I've been doing some USB benchmarking on a PS3 running 2.6.16
(Fedora Core 5).
I'm measuring bulk throughput to/from the Universal Software Radio
Peripheral (USRP) (http://www.ettus.com) used with the GNU Radio
software radio package.
On pretty much every contemporary machine we are able to susta
On Sat, Dec 30, 2006 at 10:43:43AM +0100, Ingo Molnar wrote:
> > > The bad news is, the USB keyboard is still not functioning, but the
> > > mice plugged in the keyboard hub are working.
>
> ok, but this likely a separate bug, correct? Do the USB people know
> about it - and/or is there any othe
On Fri, 29 Dec 2006, Marcelo Tosatti wrote:
> How about this, then:
>
>
> diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c
> index 025d333..5c90bde 100644
> --- a/drivers/usb/host/ehci-hcd.c
> +++ b/drivers/usb/host/ehci-hcd.c
> @@ -496,6 +496,9 @@ static int ehci_run (stru
Am Samstag, 30. Dezember 2006 17:14 schrieb Florin Iucha:
> I suspect it has something to do with sysfs or usbfs. This USB
> keyboard hang has not happened before (and I have the same hardware in
> this box for over a year, and I compile and test most of the rcX
> kernels). My workloads are CPU
On Sat, Dec 30, 2006 at 10:34:01PM +0100, Oliver Neukum wrote:
> > I suspect it has something to do with sysfs or usbfs. This USB
> > keyboard hang has not happened before (and I have the same hardware in
> > this box for over a year, and I compile and test most of the rcX
> > kernels). My worklo
On Sat, 30 Dec 2006, Florin Iucha wrote:
> > >After copying over 60 GB of data over NFS4, the USB keyboard becomes
> > >unresponsive, but the mice connected to the hub built into the
> > >keyboard are functioning normally. The kernel is 2.6.20-rc2 with a
> > >patch from Ingo Mol
On Thu, 21 Dec 2006, matthieu castet wrote:
> >> I tried the device on a computer using 2.6.19 and a NEC EHCI controller
> >> (upd720102), and I got the same problem.
> >>
> >> Seems the bug is not in the usb host hardware.
> >
> > Can you provide more detailed debugging information for the NEC
This patch adds a new, "binary" API in addition to the old, text API usbmon
had before. The new API allows for less CPU use, and it allows to capture
all data from a packet where old API only captured 32 bytes at most. There
are some limitations and conditions to this, e.g. in case someone construc
On Fri, Dec 29, at 10:31AM(-0500), Alan Stern wrote:
> On Thu, 28 Dec 2006, Vermont wrote:
> > To trigger the disconnect, I run
> >
> > dd if=/dev/sda bs=4096 count=276072 | cat | wc -c
>
> Just out of curiosity, why do you include the "cat" in this pipeline? Not
> that it's wrong, but it does