i'm getting
1+0 records in
1+0 records out
65536 bytes transferred in 54.169996 secs (12098210 bytes/sec)
running r191076.
alex
___
freebsd-usb@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-usb
To unsubscribe, s
> FreeBSD-8-Current
>
> dd if=/dev/da0 of=/dev/null bs=64k count=1
> 1+0 records in
> 1+0 records out
> 65536 bytes transferred in 60.320972 secs (10864546 bytes/sec)
>
Fantastic!!! Today result:
dd if=/dev/da0 of=/dev/null bs=64k count=1
1+0 records in
1+0 records o
On Wednesday 04 March 2009, Steve Calfee wrote:
> On Wed, Mar 4, 2009 at 12:22 AM, Hans Petter Selasky
wrote:
> > Hi Steve,
> >
> > On Tuesday 03 March 2009, Steve Calfee wrote:
> >> > I think the reduced performance can be explained by a clamp on the
> >> > interrupt rate around 1000 interrupts
On Thursday 05 March 2009, Wilkinson, Alex wrote:
> Excuse my ignorance, but what is meant by "doorbell" ?
>
Doorbell is an EHCI callback interrupt when the Async queue has been checked
for removed QH's.
--HPS
___
freebsd-usb@freebsd.org mailing list
h
Excuse my ignorance, but what is meant by "doorbell" ?
-aW
0n Thu, Mar 05, 2009 at 12:09:47AM +0100, Hans Petter Selasky wrote:
>The new USB stack _is_ doing things faster than the old one. I have gone
>through a large range of tests before I landed on the doorbell trick.
>Ac
On Wednesday 04 March 2009, Steve Calfee wrote:
> On Wed, Mar 4, 2009 at 2:10 PM, Hans Petter Selasky
wrote:
> > On Wednesday 04 March 2009, Artyom Mirgorodsky wrote:
> >> I forgot to write, that a similar problem was observed in FreeBSD 7 with
> >> usb4bsd patches.
> >
> > Here is a patch which
> Here is a patch which I think will address your problem. It is EHCI hardware
> related. Different models behave differently. Try this:
>
> http://perforce.freebsd.org/chv.cgi?CH=158692
Much better:
dd if=/dev/da0 of=/dev/null bs=64k count=5000
5000+0 records in
5000+0 records out
32768 by
On Wed, Mar 4, 2009 at 2:10 PM, Hans Petter Selasky wrote:
> On Wednesday 04 March 2009, Artyom Mirgorodsky wrote:
>> I forgot to write, that a similar problem was observed in FreeBSD 7 with
>> usb4bsd patches.
>
> Here is a patch which I think will address your problem. It is EHCI hardware
> rela
On Wed, Mar 4, 2009 at 12:22 AM, Hans Petter Selasky wrote:
> Hi Steve,
>
> On Tuesday 03 March 2009, Steve Calfee wrote:
>> > I think the reduced performance can be explained by a clamp on the
>> > interrupt rate around 1000 interrupts per second instead of 8000. Maybe
>> > someone has an explana
On Wednesday 04 March 2009, Artyom Mirgorodsky wrote:
> I forgot to write, that a similar problem was observed in FreeBSD 7 with
> usb4bsd patches.
Here is a patch which I think will address your problem. It is EHCI hardware
related. Different models behave differently. Try this:
http://perforce
In message: <200903042031.37578.hsela...@c2i.net>
Hans Petter Selasky writes:
: On Wednesday 04 March 2009, M. Warner Losh wrote:
: > : >
: > : > If you do change to filters then this is much easier with taskqueues as
: > : > it has a fast variant, otherwise you would need an intermedi
On Wednesday 04 March 2009, M. Warner Losh wrote:
> : >
> : > If you do change to filters then this is much easier with taskqueues as
> : > it has a fast variant, otherwise you would need an intermediate step in
> : > order to signal the existing usb threading scheme. The taskqueue
> : > changeover
In message: <200903041910.58446.hsela...@c2i.net>
Hans Petter Selasky writes:
: On Wednesday 04 March 2009, Andrew Thompson wrote:
: > On Wed, Mar 04, 2009 at 10:01:36AM +0100, Hans Petter Selasky wrote:
: > > On Wednesday 04 March 2009, M. Warner Losh wrote:
: > > > In message: <20090
On Wednesday 04 March 2009, Andrew Thompson wrote:
> On Wed, Mar 04, 2009 at 10:01:36AM +0100, Hans Petter Selasky wrote:
> > On Wednesday 04 March 2009, M. Warner Losh wrote:
> > > In message: <200903040922.48163.hsela...@c2i.net>
> > >
> > > : > I am looking at using FreeBSD in an embedded produc
On Wed, Mar 04, 2009 at 10:01:36AM +0100, Hans Petter Selasky wrote:
> On Wednesday 04 March 2009, M. Warner Losh wrote:
> > In message: <200903040922.48163.hsela...@c2i.net>
> >
> > : > I am looking at using FreeBSD in an embedded product. I have not
> > : > examined your ehci software, but I am a
In message: <200903041001.37376.hsela...@c2i.net>
Hans Petter Selasky writes:
: On Wednesday 04 March 2009, M. Warner Losh wrote:
: > In message: <200903040922.48163.hsela...@c2i.net>
: >
: > Hans Petter Selasky writes:
: > : Hi Steve,
: > :
: > : On Tuesday 03 March 2009,
On Wednesday 04 March 2009, M. Warner Losh wrote:
> In message: <200903040922.48163.hsela...@c2i.net>
>
> Hans Petter Selasky writes:
> : Hi Steve,
> :
> : On Tuesday 03 March 2009, Steve Calfee wrote:
> : > > I think the reduced performance can be explained by a clamp on the
> : > > i
In message: <200903040922.48163.hsela...@c2i.net>
Hans Petter Selasky writes:
: Hi Steve,
:
: On Tuesday 03 March 2009, Steve Calfee wrote:
: > > I think the reduced performance can be explained by a clamp on the
: > > interrupt rate around 1000 interrupts per second instead of 8000.
I forgot to write, that a similar problem was observed in FreeBSD 7 with
usb4bsd patches.
___
freebsd-usb@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-usb
To unsubscribe, send any mail to "freebsd-usb-unsubscr...@freebsd.org
Hi,
Do we have any knobs in FreeBSD to reduce the interrupt latency? I am
currently seeing performance differences between 8.x and 7.x. Anyone have any
ideas?
--HPS
On Wednesday 04 March 2009, Artyom Mirgorodsky wrote:
> >Repeat the same test using FreeBSD -current.
> >
> >a) On the machine wh
>Repeat the same test using FreeBSD -current.
>a) On the machine where it is slow.
vmstat -i ; sleep 1 ; vmstat -i
interrupt total rate
irq1: atkbd0 233 2
irq14: ata0 85 0
irq16: vgapci0
Hi Steve,
On Tuesday 03 March 2009, Steve Calfee wrote:
> > I think the reduced performance can be explained by a clamp on the
> > interrupt rate around 1000 interrupts per second instead of 8000. Maybe
> > someone has an explanation for this?
> >
> > The EHCI is being programmed to interrupt at 1
> I think the reduced performance can be explained by a clamp on the interrupt
> rate around 1000 interrupts per second instead of 8000. Maybe someone has an
> explanation for this?
>
> The EHCI is being programmed to interrupt at 125us intervals, but there seems
> to be limits other places.
>
> It
On Tuesday 03 March 2009, Artyom Mirgorodsky wrote:
> On Tuesday 03 March 2009 16:16:02 Hans Petter Selasky wrote:
> > if (usb2_get_speed(xfer->xroot->udev) == USB_SPEED_HIGH) {
> > qh_endp |= (EHCI_QH_SET_EPS(EHCI_QH_SPEED_HIGH) |
> > EHCI_QH_DTC);
> >
On Tuesday 03 March 2009 16:16:02 Hans Petter Selasky wrote:
> if (usb2_get_speed(xfer->xroot->udev) == USB_SPEED_HIGH) {
> qh_endp |= (EHCI_QH_SET_EPS(EHCI_QH_SPEED_HIGH) |
> EHCI_QH_DTC);
> if (methods != &ehci_device_intr_methods)
>
On Tuesday 03 March 2009, Artyom Mirgorodsky wrote:
> >I know that certain AMD USB controllers perform less with the new USB
> > stack, but I've not been able to track down the issue.
>
> Yes, I have AMD platform (motherboard ASUS A8N-VM-CSM)
>
> >Can you try another PC having another CPU brand?
>
>I know that certain AMD USB controllers perform less with the new USB stack,
>but I've not been able to track down the issue.
Yes, I have AMD platform (motherboard ASUS A8N-VM-CSM)
>Can you try another PC having another CPU brand?
Yes, I have already posted the result - 22 Mb/sec, much better
On Monday 02 March 2009, Artyom Mirgorodsky wrote:
> > It might be a bug with your memory stick. Are your findings consistent
> > accross other brands of memory sticks aswell, between 7.1 and 8.x ?
>
> Today, I tested another flash drive
>
> ugen1.2: at usbus1
> umass0: on usbus1
> umass0: SCSI
> It might be a bug with your memory stick. Are your findings consistent
> accross
> other brands of memory sticks aswell, between 7.1 and 8.x ?
Today, I tested another flash drive
ugen1.2: at usbus1
umass0: on usbus1
umass0: SCSI over Bulk-Only; quirks = 0x
umass0:3:0:-1: Attached to sc
This flash drive works fine on another system (Intel 6300ESB USB 2.0
controller) with FreeBSD-8-current.
May be problem in controller?
___
freebsd-usb@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-usb
To unsubscribe, send any
On Sunday 01 March 2009, Artyom Mirgorodsky wrote:
> 7.1 livefs test result:
>
> dd if=/dev/da0 of=/dev/null bs=65536
> 19534+0 records in
> 19534+0 records out
> 1280180224 bytes transferred in 64.897932 secs (197263329 bytes/sec)
>
Hi,
It might be a bug with your memory stick. Are your find
7.1 livefs test result:
dd if=/dev/da0 of=/dev/null bs=65536
19534+0 records in
19534+0 records out
1280180224 bytes transferred in 64.897932 secs (197263329 bytes/sec)
systat -vm
Disksda0
KB/t64.00
tps 301
MB/s 18.84
%busy 99
In message: <200903010038.29534.hsela...@c2i.net>
Hans Petter Selasky writes:
: On Saturday 28 February 2009, Artyom Mirgorodsky wrote:
: > Read speed when copy from usb flash drive decreased from 20MB/s (old usb
: > stack) to 12MB/s. (on windows above 22Mb/s).
: >
:
: Hi please expla
>Hi please explain what kind of test you have done. It's most likely not
>related to USB, but rather the file system.
> Maybe you can run the following test on Windows and FreeBSD to compare:
>
> dd if=/dev/da0 of=/dev/null bs=65536
I'm test on FreeBSD7-stable and FreeBSD-current and post result
On Saturday 28 February 2009, Artyom Mirgorodsky wrote:
> Read speed when copy from usb flash drive decreased from 20MB/s (old usb
> stack) to 12MB/s. (on windows above 22Mb/s).
>
Hi please explain what kind of test you have done. It's most likely not
related to USB, but rather the file system.
35 matches
Mail list logo
