Re: [Spice-devel] [spice-gtk PATCH v2] spice-channel: check message queue length

Hey, On Fri, Oct 16, 2015 at 04:02:11PM +0200, Fabiano Fidêncio wrote: > I had a private conversation with Hans (which he agreed to share in > this email) and here is his suggestion: > > 15:51 Hi, ah yes that one, I had it on my radar at one point... > 15:51 Overall I think the idea is good,

Re: [Spice-devel] [PATCH xf86-video-qxl 4/4] Document a number of missing XSpice vdagent options in the example xorg.conf.

On Fri, Oct 16, 2015 at 04:07:43PM -0500, Jeremy White wrote: > On 10/16/2015 11:04 AM, Christophe Fergeau wrote: > > On Fri, Oct 02, 2015 at 11:21:03AM -0500, Jeremy White wrote: > >> Signed-off-by: Jeremy White --- > >> examples/spiceqxl.xorg.conf.example | 24 > >>

Re: [Spice-devel] [PATCH xf86-video-qxl 2/4] Correct the XSpice AgentMouse configuration options.

On Mon, Oct 19, 2015 at 01:56:07PM +0200, Christophe Fergeau wrote: > On Fri, Oct 02, 2015 at 11:21:01AM -0500, Jeremy White wrote: > > Since the introduction of vdagent support in commit > > 294daff7ea930f338ec6d77ec3465b6f727c168e, the > > SpiceAgentMouse setting has not really operated

Re: [Spice-devel] [PATCH 05/10] syntax-check: Don't use tabs for indentation

On Thu, Oct 08, 2015 at 07:48:03AM +0200, Victor Toso wrote: > Would make sense to remove other tabs as well? > > > grep -nrIP '\t' server/ --exclude=Makefile* > server/mjpeg_encoder.c:277: unsigned char ** outbuffer;/* target > buffer */ > server/mjpeg_encoder.c:279: uint8_t *

Re: [Spice-devel] [usbredir PATCH v3] usbredir memory leak

p.s. One more thing, when I reproduced this myself I noticed a couple of other worrisome things: 1) When I stop the webcam-app inside the guest, the camera does not stop streaming on the client side. As if the control request being send from the client to stop the stream never reaches the

Re: [Spice-devel] [PATCH xf86-video-qxl 2/4] Correct the XSpice AgentMouse configuration options.

On Fri, Oct 02, 2015 at 11:21:01AM -0500, Jeremy White wrote: > Since the introduction of vdagent support in commit > 294daff7ea930f338ec6d77ec3465b6f727c168e, the > SpiceAgentMouse setting has not really operated correctly. > > That is, if SpiceVdagentEnabled was true, the value > of

Re: [Spice-devel] [usbredir PATCH v3] usbredir memory leak

Hi, On 19-10-15 14:25, Fabiano Fidêncio wrote: Adding Hans ... Thanks, I missed the original patch, Victor, can you please Cc me when sending this patches, and perhaps resend v3 with me in the Cc ? On Mon, Oct 19, 2015 at 12:07 PM, Victor Toso wrote: This is already

Re: [Spice-devel] [usbredir PATCH v3] usbredir memory leak

Hi, On Mon, Oct 19, 2015 at 02:35:02PM +0200, Hans de Goede wrote: > Hi, > > On 19-10-15 14:25, Fabiano Fidêncio wrote: > >Adding Hans ... > > Thanks, I missed the original patch, Victor, can you please > Cc me when sending this patches, and perhaps resend v3 with > me in the Cc ? > Sure, I'll

Re: [Spice-devel] [PATCH xf86-video-qxl v2 (really)] Bug fix: libcacard headers and libraries were not correctly used.

ACK. On Fri, Oct 16, 2015 at 03:47:14PM -0500, Jeremy White wrote: > Also shift to libspiceccid_la_CFLAGS. > > Signed-off-by: Jeremy White > --- > src/spiceccid/Makefile.am | 5 ++--- > src/spiceccid/spiceccid.c | 2 +- > 2 files changed, 3 insertions(+), 4 deletions(-)

Re: [Spice-devel] [PATCH xf86-video-qxl 3/4] Reopen the uinput socket after a close.

On Fri, Oct 16, 2015 at 04:06:28PM -0500, Jeremy White wrote: > >> > > >> static void spiceqxl_uinput_read_cb(int fd, int event, void > >> *opaque) { @@ -30,14 +31,20 @@ static void > >> spiceqxl_uinput_read_cb(int fd, int event, void *opaque) static > >> int y = -1; static int buttons_state =

Re: [Spice-devel] [PATCH] fix spelling mistakes in comments (reseting to resetting & dummym to dummy)

Acked! - Original Message - > From: "snir sheriber" > To: spice-devel@lists.freedesktop.org > Cc: "snir sheriber" > Sent: Sunday, October 18, 2015 8:51:56 AM > Subject: [Spice-devel] [PATCH] fix spelling mistakes in comments (reseting > to

Re: [Spice-devel] [usbredir PATCH v3] usbredir memory leak

Adding Hans ... On Mon, Oct 19, 2015 at 12:07 PM, Victor Toso wrote: > This is already better then v2 due the fact that we are only dropping > isochronous packets. Still need to find a way to drop the correct > amount of packages so usbredir always start sending data from

Re: [Spice-devel] [PATCH] fix spelling mistakes in comments (reseting to resetting & dummym to dummy)

Hey, This has been pushed already, but 2 cosmetic comments: - you might want to capitalize your first/last name (Snir Sheriber rather than snir sheriber) - it's better to keep the short log (first line of the commit log) short (60 characters or so) as this makes things more readable in

[Spice-devel] [PATCH] fix spelling mistakes in comments (reseting to resetting & dummym to dummy)

--- server/glz_encoder_dictionary.c | 2 +- server/red_channel.c| 4 ++-- server/red_worker.c | 2 +- server/reds.c | 2 +- 4 files changed, 5 insertions(+), 5 deletions(-) diff --git a/server/glz_encoder_dictionary.c b/server/glz_encoder_dictionary.c

[Spice-devel] [usbredir PATCH v3] usbredir memory leak

This is already better then v2 due the fact that we are only dropping isochronous packets. Still need to find a way to drop the correct amount of packages so usbredir always start sending data from start of payload (not in the middle of a frame, for instance) cheers, Victor Toso (1):

[Spice-devel] [spice-gtk PATCH v3] spice-channel: check message queue size

When channel wants to send much more data then the wire can handle, the queue grows fast. This patch does not limit the queue growth but introduces an internal API to check if queue size is too big. This internal API is used with usbredirhost_can_write_iso callback from usbredir. This way, we

[Spice-devel] [usbredir PATCH v3] usbredirhost: new callback for iso streams

For streaming devices it might be necessary from application to drop data for different reasons. This patch provides a new callback that it is called before queueing the most recent iso packages. Related: https://bugzilla.redhat.com/show_bug.cgi?id=1264156 --- usbredirhost/usbredirhost.c | 19

Re: [Spice-devel] [usbredir PATCH v3] usbredir memory leak

Hi, On Mon, Oct 19, 2015 at 02:54:41PM +0200, Hans de Goede wrote: > p.s. > > One more thing, when I reproduced this myself I noticed a > couple of other worrisome things: > > 1) When I stop the webcam-app inside the guest, the camera does > not stop streaming on the client side. As if the

[Spice-devel] [PATCH 2/2] drm/qxl: allocate objects in both video rams

If memory is not enough in the default BAR for a type try other BAR this allow better memory usage and avoid memory allocation failure if a BAR is quite small and other is quite unused. Signed-off-by: Frediano Ziglio --- drivers/gpu/drm/qxl/qxl_object.c | 11 +++ 1

[Spice-devel] [PATCH 1/2] drm/qxl: change the way slot is detected

Instead of relaying on surface type use the actual placement. This allow to have different placement for a single type of surface. Signed-off-by: Frediano Ziglio --- drivers/gpu/drm/qxl/qxl_cmd.c | 2 +- drivers/gpu/drm/qxl/qxl_drv.h | 9 - 2 files changed, 9

[Spice-devel] [PATCH 0/2] Smartly allocate memory on the two BARs

Currently a single type of surface is allocated in a specific BAR. This also changes from userspace driver to the kernel one. This way it could happen that allocation are failing even if there are plenty of space in the other BAR. For instance this can happen trying to change resolution as the old

Re: [Spice-devel] [PATCH spice-vdagent] service: simplification and accessability - reversion

For testing purposes: http://goo.gl/Gm4ffO spice/ spice-vdagent-0.16.0-5.fc24.src.rpm spice-vdagent-0.16.0-5.fc24.x86_64.rpm ___ Spice-devel mailing list Spice-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-vdagent] service: simplification and accessability - reversion

While testing, I came to the same conclusion as Marc-André - commit 1587063. Hans, why the complexity of the "rules & target", isn't "ConditionPath*" sufficient? Besides, the lack of "WantedBy=multi-user.target" - back and forth with 'graphical.target' - 'spice-vdagentd' is "lost". Obviously

Re: [Spice-devel] [PATCH spice-vdagent] service: simplification and accessability - reversion

Hi, On 19-10-15 16:55, poma wrote: While testing, I came to the same conclusion as Marc-André - commit 1587063. Hans, why the complexity of the "rules & target", isn't "ConditionPath*" sufficient? Besides, the lack of "WantedBy=multi-user.target" - back and forth with 'graphical.target' -

[Spice-devel] [PATCH xf86-video-qxl 4/4 v2] Document a number of missing XSpice vdagent options in the example xorg.conf.

Also revise the expression of the default value to be consistent. Signed-off-by: Jeremy White --- examples/spiceqxl.xorg.conf.example | 61 ++--- 1 file changed, 43 insertions(+), 18 deletions(-) diff --git

[Spice-devel] [PATCH xf86-video-qxl 2/4 v2] Only watch the uinput channel when an agent is connected.

Otherwise, you constantly spin loop, getting a return code of 0, and pegging CPU usage at 100%. Signed-off-by: Jeremy White --- src/spiceqxl_uinput.c | 13 - src/spiceqxl_uinput.h | 1 + src/spiceqxl_vdagent.c | 2 ++ 3 files changed, 15 insertions(+), 1

[Spice-devel] [PATCH xf86-video-qxl 1/4 v2] Add and remove the vdagent interface as the vdagent actually connects and disconnects.

This means that we do not attempt to use agent mouse mode if there is no connected agent. It fixes a bug which would occur if an agent disconnected after session startup which would result in having no useful mouse. Signed-off-by: Jeremy White --- src/spiceqxl_vdagent.c

[Spice-devel] [PATCH xf86-video-qxl 0/4 v2] Modest improvements to Xspice vdagent support

This patch series is a set of minor patches for XSpice around vdagent support. It improves the documentation of the options, corrects an option that did not work as expected, and handles an agent disconnection more intelligently. The code changes in this second version are substantially

[Spice-devel] [PATCH xf86-video-qxl 3/4 v2] Correct the XSpice AgentMouse configuration options.

Since the introduction of vdagent support in commit 294daff7ea930f338ec6d77ec3465b6f727c168e, the SpiceAgentMouse setting has not really operated correctly. That is, if SpiceVdagentEnabled was true, the value of SpiceAgentMouse was overridden. This patch corrects that. It also shifts the

Re: [Spice-devel] [PATCH xf86-video-qxl 2/4] Correct the XSpice AgentMouse configuration options.

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 > The code changes correspond to what this log describes, but why is > it better to defer enabling the agent to on_connect()? I've done more digging, and I think I missed the larger flaw on this issue. That is, rather than enabling/disabling the