Re: ACPI patches updated (20021111)

[Johannes Erdfelt <johannes-olwfUuQ3Q2lBDgjK7y7TUQ@public.gmane.org>]

Now that would be interesting.

There's only one interrupt line for each HC (Host Controller) and none of
the Linux USB code will poll for completions. It depends on the IRQ being
delivered.

What may be happening is that the IRQ is being misreported.

This might work because Linux thinks the HC is sharing the IRQ with
another device that works and as a result of that device receiving
interrupts, the HCD (HC Driver) will receive them as well.

JE

On Tue, Nov 19, 2002, Stian Jordet <liste-NI9zH3lwcMKHXe+LvDLADg@public.gmane.org> wrote:
> Just to let you know, this problem does not exist with usb-printers,
> they work perfect with acpi. But keyboard and mouse will only work with
> acpi turned off (!).
> 
> Regards,
> Stian Jordet
> 
> 
> tor, 2002-11-14 kl. 20:59 skrev Stian Jordet:
> > In my motherboard user-guide, it says that usb shares irq with pci-slot
> > 1 and 5. In the Bios I can change irq on slot 1/5 from auto to 5 (or
> > whatever). USB works fine whatever setting I'm using (auto or some irq),
> > as long as I'm not using acpi. With acpi everything BUT usb works
> > perfect. Hope this helps.
> > 
> > Regards,
> > Stian
> > 
> > 
> > tor, 2002-11-14 kl. 19:55 skrev Grover, Andrew:
> > > > From: Stian Jordet [mailto:liste-NI9zH3lwcMKHXe+LvDLADg@public.gmane.org] 
> > > > I did ofcourse mean that this does _not_ seem to be the 
> > > > problem without
> > > > me forcing usb to use irq 5, but perhaps the problem then is that acpi
> > > > and usb use the same irq? Is acpi not supposed to share irq with
> > > > anything?
> > > 
> > > Wait, you're saying you are somehow *forcing* USB to IRQ 5? Via what method?
> > > 
> > > The ACPI interrupt can share its interrupt, no problem. Are you experiencing
> > > problems when ACPI and usb share irq 9?
> > > 
> > > Regards -- Andy
> > > 
> > > > 
> > > > Regards,
> > > > Stian Jordet
> > > > 
> > > > tor, 2002-11-14 kl. 00:28 skrev Stian Jordet:
> > > > > ons, 2002-11-13 kl. 22:30 skrev Grover, Andrew:
> > > > > > > From: Stian Jordet [mailto:liste-NI9zH3lwcMKHXe+LvDLADg@public.gmane.org] 
> > > > > > 
> > > > > > with acpi:
> > > > > > >   5:     269867     266825    IO-APIC-edge  usb-uhci, 
> > > > > > > usb-uhci, usb-uhci
> > > > > > 
> > > > > > without acpi:
> > > > > > >   5:        893        757   IO-APIC-level  Ricoh Co 
> > > > Ltd RL5c476 II
> > > > > > > (#2), usb-uhci, usb-uhci, usb-uhci, eth0, ide2
> > > > > > 
> > > > > > Well, here's the problem. IRQ 5 should be IO-APIC-level.
> > > > > > 
> > > > > > We are supposedly setting irqs used by PCI to level, but 
> > > > apparently not in
> > > > > > your case. Interesting.
> > > > > > 
> > > > > > Regards -- Andy
> > > > > 
> > > > > That seems to be the problem without me forcing usb to irq 
> > > > 5 as well.
> > > > > I set all irq-routing to auto (and it still works perfect 
> > > > without acpi),
> > > > > and get these results:
> > > > > 
> > > > > stianj@chevrolet:~$ cat /root/interrupts-acpi 
> > > > >            CPU0       CPU1       
> > > > >   0:       8587      21835    IO-APIC-edge  timer
> > > > >   1:         51         81    IO-APIC-edge  keyboard
> > > > >   2:          0          0          XT-PIC  cascade
> > > > >   3:          0          2    IO-APIC-edge  serial
> > > > >   8:          1          2    IO-APIC-edge  rtc
> > > > >   9:         93        113   IO-APIC-level  acpi, usb-uhci, 
> > > > usb-uhci, usb-uhci
> > > > >  14:         16         12    IO-APIC-edge  ide0
> > > > >  15:         10          7    IO-APIC-edge  ide1
> > > > >  17:       1123       1104   IO-APIC-level  aic7xxx, EMU10K1
> > > > >  18:         34         36   IO-APIC-level  aic7xxx, Ricoh 
> > > > Co Ltd RL5c476 II
> > > > >  19:         92         81   IO-APIC-level  Ricoh Co Ltd 
> > > > RL5c476 II (#2), eth0,   
> > > > > ide2
> > > > > NMI:          0          0 
> > > > > LOC:      30342      30339 
> > > > > ERR:          0
> > > > > MIS:          0
> > > > > stianj@chevrolet:~$ cat /root/interrupts-no-acpi 
> > > > >            CPU0       CPU1       
> > > > >   0:       3463       4239    IO-APIC-edge  timer
> > > > >   1:         57         28    IO-APIC-edge  keyboard
> > > > >   2:          0          0          XT-PIC  cascade
> > > > >   3:          2          0    IO-APIC-edge  serial
> > > > >   5:         32         38   IO-APIC-level  aic7xxx, Ricoh 
> > > > Co Ltd RL5c476 II
> > > > >   8:          2          1    IO-APIC-edge  rtc
> > > > >   9:         67         72   IO-APIC-level  Ricoh Co Ltd 
> > > > RL5c476 II (#2), usb-uhci, usb-uhci, usb-uhci, eth0, ide2            
> > > > >  10:       1103       1094   IO-APIC-level  aic7xxx, EMU10K1
> > > > >  14:         13         15    IO-APIC-edge  ide0
> > > > >  15:          7         10    IO-APIC-edge  ide1
> > > > > NMI:          0          0      
> > > > > LOC:       7618       7617      
> > > > > ERR:          0                 
> > > > > MIS:          0                 
> > > > > 
> > > > > 
> > > > > But do you think you will find a solution to this? Or is 
> > > > this something I have to learn to live with?
> > 
> > 
> > 
> > -------------------------------------------------------
> > This sf.net email is sponsored by: To learn the basics of securing 
> > your web site with SSL, click here to get a FREE TRIAL of a Thawte 
> > Server Certificate: http://www.gothawte.com/rd524.html
> > _______________________________________________
> > Acpi-devel mailing list
> > Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> > https://lists.sourceforge.net/lists/listinfo/acpi-devel
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: To learn the basics of securing 
> your web site with SSL, click here to get a FREE TRIAL of a Thawte 
> Server Certificate: http://www.gothawte.com/rd524.html
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel


-------------------------------------------------------
This sf.net email is sponsored by: To learn the basics of securing 
your web site with SSL, click here to get a FREE TRIAL of a Thawte 
Server Certificate: http://www.gothawte.com/rd524.html