Bug 2540
| Summary: | Paralell port ZIP drive 100 dosen't work with 2.6.5 | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | Peter (tuharsky) |
| Component: | Parallel | Assignee: | drivers_parallel |
| Status: | CLOSED CODE_FIX | ||
| Severity: | high | CC: | akpm, ambx1, bunk, dommi_s1 |
| Priority: | P2 | ||
| Hardware: | i386 | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.5 | Subsystem: | |
| Regression: | --- | Bisected commit-id: | |
| Attachments: |
parport-probe-rewrite.patch
2.6.4 kernel dmesg 2.6.5 kernel dmesg the kernel config file Dmesg after patch applied parport-pnp-revert.patch pnp-parport-detection-fix.patch |
||
|
Description
Peter
2004-04-18 23:46:17 UTC
Could you please send the dmesg output from 2.6.4 and from 2.6.5? There were PNP-related parport changes in 2.6.5 and perhaps that broke things. Created attachment 2628 [details]
parport-probe-rewrite.patch
I've attached my parport changes made for 2.6.5. Feel free to try reverting
them. They only affect device detection but maybe its confusing the ECP
detection code. If you haven't specifically tested 2.6.4, you may want to try
it also. Several parport changes went in then too.
Finally could you attach your .config. I'd like to see if PnP support is
enabled and also which specific parport options are selected.
Created attachment 2632 [details]
2.6.4 kernel dmesg
Here's the 2.6.4 dmesg. Maybe You're right, there are some lines more related
to paralell port than in 2.6.5
Created attachment 2633 [details]
2.6.5 kernel dmesg
Created attachment 2634 [details]
the kernel config file
Here You are -the .config
Guys, it's much worse than I expected -not only ZIP dosen't work -the WHOLE PARALELL PORT is broken with 2.6.5. My mom has similar machine -QDI KinetiZ 7E mb with KT133 chipset, VIA686 resp. VIA82C. She couldn't print with 2.6.5 while with 2.6.2 no problem. Just today I see, that father's print problems that I was not able to solve are probably related to 2.6.5 too! Father has different machine, he has ABIT KG7 mb I think, AFAIK AMD 751 or so northbridge, but I suspect he has VIA 686 resp. VIA 82C southbridge too! So please, look at it. Since now I cannot state if the problem is related only to VIA686 / VIA 82C or it affects more chipsets; if I'll test it on different chipset, I'll let You know. The severity might be normal, as far as I read the Severity gl, but non-functional printing is enough for high, isn't it? Yours is the only report of this problem. Please test the patch I sent. Created attachment 2638 [details]
Dmesg after patch applied
Yes, after applying the pnp-changes-removal patch, the paralell port works.
Here's dmesg.
I have exactely the same problem - complete parport stopped working with no error messages at all. Any data sent to my printer just never arrives there. Identical config with 2.6.4 gets the printer working again. Please see also http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=244728 My system is an AMD-751 board with 550MHz Athlon Created attachment 2675 [details]
parport-pnp-revert.patch
Can you confirm that this patch fixes it up?
Created attachment 2706 [details]
pnp-parport-detection-fix.patch
This patch should solve the problem. It prevents PnP detection if devices were
already detected by SuperIO. Any testing would be appreciated.
I have applied Adams's patch (http://bugme.osdl.org/attachment.cgi?id=2706&action=view) to the 2.6.5 parport_pc.c as distributed by the Debian project. The patch works for me (EPoX 8KTA3 (VIA KT133A)). Looking at the bug logs, it seems this issue is fixed by the patch from Comment 11. I'm closing this bug since the patch is already included in recent 2.6 kernels. |