Hi there, On Mon, 15 Mar 2004, Vanilla Riddle wrote (1):
> On Sun, Mar 14, 2004 at 06:12:12PM +0000, Ged Haywood wrote: > > Wasn't there something about a one-line patch in one of the drivers > > recently? Something that broke things around 2.4.18-9? Or was it > > 2.4.17-8? > You tell me. :-) > If it were really so, where am I to look? At the bottom of this message. :) > What about those new chipsets that I heard about, nForce or whatever it > was - any good? Well I saw one guy who fixed his nVidia problems by buying a VIA motherboard, and I just spent most of the weekend playing with nVidia X drivers for the wife's DVDs and I can't say that I'm impressed with their support for Linux. So I don't personally recommend that source of hardware although YMMV of course. ----------------------------------------------------------------------------- On Mon, 15 Mar 2004, Vanilla Riddle wrote (2): > > > 00:07.2 USB Controller: VIA Technologies, Inc. USB (rev 1a) (prog-if 00 [UHCI]) > > Ooohhhh, bad. > Possibly, but the FAQ said newer kernels had workarounds against > problems with VIA? Yes, things have improved specifically for VIA problems but maybe not for all of them... > I'll try to see if it works on my other box, celeron based. Good plan. > > VIA stuff. Have you tried usb-uhci instad of uhci? > > You mean the 2.4 host controller? Well, as I tried to explain that it > failed also under 2.4.22. Yes, I mean the host controller driver. I don't follow what you mean about it also failing under 2.4.22 but I still mean try both usb-uhci as well as uhci. Some people have reported better results with one than with the other. My impression is that usb-uhci has a better record than uhci but that may be just plain wrong. >just trying diffrent versions of the kernel seems frustrating Sorry, but I have that T-shirt too. I've crashed more kernels than you've had hot dinners. :( > wouldn't even know what I'd be doing and what to look for. I wish I > knew which part changing caused things to stop working around some > point and patch it in a recent kernel. There have been changes for better *and* for worse recently in both the 2.4 and 2.6 kernels. Moreover the code is extremely different in the two, so it's not as simple as that. If it were, I think it would have been patched by now. > if I found an older version of the kernel, which solved things, I > can't stick to it for ever - what if an upgrade were to be necessary > at some point? I'm suggesting you experiment, which might help pin down some of the things I've been seeing people report. Things like devices that worked with kernel 2.4.19 and then mysteriously stopped working, or they worked up to 2.4.20 and then never worked again, or they worked up to 2.6.1-rc1 and not with 2.6.4, or... > I'd appreciate very much someone making a statement about the > invalid ID mapping that I was told about - the usb ID list says my > 04b4 is "Cypress Semiconductor Corp." which seems obviously > wrong. Could my device be just introducing itself wrong? The manufacturer of the chipset and the manufacturer of the device can be different. I don't know in this case that your mouse doesn't have a Cypress chip in it. ----------------------------------------------------------------------------- On Mon, 15 Mar 2004, Vanilla Riddle wrote (3): > Yesterday I plugged my mouse into somebody other's box running W2K, > without installing the drivers and it worked. So I guess its > protocol _is_ a standard one for a usb pointing device. That is of > course unless Windoze has some weird code in it which makes broken > stuff work, about which I wouldn't be surprised at all. Both of those conclusions are true. 73, Ged. PS: can we try to keep it down to one message per day? :) ============================================================================= >From [EMAIL PROTECTED] Mon Mar 15 16:44:50 2004 Date: Wed, 10 Mar 2004 21:37:59 +0100 (CET) From: Guennadi Liakhovetski <[EMAIL PROTECTED]> To: Malcolm Blaney <[EMAIL PROTECTED]> Cc: [EMAIL PROTECTED] Subject: Re: [Linux-usb-users] usb device hangs On Wed, 10 Mar 2004, Malcolm Blaney wrote: > I'm happy to say usb no longer hangs on my single board computer! I Great! > thought I'd try diffing the uhci code from the kernel we had working, > (2.4.18) and kernels that don't (so far 2.4.23 and 2.4.24). I got down > to _one_ line causing the problem: > > uhci->fsbr = 0; ...in alloc_uhci(). > > Which was of course added in 2.4.19. I removed the line, since it's not > in 2.4.18, and all the devices I had trouble with worked. Is this worth > posting to the dev list? Can someone tell me the effect of setting fsbr > to 0? Strange. In 2.4.x uhci is just kmalloc'ed at that point, so, it must be initialised. Also, I guess, you didn't test this "fix" under 2.6 - there the whole struct is initialised to 0 with memset(). So, although, it is an important observation, it is unlikely to be the correct fix. Yes, please do report this to usb-devel. And, please, put "UHCI" in the subject line, when you do this:-) Thanks Guennadi --- Guennadi Liakhovetski ------------------------------------------------------- This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users ------------------------------------------------------- This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
