qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Hans de Goede <hdegoede@redhat.com>
To: Jan Kiszka <jan.kiszka@web.de>
Cc: qemu-devel <qemu-devel@nongnu.org>, Gerd Hoffmann <kraxel@redhat.com>
Subject: Re: [Qemu-devel] [PATCH] usb-redir: Allow to attach USB 2.0 devices to 1.1 host controller
Date: Mon, 17 Sep 2012 11:08:06 +0200	[thread overview]
Message-ID: <5056E876.9010400@redhat.com> (raw)
In-Reply-To: <5054AC57.205@web.de>

Hi,

On 09/15/2012 06:27 PM, Jan Kiszka wrote:
> From: Jan Kiszka <jan.kiszka@siemens.com>
>
> This follows the logic of host-linux: If a 2.0 device has no ISO
> endpoint and no interrupt endpoint with a packet size > 64, we can
> attach it also to an 1.1 host controller. In case the redir server does
> not report endpoint sizes, play safe and remove the 1.1 compatibility as
> well.
>
> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>

Interesting thanks for the patch. I like the approach you took (simple code),
but unfortunately it won't work, if you look at usbredir_device_connect(),
where you do the dev->dev.speedmask |= USB_SPEED_MASK_FULL, it also
actually attaches the device to the controller, from which point on the
guest will see the device.

What happens on the wire is that the usbredir-host sends (in order):
-ep_info + interface_info
-device_connect message

So your clearing of the speed-mask will never trigger, unless ep-info
gets repeated later (which it does under certain circumstances).

I suggest instead changing the code to set a "fullspeed_compatible" flag
in struct USBRedirDevice from usbredir_device_disconnect(), clear that
flag from usbredir_ep_info and use it to add to the mask in
usbredir_device_connect().

###

Another issue is what happens if a device "grows" incompatible endpoints
after being attached, ie a webcam could have no isoc endpoints in alt
setting 0, and then grow an isoc endpoint on a set_interface. So we
would need a check for a device becoming not fullspeed compat while
being attached at fullspeed in usbredir_ep_info(), and then call
usbredir_reject_device() when this happens.

Although not pretty I'm ok with this, since I actually want to add
similar code to allow usb-3 (superspeed) devices like a usb-3 usb-stick
to work with ehci or uhci controllers :)

Regards,

Hans





> ---
>   hw/usb/redirect.c |   10 ++++++++++
>   1 files changed, 10 insertions(+), 0 deletions(-)
>
> diff --git a/hw/usb/redirect.c b/hw/usb/redirect.c
> index 5301a69..bc36e53 100644
> --- a/hw/usb/redirect.c
> +++ b/hw/usb/redirect.c
> @@ -1098,6 +1098,9 @@ static void usbredir_device_connect(void *priv,
>       }
>
>       dev->dev.speedmask = (1 << dev->dev.speed);
> +    if (dev->dev.speed == USB_SPEED_HIGH) {
> +        dev->dev.speedmask |= USB_SPEED_MASK_FULL;
> +    }
>       dev->device_info = *device_connect;
>
>       if (usbredir_check_filter(dev)) {
> @@ -1172,7 +1175,14 @@ static void usbredir_ep_info(void *priv,
>           case usb_redir_type_invalid:
>               break;
>           case usb_redir_type_iso:
> +            dev->dev.speedmask &= ~USB_SPEED_MASK_FULL;
> +            /* Fall through */
>           case usb_redir_type_interrupt:
> +            if (!usbredirparser_peer_has_cap(dev->parser,
> +                                     usb_redir_cap_ep_info_max_packet_size) ||
> +                ep_info->max_packet_size[i] > 64) {
> +                dev->dev.speedmask &= ~USB_SPEED_MASK_FULL;
> +            }
>               if (dev->endpoint[i].interval == 0) {
>                   ERROR("Received 0 interval for isoc or irq endpoint\n");
>                   usbredir_device_disconnect(dev);
>

  reply	other threads:[~2012-09-17  9:07 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-15 16:27 [Qemu-devel] [PATCH] usb-redir: Allow to attach USB 2.0 devices to 1.1 host controller Jan Kiszka
2012-09-17  9:08 ` Hans de Goede [this message]
2012-09-17  9:18   ` Jan Kiszka
2012-09-17 14:24     ` Hans de Goede
2012-09-17 16:22       ` Jan Kiszka
2012-09-18  9:41         ` Hans de Goede
2012-09-21 11:49           ` Jan Kiszka
2012-09-21 12:21             ` Hans de Goede
2012-09-21 12:25               ` Jan Kiszka
2012-09-18 21:18       ` Anthony Liguori
2012-09-19  9:20         ` Hans de Goede
2012-09-22  9:29   ` [Qemu-devel] [PATCH v2] " Jan Kiszka
2012-10-08 17:36     ` Jan Kiszka
2012-10-08 23:05     ` [Qemu-devel] [v2] " Hans de Goede
2012-10-09  8:38       ` Hans de Goede
2012-10-09  8:40         ` Gerd Hoffmann
2012-10-09 10:05         ` Hans de Goede

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=5056E876.9010400@redhat.com \
    --to=hdegoede@redhat.com \
    --cc=jan.kiszka@web.de \
    --cc=kraxel@redhat.com \
    --cc=qemu-devel@nongnu.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).