qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Erik Rull <erik.rull@rdsoftware.de>
To: Reeted <reeted@shiftmail.org>
Cc: "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] USB 2.0 printer passthrough very slow
Date: Tue, 03 Apr 2012 17:29:57 +0200	[thread overview]
Message-ID: <4F7B1775.6050600@rdsoftware.de> (raw)
In-Reply-To: <4F7AFFFE.4020805@shiftmail.org>

Hi,

Reeted wrote:
> On 04/03/12 15:33, Erik Rull wrote:
>> Hi,
>>
>> please try to use the .cfg file from the docs/ directory,
>
> The ich9-ehci-uhci.cfg ? Yes it's the same thing I am using.
> If you look at the <qemu:commandline> section I have replicated that file
> exactly.
> I had access rights problem using the .cfg file, that's why I replicated it
> in the commandline, but it's the same thing really.

You're right.

>> using this file, the USB printer speed is really good on my systems. Well
>> it is still an emulation so you'll never get native speed, but it is far
>> better than the USB 1.x emulation we had in 0.1x versions.
>> Best thing to test the approx. transfer speed: Plug in a fast usb key,
>> copy some larger data chunks and measure the time - on the same port
>> where you attach the printer.
>
> I had tested that on a linux guest and it was 6.5MB/sec (dd streaming of
> the device! Not file copy) as I said, but you are right, it's better if I
> try it on the real windows guest. I won't have a chance to test it until
> the weekened though.
>
>>
>> Printing e.g. the Windows XP test page takes not really more time than on
>> a native system.
>
> And if you send a second job it follows the first one right away or it
> takes minutes in between?

Hm, I didn't have an eye on that, but I don't see any reasons why the USB 
speed should slow down the Windows Printing Queue.

> What printer?

Tested with an EPSON USB 2.0 and a HP with USB 1.1.

> What qemu version?

I'm not sure which version is the oldest where it worked, but the current 
git master seems to be fine...
If you have no chance to get direct GIT access you can download the master 
from the git web interface as .tgz.

> Any chance you could post here your commandline? (from ps aux )

Currently not - the system is not available this week - sorry.

> What's the guest setting in device management, is it ACPI Uniprocessor
> http://www.neowin.net/forum/uploads/post-9158-1134631266.png
> or ACPI Multiprocessor or Standard PC or other?
> Mine is ACPI Multiprocessor with 2vCPU. Other settings were almost twice
> slower in booting the system.

Standard PC - I don't think that this is related to your issue.
Best thing would be to test it on a native system.
Well - Speeding up the system with a second core should increase the speed :-)

> (Sorry for the many questions :-) )
>
> Thanks for your help
> R.
>

Best regards,

Erik

  reply	other threads:[~2012-04-03 15:30 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-04-03 12:26 [Qemu-devel] USB 2.0 printer passthrough very slow Reeted
2012-04-03 13:33 ` Erik Rull
2012-04-03 13:49   ` Reeted
2012-04-03 15:29     ` Erik Rull [this message]
2012-04-03 13:52 ` Reeted

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4F7B1775.6050600@rdsoftware.de \
    --to=erik.rull@rdsoftware.de \
    --cc=qemu-devel@nongnu.org \
    --cc=reeted@shiftmail.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).