Re: Re: Fire-wire passthrough with Linux pv-ops (2.6.31.1)

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

On Sun, Nov 01, 2009 at 10:07:41PM +0800, Mr. Teo En Ming (Zhang Enming) wrote:
> Hi All,
> 
> I have just installed Slackware64 13.0 64-bit Linux as HVM domU. Please
> refer to
> http://enmingteo.wordpress.com/2009/11/01/picture-tutorial-how-to-setup-slackware64-13-0-hvm-domu-with-xen-3-5-unstable-pv-ops-dom0-kernel-2-6-31-4-in-fedora-11-x86-64-dom0/for
> more details and screenshots.
> 
> I have tried to passthrough the firewire controller on Intel DQ45CB
> motherboard to Slackware64 13.0 HVM domU but I can't load the kernel modules
> for it.

It looks as if they are were already loaded during bootup. If you want to
re-load them, first do 'rmmod' for all of the modules associated with the card.

But looking at your 'dmesg' output it looks as if it was loaded correctly?

> 
> Please have a look at the following output within Slackware64 13.0 HVM domU.
> 
> 00:05.0 FireWire (IEEE 1394): Agere Systems FW322/323 (rev 70) (prog-if 10
> [OHCI])
>     Subsystem: Intel Corporation Device 1003
>     Flags: bus master, medium devsel, latency 128, IRQ 36

Wow. IRQ 36? Hmm, well mmaybe that is OK since you are running
under QEMU as HVM guest.

Can you provide the lspci -vv output of the Dom0 and DomU, please?

>     Memory at e3001000 (32-bit, non-prefetchable) [size=4K]
>     Memory at e3002000 (32-bit, non-prefetchable) [size=4K]
>     Capabilities: [44] Power Management version 2
>     Kernel driver in use: ohci1394
>     Kernel modules: ohci1394

... snup ..
> dm-command: hot insert pass-through pci dev
> register_real_device: Assigning real physical device 02:01.0 ...
> pt_iomul_init: Error: pt_iomul_init can't open file /dev/xen/pci_iomul: No
> such file or directory: 0x2:0x1.0x0
> pt_register_regions: IO region registered (size=0x00001000
> base_addr=0xd3801000)
> pt_register_regions: IO region registered (size=0x00000100
> base_addr=0xd3800000)
> register_real_device: Error: Binding of interrupt failed! rc=-1
> register_real_device: Real physical device 02:01.0 registered successfuly!
> IRQ type = INTx

Hmmm.. I think that message is misleading. The code shows that it
failed to bind the PCI card (02:01:0) to an interrupt and disabled
the card. Then it prints the 'registered succesfully!' which is incorrect.

Can you provide the output of:
 - lspci -vvv (from Dom0)
 - lspci -vvv (from DomU)
 - cat /proc/interrupts (from Dom0) _before_ you launch any guests or
   call bind any devices to the pcistub/pciback.
 - cat /proc/interrupts (from DomU) after you launch it and try to 
   do stuff with the firewire. For example try to pull videos from
   your camera via the firewire port.

> cirrus vga map change while on lfb mode
> pt_iomem_map: e_phys=e3001000 maddr=d3801000 type=0 len=4096 index=0
> first_map=1
> pt_iomem_map: e_phys=e3002000 maddr=d3800000 type=0 len=4096 index=1
> first_map=1
> mapping vram to e0000000 - e0400000
> platform_fixed_ioport: changed ro/rw state of ROM memory area. now is rw
> state.
> platform_fixed_ioport: changed ro/rw state of ROM memory area. now is ro
> state.
> *pt_pci_write_config: Warning: Guest attempt to set address to unused Base
> Address Register. [00:05.0][Offset:30h][Length:4]*

It does pass it throught. I believe it is OK during the probing of the
device - but it does not say what the value it tried to write, so it is
unclear what it is doing at that point.

> 
> The line immediately above looks very suspicious.
> 
> -- 
> Mr. Teo En Ming (Zhang Enming) Dip(Mechatronics) BEng(Hons)(Mechanical
> Engineering)
> Alma Maters:
> (1) Singapore Polytechnic
> (2) National University of Singapore
> My Primary Blog: http://teo-en-ming-aka-zhang-enming.blogspot.com
> My Secondary Blog: http://enmingteo.wordpress.com
> My Youtube videos: http://www.youtube.com/user/enmingteo
> Email: space.time.universe@gmail.com
> MSN: teoenming@hotmail.com
> Mobile Phone (SingTel): +65-9648-9798
> Mobile Phone (Starhub Prepaid): +65-8369-2618
> Age: 31 (as at 30 Oct 2009)
> Height: 1.78 meters
> Race: Chinese
> Dialect: Hokkien
> Street: Bedok Reservoir Road
> Country: Singapore
> 
> On Sat, Oct 24, 2009 at 2:19 AM, Mr. Teo En Ming (Zhang Enming) <
> space.time.universe@gmail.com> wrote:
> 
> > From the screenshots of the device manager in the Win XP Home domU,
> > the OHCI IEEE 1394 firewire controller appears to be working properly.
> > The Windows drivers for the firewire controller reports that the
> > firewire controller is working fine.
> >
> > However, when you connect a firewire-based digital video camera to the
> > firewire port, it is not detected in the Windows virtual machine.
> >
> > I will probably have to find time to install a Linux HVM domU and try
> > to pass through the 1394 controller to it.
> >
> > --
> > Mr. Teo En Ming (Zhang Enming) Dip(Mechatronics) BEng(Hons)(Mechanical
> > Engineering)
> > Alma Maters:
> > (1) Singapore Polytechnic
> > (2) National University of Singapore
> > Blog URL: http://teo-en-ming-aka-zhang-enming.blogspot.com
> > Email: space.time.universe@gmail.com
> > MSN: teoenming@hotmail.com
> > Mobile Phone: +65-9648-9798
> > Street: Bedok Reservoir Road
> > Republic of Singapore
> >
> >
> > On Sat, Oct 24, 2009 at 1:57 AM, Konrad Rzeszutek Wilk
> > <konrad.wilk@oracle.com> wrote:
> > > On Sat, Oct 24, 2009 at 01:47:31AM +0800, Mr. Teo En Ming (Zhang Enming)
> > wrote:
> > >> Hi,
> > >>
> > >> Could you help me look at here
> > >>
> > http://lists.xensource.com/archives/html/xen-devel/2009-10/msg01149.html
> > >> and also here
> > http://lists.xensource.com/archives/html/xen-devel/2009-10/msg01134.html
> > >
> > > Its looks to be working from the Dom0 side just fine. I am curious as
> > what is happening
> > > on the Windows driver side. I don't know enough about where Windows
> > drivers put
> > > debug information so I can't suggest where to look there. One suggestion
> > is that
> > > you install Fedora Core 11 as a HVM guest and passthrough your firewire
> > controller to it.
> > > And then include the logs from the Fedora domU - that should give more
> > insight on why
> > > it is not working properly.
> > >
> >

> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel