Re: [Qemu-devel] Preliminary patch to implement ehci

[René Rebe]

Hi Mark,

I started to add your EHCI patch to our T2 SDE KVM package:

 http://www.t2-project.org/packages/kvm.html

Which includes some re-diffing to current KVM HEAD:

 
http://svn.exactcode.de/t2/trunk/package/emulators/kvm/99-ehci.patch.disabled

It's named .disabled as it's currently not yet built by default as
it did not yet work for me.

I'll keep reviewing it as a functional USB 2 controller would
greatly help my debugging needs.

PS: The above package also contains the Mactel patches from
Alexander Graf, forward ported to current Qemu/KVM HEAD.
Cherry picking welcome.

 René

Mark Burkley wrote:
> On Sat, 01 Nov 2008 13:50:00 -0700, you wrote:
>
>   
> > I just tried EHCI patch on top of latest KVM/QEMU.
> > XP just hangs at boot.
> >
> > Here is the debug output from EHCI.
> >
> > 000.000 ehci init : registering USB ports with no device attached
> > 000.000 Alloc new timer for 0x1f26fa0
> > 000.000 ehci init : registering MMIO size 4096
> > 000.000 *** ehci init : under development 16-OCT-08 ***
> >
> > 251.048 ehci_map: region 0, addr F0000000, size 4096, s->mem 00000068
> > 395.068 ehci_map: region 0, addr F0000000, size 4096, s->mem 00000068
> > 300.331 ehci_map: region 0, addr F0000000, size 4096, s->mem 00000068
> > 002.163 mem_readl : addr=0x00000000, val=0x01000020
> > 000.117 mem_readb : addr=0x00000009, val=68
> > 000.178 mem_readl : addr=0x00000020, val=0x00080000
> > 000.020 mem_writel : USBCMD resetting ...
> > 000.010 mem_writel : USBCMD reset done, clear reset request bit
> > 000.005 mem_writel : addr=0x00000020, val=0x00080000
> > 000.015 mem_readl : addr=0x00000020, val=0x00080000
> > 000.016 mem_readl : addr=0x00000004, val=0x00000004
> > 000.017 mem_writel : addr=0x00000038, val=0x00000000
> > 000.025 mem_writel : addr=0x00000034, val=0x00000000
> > 000.111 mem_writel : addr=0x00000034, val=0x0193B000
> > 000.017 mem_writel : addr=0x00000038, val=0x0193C002
> > 000.074 mem_writel : addr=0x00000060, val=0x00000001
> > 000.016 mem_readl : addr=0x00000020, val=0x00080000
> > 000.017 mem_writel : addr=0x00000020, val=0x00010000
> > 000.017 mem_readl : addr=0x00000020, val=0x00010000
> > 000.016 mem_writel : USBCMD run, clear halt
> > 000.011 mem_writel : addr=0x00000020, val=0x00010001
> > 504.026 mem_readl : addr=0x00000020, val=0x00010001
> > 000.084 mem_readl : addr=0x00000028, val=0x0000003B
> > 510.650 mem_readl : addr=0x00000020, val=0x00010001
> > 000.077 mem_readl : addr=0x00000028, val=0x0000003B
> > 510.544 mem_readl : addr=0x00000020, val=0x00010001
> > 000.067 mem_readl : addr=0x00000028, val=0x0000003B
> > 324.302 WARNING - EHCI skipped 10 frames
> > 960.583 WARNING - EHCI skipped 4 frames
> > 019.186 WARNING - EHCI skipped 10 frames
> > 819.797 WARNING - EHCI skipped 3 frames
> > 023.364 WARNING - EHCI skipped 1 frames
> >     
>
> Sorry for delayed reply.  I tried but I can't reproduce this.  All the
> above output looks normal.  But it seems XP just stops reading the
> device for some reason.  Is your PCI configuration different to me or
> something like that?
>
> Mark
>   
--
 René Rebe - ExactCODE GmbH - Europe, Germany, Berlin
 http://exactcode.de | http://t2-project.org | http://rene.rebe.name