Thanks, now I can use libusb_handle_events().
What kind of situation the code will run into
int usbredirparser_do_write(struct usbredirparser *parser_pub)
{
...
if ((parser->flags & usbredirparser_fl_write_cb_owns_buffer) &&
w != wbuf->len)
abort();
...
}
_
Thanks, now I can use libusb_handle_events().
What kind of situation the code will run into
int usbredirparser_do_write(struct usbredirparser *parser_pub)
{
...
if ((parser->flags & usbredirparser_fl_write_cb_owns_buffer) &&
w != wbuf->len)
abort();
...
}
__
On Thu, Dec 20, 2012 at 09:18:17PM +0100, Hans de Goede wrote:
> Hi All,
>
> I'm happy to announce a new bugfix (+ a small feature) release
> in the stable spice-0.12.x series.
>
> Changes in spice-0.12.2:
>
> * Skipped 0.12.1, it existed in git but was never released
> *
On Wed, Dec 19, 2012 at 01:05:04PM +0800, Rozen Lin wrote:
> Hi,All,
> Currently I'm trying the multi-client feature of spice, the server
> side is spice-server-0.10 and qemu-kvm-1.1, if I use only from Linux
> machine the spicec-0.10 or spice-gtk-0.10 as clients, the multi-clients
> works fine,
The commit fcbbc248a8f885f9a9a6e7c47d7aae0c1ab3cd1b changed the way a
channel coroutine is exiting. In particular, it was going through the
coroutine main cleanup (finishing in main coroutine) while switching
to TLS is recycling the channel. That part of the code is a bit
difficult to grasp, but wi
An explicit yield back to the channel coroutine when the idle function
is still pending will leave it in the background, referencing objects
that may no longer exist. Make sure we remove it when
channel_open_host() is resumed.
---
gtk/spice-session.c | 10 --
1 file changed, 4 insertions(+
Hey,
I've been trying to connect to oVirt instances using spice-gtk 0.15, and
this patch has been causing me pain (ie connection failures that go away
when I switch back to 0.14).
On Fri, Nov 30, 2012 at 09:12:03PM +0100, Marc-André Lureau wrote:
> diff --git a/gtk/spice-channel.c b/gtk/spice-cha
On 12/21/2012 12:28 PM, Christophe Fergeau wrote:
Aliasing _ftime_s to _ftime when OLDMSVCRT is defined makes it
possible to remove the timestamp-less implementation of LOG()
while still getting a binary working on Windows XP. I've tested
this with a MinGW build, hopefully this won't break VC++ b
On 12/21/2012 12:28 PM, Christophe Fergeau wrote:
vdlog.cpp has an #ifdef check for OLDMSVC to choose between using
swprintf or swprintf_s. Since vdcommon.h provides a compat #define
for swprintf_s when OLDMSVC is check, it's better to use this one.
---
common/vdlog.cpp | 5 +
1 file chang
What opportunities exist within the protocol to enhance the experience
through hardware acceleration? MJPEG seems an obvious place to start. What
about compression or acceleration of client commands?
-j
___
Spice-devel mailing list
Spice-devel@lists.fre
10 matches
Mail list logo
