Re: [linux-usb-devel] done_head LSB broken in SA1111 ohci implementation

2005-04-14 Thread Jaroslav Flidr
On Apr 14, 2005, at 12:52 PM, David Brownell wrote: On Wednesday 13 April 2005 1:30 pm, Jaroslav Flidr wrote: Well, here is the patch I am using. Quite simple, really... (2.6.11 kernel): Yes, thanks. It got line wrapped, but I can cope. Just curious: with the exception of PXA27x and possibly iPaq

Re: [linux-usb-devel] done_head LSB broken in SA1111 ohci implementation

2005-04-14 Thread David Brownell
On Wednesday 13 April 2005 1:30 pm, Jaroslav Flidr wrote: > Well, here is the patch I am using. Quite simple, really... (2.6.11 > kernel): Yes, thanks. It got line wrapped, but I can cope. > Just curious: with the exception of PXA27x and possibly iPaq who else > is using ohci? On PCs, names

Re: [linux-usb-devel] done_head LSB broken in SA1111 ohci implementation

2005-04-13 Thread Jaroslav Flidr
Well, here is the patch I am using. Quite simple, really... (2.6.11 kernel): --- a/drivers/usb/host/ohci-hcd.c 2005-04-13 16:10:34.0 -0400 +++ b/drivers/usb/host/ohci-hcd.c 2005-04-13 16:10:50.0 -0400 @@ -693,13 +693,16 @@ /* we can eliminate a (slow) ohci_r

Re: [linux-usb-devel] done_head LSB broken in SA1111 ohci implementation

2005-04-13 Thread David Brownell
On Tuesday 12 April 2005 3:18 pm, Jaroslav Flidr wrote: > For those who are interested and still using SA-based architecture > (e.g. ziti): the ohci-recommended way of handling the WDH interrupt - > reading the done_head pointer and testing its LSB without reading the > interrupt status regi

[linux-usb-devel] done_head LSB broken in SA1111 ohci implementation

2005-04-13 Thread Jaroslav Flidr
For those who are interested and still using SA-based architecture (e.g. ziti): the ohci-recommended way of handling the WDH interrupt - reading the done_head pointer and testing its LSB without reading the interrupt status register must not be used. It appears that the SA host controll