On Thu, 3 Jul 2003, Antonio Fiol Bonn�n wrote:
First of all, thank you very much for your replies,
I have tried 2.5.70 and:
- I get far less frequent wakeup_hc and suspend_hc messages ...
- ... but still a couple every few seconds seems too much.
Do you have any USB devices plugged in to your computer? If you do then
there shouldn't be any suspend_hc messages at all. If you don't then there shouldn't be any wakeup_hc messages.
It does not matter. I get the same in both cases. You should've received my e-mail with a log.
Log starts with nothing but a plain cable connected to one USB port at the back of the computer.
The cable allows me to plug a device without hiding under my desk ;-)
Suspend and wakeup messages alternate whether there is something plugged or not. I have not tried to boot with the device plugged in, but that is not the point of USB, is it?.
I've seen things like this happen with kernels ealier than 2.5.72, but always it was after some sort of oops.
No oops here before registering the device. As you will see on the log, there is one AFTER some time.
Please note that 2.4 series kernel does not show the problem. However, there must be something strange about my controller, as a 2.4.something (how is that I cannot remember if it was 17, 18 or 19) hang my computer completely when plugging the stick in. Kernel 2.4.21 is working just fine.
Mr. Fulghum said that I do not have the affected hardware (PIIX4 USB controller). Well, that is possible. I have, however, some affected hardware, as I can see the problem appear on my system.
The hard is (at least lspci says so): USB Controller: Intel Corp. 82801BA/BAM USB (Hub #1) (rev 12)
It's possible that your controller has an error similar to the PIIX4.
2.4.21 says: after detecting the hubs: "d800: host controller halted. very bad"
Just in case that may help you somehow.
It works no better. In fact, I reckon 2.5.70 did not throw the oops at me...Do you advise me to try 2.5.74?
It won't hurt to try. And it might end up working better than 2.5.70.
Hope my other message is helpful to debug the problem.
Antonio Fiol
smime.p7s
Description: S/MIME Cryptographic Signature
