x4 Registers: XXX----- Tuple #12, code = 0x1b (Configuration entry), length = 22 000: c1 41 9a 37 55 4d 5d 36 46 21 8e 7d 55 aa 60 f8 010: 03 07 30 b8 86 08 Config index = 0x1(default) Interface byte = 0x41 (I/O) +RDY/-BSY active Vcc pwr: Nominal operating supply voltage: 5 x 1V Minimum operating supply voltage: 4.5 x 1V Maximum operating supply voltage: 5.5 x 1V Max current average over 1 second: 3 x 100mA Max current average over 10 ms: 4 x 100mA Vpp pwr: Nominal operating supply voltage: 1.2 x 10V, ext = 0x7d Max current average over 10 ms: 5 x 10mA Card decodes 10 address lines, 8 Bit I/O only I/O address # 1: block start = 0x3f8 block length = 0x8 IRQ modes: Level IRQs: 4 5 11 12 13 15 Max twin cards = 0 Misc attr: (Audio-BVD2) Tuple #13, code = 0x1b (Configuration entry), length = 11 000: 83 41 18 aa 60 f8 02 07 30 b8 86 Config index = 0x3 Interface byte = 0x41 (I/O) +RDY/-BSY active Card decodes 10 address lines, 8 Bit I/O only I/O address # 1: block start = 0x2f8 block length = 0x8 IRQ modes: Level IRQs: 4 5 11 12 13 15 Tuple #14, code = 0x1b (Configuration entry), length = 11 000: 82 41 18 aa 60 e8 03 07 30 b8 86 Config index = 0x2 Interface byte = 0x41 (I/O) +RDY/-BSY active Card decodes 10 address lines, 8 Bit I/O only I/O address # 1: block start = 0x3e8 block length = 0x8 IRQ modes: Level IRQs: 4 5 11 12 13 15 Tuple #15, code = 0x1b (Configuration entry), length = 11 000: 84 41 18 aa 60 e8 02 07 30 b8 86 Config index = 0x4 Interface byte = 0x41 (I/O) +RDY/-BSY active Card decodes 10 address lines, 8 Bit I/O only I/O address # 1: block start = 0x2e8 block length = 0x8 IRQ modes: Level IRQs: 4 5 11 12 13 15 Tuple #16, code = 0x14 (No link), length = 0 Tuple #17, code = 0xff (Terminator), length = 0 2 slots found For the 3Com card (a 3c589c): Configuration data for card in slot 0 Tuple #1, code = 0x1 (Common memory descriptor), length = 2 000: 00 ff Common memory device information: Device number 1, type No device, WPS = OFF Speed = No speed, Memory block size = reserved, 32 units Tuple #2, code = 0x17 (Attribute memory descriptor), length = 3 000: 43 02 ff Attribute memory device information: Device number 1, type EEPROM, WPS = OFF Speed = 150nS, Memory block size = 8Kb, 1 units Tuple #3, code = 0x20 (Manufacturer ID), length = 4 000: 01 01 89 05 PCMCIA ID = 0x101, OEM ID = 0x589 Tuple #4, code = 0x21 (Functional ID), length = 2 000: 06 00 Network/LAN adapter Tuple #5, code = 0x15 (Version 1 info), length = 57 000: 04 01 33 43 6f 6d 20 43 6f 72 70 6f 72 61 74 69 010: 6f 6e 00 33 43 35 38 39 00 54 50 2f 42 4e 43 20 020: 4c 41 4e 20 43 61 72 64 20 56 65 72 2e 20 32 61 030: 00 30 30 30 30 30 32 00 ff Version = 4.1, Manuf = [3Com Corporation],card vers = [3C589] Addit. info = [TP/BNC LAN Card Ver. 2a],[000002] Tuple #6, code = 0x1a (Configuration map), length = 6 000: 02 03 00 00 01 03 Reg len = 3, config register addr = 0x10000, last config = 0x3 Registers: XX------ Tuple #7, code = 0x1b (Configuration entry), length = 15 000: c1 01 1d 71 55 35 55 54 e0 72 5d 64 30 ff ff Config index = 0x1(default) Interface byte = 0x1 (I/O) Vcc pwr: Nominal operating supply voltage: 5 x 1V Max current average over 1 second: 3 x 10mA Max current average over 10 ms: 5 x 10mA Power down supply current: 5 x 1mA Wait scale Speed = 7.0 x 100 ns RDY/BSY scale Speed = 7.0 x 100 ns Card decodes 18 address lines, full 8/16 Bit I/O IRQ modes: Level, Pulse IRQs: IOCK 1 4 5 6 8 10 11 12 14 Tuple #8, code = 0x1b (Configuration entry), length = 7 000: 03 01 71 55 26 26 54 Config index = 0x3 Vcc pwr: Nominal operating supply voltage: 5 x 1V Max current average over 1 second: 2 x 100mA Max current average over 10 ms: 2 x 100mA Power down supply current: 5 x 1mA Tuple #9, code = 0x19 (JEDEC descr for attribute memory), length = 3 000: 00 00 ff Tuple #10, code = 0x14 (No link), length = 0 Tuple #11, code = 0x10 (Checksum), length = 5 000: 89 ff 80 00 00 Checksum from offset -119, length 128, value is 0x0 Tuple #12, code = 0xff (Terminator), length = 218 000: 00 03 00 00 00 00 00 00 43 32 20 54 45 41 4d 2d 010: 45 61 72 6c 20 41 6c 64 72 69 64 67 65 2c 20 52 020: 65 78 20 41 6c 6c 65 72 73 2c 20 52 75 73 73 65 030: 6c 6c 20 43 68 61 6e 67 2c 20 44 61 76 69 64 20 040: 43 68 69 68 2c 20 48 65 72 6d 61 6e 20 4c 61 6f 050: 2c 20 47 65 6e 65 20 4c 69 6e 2c 20 4c 69 61 6e 060: 67 20 4c 69 6e 2c 20 48 69 65 6e 20 4e 67 75 79 070: 65 6e 2c 20 4b 61 70 20 53 6f 68 2c 20 47 61 72 080: 79 20 57 61 6e 67 00 ff ff ff ff ff ff ff ff ff 090: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 0a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 0b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 0c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 0d0: ff ff ff ff ff ff ff ff ff ff 2 slots found The link light lights on the 3com card. When I insert it (or the modem), the following appear in the log: Oct 23 13:18:00 socrates cardd[264]: driver allocation failed for 3Com Corporation Oct 23 13:18:09 socrates su: ingham to root on /dev/ttyp3 Oct 23 14:06:37 socrates /kernel: Card removed, slot 0 Oct 23 14:06:45 socrates /kernel: Card inserted, slot 0 Oct 23 14:06:51 socrates /kernel: Card removed, slot 0 Oct 23 14:06:56 socrates cardd[264]: ioctl (PIOCSIO): Device not configured Oct 23 14:06:58 socrates /kernel: Card inserted, slot 1 Oct 23 14:06:58 socrates /kernel: Card inserted, slot 0 Oct 23 14:07:08 socrates cardd[264]: driver allocation failed for USRobotics Oct 23 14:07:18 socrates cardd[264]: driver allocation failed for 3Com Corporation (The dreaded ``driver allocation failed'' message! :-) One thing that might be useful in troubleshooting is to know what device is not configured in the message above... Thanks for putting up with what is probably a duplicate question to one asked recently. -- Kenneth Ingham ingham@i-pi.com From owner-freebsd-mobile Thu Oct 23 15:32:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA02655 for mobile-outgoing; Thu, 23 Oct 1997 15:32:04 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA02600 for ; Thu, 23 Oct 1997 15:32:00 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id QAA12683; Thu, 23 Oct 1997 16:31:58 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id QAA17227; Thu, 23 Oct 1997 16:31:56 -0600 (MDT) Date: Thu, 23 Oct 1997 16:31:56 -0600 (MDT) Message-Id: <199710232231.QAA17227@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Kenneth Ingham Cc: mobile@freebsd.org Subject: Re: 2.2.5-RELEASE and PC Cards In-Reply-To: <19971023142754.64160@fatman.i-pi.com> References: <19971023142754.64160@fatman.i-pi.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I just upgraded my laptop which used to run 2.2.2+PAO to 2.2.5-RELEASE. > The upgrade went OK except for the PCCArds no longer work. I built a > new kernel with the ep, crd, and pcic drivers and booted it. So far so good. > I've verified with pccardc dumpcis that the manufacturer and version > match. This is not the PAO pccard.conf file that I was using (but > it didn't work either). In fact, here's my pccard.conf file (I > removed everything except the info for the cards I am trying to > use (hmmm, the date in this is old, but I didn't see a newer one > in the upgrade files): It's the same in 2.2.5. > # Removing all IRQ conflicts from this file can't be done because of some > # IRQ-selfish PC-cards. So if you want to use some of these cards in > # your machine, you will be forced to modify their IRQ parameters from > # the following list. > # > # IRQ == 0 means "allocate free IRQ from IRQ pool" > # IRQ == 16 means "do not use IRQ (e.g. PIO mode)" > # > # $Id: pccard.conf.sample,v 1.4 1996/06/19 01:28:07 nate Exp $ > > # Generally available IO ports > io 0x240-0x360 > # Generally available IRQs (Built-in sound-card owners remove 5) > irq 3 5 10 11 13 15 > # Available memory slots > memory 0xd4000 96k > > # 3Com Etherlink III 3C589B, 3C589C > card "3Com Corporation" "3C589" > config 0x1 "ep0" ? ^ Change this to 10 or 11. (I know, humor me.) > insert echo 3Com Etherlink III inserted > insert /etc/pccard_ether ep0 -link0 link1 > # insert /etc/pccard_ether ep0 link0 -link1 > remove echo 3Com Etherlink III removed > remove /sbin/ifconfig ep0 delete > # US Robotics Sportster PCMCIA V.34 > # US Robotics COURIER PCMCIA V.34 > card "USRobotics" "PCMCIA 28800 Data/Fax" > config 0x3 "sio2" 3 > insert echo USRobotics PCMCIA Modem inserted > remove echo USRobotics PCMCIA Modem removed OK, from above: > sio2 not found at 0x3e8 ^^^^^ > Tuple #13, code = 0x1b (Configuration entry), length = 11 > 000: 83 41 18 aa 60 f8 02 07 30 b8 86 > Config index = 0x3 ^^^^^^^^^^^ > Interface byte = 0x41 (I/O) +RDY/-BSY active > Card decodes 10 address lines, 8 Bit I/O only > I/O address # 1: block start = 0x2f8 block length = 0x8 ^^^^^ > I/O address # 1: block start = 0x2f8 block length = 0x8 > IRQ modes: Level > IRQs: 4 5 11 12 13 15 ^^^^^^^^^^^^^^^ The config index and the port differ. Also, the card "doesn't support" IRQ 3 (although that shouldn't mattter, but it does due to a bug I think.) So, replace the above config line with: config 0x3 "sio1" 5 (Use IRQ 5 or 11, again humor me.) > And, I used the undocumented pccardc dumpcis on the cards and got the > following. It's documented as well as most things are documented. :) :) [ Dumps deleted ] > The link light lights on the 3com card. When I insert it (or the modem), > the following appear in the log: The link light is the 'power light'. > Oct 23 13:18:00 socrates cardd[264]: driver allocation failed for 3Com Corporation Something must have happened before this. > Oct 23 13:18:09 socrates su: ingham to root on /dev/ttyp3 > Oct 23 14:06:37 socrates /kernel: Card removed, slot 0 > Oct 23 14:06:45 socrates /kernel: Card inserted, slot 0 > Oct 23 14:06:51 socrates /kernel: Card removed, slot 0 > Oct 23 14:06:56 socrates cardd[264]: ioctl (PIOCSIO): Device not configured This error is due to the card being swapped in/out quickly, so the device driver never was fully loaded/unloaded. > Oct 23 14:06:58 socrates /kernel: Card inserted, slot 1 > Oct 23 14:06:58 socrates /kernel: Card inserted, slot 0 > Oct 23 14:07:08 socrates cardd[264]: driver allocation failed for USRobotics > Oct 23 14:07:18 socrates cardd[264]: driver allocation failed for 3Com Corporation > > (The dreaded ``driver allocation failed'' message! :-) See above how to correct them. > One thing that might be useful in troubleshooting is to know what > device is not configured in the message above... The card was inserted/removed too quickly. Nate From owner-freebsd-mobile Thu Oct 23 17:04:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id RAA07467 for mobile-outgoing; Thu, 23 Oct 1997 17:04:34 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id RAA07458 for ; Thu, 23 Oct 1997 17:04:30 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id SAA13296; Thu, 23 Oct 1997 18:04:21 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id SAA17893; Thu, 23 Oct 1997 18:04:20 -0600 (MDT) Date: Thu, 23 Oct 1997 18:04:20 -0600 (MDT) Message-Id: <199710240004.SAA17893@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Nate Williams Cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710231611.KAA15468@rocky.mt.sri.com> References: <199710221610.KAA10346@rocky.mt.sri.com> <199710231344.XAA00290@word.smith.net.au> <199710231611.KAA15468@rocky.mt.sri.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Nate, hit me with them post-fishin' changes. > > Will do when I get them wrapped up. Unfortunately, they didn't work out. I was hoping to get suspend/resume to work without requiring the pccard daemon running. (Ie; once a driver was 'mapped' in, suspend/resume didn't require the daemon to re-register the device, which is how the apm_pccard_resume code from PAO works.) In any case, I'd like to figure out a way to have a device truly 'shutdown', and then come back up. However, I need to look at things a bit more closely, since my 'brute force and ignorance' patch didn't seem to help. However, on a positive note, I can suspend/resume with apm_pccard_resume set to 1 on my box. :) Nate From owner-freebsd-mobile Thu Oct 23 17:49:19 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id RAA09616 for mobile-outgoing; Thu, 23 Oct 1997 17:49:19 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from tesla.i-pi.com (tesla.i-pi.com [198.49.217.4]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id RAA09605 for ; Thu, 23 Oct 1997 17:49:14 -0700 (PDT) (envelope-from ingham@tesla.i-pi.com) Received: (from ingham@localhost) by tesla.i-pi.com (8.8.7/8.8.7) id SAA11202; Thu, 23 Oct 1997 18:49:06 -0600 (MDT) Message-ID: <19971023184906.08326@i-pi.com> Date: Thu, 23 Oct 1997 18:49:06 -0600 From: Kenneth Ingham To: Nate Williams Cc: mobile@freebsd.org Subject: Re: 2.2.5-RELEASE and PC Cards References: <19971023142754.64160@fatman.i-pi.com> <199710232231.QAA17227@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.84e In-Reply-To: <199710232231.QAA17227@rocky.mt.sri.com>; from Nate Williams on Thu, Oct 23, 1997 at 04:31:56PM -0600 Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Thanks to Nate's suggested changed to /etc/pccard.conf, things are better. The ethernet card is recognized when I plug it in, and ifconfig shows it properly configured. The modem appears to work (I can talk to it), but I haven't actually tried dialing with it. However, now the power light on the connector is not on. If I pull the card out, wait a moment, and put it back in, the light lights up until I get the message that the card has been recognized. Then it goes out. It goes out right on the second beep. For reference, here is the pccard.conf file again: # Sample PCCARD configuration file # # Removing all IRQ conflicts from this file can't be done because of some # IRQ-selfish PC-cards. So if you want to use some of these cards in # your machine, you will be forced to modify their IRQ parameters from # the following list. # # IRQ == 0 means "allocate free IRQ from IRQ pool" # IRQ == 16 means "do not use IRQ (e.g. PIO mode)" # # $Id: pccard.conf.sample,v 1.4 1996/06/19 01:28:07 nate Exp $ # Generally available IO ports io 0x240-0x360 # Generally available IRQs (Built-in sound-card owners remove 5) irq 3 5 10 11 13 15 # Available memory slots memory 0xd4000 96k # 3Com Etherlink III 3C589B, 3C589C card "3Com Corporation" "3C589" config 0x1 "ep0" 11 insert echo 3Com Etherlink III inserted insert /etc/pccard_ether ep0 -link0 link1 # insert /etc/pccard_ether ep0 link0 -link1 remove echo 3Com Etherlink III removed remove /sbin/ifconfig ep0 delete # US Robotics Sportster PCMCIA V.34 # US Robotics COURIER PCMCIA V.34 card "USRobotics" "PCMCIA 28800 Data/Fax" config 0x3 "sio1" 5 insert echo USRobotics PCMCIA Modem inserted remove echo USRobotics PCMCIA Modem removed Kenneth From owner-freebsd-mobile Thu Oct 23 18:06:29 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id SAA10462 for mobile-outgoing; Thu, 23 Oct 1997 18:06:29 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from hydrogen.nike.efn.org (resnet.uoregon.edu [128.223.170.28]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id SAA10456 for ; Thu, 23 Oct 1997 18:06:23 -0700 (PDT) (envelope-from gurney_j@efn.org) Received: (from jmg@localhost) by hydrogen.nike.efn.org (8.8.7/8.8.7) id SAA18724; Thu, 23 Oct 1997 18:06:02 -0700 (PDT) Message-ID: <19971023180601.65534@hydrogen.nike.efn.org> Date: Thu, 23 Oct 1997 18:06:01 -0700 From: John-Mark Gurney To: Nate Williams Cc: mobile@FreeBSD.ORG Subject: Re: Patches from -current for -stable I'd like to commit after testing References: <199710221610.KAA10346@rocky.mt.sri.com> <199710231344.XAA00290@word.smith.net.au> <199710231611.KAA15468@rocky.mt.sri.com> <199710240004.SAA17893@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199710240004.SAA17893@rocky.mt.sri.com>; from Nate Williams on Thu, Oct 23, 1997 at 06:04:20PM -0600 Reply-To: John-Mark Gurney Organization: Cu Networking X-Operating-System: FreeBSD 2.2.1-RELEASE i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Nate Williams scribbled this message on Oct 23: > In any case, I'd like to figure out a way to have a device truly > 'shutdown', and then come back up. However, I need to look at things a > bit more closely, since my 'brute force and ignorance' patch didn't seem > to help. ahh.. thanks for telling me this.. right now I'm looking at designing a new bus/device interaction spec... and stuff like this is what I want to know about.. :) with your addition I realized that having "fixed" entries for probe/attach'ing is really stupid.. and that we should actually go on an event based system where an even is delievered to the device driver... right now it's something like: enum { DV_PROBE, DV_ATTACH, DV_DETACH, DV_SUSPEND, DV_RESUME }; int (*handler)(strcut device *dev, int event); this is still in design stage... so I'm open to suggestions... current spec is: http://resnet.uoregon.edu:6971/~jmg/FreeBSD/busdevice.html > However, on a positive note, I can suspend/resume with apm_pccard_resume > set to 1 on my box. :) I need to update current on my notebook.. I wish that I had 100mbit network with a 100mbit card.. doing an installworld over nfs just takes to darn long.. :( -- John-Mark Gurney Modem/FAX: +1 541 683 6954 Cu Networking Live in Peace, destroy Micro$oft, support free software, run FreeBSD From owner-freebsd-mobile Thu Oct 23 18:10:31 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id SAA10701 for mobile-outgoing; Thu, 23 Oct 1997 18:10:31 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from hydrogen.nike.efn.org (resnet.uoregon.edu [128.223.170.28]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id SAA10694 for ; Thu, 23 Oct 1997 18:10:27 -0700 (PDT) (envelope-from gurney_j@efn.org) Received: (from jmg@localhost) by hydrogen.nike.efn.org (8.8.7/8.8.7) id SAA18758; Thu, 23 Oct 1997 18:10:09 -0700 (PDT) Message-ID: <19971023181009.47077@hydrogen.nike.efn.org> Date: Thu, 23 Oct 1997 18:10:09 -0700 From: John-Mark Gurney To: Kenneth Ingham Cc: Nate Williams , mobile@FreeBSD.ORG Subject: Re: 2.2.5-RELEASE and PC Cards References: <19971023142754.64160@fatman.i-pi.com> <199710232231.QAA17227@rocky.mt.sri.com> <19971023184906.08326@i-pi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <19971023184906.08326@i-pi.com>; from Kenneth Ingham on Thu, Oct 23, 1997 at 06:49:06PM -0600 Reply-To: John-Mark Gurney Organization: Cu Networking X-Operating-System: FreeBSD 2.2.1-RELEASE i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Kenneth Ingham scribbled this message on Oct 23: > Thanks to Nate's suggested changed to /etc/pccard.conf, things are better. > The ethernet card is recognized when I plug it in, and ifconfig shows it > properly configured. > The modem appears to work (I can talk to it), but I haven't actually > tried dialing with it. > > However, now the power light on the connector is not on. > > If I pull the card out, wait a moment, and put it back in, the light > lights up until I get the message that the card has been recognized. > Then it goes out. It goes out right on the second beep. what type of connection is it? is it a 10bt card? if so, make sure that you have it plugged into the network (it's a link light)... if your using 10b2, then the power light is an activity light... (at least that's the way on my 3c589b)... also, the above sounds almost exactly what you should recieve.. have you tried to actually use the network? hope this helps.. ttyl.. -- John-Mark Gurney Modem/FAX: +1 541 683 6954 Cu Networking Live in Peace, destroy Micro$oft, support free software, run FreeBSD From owner-freebsd-mobile Thu Oct 23 18:17:46 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id SAA11027 for mobile-outgoing; Thu, 23 Oct 1997 18:17:46 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id SAA11022 for ; Thu, 23 Oct 1997 18:17:43 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id TAA13811; Thu, 23 Oct 1997 19:17:35 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id TAA18288; Thu, 23 Oct 1997 19:17:34 -0600 (MDT) Date: Thu, 23 Oct 1997 19:17:34 -0600 (MDT) Message-Id: <199710240117.TAA18288@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Kenneth Ingham Cc: Nate Williams , mobile@freebsd.org Subject: Re: 2.2.5-RELEASE and PC Cards In-Reply-To: <19971023184906.08326@i-pi.com> References: <19971023142754.64160@fatman.i-pi.com> <199710232231.QAA17227@rocky.mt.sri.com> <19971023184906.08326@i-pi.com> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Thanks to Nate's suggested changed to /etc/pccard.conf, things are better. > The ethernet card is recognized when I plug it in, and ifconfig shows it > properly configured. > However, now the power light on the connector is not on. This is for the ethernet card, right? That means the 10Base2 connector is selected. Change the link flags in /etc/pccard.conf. > # 3Com Etherlink III 3C589B, 3C589C > card "3Com Corporation" "3C589" > config 0x1 "ep0" 11 > insert echo 3Com Etherlink III inserted > insert /etc/pccard_ether ep0 -link0 link1 > # insert /etc/pccard_ether ep0 link0 -link1 Switch the comment character on the above two lines. Nate From owner-freebsd-mobile Thu Oct 23 19:30:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id TAA14428 for mobile-outgoing; Thu, 23 Oct 1997 19:30:55 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA14392 for ; Thu, 23 Oct 1997 19:30:23 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id LAA00527; Fri, 24 Oct 1997 11:56:58 +0930 (CST) Message-Id: <199710240226.LAA00527@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Thu, 23 Oct 1997 18:04:20 CST." <199710240004.SAA17893@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Oct 1997 11:56:58 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > Nate, hit me with them post-fishin' changes. > > > > Will do when I get them wrapped up. > > Unfortunately, they didn't work out. I was hoping to get suspend/resume > to work without requiring the pccard daemon running. (Ie; once a driver > was 'mapped' in, suspend/resume didn't require the daemon to re-register > the device, which is how the apm_pccard_resume code from PAO works.) I think that this isn't realistically achievable; the countercase is when a card is changed while the system is suspended. When you come back up it is not safe to make *any* assumptions about what's in either slot. > In any case, I'd like to figure out a way to have a device truly > 'shutdown', and then come back up. However, I need to look at things a > bit more closely, since my 'brute force and ignorance' patch didn't seem > to help. Um. I would be trying to fake a removal on power down, followed by a faked insertion when we wake back up. You might be able to fool the pcic into this if you turn the power to the slots off before you go down, and then back on when you come up again - note that I haven't studied the code yet, so I could be talking out an armpit. mike From owner-freebsd-mobile Thu Oct 23 19:41:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id TAA15145 for mobile-outgoing; Thu, 23 Oct 1997 19:41:48 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from tesla.i-pi.com (tesla.i-pi.com [198.49.217.4]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA15140 for ; Thu, 23 Oct 1997 19:41:45 -0700 (PDT) (envelope-from ingham@tesla.i-pi.com) Received: (from ingham@localhost) by tesla.i-pi.com (8.8.7/8.8.7) id UAA11387; Thu, 23 Oct 1997 20:41:40 -0600 (MDT) Message-ID: <19971023204140.59508@i-pi.com> Date: Thu, 23 Oct 1997 20:41:40 -0600 From: Kenneth Ingham To: Nate Williams Cc: mobile@freebsd.org Subject: Re: 2.2.5-RELEASE and PC Cards References: <19971023142754.64160@fatman.i-pi.com> <199710232231.QAA17227@rocky.mt.sri.com> <19971023184906.08326@i-pi.com> <199710240117.TAA18288@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.84e In-Reply-To: <199710240117.TAA18288@rocky.mt.sri.com>; from Nate Williams on Thu, Oct 23, 1997 at 07:17:34PM -0600 Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In my prior email, I moved a sentence and realized when I read it that it became unclear which card was giving the problem. However, the helpful people of this list were able to read my mind and addressed the real problem. Changing the line used in the pccard.conf to use the proper combinations of link0 and link1 was the solution. All cards work now. Many thanks to Nate and the others who provided suggestions! Kenneth From owner-freebsd-mobile Thu Oct 23 19:57:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id TAA15711 for mobile-outgoing; Thu, 23 Oct 1997 19:57:43 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA15703 for ; Thu, 23 Oct 1997 19:57:40 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id UAA14334; Thu, 23 Oct 1997 20:57:38 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id UAA18449; Thu, 23 Oct 1997 20:57:38 -0600 (MDT) Date: Thu, 23 Oct 1997 20:57:38 -0600 (MDT) Message-Id: <199710240257.UAA18449@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710240226.LAA00527@word.smith.net.au> References: <199710240004.SAA17893@rocky.mt.sri.com> <199710240226.LAA00527@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Unfortunately, they didn't work out. I was hoping to get suspend/resume > > to work without requiring the pccard daemon running. (Ie; once a driver > > was 'mapped' in, suspend/resume didn't require the daemon to re-register > > the device, which is how the apm_pccard_resume code from PAO works.) > > I think that this isn't realistically achievable; the countercase is > when a card is changed while the system is suspended. When you come > back up it is not safe to make *any* assumptions about what's in either > slot. True, but if you could do a mini-probe (as the current code attempts to do), then it should work. However, somehow things aren't yet 'enough' alive when we call the mini-probe, so it doesn't work. > > In any case, I'd like to figure out a way to have a device truly > > 'shutdown', and then come back up. However, I need to look at things a > > bit more closely, since my 'brute force and ignorance' patch didn't seem > > to help. > > Um. I would be trying to fake a removal on power down, followed by a > faked insertion when we wake back up. This is what the code that's enabled by apm_pccard_resume does (sort of). Except that it fakes both remove/insertion at resume time. The bad thing is that it requires that the pccard daemon be running for the 'insertion' to be correctly done. > You might be able to fool the pcic into this if you turn the power to > the slots off before you go down, and then back on when you come up > again - note that I haven't studied the code yet, so I could be > talking out an armpit. It seems to work on my box doing that now. Try enabling the sysctl and see what happens on your box. Nate From owner-freebsd-mobile Thu Oct 23 20:14:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id UAA16598 for mobile-outgoing; Thu, 23 Oct 1997 20:14:52 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id UAA16590 for ; Thu, 23 Oct 1997 20:14:39 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id MAA00751 for ; Fri, 24 Oct 1997 12:41:11 +0930 (CST) Message-Id: <199710240311.MAA00751@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Thu, 23 Oct 1997 20:57:38 CST." <199710240257.UAA18449@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Oct 1997 12:41:10 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > I think that this isn't realistically achievable; the countercase is > > when a card is changed while the system is suspended. When you come > > back up it is not safe to make *any* assumptions about what's in either > > slot. > > True, but if you could do a mini-probe (as the current code attempts to > do), then it should work. However, somehow things aren't yet 'enough' > alive when we call the mini-probe, so it doesn't work. Um, "mini-probe"? ie. "is what used to be there still there?" > > Um. I would be trying to fake a removal on power down, followed by a > > faked insertion when we wake back up. > > This is what the code that's enabled by apm_pccard_resume does (sort > of). Except that it fakes both remove/insertion at resume time. The > bad thing is that it requires that the pccard daemon be running for the > 'insertion' to be correctly done. Natch; it has to rummage the database somehow. How else would you get at this information? > > You might be able to fool the pcic into this if you turn the power to > > the slots off before you go down, and then back on when you come up > > again - note that I haven't studied the code yet, so I could be > > talking out an armpit. > > It seems to work on my box doing that now. Try enabling the sysctl and > see what happens on your box. Hmm, OK. (saves message, tries it) Works fine here. Yay! mike From owner-freebsd-mobile Thu Oct 23 20:44:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id UAA18147 for mobile-outgoing; Thu, 23 Oct 1997 20:44:12 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id UAA18142 for ; Thu, 23 Oct 1997 20:44:05 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id NAA00908; Fri, 24 Oct 1997 13:02:43 +0930 (CST) Message-Id: <199710240332.NAA00908@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: John-Mark Gurney cc: Nate Williams , mobile@FreeBSD.ORG Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Thu, 23 Oct 1997 18:06:01 MST." <19971023180601.65534@hydrogen.nike.efn.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Oct 1997 13:02:43 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > ahh.. thanks for telling me this.. right now I'm looking at designing > a new bus/device interaction spec... and stuff like this is what I want > to know about.. :) with your addition I realized that having "fixed" > entries for probe/attach'ing is really stupid.. and that we should > actually go on an event based system where an even is delievered to the > device driver... This is pretty hot. > right now it's something like: > enum { DV_PROBE, > DV_ATTACH, > DV_DETACH, > DV_SUSPEND, > DV_RESUME > }; DV_LOAD, DV_UNLOAD at least as well. (LOAD would be the first call made after the driver was linked in, UNLOAD would have to agree to being unloaded and invalidate all references to the driver.) > int (*handler)(strcut device *dev, int event); There's no support there for passing in more event information; the probe would want to take a number of parameters varying on the bus support making the probe. eg. the PCI code would want to pass in PCI ID numbers and preset resource information. Likewise the PnP code would have information derived from the PnP data, and a vanilla ISA probe would have data derived from a static configuration. mike From owner-freebsd-mobile Thu Oct 23 21:14:35 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA19955 for mobile-outgoing; Thu, 23 Oct 1997 21:14:35 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from hydrogen.nike.efn.org (resnet.uoregon.edu [128.223.170.28]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA19945 for ; Thu, 23 Oct 1997 21:14:27 -0700 (PDT) (envelope-from gurney_j@efn.org) Received: (from jmg@localhost) by hydrogen.nike.efn.org (8.8.7/8.8.7) id VAA19703; Thu, 23 Oct 1997 21:14:08 -0700 (PDT) Message-ID: <19971023211408.08161@hydrogen.nike.efn.org> Date: Thu, 23 Oct 1997 21:14:08 -0700 From: John-Mark Gurney To: Mike Smith Cc: Nate Williams , mobile@FreeBSD.ORG Subject: Re: Patches from -current for -stable I'd like to commit after testing References: <19971023180601.65534@hydrogen.nike.efn.org> <199710240332.NAA00908@word.smith.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199710240332.NAA00908@word.smith.net.au>; from Mike Smith on Fri, Oct 24, 1997 at 01:02:43PM +0930 Reply-To: John-Mark Gurney Organization: Cu Networking X-Operating-System: FreeBSD 2.2.1-RELEASE i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Mike Smith scribbled this message on Oct 24: > > > > ahh.. thanks for telling me this.. right now I'm looking at designing > > a new bus/device interaction spec... and stuff like this is what I want > > to know about.. :) with your addition I realized that having "fixed" > > entries for probe/attach'ing is really stupid.. and that we should > > actually go on an event based system where an even is delievered to the > > device driver... > > This is pretty hot. > > > right now it's something like: > > enum { DV_PROBE, > > DV_ATTACH, > > DV_DETACH, > > DV_SUSPEND, > > DV_RESUME > > }; > > DV_LOAD, DV_UNLOAD at least as well. (LOAD would be the first call > made after the driver was linked in, UNLOAD would have to agree to > being unloaded and invalidate all references to the driver.) well.. this wouldn't work to well with the module code that Rabson came up with... if you want to do that, you do something like: static moduledata_t mytst = { "mytst", mytst_handler, (void *)100}; DECLARE_MODULE(mytst,mytst,SI_SUB_PSEUDO,SI_ORDER_ANY); then mytst_handler is called like: int mytst_handler(MOD_LOAD, (void *)100); and other possibles values are MOD_UNLOAD and MOD_SHUTDOWN... in the long run, I think it's much better to seperate the load proccess from the device driver... I guess we could make a seperate bus for these sorts of things.. but in my opinion it turns into an overkill... > > int (*handler)(strcut device *dev, int event); > > There's no support there for passing in more event information; the > probe would want to take a number of parameters varying on the bus > support making the probe. eg. the PCI code would want to pass in PCI > ID numbers and preset resource information. Likewise the PnP code > would have information derived from the PnP data, and a vanilla ISA > probe would have data derived from a static configuration. well... I do provide for a key on an address bus (darn, my spec doesn't have the defintion, see next paragraph), which would cover this... there is resource information that is stored in this structure that provides the resource info... I hadn't figured out how this was going to happen, but it will... basicly, as far as I can tell, there are two types of busses... abus is an address bus... each device is able to be addressed on this bus, and you are able to uniquely get a key that identifies this device.. this currently covers eisa, pci, pnp, mca, scsi, pccard, and more... then there is bbus, which is a bus bus.. everything listens to it and you have to magicly know what devices are listening.. :) currently only isa (and possibly vlb) are under this category... they will require a hard definition somewere for them to work... I was thinking that we could make the table that is dynamicly loaded that contains the isa bus information.. then to probe some new devices, a table with that new entry is loaded, and then a bus rescan is started which will notice the new, unprobed device, and will probe it... -- John-Mark Gurney Modem/FAX: +1 541 683 6954 Cu Networking Live in Peace, destroy Micro$oft, support free software, run FreeBSD From owner-freebsd-mobile Thu Oct 23 21:31:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA20755 for mobile-outgoing; Thu, 23 Oct 1997 21:31:55 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA20736 for ; Thu, 23 Oct 1997 21:31:03 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id NAA01201; Fri, 24 Oct 1997 13:56:50 +0930 (CST) Message-Id: <199710240426.NAA01201@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: John-Mark Gurney cc: Mike Smith , Nate Williams , mobile@FreeBSD.ORG Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Thu, 23 Oct 1997 21:14:08 MST." <19971023211408.08161@hydrogen.nike.efn.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Oct 1997 13:56:48 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > and other possibles values are MOD_UNLOAD and MOD_SHUTDOWN... in the > long run, I think it's much better to seperate the load proccess from > the device driver... I guess we could make a seperate bus for these > sorts of things.. but in my opinion it turns into an overkill... I don't think that it'll ever be possible to avoid having load/unload actions in a device driver module; drivers may want to allocate local storage, and they're the last arbiter as to whether they are ready to be unloaded, so whether you overload "LOAD" as "first probe" and "UNLOAD" as "last close", or make them explicit, the same actions are relevant. If the "kernel registry" stuff ever happens, or any other form of parametric access, there is scope for references into the driver outside the domain of open/close, and you still need a way of cleaning this up. Load/unload hooks are the only way to go. > basicly, as far as I can tell, there are two types of busses... abus > is an address bus... each device is able to be addressed on this bus, > and you are able to uniquely get a key that identifies this device.. > this currently covers eisa, pci, pnp, mca, scsi, pccard, and more... > then there is bbus, which is a bus bus.. everything listens to it and > you have to magicly know what devices are listening.. :) currently only > isa (and possibly vlb) are under this category... they will require > a hard definition somewere for them to work... Hmm, so you are proposing to avoid anything bus-specific in your driver model? ie. the driver provides a 'probe' function if it supports bbus operation, and alwyas provides an 'attach' which is called directly by abus stuff and indirectly as a result of a successful bbus probe? This is very clean. There are a few nasty-ish warts; the data passed to the attach routine will have to allow for opaque registration of interrupt handlers etc. and likewise the structure provided to probe routines will have to be *very* carefully generic, but still this would be a huge improvement. > I was thinking that we could make the table that is dynamicly loaded > that contains the isa bus information.. then to probe some new devices, > a table with that new entry is loaded, and then a bus rescan is started > which will notice the new, unprobed device, and will probe it... If you go back through your archive to the discussions that Stefan, Luigi(?) and I were having on this a few months back, I think So, big question: What can we do to help you with this? mike From owner-freebsd-mobile Thu Oct 23 22:09:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA23006 for mobile-outgoing; Thu, 23 Oct 1997 22:09:13 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from hydrogen.nike.efn.org (resnet.uoregon.edu [128.223.170.28]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA23001 for ; Thu, 23 Oct 1997 22:09:09 -0700 (PDT) (envelope-from gurney_j@efn.org) Received: (from jmg@localhost) by hydrogen.nike.efn.org (8.8.7/8.8.7) id WAA19860; Thu, 23 Oct 1997 22:08:57 -0700 (PDT) Message-ID: <19971023220857.42676@hydrogen.nike.efn.org> Date: Thu, 23 Oct 1997 22:08:57 -0700 From: John-Mark Gurney To: Mike Smith Cc: Nate Williams , mobile@FreeBSD.ORG Subject: Re: Patches from -current for -stable I'd like to commit after testing References: <19971023211408.08161@hydrogen.nike.efn.org> <199710240426.NAA01201@word.smith.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199710240426.NAA01201@word.smith.net.au>; from Mike Smith on Fri, Oct 24, 1997 at 01:56:48PM +0930 Reply-To: John-Mark Gurney Organization: Cu Networking X-Operating-System: FreeBSD 2.2.1-RELEASE i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Mike Smith scribbled this message on Oct 24: > > > > and other possibles values are MOD_UNLOAD and MOD_SHUTDOWN... in the > > long run, I think it's much better to seperate the load proccess from > > the device driver... I guess we could make a seperate bus for these > > sorts of things.. but in my opinion it turns into an overkill... > > I don't think that it'll ever be possible to avoid having load/unload > actions in a device driver module; drivers may want to allocate local > storage, and they're the last arbiter as to whether they are ready to > be unloaded, so whether you overload "LOAD" as "first probe" and > "UNLOAD" as "last close", or make them explicit, the same actions are > relevant. ok... well.. I think that should be covered by the attach/detach events... > If the "kernel registry" stuff ever happens, or any other form of > parametric access, there is scope for references into the driver > outside the domain of open/close, and you still need a way of cleaning > this up. Load/unload hooks are the only way to go. hmmm... if it's part of a module that you load, you could just unload it.. and this is what I'm thinking about... > > basicly, as far as I can tell, there are two types of busses... abus > > is an address bus... each device is able to be addressed on this bus, > > and you are able to uniquely get a key that identifies this device.. > > this currently covers eisa, pci, pnp, mca, scsi, pccard, and more... > > then there is bbus, which is a bus bus.. everything listens to it and > > you have to magicly know what devices are listening.. :) currently only > > isa (and possibly vlb) are under this category... they will require > > a hard definition somewere for them to work... > > Hmm, so you are proposing to avoid anything bus-specific in your driver > model? ie. the driver provides a 'probe' function if it supports bbus > operation, and alwyas provides an 'attach' which is called directly by > abus stuff and indirectly as a result of a successful bbus probe? pretty much yes... I was still thinking about calling probe from abus code too... but this can be eliminated if we provide for a way to tell how long each key is... and probably should be the way it's done.. > This is very clean. There are a few nasty-ish warts; the data passed > to the attach routine will have to allow for opaque registration of > interrupt handlers etc. and likewise the structure provided to probe > routines will have to be *very* carefully generic, but still this would > be a huge improvement. oh.. don't worry, check out my spec... there are no resources that are global per say... you have to obtain access to the resources through the bus code... this will return the appropriate information that provides the access... > > I was thinking that we could make the table that is dynamicly loaded > > that contains the isa bus information.. then to probe some new devices, > > a table with that new entry is loaded, and then a bus rescan is started > > which will notice the new, unprobed device, and will probe it... > > If you go back through your archive to the discussions that Stefan, > Luigi(?) and I were having on this a few months back, I think yeh... I remeber that discussion.. I thought I was in it a bit... :) > So, big question: What can we do to help you with this? well.. you could take a look at my spec, and provide input.. remeber though.. it's still under work, and that my real work is on my white board in my room.. :) also... as you probably see, I have most of the code->device, and device->bus interaction.. but I haven't got much of the code->bus interaction that provides the addresses... right now I'm thinking of providing the bus a way to notify my code that an address/device has arrived... then each address on the bus will have a message delievered to my code that it exists (this fits in nicely with interrupt driven scsi probing :) ), and then the code will probe/attach the device if there is a driver available... could you attend classes and do my Physics work? :) -- John-Mark Gurney Modem/FAX: +1 541 683 6954 Cu Networking Live in Peace, destroy Micro$oft, support free software, run FreeBSD From owner-freebsd-mobile Thu Oct 23 22:43:30 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA24538 for mobile-outgoing; Thu, 23 Oct 1997 22:43:30 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA24529 for ; Thu, 23 Oct 1997 22:43:11 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id PAA01521; Fri, 24 Oct 1997 15:09:04 +0930 (CST) Message-Id: <199710240539.PAA01521@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: John-Mark Gurney cc: mobile@FreeBSD.ORG Subject: Bus arch ramblings (was Re: Patches from -current for -stable ... In-reply-to: Your message of "Thu, 23 Oct 1997 22:08:57 MST." <19971023220857.42676@hydrogen.nike.efn.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Oct 1997 15:08:59 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > > I don't think that it'll ever be possible to avoid having load/unload > > actions in a device driver module; drivers may want to allocate local > > storage, and they're the last arbiter as to whether they are ready to > > be unloaded, so whether you overload "LOAD" as "first probe" and > > "UNLOAD" as "last close", or make them explicit, the same actions are > > relevant. > > ok... well.. I think that should be covered by the attach/detach > events... You're missing the point here; attach/detach is tied to a device instance, not a module instance. We're not just talking device driver modules here, and even they may have per-module rather than per-device setup/teardown. > > If the "kernel registry" stuff ever happens, or any other form of > > parametric access, there is scope for references into the driver > > outside the domain of open/close, and you still need a way of cleaning > > this up. Load/unload hooks are the only way to go. > > hmmm... if it's part of a module that you load, you could just unload > it.. and this is what I'm thinking about... Ok, so as a part of my module startup, I create a new sysctl node, which references a parameter in my address space. When I'm unloaded, that reference becomes invalid (and lethal). Short of attaching ownership attributes to everything like this (and there are lots of things that fall into this categorty), the *only* way to handle the departure of a module is to have the module itself clean up before it allows you to remove it. > > Hmm, so you are proposing to avoid anything bus-specific in your driver > > model? ie. the driver provides a 'probe' function if it supports bbus > > operation, and alwyas provides an 'attach' which is called directly by > > abus stuff and indirectly as a result of a successful bbus probe? > > pretty much yes... I was still thinking about calling probe from abus > code too... but this can be eliminated if we provide for a way to tell > how long each key is... and probably should be the way it's done.. The driver should never see the key: - bus-specific code accesses peripheral according to bus standard, extracts identification tokens and resource information - bus-specific code looks up identification tokens in bus-specific data structure, and determines abus-specific driver identification. - bus-specific code passes abus-specific driver identification and resource information in abus format to abus code - abus code locates driver using identification supplied, calls driver attach function with resource data > > If you go back through your archive to the discussions that Stefan, > > Luigi(?) and I were having on this a few months back, I think > > yeh... I remeber that discussion.. I thought I was in it a bit... :) You and a dozen or so others. 8) > > So, big question: What can we do to help you with this? > > well.. you could take a look at my spec, and provide input.. remeber > though.. it's still under work, and that my real work is on my white > board in my room.. :) Which is at what URL? > right now I'm thinking of providing the bus a way to notify my code > that an address/device has arrived... then each address on the bus > will have a message delievered to my code that it exists (this fits > in nicely with interrupt driven scsi probing :) ), and then the code > will probe/attach the device if there is a driver available... Isn't this the whole idea? In the case of static busses like PCI, startup is a flurry of arrival events, while with things like PCCARDs and USB it's much more dynamic. > could you attend classes and do my Physics work? :) Hmm. No, but there's gotta be at least one physicist on the list. 8) mike From owner-freebsd-mobile Thu Oct 23 23:02:07 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA25794 for mobile-outgoing; Thu, 23 Oct 1997 23:02:07 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from hydrogen.nike.efn.org (resnet.uoregon.edu [128.223.170.28]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA25780 for ; Thu, 23 Oct 1997 23:02:01 -0700 (PDT) (envelope-from gurney_j@efn.org) Received: (from jmg@localhost) by hydrogen.nike.efn.org (8.8.7/8.8.7) id XAA20206; Thu, 23 Oct 1997 23:01:47 -0700 (PDT) Message-ID: <19971023230147.53852@hydrogen.nike.efn.org> Date: Thu, 23 Oct 1997 23:01:47 -0700 From: John-Mark Gurney To: Mike Smith Cc: mobile@FreeBSD.ORG Subject: Re: Bus arch ramblings (was Re: Patches from -current for -stable ... References: <19971023220857.42676@hydrogen.nike.efn.org> <199710240539.PAA01521@word.smith.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199710240539.PAA01521@word.smith.net.au>; from Mike Smith on Fri, Oct 24, 1997 at 03:08:59PM +0930 Reply-To: John-Mark Gurney Organization: Cu Networking X-Operating-System: FreeBSD 2.2.1-RELEASE i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Mike Smith scribbled this message on Oct 24: > > > I don't think that it'll ever be possible to avoid having load/unload > > > actions in a device driver module; drivers may want to allocate local > > > storage, and they're the last arbiter as to whether they are ready to > > > be unloaded, so whether you overload "LOAD" as "first probe" and > > > "UNLOAD" as "last close", or make them explicit, the same actions are > > > relevant. > > > > ok... well.. I think that should be covered by the attach/detach > > events... > > You're missing the point here; attach/detach is tied to a device > instance, not a module instance. We're not just talking device driver > modules here, and even they may have per-module rather than per-device > setup/teardown. hmm... what is a device driver module?? a module is just set of a name, a handler routine, and an arg... nothing more... > > > If the "kernel registry" stuff ever happens, or any other form of > > > parametric access, there is scope for references into the driver > > > outside the domain of open/close, and you still need a way of cleaning > > > this up. Load/unload hooks are the only way to go. > > > > hmmm... if it's part of a module that you load, you could just unload > > it.. and this is what I'm thinking about... > > Ok, so as a part of my module startup, I create a new sysctl node, > which references a parameter in my address space. When I'm unloaded, > that reference becomes invalid (and lethal). Short of attaching ok... so you add a reference to your module.. and when the unload event comes you either a) remove it, or b) revoke the removal... > ownership attributes to everything like this (and there are lots of > things that fall into this categorty), the *only* way to handle the > departure of a module is to have the module itself clean up before it > allows you to remove it. yep.. that's what the handler allows... my dummy module (the first that I used to test the new code) scheduled a timeout.. then upon unloading I used untimeout to remove it... > > > Hmm, so you are proposing to avoid anything bus-specific in your driver > > > model? ie. the driver provides a 'probe' function if it supports bbus > > > operation, and alwyas provides an 'attach' which is called directly by > > > abus stuff and indirectly as a result of a successful bbus probe? > > > > pretty much yes... I was still thinking about calling probe from abus > > code too... but this can be eliminated if we provide for a way to tell > > how long each key is... and probably should be the way it's done.. > > The driver should never see the key: > > - bus-specific code accesses peripheral according to bus standard, > extracts identification tokens and resource information agreed... > - bus-specific code looks up identification tokens in bus-specific data > structure, and determines abus-specific driver identification. > - bus-specific code passes abus-specific driver identification and > resource information in abus format to abus code > - abus code locates driver using identification supplied, calls driver > attach function with resource data before I started working on the project, I used to agree with that statement... it will still be possible to overload the two bus types that I provide (abus and bbus), and expand it.. but the reason I want to write the abus and bbus is to remove duplicate code that each of the bus-specific code would normally have to write... > > > So, big question: What can we do to help you with this? > > > > well.. you could take a look at my spec, and provide input.. remeber > > though.. it's still under work, and that my real work is on my white > > board in my room.. :) > > Which is at what URL? http://resnet.uoregon.edu:6971/~jmg/FreeBSD/busdevice.html sorry for the long url, but that's my home machine.. :) > > right now I'm thinking of providing the bus a way to notify my code > > that an address/device has arrived... then each address on the bus > > will have a message delievered to my code that it exists (this fits > > in nicely with interrupt driven scsi probing :) ), and then the code > > will probe/attach the device if there is a driver available... > > Isn't this the whole idea? In the case of static busses like PCI, > startup is a flurry of arrival events, while with things like PCCARDs > and USB it's much more dynamic. yep... I just haven't writen that part of the spec... just making sure you knew what I was intending.. -- John-Mark Gurney Modem/FAX: +1 541 683 6954 Cu Networking Live in Peace, destroy Micro$oft, support free software, run FreeBSD From owner-freebsd-mobile Thu Oct 23 23:31:01 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA26974 for mobile-outgoing; Thu, 23 Oct 1997 23:31:01 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA26920 for ; Thu, 23 Oct 1997 23:29:37 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id PAA01789; Fri, 24 Oct 1997 15:49:45 +0930 (CST) Message-Id: <199710240619.PAA01789@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: John-Mark Gurney cc: Mike Smith , mobile@FreeBSD.ORG Subject: Re: Bus arch ramblings (was Re: Patches from -current for -stable ... In-reply-to: Your message of "Thu, 23 Oct 1997 23:01:47 MST." <19971023230147.53852@hydrogen.nike.efn.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Oct 1997 15:49:41 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk (Should we be moving this off -mobile?) > > You're missing the point here; attach/detach is tied to a device > > instance, not a module instance. We're not just talking device driver > > modules here, and even they may have per-module rather than per-device > > setup/teardown. > > hmm... what is a device driver module?? a module is just set of a > name, a handler routine, and an arg... nothing more... Exactly. So why is it that a module can only know it's been loaded if it happens to be a device driver that matches some currently-present hardware? > > Ok, so as a part of my module startup, I create a new sysctl node, > > which references a parameter in my address space. When I'm unloaded, > > that reference becomes invalid (and lethal). Short of attaching > > ok... so you add a reference to your module.. and when the unload event > comes you either a) remove it, or b) revoke the removal... So now every sysctl node has to contain an optional module reference. And every callout everywhere in the entire kernel likewise. And you have to traverse *all* of these in order to remove a module? Now how about a module that's allocated memory using the kernel malloc; you have to track that as well, and walk the entire malloc pool... Isn't it clear that it's easier to just have the module track its own resources, and back them out before leaving? > > ownership attributes to everything like this (and there are lots of > > things that fall into this categorty), the *only* way to handle the > > departure of a module is to have the module itself clean up before it > > allows you to remove it. > > yep.. that's what the handler allows... my dummy module (the first that > I used to test the new code) scheduled a timeout.. then upon unloading > I used untimeout to remove it... ... which is exactly what you're arguing against above. I'm confused now. > > - bus-specific code accesses peripheral according to bus standard, > > extracts identification tokens and resource information > > agreed... > > > - bus-specific code looks up identification tokens in bus-specific data > > structure, and determines abus-specific driver identification. > > - bus-specific code passes abus-specific driver identification and > > resource information in abus format to abus code > > - abus code locates driver using identification supplied, calls driver > > attach function with resource data > > before I started working on the project, I used to agree with that > statement... it will still be possible to overload the two bus types > that I provide (abus and bbus), and expand it.. but the reason I want > to write the abus and bbus is to remove duplicate code that each of the > bus-specific code would normally have to write... There will *have* to be bus-specific code, but it should be limited to dealing with the bus hardware and translating between the bus' specific view of the world and either the abus or bbus models. > > > > So, big question: What can we do to help you with this? > > > > > > well.. you could take a look at my spec, and provide input.. remeber > > > though.. it's still under work, and that my real work is on my white > > > board in my room.. :) > > > > Which is at what URL? > > http://resnet.uoregon.edu:6971/~jmg/FreeBSD/busdevice.html I meant "where is your whiteboard"? I have that one already 8) mike From owner-freebsd-mobile Thu Oct 23 23:59:25 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA28237 for mobile-outgoing; Thu, 23 Oct 1997 23:59:25 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from hydrogen.nike.efn.org (resnet.uoregon.edu [128.223.170.28]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA28231 for ; Thu, 23 Oct 1997 23:59:21 -0700 (PDT) (envelope-from gurney_j@efn.org) Received: (from jmg@localhost) by hydrogen.nike.efn.org (8.8.7/8.8.7) id XAA20413; Thu, 23 Oct 1997 23:59:08 -0700 (PDT) Message-ID: <19971023235908.26084@hydrogen.nike.efn.org> Date: Thu, 23 Oct 1997 23:59:08 -0700 From: John-Mark Gurney To: Mike Smith Cc: mobile@FreeBSD.ORG Subject: Re: Bus arch ramblings (was Re: Patches from -current for -stable ... References: <19971023230147.53852@hydrogen.nike.efn.org> <199710240619.PAA01789@word.smith.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199710240619.PAA01789@word.smith.net.au>; from Mike Smith on Fri, Oct 24, 1997 at 03:49:41PM +0930 Reply-To: John-Mark Gurney Organization: Cu Networking X-Operating-System: FreeBSD 2.2.1-RELEASE i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Mike Smith scribbled this message on Oct 24: > (Should we be moving this off -mobile?) > > > > You're missing the point here; attach/detach is tied to a device > > > instance, not a module instance. We're not just talking device driver > > > modules here, and even they may have per-module rather than per-device > > > setup/teardown. > > > > hmm... what is a device driver module?? a module is just set of a > > name, a handler routine, and an arg... nothing more... > > Exactly. So why is it that a module can only know it's been loaded if > it happens to be a device driver that matches some currently-present > hardware? > > > > Ok, so as a part of my module startup, I create a new sysctl node, > > > which references a parameter in my address space. When I'm unloaded, > > > that reference becomes invalid (and lethal). Short of attaching > > > > ok... so you add a reference to your module.. and when the unload event > > comes you either a) remove it, or b) revoke the removal... > > So now every sysctl node has to contain an optional module reference. > And every callout everywhere in the entire kernel likewise. And you > have to traverse *all* of these in order to remove a module? > > Now how about a module that's allocated memory using the kernel malloc; > you have to track that as well, and walk the entire malloc pool... > > Isn't it clear that it's easier to just have the module track its own > resources, and back them out before leaving? > > > > ownership attributes to everything like this (and there are lots of > > > things that fall into this categorty), the *only* way to handle the > > > departure of a module is to have the module itself clean up before it > > > allows you to remove it. > > > > yep.. that's what the handler allows... my dummy module (the first that > > I used to test the new code) scheduled a timeout.. then upon unloading > > I used untimeout to remove it... > > ... which is exactly what you're arguing against above. I'm confused > now. wierd... I thought I was arguing for the module tracking it's own resources... and refusing to unload when it gets the unload event.. :) > > > - bus-specific code accesses peripheral according to bus standard, > > > extracts identification tokens and resource information > > > > agreed... > > > > > - bus-specific code looks up identification tokens in bus-specific data > > > structure, and determines abus-specific driver identification. > > > - bus-specific code passes abus-specific driver identification and > > > resource information in abus format to abus code > > > - abus code locates driver using identification supplied, calls driver > > > attach function with resource data > > > > before I started working on the project, I used to agree with that > > statement... it will still be possible to overload the two bus types > > that I provide (abus and bbus), and expand it.. but the reason I want > > to write the abus and bbus is to remove duplicate code that each of the > > bus-specific code would normally have to write... > > There will *have* to be bus-specific code, but it should be limited to > dealing with the bus hardware and translating between the bus' specific > view of the world and either the abus or bbus models. agreed... > > > > > So, big question: What can we do to help you with this? > > > > > > > > well.. you could take a look at my spec, and provide input.. remeber > > > > though.. it's still under work, and that my real work is on my white > > > > board in my room.. :) > > > > > > Which is at what URL? > > > > http://resnet.uoregon.edu:6971/~jmg/FreeBSD/busdevice.html > > I meant "where is your whiteboard"? I have that one already 8) ahh.. ok.. if someone could mail me an electronic whiteboard.. I'd gladly use that instead... :) I'll see what I can do about updating the page in the next few hours.. I have some other work to do... thanks for the input... -- John-Mark Gurney Modem/FAX: +1 541 683 6954 Cu Networking Live in Peace, destroy Micro$oft, support free software, run FreeBSD From owner-freebsd-mobile Fri Oct 24 00:19:27 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA29095 for mobile-outgoing; Fri, 24 Oct 1997 00:19:27 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (word.smith.net.au [202.0.75.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA29088 for ; Fri, 24 Oct 1997 00:19:22 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id QAA02012; Fri, 24 Oct 1997 16:45:59 +0930 (CST) Message-Id: <199710240715.QAA02012@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: John-Mark Gurney cc: Mike Smith , mobile@FreeBSD.ORG Subject: Re: Bus arch ramblings (was Re: Patches from -current for -stable ... In-reply-to: Your message of "Thu, 23 Oct 1997 23:59:08 MST." <19971023235908.26084@hydrogen.nike.efn.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Oct 1997 16:45:57 +0930 From: Mike Smith Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > yep.. that's what the handler allows... my dummy module (the first that > > > I used to test the new code) scheduled a timeout.. then upon unloading > > > I used untimeout to remove it... > > > > ... which is exactly what you're arguing against above. I'm confused > > now. > > wierd... I thought I was arguing for the module tracking it's own > resources... and refusing to unload when it gets the unload event.. :) Um, I proposed a LOAD/UNLOAD event set, and I *thought* you were arguing against them, saying that such actions belonged in the attach/ detach routines. I think we have been in violent agreements. Whoops. > > I meant "where is your whiteboard"? I have that one already 8) > > ahh.. ok.. if someone could mail me an electronic whiteboard.. I'd > gladly use that instead... :) /usr/ports/mbone/wb mike From owner-freebsd-mobile Fri Oct 24 00:45:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA00586 for mobile-outgoing; Fri, 24 Oct 1997 00:45:54 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from hydrogen.nike.efn.org (resnet.uoregon.edu [128.223.170.28]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA00575 for ; Fri, 24 Oct 1997 00:45:48 -0700 (PDT) (envelope-from gurney_j@efn.org) Received: (from jmg@localhost) by hydrogen.nike.efn.org (8.8.7/8.8.7) id AAA20594; Fri, 24 Oct 1997 00:45:39 -0700 (PDT) Message-ID: <19971024004538.08323@hydrogen.nike.efn.org> Date: Fri, 24 Oct 1997 00:45:38 -0700 From: John-Mark Gurney To: Mike Smith Cc: mobile@FreeBSD.ORG Subject: Re: Bus arch ramblings (was Re: Patches from -current for -stable ... References: <19971023235908.26084@hydrogen.nike.efn.org> <199710240715.QAA02012@word.smith.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199710240715.QAA02012@word.smith.net.au>; from Mike Smith on Fri, Oct 24, 1997 at 04:45:57PM +0930 Reply-To: John-Mark Gurney Organization: Cu Networking X-Operating-System: FreeBSD 2.2.1-RELEASE i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Mike Smith scribbled this message on Oct 24: > > > > yep.. that's what the handler allows... my dummy module (the first that > > > > I used to test the new code) scheduled a timeout.. then upon unloading > > > > I used untimeout to remove it... > > > > > > ... which is exactly what you're arguing against above. I'm confused > > > now. > > > > wierd... I thought I was arguing for the module tracking it's own > > resources... and refusing to unload when it gets the unload event.. :) > > Um, I proposed a LOAD/UNLOAD event set, and I *thought* you were > arguing against them, saying that such actions belonged in the attach/ > detach routines. > > I think we have been in violent agreements. Whoops. yep... :) oh well.. mistakes happen... > > > I meant "where is your whiteboard"? I have that one already 8) > > > > ahh.. ok.. if someone could mail me an electronic whiteboard.. I'd > > gladly use that instead... :) > > /usr/ports/mbone/wb only problem is that it's kinda hard to hang on my wall... but I do happen to live behind a slirp connection.. I tried tunneling to the machine and using mrouted to forward packets over, but all the good broadcast packets (the ones requesting what's up in the mbone world) all went over the ethernet wire instead of down the p2p ppp link that I had set up (it was also on a private set of ip's which might cause a few problems)... any ideas? I've updated the page to list the new event handler (gee, it removed two entries from the gendevice structure :) ) and after looking at it, turns out that the document was actually newer than my whiteboard (not really, just that I had a thing or two on the whiteboard listed as not added to the document when it really was, oops)... -- John-Mark Gurney Modem/FAX: +1 541 683 6954 Cu Networking Live in Peace, destroy Micro$oft, support free software, run FreeBSD From owner-freebsd-mobile Fri Oct 24 05:24:35 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id FAA14008 for mobile-outgoing; Fri, 24 Oct 1997 05:24:35 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from out1.ibm.net (out1.ibm.net [165.87.194.252]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id FAA14003 for ; Fri, 24 Oct 1997 05:24:33 -0700 (PDT) (envelope-from SimsS@IBM.Net) Received: from Elvis.RatsNest.VaBeach.Va.Us (slip166-72-229-110.va.us.ibm.net [166.72.229.110]) by out1.ibm.net (8.8.5/8.6.9) with SMTP id MAA106798; Fri, 24 Oct 1997 12:23:21 GMT Received: by localhost with Microsoft MAPI; Fri, 24 Oct 1997 08:23:27 -0400 Message-ID: <01BCE056.1A43EE00.SimsS@IBM.Net> From: Steve Sims Reply-To: "SimsS@IBM.Net" To: Kenneth Ingham Cc: "mobile@FreeBSD.ORG" Subject: RE: 2.2.5-RELEASE and PC Cards Date: Fri, 24 Oct 1997 08:23:26 -0400 X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4128 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk There's not, by chance, Win'95 involved here, is there? I've learned that if I multi-boot into Win'95 and it weaves its voodoo on my 3C589D then the only way to get FreeBSD 2.2.x to work with it is to either power-cycle the laptop or to pull the card, wait, and re-insert it. My assumption is that the Win'95 PnP code puts the 3C589D into some bogus state that can't easily be sorted out without a hard re-init of the card. An irritation, but also a compelling reason to minimize the number of times I actually boot a Micro$oft product ;-) An aside: Under Win'95 I get, like "best-case" ~60K/sec (FTP) between my laptop and my FreeBSD server. Under FreeBSD on the laptop, I consistently get > 700K/sec. (This on a relatively saturated 10b2 network.) I think Unca Bill's people have some work to do. ...sjs... On Thursday, October 23, 1997 9:10 PM, John-Mark Gurney [SMTP:gurney_j@efn.org] wrote: > Kenneth Ingham scribbled this message on Oct 23: > > Thanks to Nate's suggested changed to /etc/pccard.conf, things are better. > > The ethernet card is recognized when I plug it in, and ifconfig shows it > > properly configured. > > The modem appears to work (I can talk to it), but I haven't actually > > tried dialing with it. > > > > However, now the power light on the connector is not on. > > > > If I pull the card out, wait a moment, and put it back in, the light > > lights up until I get the message that the card has been recognized. > > Then it goes out. It goes out right on the second beep. > > what type of connection is it? is it a 10bt card? if so, make sure > that you have it plugged into the network (it's a link light)... if > your using 10b2, then the power light is an activity light... (at least > that's the way on my 3c589b)... also, the above sounds almost exactly > what you should recieve.. have you tried to actually use the network? > > hope this helps.. ttyl.. > > -- > John-Mark Gurney Modem/FAX: +1 541 683 6954 > Cu Networking > > Live in Peace, destroy Micro$oft, support free software, run FreeBSD From owner-freebsd-mobile Fri Oct 24 07:19:14 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id HAA22102 for mobile-outgoing; Fri, 24 Oct 1997 07:19:14 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from merit.edu (merit.edu [198.108.1.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id HAA22088 for ; Fri, 24 Oct 1997 07:19:11 -0700 (PDT) (envelope-from web@merit.edu) Received: from ohm.merit.edu (ohm.merit.edu [198.108.60.65]) by merit.edu (8.8.7/8.8.5) with ESMTP id KAA04173 for ; Fri, 24 Oct 1997 10:18:40 -0400 (EDT) From: William Bulley Received: (web@localhost) by ohm.merit.edu (8.6.9/8.6.5) id KAA04444 for mobile@FreeBSD.ORG; Fri, 24 Oct 1997 10:18:46 -0400 Message-Id: <199710241418.KAA04444@ohm.merit.edu> Subject: new topic! IR support To: mobile@FreeBSD.ORG Date: Fri, 24 Oct 1997 10:18:46 -0400 (EDT) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Does FreeBSD have any support for the IR hardware found on several (most?) modern laptop units? I will admit publicly to not having RTFM'd for this answer and I apologize in advance for wasting the bandwidth if this turns out to be a simple RTFM issue... :-) Regards, web... -- William Bulley, N8NXN Senior Systems Research Programmer Merit Network, Inc. Email: web@merit.edu 4251 Plymouth Road, Suite C Phone: (313) 764-9993 Ann Arbor, Michigan 48105-2785 Fax: (313) 647-3185 [ What's all this fuss over the end of the century and mission critial ] [ programs failing due to dates? If people simply started using Roman ] [ Numerials, the problem goes away! MCM = 1900 MIM = 1999 MM = 2000 ] From owner-freebsd-mobile Fri Oct 24 08:11:53 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id IAA25926 for mobile-outgoing; Fri, 24 Oct 1997 08:11:53 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from coconut.itojun.org (root@coconut.itojun.org [210.160.95.97]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA25921 for ; Fri, 24 Oct 1997 08:11:49 -0700 (PDT) (envelope-from itojun@itojun.org) Received: from localhost (itojun@localhost.itojun.org [127.0.0.1]) by coconut.itojun.org (8.8.5/3.6Wbeta6) with ESMTP id AAA22064; Sat, 25 Oct 1997 00:11:34 +0900 (JST) To: William Bulley cc: mobile@FreeBSD.ORG In-reply-to: web's message of Fri, 24 Oct 1997 10:18:46 -0400. <199710241418.KAA04444@ohm.merit.edu> X-Template-Reply-To: itojun@itojun.org X-Template-Return-Receipt-To: itojun@itojun.org X-PGP-Fingerprint: F8 24 B4 2C 8C 98 57 FD 90 5F B4 60 79 54 16 E2 Subject: Re: new topic! IR support From: Jun-ichiro itojun Itoh Date: Sat, 25 Oct 1997 00:11:34 +0900 Message-ID: <22060.877705894@coconut.itojun.org> Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >Does FreeBSD have any support for the IR hardware found on several >(most?) modern laptop units? I will admit publicly to not having >RTFM'd for this answer and I apologize in advance for wasting the >bandwidth if this turns out to be a simple RTFM issue... :-) If you configure BIOS setup and kernel config file properly, you can use IR port port as half-duplex serial port, on most machines (NOTE: baudrate for the IR port must be <= 1Mbps). However, the IR port has very strange behavior if we see it as a serial port: if the both party transmit a character at the same time, the character will be lost due to the shared medium nature. (yes, the air) Using IR port as over-1Mbps device (such as 4Mbps) requires some hack. In that case IR port must be used as HDLC device (varies by chipset, I believe). itojun From owner-freebsd-mobile Fri Oct 24 08:52:45 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id IAA29227 for mobile-outgoing; Fri, 24 Oct 1997 08:52:45 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA29222 for ; Fri, 24 Oct 1997 08:52:42 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id JAA18824; Fri, 24 Oct 1997 09:52:40 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id JAA20521; Fri, 24 Oct 1997 09:52:38 -0600 (MDT) Date: Fri, 24 Oct 1997 09:52:38 -0600 (MDT) Message-Id: <199710241552.JAA20521@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710240311.MAA00751@word.smith.net.au> References: <199710240257.UAA18449@rocky.mt.sri.com> <199710240311.MAA00751@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [ Mike, can you try and make sure I'm in the Cc list. I'm getting lost in freebsd mail, so keeping me in the Cc list helps me stay on top of things I want to be involved with. Thanks! ] > > > I think that this isn't realistically achievable; the countercase is > > > when a card is changed while the system is suspended. When you come > > > back up it is not safe to make *any* assumptions about what's in either > > > slot. > > > > True, but if you could do a mini-probe (as the current code attempts to > > do), then it should work. However, somehow things aren't yet 'enough' > > alive when we call the mini-probe, so it doesn't work. > > Um, "mini-probe"? ie. "is what used to be there still there?" Yep. epinit(slot, int), where the int says 'is this the first time for this device'. If the second parameter is 0, then it does a 'mini-probe'. > > > Um. I would be trying to fake a removal on power down, followed by a > > > faked insertion when we wake back up. > > > > This is what the code that's enabled by apm_pccard_resume does (sort > > of). Except that it fakes both remove/insertion at resume time. The > > bad thing is that it requires that the pccard daemon be running for the > > 'insertion' to be correctly done. > > Natch; it has to rummage the database somehow. How else would you get > at this information? It's already all saved in the kernel now from when the card was initially inserted. Nate From owner-freebsd-mobile Fri Oct 24 09:01:32 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA29800 for mobile-outgoing; Fri, 24 Oct 1997 09:01:32 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (ppp20.portal.net.au [202.12.71.120]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA29789 for ; Fri, 24 Oct 1997 09:01:27 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id BAA00950; Sat, 25 Oct 1997 01:28:03 +0930 (CST) Message-Id: <199710241558.BAA00950@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Fri, 24 Oct 1997 09:52:38 CST." <199710241552.JAA20521@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 25 Oct 1997 01:27:59 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > > > > True, but if you could do a mini-probe (as the current code attempts to > > > do), then it should work. However, somehow things aren't yet 'enough' > > > alive when we call the mini-probe, so it doesn't work. > > > > Um, "mini-probe"? ie. "is what used to be there still there?" > > Yep. epinit(slot, int), where the int says 'is this the first time for > this device'. If the second parameter is 0, then it does a > 'mini-probe'. Not good enough. I take one card out and replace with a similar-but-different one (eg. two 3c589's, two NE2000's, etc.) The miniprobe will say "yes, it's the same", even though it's not. Yes, I'm being pathalogial about this. 8) > > > This is what the code that's enabled by apm_pccard_resume does (sort > > > of). Except that it fakes both remove/insertion at resume time. The > > > bad thing is that it requires that the pccard daemon be running for the > > > 'insertion' to be correctly done. > > > > Natch; it has to rummage the database somehow. How else would you get > > at this information? > > It's already all saved in the kernel now from when the card was > initially inserted. Only if the miniprobe succeeds. If it fails, you still need pccardd to handle a possible new device. I appreciate that the miniprobe is meant as a speed optimisation; I'm just not sure it's robust enough. mike From owner-freebsd-mobile Fri Oct 24 09:06:47 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA00266 for mobile-outgoing; Fri, 24 Oct 1997 09:06:47 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA00252 for ; Fri, 24 Oct 1997 09:06:41 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id KAA18909; Fri, 24 Oct 1997 10:06:38 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id KAA20591; Fri, 24 Oct 1997 10:06:37 -0600 (MDT) Date: Fri, 24 Oct 1997 10:06:37 -0600 (MDT) Message-Id: <199710241606.KAA20591@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710241558.BAA00950@word.smith.net.au> References: <199710241552.JAA20521@rocky.mt.sri.com> <199710241558.BAA00950@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > > True, but if you could do a mini-probe (as the current code attempts to > > > > do), then it should work. However, somehow things aren't yet 'enough' > > > > alive when we call the mini-probe, so it doesn't work. > > > > > > Um, "mini-probe"? ie. "is what used to be there still there?" > > > > Yep. epinit(slot, int), where the int says 'is this the first time for > > this device'. If the second parameter is 0, then it does a > > 'mini-probe'. > > Not good enough. I take one card out and replace with a > similar-but-different one (eg. two 3c589's, two NE2000's, etc.) The > miniprobe will say "yes, it's the same", even though it's not. But, it will work the same, since we don't program the card to do anything, just the controller. If two 'fairly identical' cards can be probed the same, they also better act the same given the same conditions. I think this is a non-issue. > > > > This is what the code that's enabled by apm_pccard_resume does (sort > > > > of). Except that it fakes both remove/insertion at resume time. The > > > > bad thing is that it requires that the pccard daemon be running for the > > > > 'insertion' to be correctly done. > > > > > > Natch; it has to rummage the database somehow. How else would you get > > > at this information? > > > > It's already all saved in the kernel now from when the card was > > initially inserted. > > Only if the miniprobe succeeds. If it fails, you still need pccardd to > handle a possible new device. I appreciate that the miniprobe is meant > as a speed optimisation; I'm just not sure it's robust enough. I think it is. What I'd like to do is be able to move some of the configuration into the kernel, so we can *rely* on certain things being there for embedded applications. For example, if I have a machine that is tight on memory, I want pccardd to run until it gets both cards configured, and then die to free up memory. You may ask why this is important, and who would ever do something that silly, and I'll answer 'because it is'. Nate From owner-freebsd-mobile Fri Oct 24 09:38:24 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA02379 for mobile-outgoing; Fri, 24 Oct 1997 09:38:24 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (ppp20.portal.net.au [202.12.71.120]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA02367 for ; Fri, 24 Oct 1997 09:38:06 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id CAA01177; Sat, 25 Oct 1997 02:04:28 +0930 (CST) Message-Id: <199710241634.CAA01177@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Fri, 24 Oct 1997 10:06:37 CST." <199710241606.KAA20591@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 25 Oct 1997 02:04:25 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > Yep. epinit(slot, int), where the int says 'is this the first time for > > > this device'. If the second parameter is 0, then it does a > > > 'mini-probe'. > > > > Not good enough. I take one card out and replace with a > > similar-but-different one (eg. two 3c589's, two NE2000's, etc.) The > > miniprobe will say "yes, it's the same", even though it's not. > > But, it will work the same, since we don't program the card to do > anything, just the controller. If two 'fairly identical' cards can be > probed the same, they also better act the same given the same > conditions. I think this is a non-issue. Let's just make sure I understand what the 'mini-probe' entails, as I may be misunderstanding this. Before the mini-probe runs, is the device detached? ie. the mini-probe is basically going to run the probe and then attach routines again? (Do you have to do this anyway, to get the PCCARD back to a known state?) > I think it is. What I'd like to do is be able to move some of the > configuration into the kernel, so we can *rely* on certain things being > there for embedded applications. For example, if I have a machine that > is tight on memory, I want pccardd to run until it gets both cards > configured, and then die to free up memory. You may ask why this is > important, and who would ever do something that silly, and I'll answer > 'because it is'. I understand all about that; I've always wanted to be able to stuff the card recognition data (or some subset therof) into the kernel. mike From owner-freebsd-mobile Fri Oct 24 09:43:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA02669 for mobile-outgoing; Fri, 24 Oct 1997 09:43:50 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA02660 for ; Fri, 24 Oct 1997 09:43:46 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id KAA19149; Fri, 24 Oct 1997 10:43:43 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id KAA20805; Fri, 24 Oct 1997 10:43:42 -0600 (MDT) Date: Fri, 24 Oct 1997 10:43:42 -0600 (MDT) Message-Id: <199710241643.KAA20805@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710241634.CAA01177@word.smith.net.au> References: <199710241606.KAA20591@rocky.mt.sri.com> <199710241634.CAA01177@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [ Mini-probe ] > Let's just make sure I understand what the 'mini-probe' entails, as I > may be misunderstanding this. > > Before the mini-probe runs, is the device detached? No. It was 'suspended', which basically means the power was pulled from the card slot (and hence the card.) > ie. the mini-probe > is basically going to run the probe and then attach routines again? Just the probe, not the attach. > (Do you have to do this anyway, to get the PCCARD back to a known > state?) Well, there's the issue, and the answer is 'maybe so, I'm not sure'. I don't *think* so, but it may require it. I'm playing with some code to try and not require it. I know the linux code doesn't try to save the state, and instead does what the apm_pccard_resume code does. (Which isn't necessarily a bad thing.) However, I'm not sure what the other OS's do (NetBSD for example). Win95 appears to 'suspend/resume' the card, although it may just be the 'appearances', and not how it's actually implemented under the hood. Nate From owner-freebsd-mobile Fri Oct 24 18:44:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id SAA04076 for mobile-outgoing; Fri, 24 Oct 1997 18:44:26 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from jli.com (jli.com [199.2.111.1]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id SAA04067 for ; Fri, 24 Oct 1997 18:44:19 -0700 (PDT) (envelope-from uucp@jli.com) Received: by jli.com (Smail3.1.29.1 #3) id m0xOvGU-0002VsC; Fri, 24 Oct 97 18:43 PDT Message-Id: Received: (qmail 23181 invoked from network); 25 Oct 1997 01:42:28 -0000 Received: from softdnserror (127.0.0.1) by softdnserror with SMTP; 25 Oct 1997 01:42:28 -0000 To: mobile@freebsd.org Subject: laptop review MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <23171.877743717.1@cloud.rain.com> Date: Fri, 24 Oct 1997 18:41:57 -0700 From: Bill Trost Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Since it has come up recently: I ran across an "ultraportable" laptop review at http://www.cnet.com/Content/Reviews/Compare/Ultportables/index.html that I thought you (we?) freebsd-mobilers would be interested in. One thing really cool about the page is they actually tell you what kind of display hardware the machine uses, which is lots easier than poking around in the store on a machine with a mangled Control Panel. Now, for editorial comments: The page is a bit cutesy, but not bad. They trashed the Portege, which was a little surprising to me (not that I am any fan of Toshibas). I think I'm going to take courage in hand and get a Fujitsu Lifebook 635 or so. I hope to get an XF86Config for it. Wish me luck -- if there's interest, I'll let you know how it goes. From owner-freebsd-mobile Fri Oct 24 20:34:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id UAA09223 for mobile-outgoing; Fri, 24 Oct 1997 20:34:54 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id UAA09215 for ; Fri, 24 Oct 1997 20:34:51 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id VAA22999; Fri, 24 Oct 1997 21:34:49 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id VAA22881; Fri, 24 Oct 1997 21:34:46 -0600 (MDT) Date: Fri, 24 Oct 1997 21:34:46 -0600 (MDT) Message-Id: <199710250334.VAA22881@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Bill Trost Cc: mobile@freebsd.org Subject: Re: laptop review In-Reply-To: References: X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Since it has come up recently: I ran across an "ultraportable" laptop > review at > http://www.cnet.com/Content/Reviews/Compare/Ultportables/index.html > that I thought you (we?) freebsd-mobilers would be interested in. One > thing really cool about the page is they actually tell you what kind > of display hardware the machine uses, which is lots easier than poking > around in the store on a machine with a mangled Control Panel. I was suprised they didn't have the DEC HiNote Ultra II in the list, but apparently DEC is in the middle of changing model lines. What was no suprise was the high marks for the 560. If you can afford it, there is *NO* competition for it in the laptop market (yet). I'm hoping to see some soon, cause I *really* want to upgrade my old NEC box and I want something with a bit more kick than the current models. > I think I'm going to take courage in hand and get a Fujitsu Lifebook > 635 or so. I hope to get an XF86Config for it. Wish me luck -- if > there's interest, I'll let you know how it goes. Please, I'd like to know. Nate From owner-freebsd-mobile Sat Oct 25 00:39:56 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA19901 for mobile-outgoing; Sat, 25 Oct 1997 00:39:56 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from word.smith.net.au (ppp20.portal.net.au [202.12.71.120]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA19888 for ; Sat, 25 Oct 1997 00:39:42 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id RAA00394; Sat, 25 Oct 1997 17:05:50 +0930 (CST) Message-Id: <199710250735.RAA00394@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Nate Williams cc: Mike Smith , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-reply-to: Your message of "Fri, 24 Oct 1997 10:43:42 CST." <199710241643.KAA20805@rocky.mt.sri.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 25 Oct 1997 17:05:46 +0930 From: Mike Smith Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > ie. the mini-probe > > is basically going to run the probe and then attach routines again? > > Just the probe, not the attach. Then my basic gripe remains; in the ethernet case, if I pull card A and replace with card B of the same type, the arp code will be confused (wrong MAC address). > > (Do you have to do this anyway, to get the PCCARD back to a known > > state?) > > Well, there's the issue, and the answer is 'maybe so, I'm not sure'. I > don't *think* so, but it may require it. I'm playing with some code to > try and not require it. I know the linux code doesn't try to save the > state, and instead does what the apm_pccard_resume code does. (Which > isn't necessarily a bad thing.) However, I'm not sure what the other > OS's do (NetBSD for example). Win95 appears to 'suspend/resume' the > card, although it may just be the 'appearances', and not how it's > actually implemented under the hood. I think that caching "what was in the slot" at a lower level would be good, but that means moving the CIS parser inside the kernel. mike From owner-freebsd-mobile Sat Oct 25 01:29:14 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id BAA22114 for mobile-outgoing; Sat, 25 Oct 1997 01:29:14 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from dcarmich.pr.mcs.net (dcarmich.pr.mcs.net [204.95.63.202]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id BAA22109; Sat, 25 Oct 1997 01:29:08 -0700 (PDT) (envelope-from dcarmich@dcarmich.pr.mcs.net) Received: (from dcarmich@localhost) by dcarmich.pr.mcs.net (8.8.7/8.8.7) id DAA00245; Sat, 25 Oct 1997 03:28:38 -0500 (CDT) (envelope-from dcarmich) From: Douglas Carmichael Message-Id: <199710250828.DAA00245@dcarmich.pr.mcs.net> Subject: 2.2.5-RELEASE installs fine, but can't detect TI PCI-1130 CardBus controller To: freebsd-hackers@freebsd.org, freebsd-mobile@freebsd.org Date: Sat, 25 Oct 1997 03:28:37 -0500 (CDT) X-Mailer: ELM [version 2.4ME+ PL31H (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I installed the 2.2.5-RELEASE bindist and srcdist from ftp2.freebsd.org and they installed fine over my existing 2.2.2-RELEASE/PAO-970616 system, but 2.2.5-RELEASE does not detect my NEC Versa 6050MH laptop's TI PCI-1130 CardBus PCMCIA controller when 2.2.2-RELEASE correctly detected it. Here's my kernel configuration file: # My new kernel configuration (10/3/97) machine "i386" cpu "I586_CPU" ident NECVERSA maxusers 60 options INET #InterNETworking options FFS #Berkeley Fast Filesystem options MFS #Memory Filesystem options PROCFS #Process filesystem options "COMPAT_43" #Compatible with BSD 4.3 [KEEP THIS!] options UCONSOLE #Allow users to grab the console options SYSVSHM options SYSVSEM options SYSVMSG # kernel tracing options KTRACE # laptop-specific configuration options LAPTOP # If your laptop have not had Windoze95-Ready BIOS, please update it. # Such old BIOS'es sometimes have critical bugs at 32-bit protected # mode APM BIOS interface (which have not used by Windoze 3.1). # PC-card suspend/resume support (experimental) options APM_PCCARD_RESUME options PCIC_RESUME_RESET # Keep power for serial cards when the system suspends # (If your machine hangs up when you try to suspend the system with # FAX/Modem PCMCIA card, uncomment this option). #options SIO_SUSP_KEEP_PWR # Detach SCSI devices when the SCSI card is removed options SCSI_DETACH # Don't suspend the system immediately before the system is resumed # from suspended mode (Default 3 seconds) options "APM_NOSUSPEND_IMMEDIATE=3" config kernel root on wd0 controller isa0 controller pci0 controller crd0 device pcic0 at crd? device pcic1 at crd? controller fdc0 at isa? port "IO_FD1" bio irq 6 drq 2 vector fdintr disk fd0 at fdc0 drive 0 controller wdc0 at isa? port "IO_WD1" bio irq 14 vector wdintr disk wd0 at wdc0 drive 0 flags 0x80ff options ATAPI #Enable ATAPI support for IDE bus options ATAPI_STATIC #Don't do it as an LKM device wcd0 #IDE CD-ROM device sc0 at isa? port "IO_KBD" tty irq 1 vector scintr device npx0 at isa? port "IO_NPX" irq 13 vector npxintr # # Laptop support (see LINT for more options) # device apm0 at isa? # Advanced Power Management options APM_BROKEN_STATCLOCK # Workaround some buggy APM BIOS device sio0 at isa? port "IO_COM1" tty irq 4 vector siointr device sio1 at isa? port "IO_COM2" tty irq 3 vector siointr device sio2 at isa? port "IO_COM3" tty irq 9 vector siointr device lpt0 at isa? port? tty irq 7 vector lptintr device psm0 at isa? port "IO_KBD" conflicts tty irq 12 vector psmintr # Sound devices controller snd0 device sb0 at isa? port 0x220 irq 5 drq 1 vector sbintr options SBC_IRQ=5 device sbxvi0 at isa? drq 5 device sbmidi0 at isa? port 0x330 device opl0 at isa? port 0x388 # Order is important here due to intrusive probes, do *not* alphabetize # this list of network interfaces until the probes have been fixed. # Right now it appears that the ie0 must be probed before ep0. See # revision 1.20 of this file. pseudo-device loop pseudo-device speaker pseudo-device tun 2 pseudo-device pty 16 pseudo-device gzip # Exec gzipped a.out's pseudo-device vn #Vnode driver (turns a file into a device) And dmesg output: Copyright (c) 1992-1997 FreeBSD Inc. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 2.2.5-RELEASE #0: Sat Oct 25 02:07:45 CDT 1997 dcarmich@dcarmich.pr.mcs.net:/usr/src/sys/compile/NECVERSA CPU: Pentium (150.85-MHz 586-class CPU) Origin = "GenuineIntel" Id = 0x544 Stepping=4 Features=0x8001bf real memory = 50331648 (49152K bytes) avail memory = 47636480 (46520K bytes) Probing for devices on PCI bus 0: chip0 rev 2 on pci0:0 chip1 rev 3 on pci0:1 vga0 rev 69 on pci0:2 chip2 rev 4 int a irq ?? on pci0:3:0 chip3 rev 4 int b irq ?? on pci0:3:1 Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> pccard driver sio added sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A sio2 at 0x3e8-0x3ef irq 9 on isa sio2: type 16550A lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface psm0 at 0x60-0x64 irq 12 on motherboard psm0: device ID 0 fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: FIFO enabled, 8 bytes threshold fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): , 32-bit, multi-block-16 wd0: 1376MB (2818368 sectors), 2796 cyls, 16 heads, 63 S/T, 512 B/S npx0 on motherboard npx0: INT 16 interface apm0 on isa apm: found APM BIOS version 1.1 sb0 at 0x220 irq 5 drq 1 on isa sb0: sbxvi0 at 0x0 drq 5 on isa sbxvi0: sbmidi0 at 0x330 on isa opl0 at 0x388 on isa opl0: PC-Card VLSI 82C146 (5 mem & 2 I/O windows) pcic: controller irq 10 What could be causing the problem? Is there a later PAO that I can use? A prompt response would be appreciated. From owner-freebsd-mobile Sat Oct 25 06:39:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id GAA03489 for mobile-outgoing; Sat, 25 Oct 1997 06:39:15 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from dcarmich.pr.mcs.net (dcarmich.pr.mcs.net [204.95.63.202]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id GAA03481 for ; Sat, 25 Oct 1997 06:39:11 -0700 (PDT) (envelope-from dcarmich@dcarmich.pr.mcs.net) Received: (from dcarmich@localhost) by dcarmich.pr.mcs.net (8.8.7/8.8.7) id IAA00272; Sat, 25 Oct 1997 08:42:46 -0500 (CDT) (envelope-from dcarmich) From: Douglas Carmichael Message-Id: <199710251342.IAA00272@dcarmich.pr.mcs.net> Subject: Success with PAO-970616 on 2.2.5-RELEASE (surprisingly enough!) To: hosokawa@jp.freebsd.org Date: Sat, 25 Oct 1997 08:42:45 -0500 (CDT) Cc: freebsd-mobile@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31H (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk PAO-970616 successfully patched a clean 2.2.5-RELEASE source tree with only *one* rejection in /usr/src/sys/i386/conf/files.i386. *************** *** 111,117 **** i386/isa/if_ie.c optional ie device-driver i386/isa/if_le.c optional le device-driver i386/isa/if_lnc.c optional lnc device-driver i386/isa/if_sr.c optional sr device-driver i386/isa/if_ze.c optional ze device-driver i386/isa/if_zp.c optional zp device-driver i386/isa/isa.c optional isa device-driver --- 111,119 ---- i386/isa/if_ie.c optional ie device-driver i386/isa/if_le.c optional le device-driver i386/isa/if_lnc.c optional lnc device-driver + i386/isa/if_sn.c optional sn device-driver i386/isa/if_sr.c optional sr device-driver + i386/isa/if_wlp.c optional wlp device-driver i386/isa/if_ze.c optional ze device-driver i386/isa/if_zp.c optional zp device-driver i386/isa/isa.c optional isa device-driver Here's my dmesg output: Copyright (c) 1992-1997 FreeBSD Inc. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 2.2.5-RELEASE #0: Sat Oct 25 08:19:47 CDT 1997 dcarmich@dcarmich.pr.mcs.net:/usr/src/sys/compile/NECVERSA CPU: Pentium (150.85-MHz 586-class CPU) Origin = "GenuineIntel" Id = 0x544 Stepping=4 Features=0x8001bf real memory = 50331648 (49152K bytes) avail memory = 47632384 (46516K bytes) Initializing PC-card drivers: sio wdc Probing for devices on PCI bus 0: chip0 rev 2 on pci0:0 chip1 rev 3 on pci0:1 vga0 rev 69 on pci0:2 pcic0 rev 4 int a irq ?? on pci0:3:0 CardBus: Legacy PC-card 16bit I/O address [0x3e0] pcic1 rev 4 int b irq ?? on pci0:3:1 CardBus: Legacy PC-card 16bit I/O address [0x3e0] Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A sio2 at 0x3e8-0x3ef irq 9 on isa sio2: type 16550A lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface psm0 at 0x60-0x64 irq 12 on motherboard psm0: device ID 0 fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: FIFO enabled, 8 bytes threshold fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): , 32-bit, multi-block-16 wd0: 1376MB (2818368 sectors), 2796 cyls, 16 heads, 63 S/T, 512 B/S wdc0: unit 1 (atapi): , removable, accel, dma, iordy wcd0: 1722Kb/sec, 128Kb cache, audio play, 255 volume levels, ejectable tray wcd0: no disc inside, unlocked npx0 on motherboard npx0: INT 16 interface apm0 on isa apm: found APM BIOS version 1.1 sb0 at 0x220 irq 5 drq 1 on isa sb0: sbxvi0 at 0x0 drq 5 on isa sbxvi0: sbmidi0 at 0x330 on isa opl0 at 0x388 on isa opl0: PC-Card TI PCI-1130 [i82365 compatible mode] (5 mem & 2 I/O windows) pcic: controller irq 10 Will there be a PAO distribution specific to 2.2.5-RELEASE coming? From owner-freebsd-mobile Sat Oct 25 09:47:35 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA10408 for mobile-outgoing; Sat, 25 Oct 1997 09:47:35 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA10403 for ; Sat, 25 Oct 1997 09:47:31 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id KAA28413; Sat, 25 Oct 1997 10:47:28 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id KAA24291; Sat, 25 Oct 1997 10:47:26 -0600 (MDT) Date: Sat, 25 Oct 1997 10:47:26 -0600 (MDT) Message-Id: <199710251647.KAA24291@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Mike Smith Cc: Nate Williams , mobile@freebsd.org Subject: Re: Patches from -current for -stable I'd like to commit after testing In-Reply-To: <199710250735.RAA00394@word.smith.net.au> References: <199710241643.KAA20805@rocky.mt.sri.com> <199710250735.RAA00394@word.smith.net.au> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > ie. the mini-probe > > > is basically going to run the probe and then attach routines again? > > > > Just the probe, not the attach. > > Then my basic gripe remains; in the ethernet case, if I pull card A and > replace with card B of the same type, the arp code will be confused > (wrong MAC address). UTSL. Again, it's a non-issue. > > > (Do you have to do this anyway, to get the PCCARD back to a known > > > state?) > > > > Well, there's the issue, and the answer is 'maybe so, I'm not sure'. I > > don't *think* so, but it may require it. I'm playing with some code to > > try and not require it. I know the linux code doesn't try to save the > > state, and instead does what the apm_pccard_resume code does. (Which > > isn't necessarily a bad thing.) However, I'm not sure what the other > > OS's do (NetBSD for example). Win95 appears to 'suspend/resume' the > > card, although it may just be the 'appearances', and not how it's > > actually implemented under the hood. > > I think that caching "what was in the slot" at a lower level would be > good, but that means moving the CIS parser inside the kernel. I don't think so. *However*, in the tests that I've been performing, on my box if I suspend I lose the port mapping, so I must re-initialize them on the port when we come up, then do the mini-probe (since it fails right now.) More later... Nate From owner-freebsd-mobile Sat Oct 25 10:35:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id KAA12445 for mobile-outgoing; Sat, 25 Oct 1997 10:35:48 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id KAA12440; Sat, 25 Oct 1997 10:35:44 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id LAA28680; Sat, 25 Oct 1997 11:35:42 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id LAA24441; Sat, 25 Oct 1997 11:35:41 -0600 (MDT) Date: Sat, 25 Oct 1997 11:35:41 -0600 (MDT) Message-Id: <199710251735.LAA24441@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Douglas Carmichael Cc: freebsd-hackers@freebsd.org, freebsd-mobile@freebsd.org Subject: Re: 2.2.5-RELEASE installs fine, but can't detect TI PCI-1130 CardBus controller In-Reply-To: <199710250828.DAA00245@dcarmich.pr.mcs.net> References: <199710250828.DAA00245@dcarmich.pr.mcs.net> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I installed the 2.2.5-RELEASE bindist and srcdist from > ftp2.freebsd.org and they installed fine over my existing > 2.2.2-RELEASE/PAO-970616 system, but 2.2.5-RELEASE does not detect my > NEC Versa 6050MH laptop's TI PCI-1130 CardBus PCMCIA controller when > 2.2.2-RELEASE correctly detected it. ^^^^^^^^^^^^^ There have been *NO* changes from 2.2.2 -> 2.2.5 in the pccard code, so I'm pretty sure that 2.2.2 never detected your controller w/out the PAO patches. And, there isn't (yet?) a PAO release for 2.2.5, so you'll have to wait until one is made, or port the patches from PAO to 2.2.5. But, I'm confused: > And dmesg output: > Copyright (c) 1992-1997 FreeBSD Inc. > Copyright (c) 1982, 1986, 1989, 1991, 1993 > The Regents of the University of California. All rights reserved. > > FreeBSD 2.2.5-RELEASE #0: Sat Oct 25 02:07:45 CDT 1997 > dcarmich@dcarmich.pr.mcs.net:/usr/src/sys/compile/NECVERSA ... > PC-Card VLSI 82C146 (5 mem & 2 I/O windows) > pcic: controller irq 10 It certainly looks like your controller is found to me. Nate From owner-freebsd-mobile Sat Oct 25 22:11:31 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA18382 for mobile-outgoing; Sat, 25 Oct 1997 22:11:31 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA18357; Sat, 25 Oct 1997 22:10:49 -0700 (PDT) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id XAA02488; Sat, 25 Oct 1997 23:10:44 -0600 (MDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id XAA25845; Sat, 25 Oct 1997 23:10:43 -0600 (MDT) Date: Sat, 25 Oct 1997 23:10:43 -0600 (MDT) Message-Id: <199710260510.XAA25845@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Nate Williams Cc: cvs-committers@freebsd.org, mobile@freebsd.org, cvs-sys@freebsd.org Subject: Re: cvs commit: src/sys/i386/isa aic6360.c if_ed.c if_ep.c if_fe.c sio.c src/sys/pccard pccard.c slot.h In-Reply-To: <199710260454.VAA07386@freefall.freebsd.org> References: <199710260454.VAA07386@freefall.freebsd.org> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > nate 1997/10/25 21:54:17 PDT > > Modified files: > sys/i386/isa aic6360.c if_ed.c if_ep.c if_fe.c sio.c > sys/pccard pccard.c slot.h > Log: > - Functional changes to PCCARD support. > * Kill individual drivers 'suspend' routines, since there's no simple/safe > way to suspend/resume a card w/out going through the complete probe > at initialization time. > * Default to using the apm_pccard_resume sysctl code, which basically > pretends the card was removed, and then re-inserted. Suspend/resume > is now 'emulated' with a fake insert/removal. (Hence we no longer > need the driver-specific suspend routines.) I've got 2 or 3 more things on my white-board, and I hope to get one more done this weekend. This will require re-building your kernel from scratch, and it doesn't (yet) provide a whole lot more functionality if the sysctl didn't help you. But, I was completely unable to get my PCIC controller to 'come back to life' and recognize that the card sitting in the slot was alive, except by going through all of the hoops that are done to initialize the card. So, in essence I'm now agreeing that this is the way to go for now (and given Mike Smith's arguments, it isn't a bad way to go), at least until we figure out a better solution. Nate