Re: bus_dmamap_sync() for uhci(4)

[Michael Lorenz <macallan%netbsd.org@localhost>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

On Jun 8, 2008, at 15:15, Manuel Bouyer wrote:

> On Fri, Jun 06, 2008 at 09:41:08PM +0200, Manuel Bouyer wrote:
> > Hi,
> > the attached patch adds bus_dmamap_sync() calls to uhci. I tested
> > this on a sparc64 with a via ehci and a umodem device.
> > Note that this combination works on sparc64 without the  
> > bus_dmamap_sync()
> > calls because all memory used for DMA is from bus_dmamem_alloc().
> > But bus_dmamap_sync() on this platform is not a nop, and has some
> > DIAGNOSTIC checks, so it shows the calls I added are not completely
> > brocken. Unfortunably I don't have a device to test the iso
> > transfers.
> > I'll also look at adding such calls to ehci.
>
> Here is the patch for ehci. Also tested on sparc64, with a umass  
> device.

I applied both patches and ran bonnie++ on a USB disk on my O2:

uhci0 at pci0 dev 3 function 0: VIA Technologies VT83C572 USB  
Controller (rev. 0x50)
uhci0: interrupting at crime interrupt 10
usb0 at uhci0: USB revision 1.0
uhci1 at pci0 dev 3 function 1: VIA Technologies VT83C572 USB  
Controller (rev. 0x50)
uhci1: interrupting at crime interrupt 13
usb1 at uhci1: USB revision 1.0
ehci0 at pci0 dev 3 function 2: VIA Technologies VT8237 EHCI USB  
Controller (rev. 0x51)
ehci0: interrupting at crime interrupt 14
ehci0: dropped intr workaround enabled
ehci0: EHCI version 0.95
ehci0: companion controllers, 2 ports each: uhci0 uhci1
usb2 at ehci0: USB revision 2.0
...
uhub0 at usb0: VIA Technologies UHCI root hub, class 9/0, rev  
1.00/1.00, addr 1
uhub0: 2 ports with 2 removable, self powered
uhub1 at usb1: VIA Technologies UHCI root hub, class 9/0, rev  
1.00/1.00, addr 1
uhub1: 2 ports with 2 removable, self powered
uhub2 at usb2: VIA Technologies EHCI root hub, class 9/0, rev  
2.00/1.00, addr 1
uhub2: 4 ports with 4 removable, self powered
umass0 at uhub2 port 1 configuration 1 interface 0
umass0: vendor 0x05e3 USB TO IDE, rev 2.00/0.02, addr 2
umass0: using SCSI over Bulk-Only
scsibus2 at umass0: 2 targets, 1 lun per target
...
sd2 at scsibus2 target 0 lun 0: <TOSHIBA, MK2104MAV, 0811> disk fixed
sd2: fabricating a geometry
sd2: 2067 MB, 2067 cyl, 64 head, 32 sec, 512 bytes/sect x 4233600  
sectors

a few seconds into bonnie++ I got this:
umass0: at uhub2 port 1 (addr 2) disconnected
ehci_idone: ex=0xc00e5800 is done!
sd2(umass0:0:0:0): generic HBA error

and now bonnie++ just hangs at genput.

The filesystem on the USB disk is ffsv1 without softdeps.

Should I try a kernel without ehci?

have fun
Michael


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)

iQEVAwUBSE190spnzkX8Yg2nAQKZLwgAo88iRMPKJvGOUaNfQ9po8K60cOC0v2K0
FeZPT3pYOA39ycUFF9GefwdGTmJvs36zrEQ7SWL9fqujiitNeSRN82YUfSanUcgD
geYSEazIsdQSUxEl4g6s2eM3edcjJmaUI9qpVKto+KkdQYnk5ePvBq1nKC4jhLGi
/W8Nw1CWi+sWsjBdPTmePtTHor4ONqcZMgWvQdKk5wh6qBp9FyHlVvcfOaJpuIya
ELAVg9ocGpn8ZWeNBDayzSH/egy7kuSaXvpOsLUhtDX6lX0sZKUpoJoWZ+W2frei
/3XEE6q6Pm43LCY/OieQEsthZ1EA07w2Faa4zjwxnvbVVcsZ9v9ppA==
=sUJV
-----END PGP SIGNATURE-----