Re: [Spice-devel] [PATCH spice-gtk] Ensure that file transfers get cancelled

[Pavel Grunt]

On Thu, 2016-04-07 at 15:51 -0500, Jonathon Jongsma wrote:
> When canceling a file transfer task in spicy, the client would often
> stop sending additional data, but it would not send a "CANCELLED"
> message to the guest. Because of this, the partial file would remain
> in
> the guest's downloads folder until the spice client disconnected, at
> which point the vdagent would remove the unfinshed file.
> 
> This CANCELLED status message was only being sent if the task was
> canceled during the async file read operation. If you cancel a task,
> it's quite likely that it will happen during other operations
> (e.g. file_xfer_flush_async(), etc). In order to handle these
> scenarios
> (and make sure that the file gets canceled properly), send the
> FILE_XFER_STATUS message in spice_file_transfer_task_completed().

Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  src/channel-main.c | 17 ++---
>  1 file changed, 10 insertions(+), 7 deletions(-)
> 
> diff --git a/src/channel-main.c b/src/channel-main.c
> index 4a1f58a..85353d4 100644
> --- a/src/channel-main.c
> +++ b/src/channel-main.c
> @@ -1912,13 +1912,6 @@ static void file_xfer_read_cb(GObject
> *source_object,
>    file_xfer_data_flushed_cb, self);
>  self->priv->pending = TRUE;
>  } else if (error) {
> -VDAgentFileXferStatusMessage msg = {
> -.id = self->priv->id,
> -.result = error->code == G_IO_ERROR_CANCELLED ?
> -VD_AGENT_FILE_XFER_STATUS_CANCELLED :
> VD_AGENT_FILE_XFER_STATUS_ERROR,
> -};
> -agent_msg_queue_many(self->priv->channel,
> VD_AGENT_FILE_XFER_STATUS,
> - , sizeof(msg), NULL);
>  spice_channel_wakeup(SPICE_CHANNEL(self->priv->channel),
> FALSE);
>  spice_file_transfer_task_completed(self, error);
>  }
> @@ -2944,6 +2937,16 @@ static void
> spice_file_transfer_task_completed(SpiceFileTransferTask *self,
>  self->priv->error = error;
>  }
>  
> +if (self->priv->error) {
> +VDAgentFileXferStatusMessage msg = {
> +.id = self->priv->id,
> +.result = error->code == G_IO_ERROR_CANCELLED ?
> +VD_AGENT_FILE_XFER_STATUS_CANCELLED :
> VD_AGENT_FILE_XFER_STATUS_ERROR,
> +};
> +agent_msg_queue_many(self->priv->channel,
> VD_AGENT_FILE_XFER_STATUS,
> + , sizeof(msg), NULL);
> +}
> +
>  if (self->priv->pending)
>  return;
>  
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH 06/14] FIXME/HACK: RedPipeItem/VDIPortBuf refcounting hack

[Pavel Grunt]

Hi,

On Thu, 2016-04-07 at 17:11 -0500, Jonathon Jongsma wrote:
> From: Christophe Fergeau
> 
> related to 7d375e0
> workaround a runtime check in red_pipe_item_ref()

Should the refcount be initialized by red_pipe_item_init?

Pavel

> ---
>  server/reds.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/server/reds.c b/server/reds.c
> index 64f7df8..bf39cf7 100644
> --- a/server/reds.c
> +++ b/server/reds.c
> @@ -752,7 +752,8 @@ static VDIReadBuf
> *vdi_port_get_read_buf(RedCharDeviceVDIPort *dev)
>  buf = SPICE_CONTAINEROF(item, VDIReadBuf, link.parent.link);
>  
>  g_warn_if_fail(buf->link.refcount == 0);
> -red_pipe_item_ref(buf);
> +buf->link.refcount = 1;
> +//red_pipe_item_ref(buf);
>  
>  return buf;
>  }
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [vdagent-linux v2] remove trailing spaces

[Pavel Grunt]

Hi,

git grep "[[:space:]]$" shows one more in the ChangeLog

Pavel

On Mon, 2016-04-11 at 11:33 +0200, Victor Toso wrote:
> ---
>  src/console-kit.c| 10 +-
>  src/dummy-session-info.c |  6 +++---
>  src/session-info.h   |  6 +++---
>  src/systemd-login.c  |  6 +++---
>  src/udscs.c  |  8 
>  src/udscs.h  |  8 
>  src/vdagent-file-xfers.c |  6 +++---
>  src/vdagent-file-xfers.h |  6 +++---
>  src/vdagent-virtio-port.c|  8 
>  src/vdagent-virtio-port.h|  8 
>  src/vdagent-x11-randr.c  |  6 +++---
>  src/vdagent-x11.c|  8 
>  src/vdagent-x11.h|  6 +++---
>  src/vdagent.c|  6 +++---
>  src/vdagentd-proto-strings.h |  6 +++---
>  src/vdagentd-proto.h |  6 +++---
>  src/vdagentd-uinput.c|  6 +++---
>  src/vdagentd-uinput.h|  6 +++---
>  src/vdagentd-xorg-conf.h |  6 +++---
>  src/vdagentd.c   |  8 
>  20 files changed, 68 insertions(+), 68 deletions(-)
> 
> diff --git a/src/console-kit.c b/src/console-kit.c
> index 759a81e..6e83bf2 100644
> --- a/src/console-kit.c
> +++ b/src/console-kit.c
> @@ -7,12 +7,12 @@
>  
>  This program is free software: you can redistribute it and/or
> modify
>  it under the terms of the GNU General Public License as
> published by
> -the Free Software Foundation, either version 3 of the License,
> or   
> +the Free Software Foundation, either version 3 of the License,
> or
>  (at your option) any later version.
>  
>  This program is distributed in the hope that it will be useful,
> -but WITHOUT ANY WARRANTY; without even the implied warranty of 
> -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the  
> +but WITHOUT ANY WARRANTY; without even the implied warranty of
> +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>  GNU General Public License for more details.
>  
>  You should have received a copy of the GNU General Public
> License
> @@ -59,7 +59,7 @@ struct session_info *session_info_create(int
> verbose)
>  free(ck);
>  return NULL;
>  }
> -
> +
>  if (!dbus_connection_get_unix_fd(ck->connection, >fd)) {
>  syslog(LOG_ERR, "Unable to get connection fd");
>  session_info_destroy(ck);
> @@ -77,7 +77,7 @@ struct session_info *session_info_create(int
> verbose)
>   "path='%s',member='ActiveSessionChanged'", ck->seat);
>  dbus_error_init();
>  dbus_bus_add_match(ck->connection, match, );
> -if (dbus_error_is_set()) { 
> +if (dbus_error_is_set()) {
>  syslog(LOG_ERR, "Match Error (%s)", error.message);
>  session_info_destroy(ck);
>  return NULL;
> diff --git a/src/dummy-session-info.c b/src/dummy-session-info.c
> index e188ddc..2f0ae65 100644
> --- a/src/dummy-session-info.c
> +++ b/src/dummy-session-info.c
> @@ -7,12 +7,12 @@
>  
>  This program is free software: you can redistribute it and/or
> modify
>  it under the terms of the GNU General Public License as
> published by
> -the Free Software Foundation, either version 3 of the License,
> or   
> +the Free Software Foundation, either version 3 of the License,
> or
>  (at your option) any later version.
>  
>  This program is distributed in the hope that it will be useful,
> -but WITHOUT ANY WARRANTY; without even the implied warranty of 
> -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the  
> +but WITHOUT ANY WARRANTY; without even the implied warranty of
> +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>  GNU General Public License for more details.
>  
>  You should have received a copy of the GNU General Public
> License
> diff --git a/src/session-info.h b/src/session-info.h
> index 67099de..c4f8187 100644
> --- a/src/session-info.h
> +++ b/src/session-info.h
> @@ -7,12 +7,12 @@
>  
>  This program is free software: you can redistribute it and/or
> modify
>  it under the terms of the GNU General Public License as
> published by
> -the Free Software Foundation, either version 3 of the License,
> or   
> +the Free Software Foundation, either version 3 of the License,
> or
>  (at your option) any later version.
>  
>  This program is distributed in the hope that it will be useful,
> -but WITHOUT ANY WARRANTY; without even the implied warranty of 
> -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the  
> +but WITHOUT ANY WARRANTY; without even the implied warranty of
> +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>  GNU General Public License for more details.
>  
>  You should have received a copy of the GNU General Public
> License
> diff --git a/src/systemd-login.c b/src/systemd-login.c
> index 73db37f..9ba300a 100644
> --- a/src/systemd-login.c
> +++ b/src/systemd-login.c
> @@ -7,12 +7,12 @@

Re: [Spice-devel] [vdagent-linux v3] remove trailing spaces

[Pavel Grunt]

On Mon, 2016-04-11 at 11:54 +0200, Victor Toso wrote:

Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  ChangeLog|  2 +-
>  src/console-kit.c| 10 +-
>  src/dummy-session-info.c |  6 +++---
>  src/session-info.h   |  6 +++---
>  src/systemd-login.c  |  6 +++---
>  src/udscs.c  |  8 
>  src/udscs.h  |  8 
>  src/vdagent-file-xfers.c |  6 +++---
>  src/vdagent-file-xfers.h |  6 +++---
>  src/vdagent-virtio-port.c|  8 
>  src/vdagent-virtio-port.h|  8 
>  src/vdagent-x11-randr.c  |  6 +++---
>  src/vdagent-x11.c|  8 
>  src/vdagent-x11.h|  6 +++---
>  src/vdagent.c|  6 +++---
>  src/vdagentd-proto-strings.h |  6 +++---
>  src/vdagentd-proto.h |  6 +++---
>  src/vdagentd-uinput.c|  6 +++---
>  src/vdagentd-uinput.h|  6 +++---
>  src/vdagentd-xorg-conf.h |  6 +++---
>  src/vdagentd.c   |  8 
>  21 files changed, 69 insertions(+), 69 deletions(-)
> 
> diff --git a/ChangeLog b/ChangeLog
> index 4138359..da24341 100644
> --- a/ChangeLog
> +++ b/ChangeLog
> @@ -26,7 +26,7 @@ spice-vdagent-0.12.1
>  
>  spice-vdagent-0.12.0
>  
> -* Full multi-monitor and arbritary resolution support, this requires
> a new   
> +* Full multi-monitor and arbritary resolution support, this requires
> a new
>    enough xorg-x11-drv-qxl driver, as well as a new enough host
>  * systemd service support, using systemd hardware acivation
>  * Use syslog for logging, rather then logging to private log files
> diff --git a/src/console-kit.c b/src/console-kit.c
> index 759a81e..6e83bf2 100644
> --- a/src/console-kit.c
> +++ b/src/console-kit.c
> @@ -7,12 +7,12 @@
>  
>  This program is free software: you can redistribute it and/or
> modify
>  it under the terms of the GNU General Public License as
> published by
> -the Free Software Foundation, either version 3 of the License,
> or   
> +the Free Software Foundation, either version 3 of the License,
> or
>  (at your option) any later version.
>  
>  This program is distributed in the hope that it will be useful,
> -but WITHOUT ANY WARRANTY; without even the implied warranty of 
> -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the  
> +but WITHOUT ANY WARRANTY; without even the implied warranty of
> +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>  GNU General Public License for more details.
>  
>  You should have received a copy of the GNU General Public
> License
> @@ -59,7 +59,7 @@ struct session_info *session_info_create(int
> verbose)
>  free(ck);
>  return NULL;
>  }
> -
> +
>  if (!dbus_connection_get_unix_fd(ck->connection, >fd)) {
>  syslog(LOG_ERR, "Unable to get connection fd");
>  session_info_destroy(ck);
> @@ -77,7 +77,7 @@ struct session_info *session_info_create(int
> verbose)
>   "path='%s',member='ActiveSessionChanged'", ck->seat);
>  dbus_error_init();
>  dbus_bus_add_match(ck->connection, match, );
> -if (dbus_error_is_set()) { 
> +if (dbus_error_is_set()) {
>  syslog(LOG_ERR, "Match Error (%s)", error.message);
>  session_info_destroy(ck);
>  return NULL;
> diff --git a/src/dummy-session-info.c b/src/dummy-session-info.c
> index e188ddc..2f0ae65 100644
> --- a/src/dummy-session-info.c
> +++ b/src/dummy-session-info.c
> @@ -7,12 +7,12 @@
>  
>  This program is free software: you can redistribute it and/or
> modify
>  it under the terms of the GNU General Public License as
> published by
> -the Free Software Foundation, either version 3 of the License,
> or   
> +the Free Software Foundation, either version 3 of the License,
> or
>  (at your option) any later version.
>  
>  This program is distributed in the hope that it will be useful,
> -but WITHOUT ANY WARRANTY; without even the implied warranty of 
> -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the  
> +but WITHOUT ANY WARRANTY; without even the implied warranty of
> +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>  GNU General Public License for more details.
>  
>  You should have received a copy of the GNU General Public
> License
> diff --git a/src/session-info.h b/src/session-info.h
> index 67099de..c4f8187 100644
> --- a/src/session-info.h
> +++ b/src/session-info.h
> @@ -7,12 +7,12 @@
>  
>  This program is free s

[Spice-devel] [PATCH spice-gtk v2 3/6] Use GMutex instead of GStaticMutex

[Pavel Grunt]

Since GLib 2.32 GMutex can be statically allocated, so GStaticMutex has
been deprecated.

Acked-by: Christophe Fergeau 
---
 src/desktop-integration.c |  6 +++---
 src/spice-gtk-session.c   |  6 +++---
 src/spice-session.c   | 12 ++--
 src/usbutil.c |  6 +++---
 4 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/src/desktop-integration.c b/src/desktop-integration.c
index 01300e8..529fb05 100644
--- a/src/desktop-integration.c
+++ b/src/desktop-integration.c
@@ -196,17 +196,17 @@ static void 
spice_desktop_integration_class_init(SpiceDesktopIntegrationClass *k
 SpiceDesktopIntegration *spice_desktop_integration_get(SpiceSession *session)
 {
 SpiceDesktopIntegration *self;
-static GStaticMutex mutex = G_STATIC_MUTEX_INIT;
+static GMutex mutex;
 
 g_return_val_if_fail(session != NULL, NULL);
 
-g_static_mutex_lock();
+g_mutex_lock();
 self = g_object_get_data(G_OBJECT(session), "spice-desktop");
 if (self == NULL) {
 self = g_object_new(SPICE_TYPE_DESKTOP_INTEGRATION, NULL);
 g_object_set_data_full(G_OBJECT(session), "spice-desktop", self, 
g_object_unref);
 }
-g_static_mutex_unlock();
+g_mutex_unlock();
 
 return self;
 }
diff --git a/src/spice-gtk-session.c b/src/spice-gtk-session.c
index 380b0bb..7370599 100644
--- a/src/spice-gtk-session.c
+++ b/src/spice-gtk-session.c
@@ -1133,15 +1133,15 @@ SpiceGtkSession *spice_gtk_session_get(SpiceSession 
*session)
 g_return_val_if_fail(SPICE_IS_SESSION(session), NULL);
 
 SpiceGtkSession *self;
-static GStaticMutex mutex = G_STATIC_MUTEX_INIT;
+static GMutex mutex;
 
-g_static_mutex_lock();
+g_mutex_lock();
 self = g_object_get_data(G_OBJECT(session), "spice-gtk-session");
 if (self == NULL) {
 self = g_object_new(SPICE_TYPE_GTK_SESSION, "session", session, NULL);
 g_object_set_data_full(G_OBJECT(session), "spice-gtk-session", self, 
g_object_unref);
 }
-g_static_mutex_unlock();
+g_mutex_unlock();
 
 return SPICE_GTK_SESSION(self);
 }
diff --git a/src/spice-session.c b/src/spice-session.c
index 6a0edae..9bf3f40 100644
--- a/src/spice-session.c
+++ b/src/spice-session.c
@@ -2641,18 +2641,18 @@ SpiceURI *spice_session_get_proxy_uri(SpiceSession 
*session)
  **/
 SpiceAudio *spice_audio_get(SpiceSession *session, GMainContext *context)
 {
-static GStaticMutex mutex = G_STATIC_MUTEX_INIT;
+static GMutex mutex;
 SpiceAudio *self;
 
 g_return_val_if_fail(SPICE_IS_SESSION(session), NULL);
 
-g_static_mutex_lock();
+g_mutex_lock();
 self = session->priv->audio_manager;
 if (self == NULL) {
 self = spice_audio_new(session, context, NULL);
 session->priv->audio_manager = self;
 }
-g_static_mutex_unlock();
+g_mutex_unlock();
 
 return self;
 }
@@ -2675,19 +2675,19 @@ SpiceUsbDeviceManager 
*spice_usb_device_manager_get(SpiceSession *session,
 GError **err)
 {
 SpiceUsbDeviceManager *self;
-static GStaticMutex mutex = G_STATIC_MUTEX_INIT;
+static GMutex mutex;
 
 g_return_val_if_fail(SPICE_IS_SESSION(session), NULL);
 g_return_val_if_fail(err == NULL || *err == NULL, NULL);
 
-g_static_mutex_lock();
+g_mutex_lock();
 self = session->priv->usb_manager;
 if (self == NULL) {
 self = g_initable_new(SPICE_TYPE_USB_DEVICE_MANAGER, NULL, err,
   "session", session, NULL);
 session->priv->usb_manager = self;
 }
-g_static_mutex_unlock();
+g_mutex_unlock();
 
 return self;
 }
diff --git a/src/usbutil.c b/src/usbutil.c
index 24330b2..6cd8148 100644
--- a/src/usbutil.c
+++ b/src/usbutil.c
@@ -51,7 +51,7 @@ typedef struct _usb_vendor_info {
 char name[VENDOR_NAME_LEN];
 } usb_vendor_info;
 
-static GStaticMutex usbids_load_mutex = G_STATIC_MUTEX_INIT;
+static GMutex usbids_load_mutex;
 static int usbids_vendor_count = 0; /* < 0: failed, 0: empty, > 0: loaded */
 static usb_vendor_info *usbids_vendor_info = NULL;
 
@@ -215,7 +215,7 @@ static gboolean spice_usbutil_load_usbids(void)
 {
 gboolean success = FALSE;
 
-g_static_mutex_lock(_load_mutex);
+g_mutex_lock(_load_mutex);
 if (usbids_vendor_count) {
 success = usbids_vendor_count > 0;
 goto leave;
@@ -242,7 +242,7 @@ static gboolean spice_usbutil_load_usbids(void)
 #endif
 
 leave:
-g_static_mutex_unlock(_load_mutex);
+g_mutex_unlock(_load_mutex);
 return success;
 }
 
-- 
2.8.0

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk v2 4/6] Use g_object_unref instead of gdk_cursor_unref

[Pavel Grunt]

GdkCursor is GObject and gdk_cursor_unref has been deprecated since 3.0

Acked-by: Eduardo Lima (Etrunko) 
---
v2: fix commit message g_object_unref -> gdk_cursor_unref
---
 src/spice-widget.c | 12 ++--
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/src/spice-widget.c b/src/spice-widget.c
index f605439..8ca8631 100644
--- a/src/spice-widget.c
+++ b/src/spice-widget.c
@@ -440,12 +440,12 @@ static void spice_display_finalize(GObject *obj)
 d->activeseq = NULL;
 
 if (d->show_cursor) {
-gdk_cursor_unref(d->show_cursor);
+g_object_unref(d->show_cursor);
 d->show_cursor = NULL;
 }
 
 if (d->mouse_cursor) {
-gdk_cursor_unref(d->mouse_cursor);
+g_object_unref(d->mouse_cursor);
 d->mouse_cursor = NULL;
 }
 
@@ -987,7 +987,7 @@ static GdkGrabStatus do_pointer_grab(SpiceDisplay *display)
 }
 
 end:
-gdk_cursor_unref(blank);
+g_object_unref(blank);
 return status;
 }
 
@@ -2368,7 +2368,7 @@ static void cursor_set(SpiceCursorChannel *channel,
 #endif
 if (d->show_cursor) {
 /* unhide */
-gdk_cursor_unref(d->show_cursor);
+g_object_unref(d->show_cursor);
 d->show_cursor = NULL;
 if (d->mouse_mode == SPICE_MOUSE_MODE_SERVER) {
 /* keep a hidden cursor, will be shown in cursor_move() */
@@ -2377,7 +2377,7 @@ static void cursor_set(SpiceCursorChannel *channel,
 }
 }
 
-gdk_cursor_unref(d->mouse_cursor);
+g_object_unref(d->mouse_cursor);
 d->mouse_cursor = cursor;
 
 update_mouse_pointer(display);
@@ -2493,7 +2493,7 @@ static void cursor_move(SpiceCursorChannel *channel, gint 
x, gint y, gpointer da
 
 /* apparently we have to restore cursor when "cursor_move" */
 if (d->show_cursor != NULL) {
-gdk_cursor_unref(d->mouse_cursor);
+g_object_unref(d->mouse_cursor);
 d->mouse_cursor = d->show_cursor;
 d->show_cursor = NULL;
 update_mouse_pointer(display);
-- 
2.8.0

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk v2 6/6] Use g_clear_pointer if possible

[Pavel Grunt]

---
v2: removed extra curly braces
fix wrong calls
diff to v1 - http://paste.fedoraproject.org/350384/14599492/
---
 src/bio-gio.c   |  3 +--
 src/channel-display-mjpeg.c |  3 +--
 src/channel-display.c   | 10 +++---
 src/channel-main.c  |  6 ++
 src/channel-playback.c  |  3 +--
 src/channel-record.c|  9 +++--
 src/channel-smartcard.c | 28 ++-
 src/channel-usbredir.c  |  9 +++--
 src/decode-glz.c|  3 +--
 src/smartcard-manager.c | 13 +++--
 src/spice-audio.c   |  5 +
 src/spice-channel.c | 26 ++---
 src/spice-gstaudio.c| 32 +--
 src/spice-gtk-session.c |  6 ++
 src/spice-option.c  |  3 +--
 src/spice-pulse.c   | 46 +
 src/spice-session.c | 12 
 src/spice-widget-cairo.c| 11 +++
 src/spice-widget.c  | 16 
 src/usb-acl-helper.c| 11 ++-
 src/wocky-http-proxy.c  |  3 +--
 21 files changed, 70 insertions(+), 188 deletions(-)

diff --git a/src/bio-gio.c b/src/bio-gio.c
index 108ac1a..b310c97 100644
--- a/src/bio-gio.c
+++ b/src/bio-gio.c
@@ -76,8 +76,7 @@ static int bio_gio_destroy(BIO *bio)
 return 0;
 
 SPICE_DEBUG("bio gsocket destroy");
-g_free(bio->method);
-bio->method = NULL;;
+g_clear_pointer(>method, g_free);
 
 return 1;
 }
diff --git a/src/channel-display-mjpeg.c b/src/channel-display-mjpeg.c
index 95d5b33..5ea3371 100644
--- a/src/channel-display-mjpeg.c
+++ b/src/channel-display-mjpeg.c
@@ -151,6 +151,5 @@ G_GNUC_INTERNAL
 void stream_mjpeg_cleanup(display_stream *st)
 {
 jpeg_destroy_decompress(>mjpeg_cinfo);
-g_free(st->out_frame);
-st->out_frame = NULL;
+g_clear_pointer(>out_frame, g_free);
 }
diff --git a/src/channel-display.c b/src/channel-display.c
index 431c2e7..b6c79e0 100644
--- a/src/channel-display.c
+++ b/src/channel-display.c
@@ -831,11 +831,8 @@ static void destroy_canvas(display_surface *surface)
 zlib_decoder_destroy(surface->zlib_decoder);
 jpeg_decoder_destroy(surface->jpeg_decoder);
 
-g_free(surface->data);
-surface->data = NULL;
-
-surface->canvas->ops->destroy(surface->canvas);
-surface->canvas = NULL;
+g_clear_pointer(>data, g_free);
+g_clear_pointer(>canvas, surface->canvas->ops->destroy);
 }
 
 static display_surface *find_surface(SpiceDisplayChannelPrivate *c, guint32 
surface_id)
@@ -1597,8 +1594,7 @@ static void clear_streams(SpiceChannel *channel)
 for (i = 0; i < c->nstreams; i++) {
 destroy_stream(channel, i);
 }
-g_free(c->streams);
-c->streams = NULL;
+g_clear_pointer(>streams, g_free);
 c->nstreams = 0;
 }
 
diff --git a/src/channel-main.c b/src/channel-main.c
index 4a1f58a..955b358 100644
--- a/src/channel-main.c
+++ b/src/channel-main.c
@@ -449,8 +449,7 @@ static void spice_main_channel_reset_agent(SpiceMainChannel 
*channel)
 c->agent_caps_received = FALSE;
 c->agent_display_config_sent = FALSE;
 c->agent_msg_pos = 0;
-g_free(c->agent_msg_data);
-c->agent_msg_data = NULL;
+g_clear_pointer(>agent_msg_data, g_free);
 c->agent_msg_size = 0;
 
 tasks = g_hash_table_get_values(c->file_xfer_tasks);
@@ -916,8 +915,7 @@ static void agent_free_msg_queue(SpiceMainChannel *channel)
 spice_msg_out_unref(out);
 }
 
-g_queue_free(c->agent_msg_queue);
-c->agent_msg_queue = NULL;
+g_clear_pointer(>agent_msg_queue, g_queue_free);
 }
 
 static gboolean flush_foreach_remove(gpointer key G_GNUC_UNUSED,
diff --git a/src/channel-playback.c b/src/channel-playback.c
index c97afd8..e931727 100644
--- a/src/channel-playback.c
+++ b/src/channel-playback.c
@@ -114,8 +114,7 @@ static void spice_playback_channel_finalize(GObject *obj)
 
 snd_codec_destroy(>codec);
 
-g_free(c->volume);
-c->volume = NULL;
+g_clear_pointer(>volume, g_free);
 
 if (G_OBJECT_CLASS(spice_playback_channel_parent_class)->finalize)
 G_OBJECT_CLASS(spice_playback_channel_parent_class)->finalize(obj);
diff --git a/src/channel-record.c b/src/channel-record.c
index 08269c9..09598b7 100644
--- a/src/channel-record.c
+++ b/src/channel-record.c
@@ -108,13 +108,11 @@ static void spice_record_channel_finalize(GObject *obj)
 {
 SpiceRecordChannelPrivate *c = SPICE_RECORD_CHANNEL(obj)->priv;
 
-g_free(c->last_frame);
-c->last_frame = NULL;
+g_clear_pointer(>last_frame, g_free);
 
 snd_codec_destroy(>codec);
 
-g_free(c->volume);
-c->volume = NULL;
+g_clear_pointer(>volume, g_free);
 
 if (G_OBJECT_CLASS(spice_record_channel_parent_class)->finalize)
 G_OBJECT_CLASS(spice_record_channel_parent_class)->finalize(obj);
@@ -166,8 +164,7 @@ static void channel_reset(SpiceChannel *channel, gboolean 
migrating)
 {
 SpiceRecordChannelPrivate *c = 

[Spice-devel] [PATCH spice-gtk v2 0/6] Cleanups for new gtk and glib

[Pavel Grunt]

Hi,

since we bumped the glib and gtk dependencies, it's possible to clean up
the code little bit. Patches silence some deprecated warnings and change
our code to use more g_clear_pointer and g_clear_object.

Thanks,
Pavel

v2 adds commits for checking version of glib and gtk (works when deprecated 
warnings are enabled), 
fixes the commit message in the PATCH 4, fixes wrong usage of g_clear_pointer 
in the PATCH 6

Pavel Grunt (6):
  Use GLIB_VERSION_MAX_ALLOWED
  Use GDK_VERSION_MAX_ALLOWED
  Use GMutex instead of GStaticMutex
  Use g_object_unref instead of gdk_cursor_unref
  Use g_clear_object if possible
  Use g_clear_pointer if possible

 configure.ac| 13 +++--
 src/bio-gio.c   |  3 +--
 src/channel-display-mjpeg.c |  3 +--
 src/channel-display.c   | 10 +++---
 src/channel-main.c  |  6 ++
 src/channel-playback.c  |  3 +--
 src/channel-record.c|  9 +++--
 src/channel-smartcard.c | 28 ++-
 src/channel-usbredir.c  |  9 +++--
 src/decode-glz.c|  3 +--
 src/desktop-integration.c   |  6 +++---
 src/smartcard-manager.c | 13 +++--
 src/spice-audio.c   |  5 +
 src/spice-channel.c | 37 
 src/spice-gstaudio.c| 32 +--
 src/spice-gtk-session.c | 12 +---
 src/spice-option.c  |  3 +--
 src/spice-pulse.c   | 46 +
 src/spice-session.c | 24 ++-
 src/spice-widget-cairo.c| 11 +++
 src/spice-widget.c  | 44 +++
 src/usb-acl-helper.c| 11 ++-
 src/usbutil.c   |  6 +++---
 src/wocky-http-proxy.c  |  6 ++
 24 files changed, 106 insertions(+), 237 deletions(-)

-- 
2.8.0

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk v2 5/6] Use g_clear_object if possible

[Pavel Grunt]

Acked-by: Eduardo Lima (Etrunko) 
---
kept separated per Christophe's suggestion
---
 src/spice-channel.c| 11 ++-
 src/spice-widget.c | 22 --
 src/wocky-http-proxy.c |  3 +--
 3 files changed, 7 insertions(+), 29 deletions(-)

diff --git a/src/spice-channel.c b/src/spice-channel.c
index 8ae0e4d..e9c5a1b 100644
--- a/src/spice-channel.c
+++ b/src/spice-channel.c
@@ -155,10 +155,7 @@ static void spice_channel_dispose(GObject *gobject)
 
 spice_channel_disconnect(channel, SPICE_CHANNEL_CLOSED);
 
-if (c->session) {
- g_object_unref(c->session);
- c->session = NULL;
-}
+g_clear_object(>session);
 
 g_clear_error(>error);
 
@@ -2718,11 +2715,7 @@ static void channel_reset(SpiceChannel *channel, 
gboolean migrating)
 c->ctx = NULL;
 }
 
-if (c->conn) {
-g_object_unref(c->conn);
-c->conn = NULL;
-}
-
+g_clear_object(>conn);
 g_clear_object(>sock);
 
 c->fd = -1;
diff --git a/src/spice-widget.c b/src/spice-widget.c
index 8ca8631..72a0355 100644
--- a/src/spice-widget.c
+++ b/src/spice-widget.c
@@ -439,20 +439,9 @@ static void spice_display_finalize(GObject *obj)
 g_free(d->activeseq);
 d->activeseq = NULL;
 
-if (d->show_cursor) {
-g_object_unref(d->show_cursor);
-d->show_cursor = NULL;
-}
-
-if (d->mouse_cursor) {
-g_object_unref(d->mouse_cursor);
-d->mouse_cursor = NULL;
-}
-
-if (d->mouse_pixbuf) {
-g_object_unref(d->mouse_pixbuf);
-d->mouse_pixbuf = NULL;
-}
+g_clear_object(>show_cursor);
+g_clear_object(>mouse_cursor);
+g_clear_object(>mouse_pixbuf);
 
 G_OBJECT_CLASS(spice_display_parent_class)->finalize(obj);
 }
@@ -2343,10 +2332,7 @@ static void cursor_set(SpiceCursorChannel *channel,
 
 cursor_invalidate(display);
 
-if (d->mouse_pixbuf) {
-g_object_unref(d->mouse_pixbuf);
-d->mouse_pixbuf = NULL;
-}
+g_clear_object(>mouse_pixbuf);
 
 if (rgba != NULL) {
 d->mouse_pixbuf = gdk_pixbuf_new_from_data(g_memdup(rgba, width * 
height * 4),
diff --git a/src/wocky-http-proxy.c b/src/wocky-http-proxy.c
index a1a030a..33d57d8 100644
--- a/src/wocky-http-proxy.c
+++ b/src/wocky-http-proxy.c
@@ -222,8 +222,7 @@ wocky_http_proxy_connect (GProxy *proxy,
   g_free (buffer);
   buffer = g_data_input_stream_read_until (data_in, HTTP_END_MARKER, NULL,
   cancellable, error);
-  g_object_unref (data_in);
-  data_in = NULL;
+  g_clear_object(_in);
 
   if (buffer == NULL)
 {
-- 
2.8.0

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk v2 1/6] Use GLIB_VERSION_MAX_ALLOWED

[Pavel Grunt]

In order to avoid using a too new GLib API.

Taken from virt-viewer 96d120903f0b95d49642e58d4f796b4f62aa8b20
---
v2: new patch
---
 configure.ac | 8 +++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/configure.ac b/configure.ac
index 4227fd8..817aaab 100644
--- a/configure.ac
+++ b/configure.ac
@@ -166,7 +166,13 @@ fi
 
 AC_CHECK_FUNCS(clearenv strtok_r)
 
-PKG_CHECK_MODULES(GLIB2, glib-2.0)
+# Keep these two definitions in agreement.
+GLIB2_REQUIRED="2.36"
+GLIB2_ENCODED_VERSION="GLIB_VERSION_2_36"
+
+PKG_CHECK_MODULES(GLIB2, glib-2.0 >= $GLIB2_REQUIRED)
+GLIB2_CFLAGS="$GLIB2_CFLAGS -DGLIB_VERSION_MIN_REQUIRED=$GLIB2_ENCODED_VERSION 
\
+  -DGLIB_VERSION_MAX_ALLOWED=$GLIB2_ENCODED_VERSION"
 AC_SUBST(GLIB2_CFLAGS)
 AC_SUBST(GLIB2_LIBS)
 
-- 
2.8.0

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH 04/10] Rename SmartCardDeviceState to RedCharDeviceSmartcard

[Pavel Grunt]

On Fri, 2016-04-01 at 15:51 -0500, Jonathon Jongsma wrote:
> More consistent with naming conventions, and prepares for converting
> this type to a GObject which inherits from RedCharDevice.

Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  server/smartcard.c | 260 ++-
> --
>  1 file changed, 130 insertions(+), 130 deletions(-)
> 
> diff --git a/server/smartcard.c b/server/smartcard.c
> index 03782e2..d138ea7 100644
> --- a/server/smartcard.c
> +++ b/server/smartcard.c
> @@ -49,11 +49,11 @@
>  // Maximal length of APDU
>  #define APDUBufSize 270
>  
> -typedef struct SmartCardDeviceState SmartCardDeviceState;
> +typedef struct RedCharDeviceSmartcard RedCharDeviceSmartcard;
>  
>  typedef struct SmartCardChannelClient {
>  RedChannelClient base;
> -SmartCardDeviceState *smartcard_state;
> +RedCharDeviceSmartcard *smartcard;
>  
>  /* read_from_client/write_to_device buffer.
>   * The beginning of the buffer should always be VSCMsgHeader*/
> @@ -75,7 +75,7 @@ struct RedCharDeviceSmartcardPrivate {
>  int  reader_added; // has reader_add been sent
> to the device
>  };
>  
> -struct SmartCardDeviceState {
> +struct RedCharDeviceSmartcard {
>  struct RedCharDeviceSmartcardPrivate priv[1];
>  };
>  
> @@ -120,52 +120,52 @@ static void
> smartcard_char_device_attach_client(
>  static void
> smartcard_channel_write_to_reader(RedCharDeviceWriteBuffer
> *write_buf);
>  
>  static MsgItem *smartcard_char_device_on_message_from_device(
> -SmartCardDeviceState *state, VSCMsgHeader *header);
> -static SmartCardDeviceState *smartcard_device_state_new(RedsState
> *reds, SpiceCharDeviceInstance *sin);
> -static void smartcard_device_state_free(SmartCardDeviceState* st);
> +RedCharDeviceSmartcard *dev, VSCMsgHeader *header);
> +static RedCharDeviceSmartcard *smartcard_device_new(RedsState *reds,
> SpiceCharDeviceInstance *sin);
> +static void smartcard_device_free(RedCharDeviceSmartcard* dev);
>  static void smartcard_init(RedsState *reds);
>  
> -static void smartcard_read_buf_prepare(SmartCardDeviceState *state,
> VSCMsgHeader *vheader)
> +static void smartcard_read_buf_prepare(RedCharDeviceSmartcard *dev,
> VSCMsgHeader *vheader)
>  {
>  uint32_t msg_len;
>  
>  msg_len = ntohl(vheader->length);
> -if (msg_len > state->priv->buf_size) {
> -state->priv->buf_size = MAX(state->priv->buf_size * 2,
> msg_len + sizeof(VSCMsgHeader));
> -state->priv->buf = spice_realloc(state->priv->buf, state-
> >priv->buf_size);
> +if (msg_len > dev->priv->buf_size) {
> +dev->priv->buf_size = MAX(dev->priv->buf_size * 2, msg_len +
> sizeof(VSCMsgHeader));
> +dev->priv->buf = spice_realloc(dev->priv->buf, dev->priv-
> >buf_size);
>  }
>  }
>  
>  static RedCharDeviceMsgToClient
> *smartcard_read_msg_from_device(SpiceCharDeviceInstance *sin,
>  void
> *opaque)
>  {
> -SmartCardDeviceState *state = opaque;
> +RedCharDeviceSmartcard *dev = opaque;
>  SpiceCharDeviceInterface *sif =
> spice_char_device_get_interface(sin);
> -VSCMsgHeader *vheader = (VSCMsgHeader*)state->priv->buf;
> +VSCMsgHeader *vheader = (VSCMsgHeader*)dev->priv->buf;
>  int n;
>  int remaining;
>  int actual_length;
>  
> -while ((n = sif->read(sin, state->priv->buf_pos, state->priv-
> >buf_size - state->priv->buf_used)) > 0) {
> +while ((n = sif->read(sin, dev->priv->buf_pos, dev->priv-
> >buf_size - dev->priv->buf_used)) > 0) {
>  MsgItem *msg_to_client;
>  
> -state->priv->buf_pos += n;
> -state->priv->buf_used += n;
> -if (state->priv->buf_used < sizeof(VSCMsgHeader)) {
> +dev->priv->buf_pos += n;
> +dev->priv->buf_used += n;
> +if (dev->priv->buf_used < sizeof(VSCMsgHeader)) {
>  continue;
>  }
> -smartcard_read_buf_prepare(state, vheader);
> +smartcard_read_buf_prepare(dev, vheader);
>  actual_length = ntohl(vheader->length);
> -if (state->priv->buf_used - sizeof(VSCMsgHeader) <
> actual_length) {
> +if (dev->priv->buf_used - sizeof(VSCMsgHeader) <
> actual_length) {
>  continue;
>  }
> -msg_to_client =
> smartcard_char_device_on_message_from_device(state, vheader);
> -remaining = state->priv->buf_used - sizeof(VSCMsg

[Spice-devel] [PATCH spice-gtk] Do not use deprecated GtkStock

[Pavel Grunt]

Use labels or named icons

Deprecated since Gtk 3.10
---
 src/spicy.c | 23 +++
 src/usb-device-widget.c | 12 ++--
 2 files changed, 17 insertions(+), 18 deletions(-)

diff --git a/src/spicy.c b/src/spicy.c
index ca9db84..0475530 100644
--- a/src/spicy.c
+++ b/src/spicy.c
@@ -143,9 +143,9 @@ static int ask_user(GtkWidget *parent, char *title, char 
*message,
 dialog = gtk_dialog_new_with_buttons(title,
  parent ? GTK_WINDOW(parent) : NULL,
  GTK_DIALOG_DESTROY_WITH_PARENT,
- GTK_STOCK_OK,
+ "_OK",
  GTK_RESPONSE_ACCEPT,
- GTK_STOCK_CANCEL,
+ "_Cancel",
  GTK_RESPONSE_REJECT,
  NULL);
 gtk_dialog_set_default_response(GTK_DIALOG(dialog), GTK_RESPONSE_ACCEPT);
@@ -370,7 +370,7 @@ static void menu_cb_select_usb_devices(GtkAction *action, 
void *data)
 "Select USB devices for redirection",
 GTK_WINDOW(win->toplevel),
 GTK_DIALOG_MODAL | GTK_DIALOG_DESTROY_WITH_PARENT,
-GTK_STOCK_CLOSE, GTK_RESPONSE_ACCEPT,
+"_Close", GTK_RESPONSE_ACCEPT,
 NULL);
 gtk_dialog_set_default_response(GTK_DIALOG(dialog), GTK_RESPONSE_ACCEPT);
 gtk_container_set_border_width(GTK_CONTAINER(dialog), 12);
@@ -462,7 +462,7 @@ static void menu_cb_about(GtkAction *action, void *data)
   "authors", authors,
   "comments",comments,
   "copyright",   copyright,
-  "logo-icon-name",  GTK_STOCK_ABOUT,
+  "logo-icon-name",  "help-about",
   "website", website,
   "version", PACKAGE_VERSION,
   "license", "LGPLv2.1",
@@ -640,12 +640,12 @@ static const GtkActionEntry entries[] = {
 
 /* File menu */
 .name= "Connect",
-.stock_id= GTK_STOCK_CONNECT,
+.stock_id= "_Connect",
 .label   = "_Connect ...",
 .callback= G_CALLBACK(menu_cb_connect),
 },{
 .name= "Close",
-.stock_id= GTK_STOCK_CLOSE,
+.stock_id= "window-close",
 .label   = "_Close",
 .callback= G_CALLBACK(menu_cb_close),
 .accelerator = "", /* none (disable default "W") */
@@ -653,13 +653,13 @@ static const GtkActionEntry entries[] = {
 
 /* Edit menu */
 .name= "CopyToGuest",
-.stock_id= GTK_STOCK_COPY,
+.stock_id= "edit-copy",
 .label   = "_Copy to guest",
 .callback= G_CALLBACK(menu_cb_copy),
 .accelerator = "", /* none (disable default "C") */
 },{
 .name= "PasteFromGuest",
-.stock_id= GTK_STOCK_PASTE,
+.stock_id= "edit-paste",
 .label   = "_Paste from guest",
 .callback= G_CALLBACK(menu_cb_paste),
 .accelerator = "", /* none (disable default "V") */
@@ -667,7 +667,7 @@ static const GtkActionEntry entries[] = {
 
 /* View menu */
 .name= "Fullscreen",
-.stock_id= GTK_STOCK_FULLSCREEN,
+.stock_id= "view-fullscreen",
 .label   = "_Fullscreen",
 .callback= G_CALLBACK(menu_cb_fullscreen),
 .accelerator = "F11",
@@ -695,7 +695,7 @@ static const GtkActionEntry entries[] = {
 
 /* Help menu */
 .name= "About",
-.stock_id= GTK_STOCK_ABOUT,
+.stock_id= "help-about",
 .label   = "_About ...",
 .callback= G_CALLBACK(menu_cb_about),
 }
@@ -1486,8 +1486,7 @@ TransferTaskWidgets 
*transfer_task_widgets_new(SpiceFileTransferTask *task)
 
 widgets->vbox = gtk_box_new(GTK_ORIENTATION_VERTICAL, 0);
 widgets->hbox = gtk_box_new(GTK_ORIENTATION_HORIZONTAL, 6);
-widgets->cancel = gtk_button_new_from_icon_name(GTK_STOCK_CANCEL,
-
GTK_ICON_SIZE_SMALL_TOOLBAR);
+widgets->cancel = gtk_button_new_with_label("_Cancel");
 
 widgets->progress = gtk_progress_bar_new();
 widgets->label = 
gtk_label_new(spice_file_transfer_task_get_filename(task));
diff --git a/src/usb-device-widget.c b/src/usb-device-widget.c
index d3ea79a..05c7054 100644
--- a/src/usb-device-widget.c
+++ b/src/usb-device-widget.c
@@ -166,8 +166,8 @@ spice_usb_device_widget_show_info_bar(SpiceUsbDeviceWidget 
*self,
 hbox = gtk_box_new(GTK_ORIENTATION_HORIZONTAL, 12);
 gtk_container_add(GTK_CONTAINER(content_area), hbox);
 
-widget = 

Re: [Spice-devel] [PATCH 1/2] define SPICE_CONSTRUCTOR_FUNC and SPICE_DESTRUCTOR_FUNC macros

[Pavel Grunt]

On Wed, 2016-03-16 at 11:52 -0400, Frediano Ziglio wrote:
> > 
> > 
> > Hi,
> > 
> > On Mon, Feb 29, 2016 at 10:36:53AM +, Frediano Ziglio wrote:
> > > 
> > > Allow to define functions executed at program/shared object
> > > initialization
> > > or close.
> > > 
> > > Signed-off-by: Frediano Ziglio 
> > > ---
> > >  common/macros.h | 22 ++
> > >  1 file changed, 22 insertions(+)
> > > 
> > > diff --git a/common/macros.h b/common/macros.h
> > > index 47289be..fe36929 100644
> > > --- a/common/macros.h
> > > +++ b/common/macros.h
> > > @@ -29,5 +29,27 @@
> > >  #define SPICE_ATTR_PRINTF
> > >  #endif /* __GNUC__ */
> > >  
> > > +#ifdef __GNUC__
> > > +#define SPICE_CONSTRUCTOR_FUNC(func_name) \
> > > +static void __attribute__((constructor)) func_name(void)
> > > +#define SPICE_DESTRUCTOR_FUNC(func_name) \
> > > +static void __attribute__((destructor)) func_name(void)
> > > +#elif defined(_MSC_VER)
> > > +#define SPICE_CONSTRUCTOR_FUNC(func_name) \
> > > +static void func_name(void); \
> > > +static int func_name ## _wrapper(void) { func_name(); return
> > > 0; } \
> > > +__pragma(section(".CRT$XCU",read)) \
> > > +__declspec(allocate(".CRT$XCU")) static int (* _array ##
> > > func_name)(void) = func_name ## _wrapper; \
> > > +static void func_name(void)
> > > +#define SPICE_DESTRUCTOR_FUNC(func_name) \
> > > +static void func_name(void); \
> > > +static int func_name ## _wrapper(void) { func_name(); return
> > > 0; } \
> > > +__pragma(section(".CRT$XPU",read)) \
> > > +__declspec(allocate(".CRT$XPU")) static int (* _array ##
> > > func_name)(void) = func_name ## _wrapper; \
> > > +static void func_name(void)
> > > +#else
> > > +#error Please implement SPICE_CONSTRUCTOR_FUNC and
> > > SPICE_DESTRUCTOR_FUNC
> > > for this compiler
> > > +#endif
> > > +
> > >  
> > >  #endif /* __MACROS_H */
> > > --
> > > 2.5.0
> > Testes, works without issues.
> > 
> > I'm wondering why not to include the whole header from glib [0] and
> > maybe file a bug to ask them to export in the future.
> > 
> > [0] https://git.gnome.org/browse/glib/tree/glib/gconstructor.h
> > 
> > Then, in case G_HAS_CONSTRUCTORS is not defined we can include
> > different
> > arch constructors. What do you think?
> > 
> > PS: while looking for an open bug:
> > aix: https://bugzilla.gnome.org/show_bug.cgi?id=763560
> > hp-ux/ia: https://bugzilla.gnome.org/show_bug.cgi?id=763466
> > 
> > cheers,
> >   toso
> > 
> Do we want to support HP-UX or AIX? I did work on HP-UX and from what
> I
> can see not even HP is supporting that much this OS.
> Personally I don't like Gnome implementation, particularly the atexit
> call
> on Windows.
> I don't understand the g_slist_find call either.
> 
> Frediano


What about clang ? Does it work there?

Pavel
> ___
> Spice-devel mailing list
> Spice-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/spice-devel
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 5/5] Adjust to Gtk+ 3.10

[Pavel Grunt]

Dependency since f9a1aad85fcc76dd76c454b51fbce5e07c9b145f

Remove unneeded GTK_CHECK_VERSION guards
---
 src/gtk-compat.h|  2 --
 src/spicy.c | 16 
 src/usb-device-widget.c |  9 -
 3 files changed, 27 deletions(-)

diff --git a/src/gtk-compat.h b/src/gtk-compat.h
index b2cf6d6..39438f9 100644
--- a/src/gtk-compat.h
+++ b/src/gtk-compat.h
@@ -22,12 +22,10 @@
 
 #include 
 
-#if GTK_CHECK_VERSION (2, 91, 0)
 static inline void gdk_drawable_get_size(GdkWindow *w, gint *ww, gint *wh)
 {
 *ww = gdk_window_get_width(w);
 *wh = gdk_window_get_height(w);
 }
-#endif
 
 #endif /* GTK_COMPAT_H */
diff --git a/src/spicy.c b/src/spicy.c
index c3564f0..6ca014b 100644
--- a/src/spicy.c
+++ b/src/spicy.c
@@ -1001,11 +1001,7 @@ static SpiceWindow *create_spice_window(spice_connection 
*conn, SpiceChannel *ch
  G_CALLBACK(grab_keys_pressed_cb), win);
 
 /* status line */
-#if GTK_CHECK_VERSION(3,0,0)
 win->statusbar = gtk_box_new(GTK_ORIENTATION_HORIZONTAL, 1);
-#else
-win->statusbar = gtk_hbox_new(FALSE, 1);
-#endif
 
 win->status = gtk_label_new("status line");
 gtk_misc_set_alignment(GTK_MISC(win->status), 0, 0.5);
@@ -1025,11 +1021,7 @@ static SpiceWindow *create_spice_window(spice_connection 
*conn, SpiceChannel *ch
 }
 
 /* Make a vbox and put stuff in */
-#if GTK_CHECK_VERSION(3,0,0)
 vbox = gtk_box_new(GTK_ORIENTATION_VERTICAL, 1);
-#else
-vbox = gtk_vbox_new(FALSE, 1);
-#endif
 gtk_container_set_border_width(GTK_CONTAINER(vbox), 0);
 gtk_container_add(GTK_CONTAINER(win->toplevel), vbox);
 gtk_box_pack_start(GTK_BOX(vbox), win->menubar, FALSE, FALSE, 0);
@@ -1510,28 +1502,20 @@ TransferTaskWidgets 
*transfer_task_widgets_new(SpiceFileTransferTask *task)
 {
 TransferTaskWidgets *widgets = g_new0(TransferTaskWidgets, 1);
 
-#if GTK_CHECK_VERSION(3,0,0)
 widgets->vbox = gtk_box_new(GTK_ORIENTATION_VERTICAL, 0);
 widgets->hbox = gtk_box_new(GTK_ORIENTATION_HORIZONTAL, 6);
 widgets->cancel = gtk_button_new_from_icon_name(GTK_STOCK_CANCEL,
 
GTK_ICON_SIZE_SMALL_TOOLBAR);
-#else
-widgets->vbox = gtk_vbox_new(FALSE, 0);
-widgets->hbox = gtk_hbox_new(FALSE, 6);
-widgets->cancel = gtk_button_new_from_stock(GTK_STOCK_CANCEL);
-#endif
 
 widgets->progress = gtk_progress_bar_new();
 widgets->label = 
gtk_label_new(spice_file_transfer_task_get_filename(task));
 
-#if GTK_CHECK_VERSION(3,0,0)
 gtk_widget_set_halign(widgets->label, GTK_ALIGN_START);
 gtk_widget_set_valign(widgets->label, GTK_ALIGN_BASELINE);
 gtk_widget_set_valign(widgets->progress, GTK_ALIGN_CENTER);
 gtk_widget_set_hexpand(widgets->progress, TRUE);
 gtk_widget_set_valign(widgets->cancel, GTK_ALIGN_CENTER);
 gtk_widget_set_hexpand(widgets->progress, FALSE);
-#endif
 
 gtk_box_pack_start(GTK_BOX(widgets->hbox), widgets->progress,
TRUE, TRUE, 0);
diff --git a/src/usb-device-widget.c b/src/usb-device-widget.c
index 8f85ae9..3b060e9 100644
--- a/src/usb-device-widget.c
+++ b/src/usb-device-widget.c
@@ -94,12 +94,7 @@ struct _SpiceUsbDeviceWidgetPrivate {
 
 static guint signals[LAST_SIGNAL] = { 0, };
 
-#if GTK_CHECK_VERSION(3,0,0)
 G_DEFINE_TYPE(SpiceUsbDeviceWidget, spice_usb_device_widget, GTK_TYPE_BOX);
-#else
-G_DEFINE_TYPE(SpiceUsbDeviceWidget, spice_usb_device_widget, GTK_TYPE_VBOX);
-#endif
-
 
 static void spice_usb_device_widget_get_property(GObject *gobject,
  guintprop_id,
@@ -168,11 +163,7 @@ spice_usb_device_widget_show_info_bar(SpiceUsbDeviceWidget 
*self,
 gtk_info_bar_set_message_type(GTK_INFO_BAR(info_bar), message_type);
 
 content_area = gtk_info_bar_get_content_area(GTK_INFO_BAR(info_bar));
-#if GTK_CHECK_VERSION(3,0,0)
 hbox = gtk_box_new(GTK_ORIENTATION_HORIZONTAL, 12);
-#else
-hbox = gtk_hbox_new(FALSE, 12);
-#endif
 gtk_container_add(GTK_CONTAINER(content_area), hbox);
 
 widget = gtk_image_new_from_stock(stock_icon_id,
-- 
2.7.3

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 3/3] spicy: s/Automagic/Automatic/

[Pavel Grunt]

---
It's A Kind of Magic
---
 src/spicy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/spicy.c b/src/spicy.c
index 12ae653..a3f357a 100644
--- a/src/spicy.c
+++ b/src/spicy.c
@@ -738,7 +738,7 @@ static const GtkToggleActionEntry tentries[] = {
 .callback= G_CALLBACK(menu_cb_bool_prop),
 },{
 .name= "auto-clipboard",
-.label   = "Automagic clipboard sharing between host and guest",
+.label   = "Automatic clipboard sharing between host and guest",
 .callback= G_CALLBACK(menu_cb_bool_prop),
 },{
 .name= "auto-usbredir",
-- 
2.7.3

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 4/5] Adjust to GLib 2.36

[Pavel Grunt]

Dependency since 8693e7d3f7de1ff102082212fa6e35fb1a252ef7

Remove glib-compat files and most of GLIB_CHECK_VERSION guards
---
 doc/reference/Makefile.am  |   1 -
 po/POTFILES.skip   |   1 -
 src/Makefile.am|   5 --
 src/channel-cursor.c   |   1 -
 src/channel-display.c  |   1 -
 src/channel-main.c |   1 -
 src/channel-port.c |   1 -
 src/channel-usbredir.c |   9 ---
 src/channel-webdav.c   |   1 -
 src/controller/dump.c  |   3 -
 src/controller/test.c  |   3 -
 src/desktop-integration.c  |   1 -
 src/glib-compat.c  | 112 -
 src/glib-compat.h  |  65 ---
 src/smartcard-manager.c|   2 -
 src/spice-audio.c  |   2 -
 src/spice-channel.c|   1 -
 src/spice-client-glib-usb-acl-helper.c |   6 --
 src/spice-option.c |   1 -
 src/spice-pulse.c  |   1 -
 src/spice-session.c|   1 -
 src/spice-uri.c|   1 -
 src/spice-util-priv.h  |   8 ---
 src/spice-util.c   |   2 -
 src/spice-util.h   |   2 +-
 src/spice-widget.c |   1 -
 src/spicy-screenshot.c |   3 -
 src/spicy-stats.c  |   3 -
 src/spicy.c|   7 ---
 src/usb-acl-helper.c   |   1 -
 src/usb-device-manager.c   |   7 ---
 src/usb-device-widget.c|   1 -
 src/usbutil.c  |   2 -
 src/vmcstream.c|   1 -
 src/wocky-http-proxy.c |   1 -
 src/wocky-http-proxy.h |   2 -
 tests/coroutine.c  |   4 --
 tests/mock-acl-helper.c|   4 --
 38 files changed, 1 insertion(+), 268 deletions(-)
 delete mode 100644 src/glib-compat.c
 delete mode 100644 src/glib-compat.h

diff --git a/doc/reference/Makefile.am b/doc/reference/Makefile.am
index b76a9cf..cb01f97 100644
--- a/doc/reference/Makefile.am
+++ b/doc/reference/Makefile.am
@@ -32,7 +32,6 @@ IGNORE_HFILES=\
decode.h\
display \
gio-coroutine.h \
-   glib-compat.h   \
smartcard-manager-priv.h\
spice-audio-priv.h  \
spice-channel-cache.h   \
diff --git a/po/POTFILES.skip b/po/POTFILES.skip
index 4da36d0..0dd1912 100644
--- a/po/POTFILES.skip
+++ b/po/POTFILES.skip
@@ -1,3 +1,2 @@
 spice-common/python_modules/spice_parser.py
 spice-common/spice_codegen.py
-src/glib-compat.c
diff --git a/src/Makefile.am b/src/Makefile.am
index 2e7aa3b..e393913 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -122,7 +122,6 @@ SPICE_GTK_LIBADD_COMMON =   \
$(NULL)
 
 SPICE_GTK_SOURCES_COMMON = \
-   glib-compat.h   \
gtk-compat.h\
spice-util.c\
spice-util-priv.h   \
@@ -214,8 +213,6 @@ endif
 libspice_client_glib_2_0_la_SOURCES =  \
bio-gio.c   \
bio-gio.h   \
-   glib-compat.c   \
-   glib-compat.h   \
spice-audio.c   \
spice-audio-priv.h  \
spice-common.h  \
@@ -389,8 +386,6 @@ spicy_CPPFLAGS =\
 
 if WITH_POLKIT
 spice_client_glib_usb_acl_helper_SOURCES = \
-   glib-compat.c   \
-   glib-compat.h   \
spice-client-glib-usb-acl-helper.c  \
$(NULL)
 
diff --git a/src/channel-cursor.c b/src/channel-cursor.c
index 0765209..2ae8f86 100644
--- a/src/channel-cursor.c
+++ b/src/channel-cursor.c
@@ -17,7 +17,6 @@
 */
 #include "config.h"
 
-#include "glib-compat.h"
 #include "spice-client.h"
 #include "spice-common.h"
 
diff --git a/src/channel-display.c b/src/channel-display.c
index f576679..2c54df7 100644
--- a/src/channel-display.c
+++ b/src/channel-display.c
@@ -21,7 +21,6 @@
 #include 
 #endif
 
-#include "glib-compat.h"
 #include "spice-client.h"
 #include "spice-common.h"
 
diff --git a/src/channel-main.c b/src/channel-main.c
index 276799f..1c19de1 100644
--- a/src/channel-main.c
+++ b/src/channel-main.c
@@ -21,7 +21,6 @@
 #include 
 #include 
 
-#include "glib-compat.h"
 #include "spice-client.h"
 #include "spice-common.h"
 #include "spice-marshal.h"
diff --git 

Re: [Spice-devel] [PATCH spice-gtk 3/3] spicy: s/Automagic/Automatic/

[Pavel Grunt]

On Wed, 2016-03-16 at 17:16 -0300, Eduardo Lima (Etrunko) wrote:
> On 03/16/2016 10:46 AM, Pavel Grunt wrote:
> > 
> > ---
> > It's A Kind of Magic
> > ---
> >  src/spicy.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/src/spicy.c b/src/spicy.c
> > index 12ae653..a3f357a 100644
> > --- a/src/spicy.c
> > +++ b/src/spicy.c
> > @@ -738,7 +738,7 @@ static const GtkToggleActionEntry tentries[] =
> > {
> >  .callback= G_CALLBACK(menu_cb_bool_prop),
> >  },{
> >  .name= "auto-clipboard",
> > -.label   = "Automagic clipboard sharing between host
> > and guest",
> > +.label   = "Automatic clipboard sharing between host
> > and guest",
> >  .callback= G_CALLBACK(menu_cb_bool_prop),
> >  },{
> >  .name= "auto-usbredir",
> > 
> Acked-by: Eduardo Lima (Etrunko) <etru...@redhat.com>
> 
> 

Thanks, pushed

Pavel

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 3/5] Move strtok_r from glib-compat to spice-uri

[Pavel Grunt]

It is not glib specific and it is only used in spice-uri
---
 src/glib-compat.c | 31 ---
 src/glib-compat.h |  4 
 src/spice-uri.c   | 29 +
 3 files changed, 29 insertions(+), 35 deletions(-)

diff --git a/src/glib-compat.c b/src/glib-compat.c
index 41a7f52..622620b 100644
--- a/src/glib-compat.c
+++ b/src/glib-compat.c
@@ -110,34 +110,3 @@ g_queue_free_full (GQueue*queue,
   g_queue_free (queue);
 }
 #endif
-
-
-#ifndef HAVE_STRTOK_R
-G_GNUC_INTERNAL
-char *strtok_r(char *s, const char *delim, char **save_ptr)
-{
-char *token;
-
-if (s == NULL)
-s = *save_ptr;
-
-/* Scan leading delimiters. */
-s += strspn (s, delim);
-if (*s == '\0')
-return NULL;
-
-/* Find the end of the token. */
-token = s;
-s = strpbrk (token, delim);
-if (s == NULL)
-/* This token finishes the string. */
-*save_ptr = strchr (token, '\0');
-else
-{
-/* Terminate the token and make *SAVE_PTR point past it. */
-*s = '\0';
-*save_ptr = s + 1;
-}
-return token;
-}
-#endif
diff --git a/src/glib-compat.h b/src/glib-compat.h
index 512ea55..51eef5c 100644
--- a/src/glib-compat.h
+++ b/src/glib-compat.h
@@ -62,8 +62,4 @@ g_queue_free_full (GQueue*queue,
   } G_STMT_END
 #endif
 
-#ifndef HAVE_STRTOK_R
-char* strtok_r(char *s, const char *delim, char **save_ptr);
-#endif
-
 #endif /* GLIB_COMPAT_H */
diff --git a/src/spice-uri.c b/src/spice-uri.c
index e2c5c9a..d925973 100644
--- a/src/spice-uri.c
+++ b/src/spice-uri.c
@@ -60,6 +60,35 @@ enum  {
 SPICE_URI_PORT
 };
 
+#ifndef HAVE_STRTOK_R
+static char *strtok_r(char *s, const char *delim, char **save_ptr)
+{
+char *token;
+
+if (s == NULL)
+s = *save_ptr;
+
+/* Scan leading delimiters. */
+s += strspn (s, delim);
+if (*s == '\0')
+return NULL;
+
+/* Find the end of the token. */
+token = s;
+s = strpbrk (token, delim);
+if (s == NULL)
+/* This token finishes the string. */
+*save_ptr = strchr (token, '\0');
+else
+{
+/* Terminate the token and make *SAVE_PTR point past it. */
+*s = '\0';
+*save_ptr = s + 1;
+}
+return token;
+}
+#endif
+
 G_GNUC_INTERNAL
 SpiceURI* spice_uri_new(void)
 {
-- 
2.7.3

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 2/5] spice-util: Fix alignment

[Pavel Grunt]

---
 src/spice-util.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/spice-util.c b/src/spice-util.c
index fd97ee7..d2bcfbf 100644
--- a/src/spice-util.c
+++ b/src/spice-util.c
@@ -47,7 +47,7 @@ static void spice_util_enable_debug_messages(void)
 if (!doms) {
 g_setenv("G_MESSAGES_DEBUG", G_LOG_DOMAIN, 1);
 } else if (g_str_equal(doms, "all")) {
-   return;
+return;
 } else if (!strstr(doms, G_LOG_DOMAIN)) {
 gchar *newdoms = g_strdup_printf("%s %s", doms, G_LOG_DOMAIN);
 g_setenv("G_MESSAGES_DEBUG", newdoms, 1);
-- 
2.7.3

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 2/3] spice-display: Remove auto-clipboard notify

[Pavel Grunt]

It was used to notify the (deprecated) auto-clipboard property of
the SpiceDisplay everytime the SpiceGtkSession emits its auto-clipboard
notify.

The commit 70a00b8fcc2ca85d3601f6ac4895f906f9032e77 removed the property
from SpiceDisplay causing a runtime warning:

GLib-GObject-WARNING **: g_object_notify: object class 'SpiceDisplay' has no 
property named 'auto-clipboard'
---
 src/spice-widget.c | 12 
 1 file changed, 12 deletions(-)

diff --git a/src/spice-widget.c b/src/spice-widget.c
index f86054f..30426be 100644
--- a/src/spice-widget.c
+++ b/src/spice-widget.c
@@ -390,15 +390,6 @@ static void spice_display_set_property(GObject  
*object,
 }
 }
 
-static void gtk_session_property_changed(GObject*gobject,
- GParamSpec *pspec,
- gpointeruser_data)
-{
-SpiceDisplay *display = user_data;
-
-g_object_notify(G_OBJECT(display), g_param_spec_get_name(pspec));
-}
-
 static void session_inhibit_keyboard_grab_changed(GObject*gobject,
   GParamSpec *pspec,
   gpointeruser_data)
@@ -689,9 +680,6 @@ spice_display_constructor(GType  gtype,
 }
 g_list_free(list);
 
-spice_g_signal_connect_object(d->gtk_session, "notify::auto-clipboard",
-  G_CALLBACK(gtk_session_property_changed), 
display, 0);
-
 spice_g_signal_connect_object(d->session, "notify::inhibit-keyboard-grab",
   
G_CALLBACK(session_inhibit_keyboard_grab_changed),
   display, 0);
-- 
2.7.3

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-gtk] channel-usbredir: Fix crash due to a Task returning earlier than expected

[Pavel Grunt]

 n_params=, param_types=0x644d40) at gclosure.c:864
>  #33 0x724d1dd3 in g_signal_emit_valist (instance=0x80faa0,
> signal_id=, detail=0, var_args=var_args@entry=0x7f
> ffd9d0) at gsignal.c:3292
>  #34 0x724d28ff in g_signal_emit (instance=,
> signal_id=, detail=) at gsignal.c:3439
>  #35 0x73e6e4bc in gtk_widget_event_internal () at
> /lib64/libgtk-3.so.0
>  #36 0x73d2e34e in propagate_event () at /lib64/libgtk-3.so.0
>  #37 0x73d300fc in gtk_main_do_event () at /lib64/libgtk-
> 3.so.0
>  #38 0x738a8e92 in gdk_event_source_dispatch () at
> /lib64/libgdk-3.so.0
>  #39 0x721b8e3a in g_main_context_dispatch (context=0x647390)
> at gmain.c:3154
>  #40 0x721b8e3a in g_main_context_dispatch (context=context@e
> ntry=0x647390) at gmain.c:3769
>  #41 0x721b91d0 in g_main_context_iterate (context=0x647390, 
> block=block@entry=1, dispatch=dispatch@entry=1, self=)
> at gmain.c:3840
>  #42 0x721b94f2 in g_main_loop_run (loop=0x6e2730) at
> gmain.c:4034
>  #43 0x0040b2f9 in main (argc=1, argv=0x7fffde48) at
> spicy.c:1920
> 
> Signed-off-by: Fabiano Fidêncio <fiden...@redhat.com>

Tested-by: Pavel Grunt <pgr...@redhat.com>
Acked-by: Pavel Grunt <pgr...@redhat.com>

Thanks,
Pavel

> ---
>  src/channel-usbredir.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/channel-usbredir.c b/src/channel-usbredir.c
> index d95a6c5..dca8455 100644
> --- a/src/channel-usbredir.c
> +++ b/src/channel-usbredir.c
> @@ -297,12 +297,12 @@ static void spice_usbredir_channel_open_acl_cb(
>  spice_usbredir_channel_open_device(channel, );
>  }
>  if (err) {
> -g_task_return_error(priv->task, err);
>  libusb_unref_device(priv->device);
>  priv->device = NULL;
>  g_boxed_free(spice_usb_device_get_type(), priv-
> >spice_device);
>  priv->spice_device = NULL;
>  priv->state  = STATE_DISCONNECTED;
> +g_task_return_error(priv->task, err);
>  } else {
>  g_task_return_boolean(priv->task, TRUE);
>  }
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [phodav PATCH 5/7 v3] spice-webdavd-windows: Automount shared folder

[Pavel Grunt]

Hi Lukas,

On Thu, 2016-03-17 at 14:47 +0100, Lukas Venhoda wrote:
> Try to connect to shared folder automatically on Windows.
> 
> On each loop of run_service(), run a GTask, that waits for 0.5s.
> If sharing is disabled, read_thread will immediately return and cause
> the main_loop to stop.
> This in turn causes the cancellable object to be canceled and the
> loop stops.
> 
> If nothing happens for 0.5s (no cancel is called), it means, that
> sharing is enabled,
> because read_thread (and main loop) is blocked, and we can call
> map_drive.
> 
> map_drive then enumerates free drive letters, and maps Spice Folder
> to highest possible
> (from Z: to A:).
> 
> If all drive letters are already assigned, show a critical error, but
> don't stop the service. The folder can still be accessed trough other
> means.
> ---
> Changes since v2:
>  - Changed for loop to gpoll
>  - split map_drive into 2 functions
>  - added enum for return values of map_drive
>  - added TODO for renaming the drive
>  - added explanation for the 0.5 delay into commit log
> 
> Changes since v1:
>  - Changed GThread to a GTask
>  - Only wait half a second, instead of 5
> 
> Changes since RFC:
>  - Calling WNetAddConnection2() blindly was slow and caause many
> threads to spawn.
> - Now only call it once, when it is sure, that it will connect.
>  - Now connects to a * drive, instead of always Z:
>  - Thread is now spawned and joined every time run_service() is
> called.
> ---
>  Makefile.am   |   4 ++
>  spice/spice-webdavd.c | 112
> ++
>  2 files changed, 116 insertions(+)
> 
> diff --git a/Makefile.am b/Makefile.am
> index 6127f93..d8e2d53 100644
> --- a/Makefile.am
> +++ b/Makefile.am
> @@ -87,6 +87,10 @@ spice_webdavd_LDADD =  \
>   $(PIE_LDFLAGS)  \
>   $(NULL)
> 
> +if OS_WIN32
> +spice_webdavd_LDADD += -lnetapi32 -lmpr
> +endif
> +
>  deps.txt:
>   $(AM_V_GEN)rpm -qa | grep $(host_os) | sort | unix2dos > $@
> 
> diff --git a/spice/spice-webdavd.c b/spice/spice-webdavd.c
> index f9c5cf1..49b1884 100644
> --- a/spice/spice-webdavd.c
> +++ b/spice/spice-webdavd.c
> @@ -737,11 +737,118 @@ open_mux_path (const char *path)
>    mux_queue = output_queue_new (G_OUTPUT_STREAM (mux_ostream));
>  }
> 
> +#ifdef G_OS_WIN32
> +typedef enum Tmap_drive_enum
> +{
> +  MAP_DRIVE_OK,
> +  MAP_DRIVE_TRY_AGAIN,
> +  MAP_DRIVE_ERROR
> +} map_drive_enum;
> +
> +gchar drive_letter;
You don't need this variable. map_drive can accept const gchar instead
of void.

> +
> +static gchar
> +get_free_drive_letter(void)
> +{
> +  const guint32 max_mask = 1 << 25;
> +  guint32 drives;
> +  guint32 mask;
> +  gint i;
> +
> +  drives = GetLogicalDrives ();
> +
> +  for (i = 0; i < 26; i++)
> +  {
> +  mask = max_mask >> i;
> +  if ((drives & mask) == 0)
> +return 'z' - i;
> +  }
> +
> +  return 0;
> +}
> +
> +static map_drive_enum
> +map_drive(void)
> +{
> +  gchar local_name[3];
> +  gchar remote_name[] = "http://localhost:9843/;;
it is only used in the net_resource
> +  NETRESOURCE net_resource;
> +  guint32 retval;
> +
> +  local_name[0] = drive_letter;
drive_letter should be parameter to this function
> +  local_name[1] = ':';
> +  local_name[2] = 0;
I would prefer some sprintf-like function instead
> +
> +  net_resource.dwType = RESOURCETYPE_DISK;
> +  net_resource.lpLocalName = local_name;
> +  net_resource.lpRemoteName = remote_name;
> +  net_resource.lpProvider = NULL;

In my opinion setting up net_resource should go to a separate
function,. You don't need 'local_name' and 'remote_name' to map the
drive, you need the net_resource.
> +
> +  retval = WNetAddConnection2 (_resource, NULL, NULL,
> CONNECT_TEMPORARY);
> +
> +  if (retval == NO_ERROR) {
> +g_debug ("map_drive ok");
> +return MAP_DRIVE_OK;
> +  } else if (retval == ERROR_ALREADY_ASSIGNED) {
> +g_debug ("map_drive already asigned");
> +return MAP_DRIVE_TRY_AGAIN;
> +  } else {
> +g_critical ("map_drive error %d", retval);
Is it critical enough to exit the program? I would use g_warning
> +return MAP_DRIVE_ERROR;
> +  }
> +}
> +
> +static void
> +map_drive_cb(GTask *task,
> +  gpointer source_object,
> +  gpointer task_data,
> +  GCancellable *cancellable)
> +{
> +  guint32 timeout = 500; //half a second
> +  GCancellable * cancel_map = task_data;
> +  GPollFD cancel_pollfd;
> +
> +  if (!g_cancellable_make_pollfd (cancel_map, _pollfd))
> +  {
> +g_critical ("GPollFD failed to create.");
critical/warning
> +return;
> +  }
> +
> +  if (g_poll (_pollfd, 1, timeout) <= 0)
> +  {
> +while (TRUE)
> +{
> +  if ((drive_letter = get_free_drive_letter ()) == 0)
I would move assignment from the condition
> +  {
> +g_critical ("all drive letters already assigned.");
critical/warning
> +break;
> +  }
> +
> +  if (map_drive () != MAP_DRIVE_TRY_AGAIN)
> +break;
> +}
> +
> +//TODO: Rename network 

Re: [Spice-devel] [PATCH] usbredir_handle_msg: rename data to err_data

[Pavel Grunt]

On Mon, 2016-03-21 at 12:07 +0100, Victor Toso wrote:
> On Sun, Mar 20, 2016 at 03:03:41PM +0200, Snir Sheriber wrote:
> > 
> > To better reflect what this variable is used for and change the
> > declare
> > location such that it will be declared only if it should
> Acked-by: Victor Toso 
> 

Pushed

> > 
> > ---
> >  src/channel-usbredir.c | 14 +++---
> >  1 file changed, 7 insertions(+), 7 deletions(-)
> > 
> > diff --git a/src/channel-usbredir.c b/src/channel-usbredir.c
> > index d04267f..0f76126 100644
> > --- a/src/channel-usbredir.c
> > +++ b/src/channel-usbredir.c
> > @@ -629,7 +629,6 @@ static void usbredir_handle_msg(SpiceChannel
> > *c, SpiceMsgIn *in)
> >  {
> >  SpiceUsbredirChannel *channel = SPICE_USBREDIR_CHANNEL(c);
> >  SpiceUsbredirChannelPrivate *priv = channel->priv;
> > -device_error_data data;
> >  int r, size;
> >  uint8_t *buf;
> >  
> > @@ -645,6 +644,7 @@ static void usbredir_handle_msg(SpiceChannel
> > *c, SpiceMsgIn *in)
> >  r = usbredirhost_read_guest_data(priv->host);
> >  if (r != 0) {
> >  SpiceUsbDevice *spice_device = priv->spice_device;
> > +device_error_data err_data;
> >  gchar *desc;
> >  GError *err;
> >  
> > @@ -674,14 +674,14 @@ static void usbredir_handle_msg(SpiceChannel
> > *c, SpiceMsgIn *in)
> >  
> >  CHANNEL_DEBUG(c, "%s", err->message);
> >  
> > -data.channel = channel;
> > -data.caller = coroutine_self();
> > -data.spice_device =
> > g_boxed_copy(spice_usb_device_get_type(), spice_device);
> > -data.error = err;
> > -g_idle_add(device_error, );
> > +err_data.channel = channel;
> > +err_data.caller = coroutine_self();
> > +err_data.spice_device =
> > g_boxed_copy(spice_usb_device_get_type(), spice_device);
> > +err_data.error = err;
> > +g_idle_add(device_error, _data);
> >  coroutine_yield(NULL);
> >  
> > -g_boxed_free(spice_usb_device_get_type(),
> > data.spice_device);
> > +g_boxed_free(spice_usb_device_get_type(),
> > err_data.spice_device);
> >  
> >  g_error_free(err);
> >  }
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH] char-device: Add helpers for SpiceCharDeviceCallbacks vfuncs

[Pavel Grunt]

On Mon, 2016-03-21 at 13:23 +, Frediano Ziglio wrote:
> From: Christophe Fergeau <cferg...@redhat.com>
> 
> Add helper functions wrapping calls to the virtual functions
> defined in SpiceCharDeviceCallbacks.
> ---
>  server/char-device.c | 78 +-
> --
>  1 file changed, 62 insertions(+), 16 deletions(-)
> 
> Changes:
> - removed excessive checks (proposed by Pavel Grunt).
> 
> I would ack it. Not clear if I can ack after I did the change (even
> if
> suggested by Pavel).

I would ack it as well

Pavel

> 
> diff --git a/server/char-device.c b/server/char-device.c
> index d3af960..ac25b01 100644
> --- a/server/char-device.c
> +++ b/server/char-device.c
> @@ -94,6 +94,56 @@ typedef struct SpiceCharDeviceMsgToClientItem {
>  SpiceCharDeviceMsgToClient *msg;
>  } SpiceCharDeviceMsgToClientItem;
>  
> +static SpiceCharDeviceMsgToClient *
> +spice_char_device_read_one_msg_from_device(SpiceCharDeviceState
> *dev)
> +{
> +   return dev->cbs.read_one_msg_from_device(dev->sin, dev->opaque);
> +}
> +
> +static SpiceCharDeviceMsgToClient *
> +spice_char_device_ref_msg_to_client(SpiceCharDeviceState *dev,
> +SpiceCharDeviceMsgToClient *msg)
> +{
> +   return dev->cbs.ref_msg_to_client(msg, dev->opaque);
> +}
> +
> +static void
> +spice_char_device_unref_msg_to_client(SpiceCharDeviceState *dev,
> +  SpiceCharDeviceMsgToClient
> *msg)
> +{
> +   dev->cbs.unref_msg_to_client(msg, dev->opaque);
> +}
> +
> +static void
> +spice_char_device_send_msg_to_client(SpiceCharDeviceState *dev,
> + SpiceCharDeviceMsgToClient
> *msg,
> + RedClient *client)
> +{
> +   dev->cbs.send_msg_to_client(msg, client, dev->opaque);
> +}
> +
> +static void
> +spice_char_device_send_tokens_to_client(SpiceCharDeviceState *dev,
> +RedClient *client,
> +uint32_t tokens)
> +{
> +   dev->cbs.send_tokens_to_client(client, tokens, dev->opaque);
> +}
> +
> +static void
> +spice_char_device_on_free_self_token(SpiceCharDeviceState *dev)
> +{
> +   if (dev->cbs.on_free_self_token != NULL) {
> +   dev->cbs.on_free_self_token(dev->opaque);
> +   }
> +}
> +
> +static void
> +spice_char_device_remove_client(SpiceCharDeviceState *dev, RedClient
> *client)
> +{
> +   dev->cbs.remove_client(client, dev->opaque);
> +}
> +
>  static void
> spice_char_device_write_buffer_free(SpiceCharDeviceWriteBuffer *buf)
>  {
>  if (buf == NULL)
> @@ -143,7 +193,7 @@ static void
> spice_char_device_client_send_queue_free(SpiceCharDeviceState *dev,
>  
>  link);
>  
>  ring_remove(item);
> -dev->cbs.unref_msg_to_client(msg_item->msg, dev->opaque);
> +spice_char_device_unref_msg_to_client(dev, msg_item->msg);
>  free(msg_item);
>  }
>  dev_client->num_send_tokens += dev_client->send_queue_size;
> @@ -189,7 +239,7 @@ static void
> spice_char_device_handle_client_overflow(SpiceCharDeviceClientState
>  {
>  SpiceCharDeviceState *dev = dev_client->dev;
>  spice_printerr("dev %p client %p ", dev, dev_client);
> -dev->cbs.remove_client(dev_client->client, dev->opaque);
> +spice_char_device_remove_client(dev, dev_client->client);
>  }
>  
>  static SpiceCharDeviceClientState
> *spice_char_device_client_find(SpiceCharDeviceState *dev,
> @@ -258,7 +308,7 @@ static void
> spice_char_device_add_msg_to_client_queue(SpiceCharDeviceClientState
>  }
>  
>  msg_item = spice_new0(SpiceCharDeviceMsgToClientItem, 1);
> -msg_item->msg = dev->cbs.ref_msg_to_client(msg, dev->opaque);
> +msg_item->msg = spice_char_device_ref_msg_to_client(dev, msg);
>  ring_add(_client->send_queue, _item->link);
>  dev_client->send_queue_size++;
>  if (!dev_client->wait_for_tokens_started) {
> @@ -280,7 +330,7 @@ static void
> spice_char_device_send_msg_to_clients(SpiceCharDeviceState *dev,
>  if (spice_char_device_can_send_to_client(dev_client)) {
>  dev_client->num_send_tokens--;
>  spice_assert(ring_is_empty(_client->send_queue));
> -dev->cbs.send_msg_to_client(msg, dev_client->client,
> dev->opaque);
> +spice_char_device_send_msg_to_client(dev, msg,
> dev_client->client);
>  
>  /* don't refer to d

Re: [Spice-devel] [PATCH spice-gtk 1/7] widget: remove superflous WIN32 check

[Pavel Grunt]

On Mon, 2016-03-21 at 19:14 +0100, Marc-André Lureau wrote:
> The egl call is already inside a X11 conditional block.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  src/spice-widget.c | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index e329809..c8bb72c 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -580,13 +580,11 @@ drawing_area_realize(GtkWidget *area, gpointer
> user_data)
>  g_clear_error();
>  }
>  
> -#ifndef G_OS_WIN32
>  if (!spice_egl_realize_display(display,
> gtk_widget_get_window(area), )) {
>  g_critical("egl realize failed: %s", err->message);
>  g_clear_error();
>  }
>  #endif
> -#endif
>  }
>  
>  static void spice_display_init(SpiceDisplay *display)
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-gtk] vmcstream, gtask: Fix crash when trying to use webdav

[Pavel Grunt]

Ack. 

Pavel

On Mon, 2016-03-21 at 16:02 +0100, Fabiano Fidêncio wrote:
> Don't try to get the cancellable from a task that is, for sure
> (ensured
> by us), NULL. See the backtrace:
> 
>  #0  0x729cf250 in g_task_get_cancellable () at
> /lib64/libgio-2.0.so.0
>  #1  0x778a9012 in spice_vmc_input_stream_read_all_async
> (stream=, buffer=0x7070f8, count=8,
> io_priority=, cancellable=0x7fffcc00a470,
> callback=0x7788c190 , user_data=0x707b30) at
> vmcstream.c:187
>  #2  0x7788c0ab in start_demux (self=0x707b30) at channel-
> webdav.c:498
>  #3  0x72700578 in g_closure_invoke () at /lib64/libgobject-
> 2.0.so.0
>  #4  0x727135f0 in signal_emit_unlocked_R () at
> /lib64/libgobject-2.0.so.0
>  #5  0x7271c43c in g_signal_emit_valist () at
> /lib64/libgobject-2.0.so.0
>  #6  0x7788acca in emit_main_context (opaque=0x7fffbfffe880)
> at gio-coroutine.c:200
>  #7  0x7242bab3 in g_main_context_dispatch () at
> /lib64/libglib-2.0.so.0
>  #8  0x7242be60 in g_main_context_iterate.isra () at
> /lib64/libglib-2.0.so.0
>  #9  0x7242c182 in g_main_loop_run () at /lib64/libglib-
> 2.0.so.0
>  #10 0x00405df6 in main (argc=,
> argv=) at spicy.c:1897
> 
> It was pointed out during the review and, somehow, missed :-\
> 
> Signed-off-by: Fabiano Fidêncio <fiden...@redhat.com>
> Signed-off-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  src/vmcstream.c | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/src/vmcstream.c b/src/vmcstream.c
> index 48f9513..8a31a47 100644
> --- a/src/vmcstream.c
> +++ b/src/vmcstream.c
> @@ -184,7 +184,6 @@
> spice_vmc_input_stream_read_all_async(GInputStream*stream,
>  
>  /* no concurrent read permitted by ginputstream */
>  g_return_if_fail(self->task == NULL);
> -g_return_if_fail(g_task_get_cancellable(self->task) == NULL);
>  self->all = TRUE;
>  self->buffer = buffer;
>  self->count = count;
> @@ -236,7 +235,6 @@
> spice_vmc_input_stream_read_async(GInputStream*stream,
>  
>  /* no concurrent read permitted by ginputstream */
>  g_return_if_fail(self->task == NULL);
> -g_return_if_fail(g_task_get_cancellable(self->task) == NULL);
>  self->all = FALSE;
>  self->buffer = buffer;
>  self->count = count;
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-server] doc: Add virgl documentation

[Pavel Grunt]

On Wed, 2016-03-23 at 17:45 +0100, Christophe Fergeau wrote:
> On Wed, Mar 23, 2016 at 05:36:04PM +0100, Pavel Grunt wrote:
> > 
> > Hi Christophe,
> > 
> > On Wed, 2016-03-23 at 17:18 +0100, Christophe Fergeau wrote:
> > > 
> > > On Mon, Mar 21, 2016 at 03:28:41PM +0100, Christophe Fergeau
> > > wrote:
> > > > 
> > > > 
> > > > On Mon, Mar 21, 2016 at 02:57:18PM +0100, Pavel Grunt wrote:
> > > > > 
> > > > > 
> > > > > Thanks!
> > > > > 
> > > > > Acked-by: Pavel Grunt <pgr...@redhat.com>
> > > > > 
> > > > > On Fri, 2016-03-18 at 17:12 +0100, Christophe Fergeau wrote:
> > > > > > 
> > > > > > 
> > > > > > +GL acceleration (virgl)
> > > > > > +===
> > > > > > +
> > > > > > +OpenGL acceleration is currently local only (it has to go
> > > > > > through a
> > > > > > Unix socket)
> > > > > > +and it needs guest support. It's currently limited to
> > > > > > recent
> > > > > > linux
> > > > > > +distributions (for example up to date Fedora 23).
> > > > > > +
> > > > After running more tests, I realized that on a f23 guest, you
> > > > need
> > > > to
> > > > update mesa from Gerd's copr, I'll have to adjust the wording
> > > > here.
> > > I"ll change it to
> > > 
> > > -distributions (for example up to date Fedora 23).
> > > +distributions (for example up to date Fedora 23 + updated mesa
> > > packages from
> > > +copr:kraxel).
> > I would keep up to date Fedora 23. Mesa should have virgl support -
> > see http://koji.fedoraproject.org/koji/buildinfo?buildID=739036 
> But it seems these are good enough, when I tested this, Xorg was
> picking
> the QXL driver, which gave me a blackscreen with the f23 packages.
> With
> packages from kraxel's repository, the correct driver was picked up,
> and
> I could get a working guest display. Which is why I want to mention
> them
> now ;)
> 
ok maybe it deserves a bug report. I already acked the patch, but would
you mind adding a link to the copr repo [0] ?

Pavel

[0] https://copr.fedorainfracloud.org/coprs/kraxel/virgl/


___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 8/8] widget: modify update-area to not require primary surface

[Pavel Grunt]

Hi,

as we discussed, I prefer this patch to go in after 9/8 to avoid some
incorrect resizes

On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> If egl display is enabled, use GL scanout geometry to check
> intersection
> with the monitor area. This solves displaying GL display without
> software canvas.
What are the issues?

Pavel
> 
> Signed-off-by: Marc-André Lureau 
> ---
>  src/spice-widget.c | 41 -
>  1 file changed, 28 insertions(+), 13 deletions(-)
> 
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index e16126a..4b7e202 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -2190,32 +2190,47 @@ static void update_area(SpiceDisplay
> *display,
>  gint x, gint y, gint width, gint height)
>  {
>  SpiceDisplayPrivate *d = display->priv;
> -GdkRectangle primary = {
> -.x = 0,
> -.y = 0,
> -.width = d->width,
> -.height = d->height
> -};
> -GdkRectangle area = {
> +GdkRectangle primary;
> +
> +SPICE_DEBUG("update area +%d+%d %dx%d", x, y, width, height);
> +d->area = (GdkRectangle) {
>  .x = x,
>  .y = y,
>  .width = width,
>  .height = height
>  };
>  
> -SPICE_DEBUG("update area, primary: %dx%d, area: +%d+%d %dx%d",
> d->width, d->height, area.x, area.y, area.width, area.height);
> +#ifndef G_OS_WIN32
> +if (d->egl.enabled) {
> +primary = (GdkRectangle) {
> +.width = d->egl.scanout.width,
> +.height = d->egl.scanout.height
> +};
> +} else
> +#endif
> +{
> +primary = (GdkRectangle) {
> +.width = d->width,
> +.height = d->height
> +};
> +}
>  
> -if (!gdk_rectangle_intersect(, , )) {
> +SPICE_DEBUG("primary: %dx%d", primary.width, primary.height);
> +if (!gdk_rectangle_intersect(, >area, >area)) {
>  SPICE_DEBUG("The monitor area is not intersecting primary
> surface");
>  memset(>area, '\0', sizeof(d->area));
>  set_monitor_ready(display, false);
>  return;
>  }
>  
> -spicex_image_destroy(display);
> -d->area = area;
> -if (gtk_widget_get_realized(GTK_WIDGET(display)))
> -update_image(display);
> +#ifndef G_OS_WIN32
> +if (!d->egl.enabled)
> +#endif
> +{
> +spicex_image_destroy(display);
> +if (gtk_widget_get_realized(GTK_WIDGET(display)))
> +update_image(display);
> +}
>  
>  update_size_request(display);
>  
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 1/8] widget: remove superflous WIN32 check

[Pavel Grunt]

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> The egl call is already inside a X11 conditional block.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
> ---
>  src/spice-widget.c | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index e329809..c8bb72c 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -580,13 +580,11 @@ drawing_area_realize(GtkWidget *area, gpointer
> user_data)
>  g_clear_error();
>  }
>  
> -#ifndef G_OS_WIN32
>  if (!spice_egl_realize_display(display,
> gtk_widget_get_window(area), )) {
>  g_critical("egl realize failed: %s", err->message);
>  g_clear_error();
>  }
>  #endif
> -#endif
>  }
>  
>  static void spice_display_init(SpiceDisplay *display)
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 5/8] widget: make gl-scanout and update-monitor-area privately exported

[Pavel Grunt]

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> The following changes will call these functions from spice-widget-
> egl.c
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
> ---
>  src/spice-widget-priv.h |  3 +++
>  src/spice-widget.c  | 17 +++--
>  2 files changed, 14 insertions(+), 6 deletions(-)
> 
> diff --git a/src/spice-widget-priv.h b/src/spice-widget-priv.h
> index bafeb7d..95bca8a 100644
> --- a/src/spice-widget-priv.h
> +++ b/src/spice-widget-priv.h
> @@ -163,6 +163,9 @@ gboolean
> spice_egl_update_scanout(SpiceDisplay *display,
>    GError **err);
>  void spice_egl_cursor_set(SpiceDisplay
> *display);
>  
> +void spice_display_widget_gl_scanout (SpiceDisplay
> *display);
> +void spice_display_widget_update_monitor_area(SpiceDisplay
> *display);
> +
>  G_END_DECLS
>  
>  #endif
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index 198d0cd..f0f065c 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -264,7 +264,8 @@ static void set_monitor_ready(SpiceDisplay *self,
> gboolean ready)
>  update_ready(self);
>  }
>  
> -static void update_monitor_area(SpiceDisplay *display)
> +G_GNUC_INTERNAL
> +void spice_display_widget_update_monitor_area(SpiceDisplay *display)
>  {
>  SpiceDisplayPrivate *d = display->priv;
>  SpiceDisplayMonitorConfig *cfg, *c = NULL;
> @@ -339,7 +340,7 @@ static void
> spice_display_set_property(GObject  *object,
>  case PROP_MONITOR_ID:
>  d->monitor_id = g_value_get_int(value);
>  if (d->display) /* if constructed */
> -update_monitor_area(display);
> +spice_display_widget_update_monitor_area(display);
>  break;
>  case PROP_KEYBOARD_GRAB:
>  d->keyboard_grab_enable = g_value_get_boolean(value);
> @@ -2230,7 +2231,7 @@ static void primary_create(SpiceChannel
> *channel, gint format,
>  d->height = height;
>  d->data_origin = d->data = imgdata;
>  
> -update_monitor_area(display);
> +spice_display_widget_update_monitor_area(display);
>  }
>  
>  static void primary_destroy(SpiceChannel *channel, gpointer data)
> @@ -2480,7 +2481,8 @@ static void cursor_reset(SpiceCursorChannel
> *channel, gpointer data)
>  }
>  
>  #ifndef G_OS_WIN32
> -static void gl_scanout(SpiceDisplay *display)
> +G_GNUC_INTERNAL
> +void spice_display_widget_gl_scanout(SpiceDisplay *display)
>  {
>  SpiceDisplayPrivate *d = display->priv;
>  const SpiceGlScanout *scanout;
> @@ -2555,15 +2557,18 @@ static void channel_new(SpiceSession *s,
> SpiceChannel *channel, gpointer data)
>  spice_g_signal_connect_object(channel, "display-mark",
>    G_CALLBACK(mark), display,
> G_CONNECT_AFTER | G_CONNECT_SWAPPED);
>  spice_g_signal_connect_object(channel, "notify::monitors",
> -  G_CALLBACK(update_monitor_area
> ), display, G_CONNECT_AFTER | G_CONNECT_SWAPPED);
> +  G_CALLBACK(spice_display_widge
> t_update_monitor_area),
> +  display, G_CONNECT_AFTER |
> G_CONNECT_SWAPPED);
>  if (spice_display_get_primary(channel, 0, )) {
>  primary_create(channel, primary.format, primary.width,
> primary.height,
> primary.stride, primary.shmid,
> primary.data, display);
>  mark(display, primary.marked);
>  }
> +
>  #ifndef G_OS_WIN32
>  spice_g_signal_connect_object(channel, "notify::gl-scanout",
> -  G_CALLBACK(gl_scanout),
> display, G_CONNECT_SWAPPED);
> +  G_CALLBACK(spice_display_widge
> t_gl_scanout),
> +  display, G_CONNECT_SWAPPED);
>  spice_g_signal_connect_object(channel, "gl-draw",
>    G_CALLBACK(gl_draw), display,
> G_CONNECT_SWAPPED);
>  #endif
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 9/8] main: don't update display timer for unchanged config

[Pavel Grunt]

Hi,

So the timer is the main problem ?

It also solves/avoids problems with a guest running on wayland when the
"resize-guest" property is TRUE.
See bug https://bugzilla.redhat.com/show_bug.cgi?id=1266484 
It avoids destroying the primary surface when the display configuration
has not changed

I would like to hear from Jonathon what he thinks about it. There were
some concerns / discussion whether it should be compared to the "real"
state of displays in the guest.
https://lists.freedesktop.org/archives/spice-devel/2015-October/022785.
html

Maybe some check should be done on the server side because it is
"closer" to the guest.

Pavel

On Wed, 2016-03-23 at 18:02 +0100, Marc-André Lureau wrote:
> With virgl, set_monitor_ready() may be called each time the scanout
> is
> updated to set the monitor area. This will call
> spice_main_update_display(), and keep the timer postponed even if the
> monitor configuration didn't change. Treat unchanged configuration
> has a
> no-op and keep configuration timer unchanged. This fixes monitor
> autoconfig with virgl (when the display is regularly updated).
> 
> Signed-off-by: Marc-André Lureau 
> ---
>  src/channel-main.c | 29 ++---
>  1 file changed, 18 insertions(+), 11 deletions(-)
> 
> diff --git a/src/channel-main.c b/src/channel-main.c
> index 1c19de1..b4875f6 100644
> --- a/src/channel-main.c
> +++ b/src/channel-main.c
> @@ -121,6 +121,14 @@ typedef enum {
>  DISPLAY_ENABLED,
>  } SpiceDisplayState;
>  
> +typedef struct {
> +int x;
> +int y;
> +int width;
> +int height;
> +SpiceDisplayState   display_state;
> +} SpiceDisplayConfig;
> +
>  struct _SpiceMainChannelPrivate  {
>  enum SpiceMouseMode mouse_mode;
>  boolagent_connected;
> @@ -140,13 +148,7 @@ struct _SpiceMainChannelPrivate  {
>  guint   agent_msg_pos;
>  uint8_t agent_msg_size;
>  uint32_tagent_caps[VD_AGENT_CAPS_SIZE];
> -struct {
> -int x;
> -int y;
> -int width;
> -int height;
> -SpiceDisplayState   display_state;
> -} display[MAX_DISPLAY];
> +SpiceDisplayConfig  display[MAX_DISPLAY];
>  ginttimer_id;
>  GQueue  *agent_msg_queue;
>  GHashTable  *file_xfer_tasks;
> @@ -2686,10 +2688,15 @@ void
> spice_main_update_display(SpiceMainChannel *channel, int id,
>  
>  g_return_if_fail(id < SPICE_N_ELEMENTS(c->display));
>  
> -c->display[id].x  = x;
> -c->display[id].y  = y;
> -c->display[id].width  = width;
> -c->display[id].height = height;
> +SpiceDisplayConfig display = {
> +.x = x, .y = y, .width = width, .height = height,
> +.display_state = c->display[id].display_state
> +};
> +
> +if (memcmp(, >display[id],
> sizeof(SpiceDisplayConfig)) == 0)
> +return;
> +
> +c->display[id] = display;
>  
>  if (update)
>  update_display_timer(channel, 1);
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 4/8] display: return NULL in spice_display_get_gl_scanout()

[Pavel Grunt]

Hi Marc-André,

On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> If there is no valid scanout, return NULL.

When / how it can happen ? Looking at gl_scanout() there is g_return if
scanout == NULL, if NULL is a valid/expected return value then that
g_return should be changed.

Pavel
> 
> Signed-off-by: Marc-André Lureau 
> ---
>  src/channel-display.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/src/channel-display.c b/src/channel-display.c
> index 2c54df7..90d155e 100644
> --- a/src/channel-display.c
> +++ b/src/channel-display.c
> @@ -532,7 +532,7 @@ void
> spice_display_change_preferred_compression(SpiceChannel *channel,
> gint comp
>   * spice_display_get_gl_scanout:
>   * @channel: a #SpiceDisplayChannel
>   *
> - * Returns: the current GL scanout
> + * Returns: the current GL scanout, or %NULL if none or not valid
>   *
>   * Since: 0.31
>   **/
> @@ -541,7 +541,7 @@ spice_display_get_gl_scanout(SpiceDisplayChannel
> *channel)
>  {
>  g_return_val_if_fail(SPICE_IS_DISPLAY_CHANNEL(channel), NULL);
>  
> -return >priv->scanout;
> +return channel->priv->scanout.fd != -1 ? >priv->scanout 
> : NULL;
>  }
>  
>  /* ---
> --- */
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 3/8] widget: enable egl before updating scanout

[Pavel Grunt]

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> The GLArea is realized lazily, when it is made visible in
> set_egl_enabled(). The egl context is initialized once the GLArea is
> realized. Enable egl before updating the scanout.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
> ---
>  src/spice-widget.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index 8518ec6..198d0cd 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -2486,14 +2486,14 @@ static void gl_scanout(SpiceDisplay *display)
>  const SpiceGlScanout *scanout;
>  GError *err = NULL;
>  
> +SPICE_DEBUG("%s: got scanout",  __FUNCTION__);
> +set_egl_enabled(display, true);
> +
>  g_return_if_fail(d->egl.context_ready);
>  
>  scanout = spice_display_get_gl_scanout(SPICE_DISPLAY_CHANNEL(d-
> >display));
>  g_return_if_fail(scanout != NULL);
>  
> -SPICE_DEBUG("%s: got scanout",  __FUNCTION__);
> -set_egl_enabled(display, true);
> -
>  if (!spice_egl_update_scanout(display, scanout, )) {
>  g_critical("update scanout failed: %s", err->message);
>  g_clear_error();
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 2/8] egl: check context is ready

[Pavel Grunt]

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> Add a new GL status field to check if the GL context is ready. This
> helps debugging races where GL is called before the context is ready.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
> ---
>  src/spice-widget-egl.c  | 10 --
>  src/spice-widget-priv.h |  1 +
>  src/spice-widget.c  |  5 +
>  3 files changed, 14 insertions(+), 2 deletions(-)
> 
> diff --git a/src/spice-widget-egl.c b/src/spice-widget-egl.c
> index a9bea52..8ac3b90 100644
> --- a/src/spice-widget-egl.c
> +++ b/src/spice-widget-egl.c
> @@ -208,7 +208,7 @@ gboolean spice_egl_init(SpiceDisplay *display,
> GError **err)
>  d->egl.ctx = eglGetCurrentContext();
>  dpy =
> (EGLNativeDisplayType)gdk_wayland_display_get_wl_display(gdk_dpy);
>  d->egl.display = eglGetDisplay(dpy);
> -return spice_egl_init_shaders(display, err);
> +goto end;
>  }
>  #endif
>  #ifdef GDK_WINDOWING_X11
> @@ -267,7 +267,13 @@ gboolean spice_egl_init(SpiceDisplay *display,
> GError **err)
>  eglMakeCurrent(d->egl.display, EGL_NO_SURFACE, EGL_NO_SURFACE,
> d->egl.ctx);
>  
> -return spice_egl_init_shaders(display, err);
> +end:
> +if (!spice_egl_init_shaders(display, err))
> +return FALSE;
> +
> +d->egl.context_ready = TRUE;
> +
> +return TRUE;
>  }
>  
>  static gboolean spice_widget_init_egl_win(SpiceDisplay *display,
> GdkWindow *win,
> diff --git a/src/spice-widget-priv.h b/src/spice-widget-priv.h
> index 2f7e39e..bafeb7d 100644
> --- a/src/spice-widget-priv.h
> +++ b/src/spice-widget-priv.h
> @@ -129,6 +129,7 @@ struct _SpiceDisplayPrivate {
>  #endif
>  #ifndef G_OS_WIN32
>  struct {
> +gbooleancontext_ready;
>  gbooleanenabled;
>  EGLSurface  surface;
>  EGLDisplay  display;
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index c8bb72c..8518ec6 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -2486,6 +2486,8 @@ static void gl_scanout(SpiceDisplay *display)
>  const SpiceGlScanout *scanout;
>  GError *err = NULL;
>  
> +g_return_if_fail(d->egl.context_ready);
> +
>  scanout = spice_display_get_gl_scanout(SPICE_DISPLAY_CHANNEL(d-
> >display));
>  g_return_if_fail(scanout != NULL);
>  
> @@ -2504,8 +2506,11 @@ static void gl_draw(SpiceDisplay *display,
>  SpiceDisplayPrivate *d = display->priv;
>  
>  SPICE_DEBUG("%s",  __FUNCTION__);
> +
>  set_egl_enabled(display, true);
>  
> +g_return_if_fail(d->egl.context_ready);
> +
>  #if GTK_CHECK_VERSION(3,16,0)
>  GtkWidget *gl = gtk_stack_get_child_by_name(GTK_STACK(display),
> "gl-area");
>  
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 7/8] widget: update d->ready based on scanout image available

[Pavel Grunt]

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> d->ready is updated based on monitor area & canvas. In case of GL
> rendering, update it based on monitor area & scanout and add check
> before drawing update.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
> ---
>  src/spice-widget-egl.c | 6 ++
>  src/spice-widget.c | 6 +-
>  2 files changed, 7 insertions(+), 5 deletions(-)
> 
> diff --git a/src/spice-widget-egl.c b/src/spice-widget-egl.c
> index d6f436d..5fcdcaa 100644
> --- a/src/spice-widget-egl.c
> +++ b/src/spice-widget-egl.c
> @@ -396,7 +396,7 @@ void spice_egl_resize_display(SpiceDisplay
> *display, int w, int h)
>  apply_ortho(d->egl.mproj, 0, w, 0, h, -1, 1);
>  glViewport(0, 0, w, h);
>  
> -if (d->egl.image)
> +if (d->ready)
>  spice_egl_update_display(display);
>  
>  glUseProgram(prog);
> @@ -506,9 +506,7 @@ void spice_egl_update_display(SpiceDisplay
> *display)
>  gdouble tx, ty, tw, th;
>  int prog;
>  
> -g_return_if_fail(d->egl.image != NULL);
> -if (d->egl.surface == EGL_NO_SURFACE)
> -return;
> +g_return_if_fail(d->ready);
>  
>  spice_display_get_scaling(display, , , , , );
>  
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index f0f065c..e16126a 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -234,7 +234,11 @@ static void update_ready(SpiceDisplay *display)
>  SpiceDisplayPrivate *d = display->priv;
>  gboolean ready;
>  
> -ready = d->mark != 0 && d->monitor_ready;
> +ready = d->monitor_ready &&
> +#ifndef G_OS_WIN32
> +d->egl.enabled ? d->egl.image != NULL :
> +#endif
> +d->mark;
>  
>  /* If the 'resize-guest' property is set, the application
> expects spice-gtk
>   * to manage the size and state of the displays, so update the
> 'enabled'
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 6/8] egl: check and update scanout after egl init

[Pavel Grunt]

Hi,

On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> Once the egl/GL context are initialized, check if there is a scanout
> to
> associate to display widget. This solves races when scanout update is
> happening before the egl/GL context is ready.
> 
Is it common that the display channel has the scanout before the widget
is initialized ? What about adding a debug when it happens?

> Signed-off-by: Marc-André Lureau 
> ---
>  src/spice-widget-egl.c | 5 +
>  1 file changed, 5 insertions(+)
> 
> diff --git a/src/spice-widget-egl.c b/src/spice-widget-egl.c
> index 8ac3b90..d6f436d 100644
> --- a/src/spice-widget-egl.c
> +++ b/src/spice-widget-egl.c
> @@ -273,6 +273,11 @@ end:
>  
>  d->egl.context_ready = TRUE;
>  
> +if (spice_display_get_gl_scanout(SPICE_DISPLAY_CHANNEL(d-
> >display))) {
I prefer explicit comparison to NULL 
> +spice_display_widget_gl_scanout(display);
> +spice_display_widget_update_monitor_area(display);
> +}
> +
>  return TRUE;
>  }
>  

Thanks,
Pavel

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-gtk 0/2] Introduce gtask-helper.[ch]

[Pavel Grunt]

Hi Fabiano,

On Wed, 2016-03-23 at 10:48 +0100, Fabiano Fidêncio wrote:
> The first patch is only about reverting a patch that, even if it
> fixed
> the problem, it wasn't done in the proper way.
> The second patch introduces and uses new helpers for GTask, providing
> them functions that will explicitely return in idle instead of
> leaving
> the decision to the GTask heurestic (which doesn't work with
> Coroutine).
> 
> Fabiano Fidêncio (2):
>   Revert "channel-usbredir: Fix crash due to a Task returning earlier
> than expected"
>   Introduce gtask-helper.[ch]
> 

you should the order of the patches otherwise the crash would be
reintroduced

Pavel
>  m4/manywarnings.m4   |   1 -
>  src/Makefile.am  |   2 +
>  src/channel-base.c   |   4 +-
>  src/channel-main.c   |  12 ++--
>  src/channel-usbredir.c   |  12 ++--
>  src/gtask-helper.c   | 152
> +++
>  src/gtask-helper.h   |  43 
>  src/spice-channel.c  |   4 +-
>  src/spice-gstaudio.c |   6 +-
>  src/usb-acl-helper.c |  12 ++--
>  src/usb-device-manager.c |  18 ++---
>  src/vmcstream.c  |  35 ++
>  src/win-usb-driver-install.c |  21 +++---
>  13 files changed, 254 insertions(+), 68 deletions(-)
>  create mode 100644 src/gtask-helper.c
>  create mode 100644 src/gtask-helper.h
> 
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 4/8] display: return NULL in spice_display_get_gl_scanout()

[Pavel Grunt]

On Thu, 2016-03-24 at 11:59 +0100, Marc-André Lureau wrote:
> On Thu, Mar 24, 2016 at 12:39 AM, Marc-André Lureau  om> wrote:
> > 
> > > 
> > > When / how it can happen ? Looking at gl_scanout() there is
> > > g_return if
> > > scanout == NULL, if NULL is a valid/expected return value then
> > > that
> > > g_return should be changed.
> > Good point. AI think I wrote this initially thinking that no one
> > should ever call get_gl_scanout() without having first received the
> > scanout notification, but that assumption is wrong. It should be
> > fine to call get_gl_scanout() and it's easier to deal with a NULL
> > return value rather than checking the returned scanout.fd ==
> > -1.  I'll change the assert. (btw, the set_static_boxed for the
> > property should be fine)
> 
> Actually, gl_scanout() should only be called with a valid scanout,
> the
> patch doesn't need to be modified then. If you prefer, we can add a
> comment above the g_return_if_fail(scanout).
> 
Yes please

Pavel

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH 3/5] removed unused tree debugging function

[Pavel Grunt]

Removed in

*39e1cbe3211b06bca6e5d906ffb4d0582440c5cc*


I would ack it but Fabiano had some comments about it

https://lists.freedesktop.org/archives/spice-devel/2015-November/024060.html



On Thu, Mar 24, 2016 at 12:45 PM, Victor Toso  wrote:

> Hi,
>
> On Thu, Mar 24, 2016 at 08:53:00AM +, Frediano Ziglio wrote:
> > Support for this debugging was removed time ago.
> >
>
> Can you point it out when it was removed?
> Thanks,
>
> > Signed-off-by: Frediano Ziglio 
> > ---
> >  server/tree.c | 161
> --
> >  server/tree.h |   1 -
> >  2 files changed, 162 deletions(-)
> >
> > diff --git a/server/tree.c b/server/tree.c
> > index 9e5a281..3866616 100644
> > --- a/server/tree.c
> > +++ b/server/tree.c
> > @@ -24,167 +24,6 @@
> >
> >  #include "tree.h"
> >
> > -static const char *draw_type_to_str(uint8_t type)
> > -{
> > -switch (type) {
> > -case QXL_DRAW_FILL:
> > -return "QXL_DRAW_FILL";
> > -case QXL_DRAW_OPAQUE:
> > -return "QXL_DRAW_OPAQUE";
> > -case QXL_DRAW_COPY:
> > -return "QXL_DRAW_COPY";
> > -case QXL_DRAW_TRANSPARENT:
> > -return "QXL_DRAW_TRANSPARENT";
> > -case QXL_DRAW_ALPHA_BLEND:
> > -return "QXL_DRAW_ALPHA_BLEND";
> > -case QXL_COPY_BITS:
> > -return "QXL_COPY_BITS";
> > -case QXL_DRAW_BLEND:
> > -return "QXL_DRAW_BLEND";
> > -case QXL_DRAW_BLACKNESS:
> > -return "QXL_DRAW_BLACKNESS";
> > -case QXL_DRAW_WHITENESS:
> > -return "QXL_DRAW_WHITENESS";
> > -case QXL_DRAW_INVERS:
> > -return "QXL_DRAW_INVERS";
> > -case QXL_DRAW_ROP3:
> > -return "QXL_DRAW_ROP3";
> > -case QXL_DRAW_COMPOSITE:
> > -return "QXL_DRAW_COMPOSITE";
> > -case QXL_DRAW_STROKE:
> > -return "QXL_DRAW_STROKE";
> > -case QXL_DRAW_TEXT:
> > -return "QXL_DRAW_TEXT";
> > -default:
> > -return "?";
> > -}
> > -}
> > -
> > -static void show_red_drawable(RedDrawable *drawable, const char *prefix)
> > -{
> > -if (prefix) {
> > -printf("%s: ", prefix);
> > -}
> > -
> > -printf("%s effect %d bbox(%d %d %d %d)",
> > -   draw_type_to_str(drawable->type),
> > -   drawable->effect,
> > -   drawable->bbox.top,
> > -   drawable->bbox.left,
> > -   drawable->bbox.bottom,
> > -   drawable->bbox.right);
> > -
> > -switch (drawable->type) {
> > -case QXL_DRAW_FILL:
> > -case QXL_DRAW_OPAQUE:
> > -case QXL_DRAW_COPY:
> > -case QXL_DRAW_TRANSPARENT:
> > -case QXL_DRAW_ALPHA_BLEND:
> > -case QXL_COPY_BITS:
> > -case QXL_DRAW_BLEND:
> > -case QXL_DRAW_BLACKNESS:
> > -case QXL_DRAW_WHITENESS:
> > -case QXL_DRAW_INVERS:
> > -case QXL_DRAW_ROP3:
> > -case QXL_DRAW_COMPOSITE:
> > -case QXL_DRAW_STROKE:
> > -case QXL_DRAW_TEXT:
> > -break;
> > -default:
> > -spice_error("bad drawable type");
> > -}
> > -printf("\n");
> > -}
> > -
> > -static void show_draw_item(DrawItem *draw_item, const char *prefix)
> > -{
> > -if (prefix) {
> > -printf("%s: ", prefix);
> > -}
> > -printf("effect %d bbox(%d %d %d %d)\n",
> > -   draw_item->effect,
> > -   draw_item->base.rgn.extents.x1,
> > -   draw_item->base.rgn.extents.y1,
> > -   draw_item->base.rgn.extents.x2,
> > -   draw_item->base.rgn.extents.y2);
> > -}
> > -
> > -typedef struct DumpItem {
> > -int level;
> > -Container *container;
> > -} DumpItem;
> > -
> > -static void dump_item(TreeItem *item, void *data)
> > -{
> > -DumpItem *di = data;
> > -const char *item_prefix = "|--";
> > -int i;
> > -
> > -if (di->container) {
> > -while (di->container != item->container) {
> > -di->level--;
> > -di->container = di->container->base.container;
> > -}
> > -}
> > -
> > -switch (item->type) {
> > -case TREE_ITEM_TYPE_DRAWABLE: {
> > -Drawable *drawable = SPICE_CONTAINEROF(item, Drawable,
> tree_item.base);
> > -const int max_indent = 200;
> > -char indent_str[max_indent + 1];
> > -int indent_str_len;
> > -
> > -for (i = 0; i < di->level; i++) {
> > -printf("  ");
> > -}
> > -printf(item_prefix, 0);
> > -show_red_drawable(drawable->red_drawable, NULL);
> > -for (i = 0; i < di->level; i++) {
> > -printf("  ");
> > -}
> > -printf("|  ");
> > -show_draw_item(>tree_item, NULL);
> > -indent_str_len = MIN(max_indent, strlen(item_prefix) +
> di->level * 2);
> > -memset(indent_str, ' ', indent_str_len);
> > -indent_str[indent_str_len] = 0;
> > -region_dump(>rgn, indent_str);
> > -printf("\n");
> > -break;
> > -}
> > -case TREE_ITEM_TYPE_CONTAINER:
> > -di->level++;
> > -   

[Spice-devel] [spice v11 05/27] replay: Add an option to change video codec

[Pavel Grunt]

---
 server/tests/replay.c | 11 ++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/server/tests/replay.c b/server/tests/replay.c
index 7e4659b..025b20c 100644
--- a/server/tests/replay.c
+++ b/server/tests/replay.c
@@ -290,7 +290,7 @@ int main(int argc, char **argv)
 {
 GError *error = NULL;
 GOptionContext *context = NULL;
-gchar *client = NULL, **file = NULL;
+gchar *client = NULL, *codecs = NULL, **file = NULL;
 gint port = 5000, compression = SPICE_IMAGE_COMPRESSION_AUTO_GLZ;
 gint streaming = SPICE_STREAM_VIDEO_FILTER;
 gboolean wait = FALSE;
@@ -300,6 +300,7 @@ int main(int argc, char **argv)
 { "client", 'c', 0, G_OPTION_ARG_STRING, , "Client", "CMD" },
 { "compression", 'C', 0, G_OPTION_ARG_INT, , "Compression 
(default 2)", "INT" },
 { "streaming", 'S', 0, G_OPTION_ARG_INT, , "Streaming 
(default 3)", "INT" },
+{ "video-codecs", 'v', 0, G_OPTION_ARG_STRING, , "Video 
codecs", "STRING" },
 { "port", 'p', 0, G_OPTION_ARG_INT, , "Server port (default 
5000)", "PORT" },
 { "wait", 'w', 0, G_OPTION_ARG_NONE, , "Wait for client", NULL },
 { "slow", 's', 0, G_OPTION_ARG_INT, , "Slow down replay. Delays 
USEC microseconds before each command", "USEC" },
@@ -386,6 +387,14 @@ int main(int argc, char **argv)
 server = spice_server_new();
 spice_server_set_image_compression(server, compression);
 spice_server_set_streaming_video(server, streaming);
+
+if (codecs != NULL) {
+if (spice_server_set_video_codecs(server, codecs) != 0) {
+g_warning("could not set codecs: %s", codecs);
+}
+g_free(codecs);
+}
+
 spice_server_set_port(server, port);
 spice_server_set_noauth(server);
 
-- 
2.7.0

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk] usb-device-manager: Remove invalid return annotation

[Pavel Grunt]

---
 src/usb-device-manager.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/src/usb-device-manager.c b/src/usb-device-manager.c
index 417c5b2..85231a1 100644
--- a/src/usb-device-manager.c
+++ b/src/usb-device-manager.c
@@ -1568,8 +1568,6 @@ gboolean spice_usb_device_manager_connect_device_finish(
  * @device: a #SpiceUsbDevice to disconnect
  *
  * Disconnects the @device.
- *
- * Returns: %TRUE if @device has an associated USB redirection channel
  */
 void spice_usb_device_manager_disconnect_device(SpiceUsbDeviceManager *self,
 SpiceUsbDevice *device)
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-server] doc: Add virgl documentation

[Pavel Grunt]

Thanks!

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Fri, 2016-03-18 at 17:12 +0100, Christophe Fergeau wrote:
> Document how to use virgl with QEMU/libvirt
> 
> virt-manager documentation still needs to be added.
> ---
>  docs/manual/manual.txt | 55
> ++
>  1 file changed, 55 insertions(+)
> 
> diff --git a/docs/manual/manual.txt b/docs/manual/manual.txt
> index 2c0035a..a89a702 100644
> --- a/docs/manual/manual.txt
> +++ b/docs/manual/manual.txt
> @@ -901,6 +901,61 @@ folder will show up in GNOME Files network
> places (or Nautilus). It
>  can then be mounted and browsed in traditional applications thanks
> to
>  `gvfs-fuse`.
>  
> +GL acceleration (virgl)
> +===
> +
> +OpenGL acceleration is currently local only (it has to go through a
> Unix socket)
> +and it needs guest support. It's currently limited to recent linux
> +distributions (for example up to date Fedora 23).
> +
> +Configuration
> +-
> +
> +.Using libvirt
> +
> +You need to add a virtio-gpu video device to your virtual machine
> instead of QXL.
> +[source,xml]
> +
> +  
> +
> +  
> +
> +
> +Then you need to enable OpenGL on your SPICE graphics node:
> +[source,xml]
> +
> +  
> +
> +
> +You don't need any port/address as they won't be usable with
> +GL.
> +
> +.Using QEMU
> +
> +You need to add a virtio-gpu device on QEMU command line,
> +as well as enable GL with SPICE. port/tls-port/addr arguments
> +won't be used in this setup. You need to configure a Unix socket to
> +connect to the VM display.
> +
> +[source,sh]
> +-device virtio-vga,virgl=on -spice
> gl=on,unix,addr=/run/user/1000/spice.sock
> +
> +Connecting to the guest
> +---
> +
> +Connecting to the guest when virgl is in use is slightly different
> +than usual
> +
> +.If libvirt is being used
> +
> +[source,sh]
> +client$ virt-viewer -a $vmname
> +
> +.If a Unix socket has been set on QEMU command line
> +
> +[source,sh]
> +client$ remote-viewer spice+unix:///run/user/1000/spice.sock
> +
>  QEMU Spice reference
>  
>  
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk] Remove extra checks before g_cancellable_disconnect()

[Pavel Grunt]

If cancellable is NULL or handler_id is 0 the function does nothing.
---
 src/spice-pulse.c |  4 +---
 src/vmcstream.c   | 15 +++
 2 files changed, 4 insertions(+), 15 deletions(-)

diff --git a/src/spice-pulse.c b/src/spice-pulse.c
index 8d45dbe..6bc3014 100644
--- a/src/spice-pulse.c
+++ b/src/spice-pulse.c
@@ -945,9 +945,7 @@ static gboolean free_async_task(gpointer user_data)
 if (task->pa_op != NULL)
 pa_operation_unref(task->pa_op);
 
-if (task->cancel_id != 0)
-g_cancellable_disconnect(g_task_get_cancellable(task->gtask),
- task->cancel_id);
+g_cancellable_disconnect(g_task_get_cancellable(task->gtask), 
task->cancel_id);
 
 if (task->gtask)
 g_object_unref(task->gtask);
diff --git a/src/vmcstream.c b/src/vmcstream.c
index 48f9513..a206961 100644
--- a/src/vmcstream.c
+++ b/src/vmcstream.c
@@ -116,7 +116,6 @@ spice_vmc_input_stream_co_data(SpiceVmcInputStream *self,
 self->coroutine = coroutine_self();
 
 while (size > 0) {
-GCancellable *cancellable;
 SPICE_DEBUG("spicevmc co_data %p", self->task);
 if (!self->task)
 coroutine_yield(NULL);
@@ -140,9 +139,7 @@ spice_vmc_input_stream_co_data(SpiceVmcInputStream *self,
 
 g_task_return_int(self->task, self->pos);
 
-cancellable = g_task_get_cancellable(self->task);
-if (cancellable)
-g_cancellable_disconnect(cancellable, self->cancel_id);
+g_cancellable_disconnect(g_task_get_cancellable(self->task), 
self->cancel_id);
 
 g_clear_object(>task);
 }
@@ -208,16 +205,13 @@ spice_vmc_input_stream_read_all_finish(GInputStream 
*stream,
GError **error)
 {
 GTask *task = G_TASK(result);
-GCancellable *cancellable;
 SpiceVmcInputStream *self = SPICE_VMC_INPUT_STREAM(stream);
 
 g_return_val_if_fail(g_task_is_valid(task, self), -1);
 
 /* FIXME: calling _finish() is required. Disconnecting in
read_cancelled() causes a deadlock. #705395 */
-cancellable = g_task_get_cancellable(task);
-if (cancellable)
-g_cancellable_disconnect(cancellable, self->cancel_id);
+g_cancellable_disconnect(g_task_get_cancellable(task), self->cancel_id);
 
 return g_task_propagate_int(task, error);
 }
@@ -258,16 +252,13 @@ spice_vmc_input_stream_read_finish(GInputStream *stream,
GError **error)
 {
 GTask *task = G_TASK(result);
-GCancellable *cancellable;
 SpiceVmcInputStream *self = SPICE_VMC_INPUT_STREAM(stream);
 
 g_return_val_if_fail(g_task_is_valid(task, self), -1);
 
 /* FIXME: calling _finish() is required. Disconnecting in
read_cancelled() causes a deadlock. #705395 */
-cancellable = g_task_get_cancellable(task);
-if (cancellable)
-g_cancellable_disconnect(cancellable, self->cancel_id);
+g_cancellable_disconnect(g_task_get_cancellable(task), self->cancel_id);
 
 return g_task_propagate_int(task, error);
 }
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-gtk] Avoid to access array with negative numbers

[Pavel Grunt]

Hi,

according to GdkEventButton documentation [0] the 'button' is guint so
the button_gdk_to_spice() should take guint instead of int.

Pavel

[0] https://developer.gnome.org/gdk3/stable/gdk3-Event-Structures.html#
GdkEventButton

On Wed, 2016-03-23 at 15:28 +, Frediano Ziglio wrote:
> ---
>  src/spice-widget.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index 19753e7..d2fd789 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -1549,7 +1549,7 @@ static int button_gdk_to_spice(int gdk)
>  [ 5 ] = SPICE_MOUSE_BUTTON_DOWN,
>  };
>  
> -if (gdk < SPICE_N_ELEMENTS(map)) {
> +if (gdk >= 0 && gdk < SPICE_N_ELEMENTS(map)) {
>  return map [ gdk ];
>  }
>  return 0;
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH 2/2] use macro to define constructor function

[Pavel Grunt]

On Mon, 2016-02-29 at 10:36 +, Frediano Ziglio wrote:
> Avoid having to call function at runtime to inialize static.
> Old functions are defined as deprecated for compatibility.
> 
> Signed-off-by: Frediano Ziglio <fzig...@redhat.com>
Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  common/gdi_canvas.c |  2 --
>  common/gl_canvas.c  |  2 --
>  common/quic.c   | 11 ++-
>  common/quic.h   |  2 +-
>  common/rop3.c   |  8 +---
>  common/rop3.h   |  2 +-
>  common/sw_canvas.c  | 12 +---
>  common/sw_canvas.h  |  2 +-
>  8 files changed, 7 insertions(+), 34 deletions(-)
> 
> diff --git a/common/gdi_canvas.c b/common/gdi_canvas.c
> index 3c434d6..fcc5fc9 100644
> --- a/common/gdi_canvas.c
> +++ b/common/gdi_canvas.c
> @@ -1857,6 +1857,4 @@ void gdi_canvas_init(void) //unsafe global
> function
>  gdi_canvas_ops.put_image = gdi_canvas_put_image;
>  gdi_canvas_ops.clear = gdi_canvas_clear;
>  gdi_canvas_ops.destroy = gdi_canvas_destroy;
> -
> -rop3_init();
>  }
> diff --git a/common/gl_canvas.c b/common/gl_canvas.c
> index 9c5b98c..5ffb47b 100644
> --- a/common/gl_canvas.c
> +++ b/common/gl_canvas.c
> @@ -908,6 +908,4 @@ void gl_canvas_init(void) //unsafe global
> function
>  gl_canvas_ops.group_start = gl_canvas_group_start;
>  gl_canvas_ops.group_end = gl_canvas_group_end;
>  gl_canvas_ops.destroy = gl_canvas_destroy;
> -
> -rop3_init();
>  }
> diff --git a/common/quic.c b/common/quic.c
> index 88cf143..f014a2a 100644
> --- a/common/quic.c
> +++ b/common/quic.c
> @@ -1634,13 +1634,11 @@ int quic_decode(QuicContext *quic,
> QuicImageType type, uint8_t *buf, int stride)
>  return QUIC_OK;
>  }
>  
> -static int need_init = TRUE;
> -
>  QuicContext *quic_create(QuicUsrContext *usr)
>  {
>  Encoder *encoder;
>  
> -if (!usr || need_init || !usr->error || !usr->warn || !usr->info 
> || !usr->malloc ||
> +if (!usr || !usr->error || !usr->warn || !usr->info || !usr-
> >malloc ||
>  !usr->free || !usr->more_space || !usr->more_lines) {
>  return NULL;
>  }
> @@ -1671,13 +1669,8 @@ void quic_destroy(QuicContext *quic)
>  encoder->usr->free(encoder->usr, encoder);
>  }
>  
> -void quic_init(void)
> +SPICE_CONSTRUCTOR_FUNC(quic_global_init)
>  {
> -if (!need_init) {
> -return;
> -}
> -need_init = FALSE;
> -
>  family_init(_8bpc, 8, DEFmaxclen);
>  family_init(_5bpc, 5, DEFmaxclen);
>  }
> diff --git a/common/quic.h b/common/quic.h
> index ee5fb17..97622b3 100644
> --- a/common/quic.h
> +++ b/common/quic.h
> @@ -63,7 +63,7 @@ int quic_decode(QuicContext *quic, QuicImageType
> type, uint8_t *buf, int stride)
>  QuicContext *quic_create(QuicUsrContext *usr);
>  void quic_destroy(QuicContext *quic);
>  
> -void quic_init(void);
> +static inline SPICE_GNUC_DEPRECATED void quic_init(void) {}
>  
>  SPICE_END_DECLS
>  
> diff --git a/common/rop3.c b/common/rop3.c
> index 1a18fe9..2a0a167 100644
> --- a/common/rop3.c
> +++ b/common/rop3.c
> @@ -376,16 +376,10 @@ ROP3_HANDLERS(DPSoo, *src | *pat | *dest,
> 0xfe);
>  rop3_test_handlers_32[index] = rop3_test32_##op; \
>  rop3_test_handlers_16[index] = rop3_test16_##op;
>  
> -void rop3_init(void)
> +SPICE_CONSTRUCTOR_FUNC(rop3_global_init)
>  {
> -static int need_init = 1;
>  int i;
>  
> -if (!need_init) {
> -return;
> -}
> -need_init = 0;
> -
>  for (i = 0; i < ROP3_NUM_OPS; i++) {
>  rop3_with_pattern_handlers_32[i] =
> default_rop3_with_pattern_handler;
>  rop3_with_pattern_handlers_16[i] =
> default_rop3_with_pattern_handler;
> diff --git a/common/rop3.h b/common/rop3.h
> index 7e00499..20e100c 100644
> --- a/common/rop3.h
> +++ b/common/rop3.h
> @@ -32,7 +32,7 @@ void do_rop3_with_pattern(uint8_t rop3,
> pixman_image_t *d, pixman_image_t *s, Sp
>  void do_rop3_with_color(uint8_t rop3, pixman_image_t *d,
> pixman_image_t *s, SpicePoint *src_pos,
>  uint32_t rgb);
>  
> -void rop3_init(void);
> +static inline SPICE_GNUC_DEPRECATED void rop3_init(void) {}
>  
>  SPICE_END_DECLS
>  
> diff --git a/common/sw_canvas.c b/common/sw_canvas.c
> index a43cee4..e8b0f3b 100644
> --- a/common/sw_canvas.c
> +++ b/common/sw_canvas.c
> @@ -1188,7 +1188,6 @@ static void canvas_destroy(SpiceCanvas
> *spice_canvas)
>  free(canvas);
>  }
>  
> -static int need_init = 1;
>  static SpiceCanvasOps sw_canvas_ops;
>  
>  static SpiceCanvas *canvas_create_common(pixman_image_t *image,
&

Re: [Spice-devel] [spice-gtk v2] Avoid to access array with negative numbers

[Pavel Grunt]

On Wed, 2016-03-23 at 15:40 +, Frediano Ziglio wrote:

Acked-by: Pavel Grunt <pgr...@redhat.com>

> ---
>  src/spice-widget.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> Changes from v1:
> - make parameter guint as GdkEventButton::button is guint.
> 
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index 19753e7..b1a9423 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -1539,7 +1539,7 @@ static gboolean focus_out_event(GtkWidget
> *widget, GdkEventFocus *focus G_GNUC_U
>  return true;
>  }
>  
> -static int button_gdk_to_spice(int gdk)
> +static int button_gdk_to_spice(guint gdk)
>  {
>  static const int map[] = {
>  [ 1 ] = SPICE_MOUSE_BUTTON_LEFT,
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH] use constructor/destructor macros

[Pavel Grunt]

On Wed, 2016-03-23 at 15:23 +, Frediano Ziglio wrote:
> Initialize global images at load time:
> - make sure these stuff are initialized without using runtime
> resources;
> - make sure stuff are initialize really earlier.
> Do not use atexit function but a destructor.
> 
> Signed-off-by: Frediano Ziglio <fzig...@redhat.com>
Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  server/red-qxl.c | 7 ---
>  server/reds.c| 7 +--
>  spice-common | 2 +-
>  3 files changed, 2 insertions(+), 14 deletions(-)
> 
> diff --git a/server/red-qxl.c b/server/red-qxl.c
> index 3fd5c12..fbce510 100644
> --- a/server/red-qxl.c
> +++ b/server/red-qxl.c
> @@ -961,13 +961,6 @@ void red_qxl_init(RedsState *reds, QXLInstance
> *qxl)
>  
>  spice_return_if_fail(qxl != NULL);
>  
> -static gsize initialized = FALSE;
> -if (g_once_init_enter()) {
> -quic_init();
> -sw_canvas_init();
> -g_once_init_leave(, TRUE);
> -}
> -
>  qxl_state = spice_new0(QXLState, 1);
>  qxl_state->reds = reds;
>  qxl_state->qxl = qxl;
> diff --git a/server/reds.c b/server/reds.c
> index 254840b..38a369c 100644
> --- a/server/reds.c
> +++ b/server/reds.c
> @@ -2801,7 +2801,7 @@ static void reds_cleanup(RedsState *reds)
>  #endif
>  }
>  
> -static void reds_exit(void)
> +SPICE_DESTRUCTOR_FUNC(reds_exit)
>  {
>  GList *l;
>  
> @@ -3335,7 +3335,6 @@ static void
> reds_init_vd_agent_resources(RedsState *reds)
>  
>  static int do_spice_init(RedsState *reds, SpiceCoreInterface
> *core_interface)
>  {
> -static gboolean first = TRUE;
>  spice_info("starting %s", VERSION);
>  
>  if (core_interface->base.major_version !=
> SPICE_INTERFACE_CORE_MAJOR) {
> @@ -3416,10 +3415,6 @@ static int do_spice_init(RedsState *reds,
> SpiceCoreInterface *core_interface)
>  if (reds->allow_multiple_clients) {
>  spice_warning("spice: allowing multiple client
> connections");
>  }
> -if (first) {
> -atexit(reds_exit);
> -first = FALSE;
> -}
>  servers = g_list_prepend(servers, reds);
>  return 0;
>  
> diff --git a/spice-common b/spice-common
> index 00db440..5b6be16 16
> --- a/spice-common
> +++ b/spice-common
> @@ -1 +1 @@
> -Subproject commit 00db44087807335e74c8f65ea3e6fde0b684e7db
> +Subproject commit 5b6be16b37370bb774aa3c2a7d5c41791fdc3a1e
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH v2 spice-gtk 2/2] widget: use allocation instead of window size

[Pavel Grunt]

On Wed, 2016-03-23 at 12:25 +0100, Marc-André Lureau wrote:
> Since spice-gtk 57df040cc, the SpiceDisplay no longer forces its own
> window. Get allocation instead of window size.  This fixes a "top
> border" issue in spicy with gtk+ >= 3.19.8.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
Acked-by: Pavel Grunt <pgr...@redhat.com>

Thanks,
Pavel
> ---
>  src/Makefile.am  |  1 -
>  src/gtk-compat.h | 31 ---
>  src/spice-gtk-session.c  |  1 -
>  src/spice-widget-cairo.c |  4 ++--
>  src/spice-widget-egl.c   |  1 -
>  src/spice-widget.c   |  9 -
>  src/vncdisplaykeymap.c   |  1 -
>  7 files changed, 6 insertions(+), 42 deletions(-)
>  delete mode 100644 src/gtk-compat.h
> 
> diff --git a/src/Makefile.am b/src/Makefile.am
> index 66ba58b..330c85d 100644
> --- a/src/Makefile.am
> +++ b/src/Makefile.am
> @@ -122,7 +122,6 @@ SPICE_GTK_LIBADD_COMMON = \
>   $(NULL)
>  
>  SPICE_GTK_SOURCES_COMMON =   \
> - gtk-compat.h\
>   spice-util.c\
>   spice-util-priv.h   \
>   spice-gtk-session.c \
> diff --git a/src/gtk-compat.h b/src/gtk-compat.h
> deleted file mode 100644
> index 39438f9..000
> --- a/src/gtk-compat.h
> +++ /dev/null
> @@ -1,31 +0,0 @@
> -/* -*- Mode: C; c-basic-offset: 4; indent-tabs-mode: nil -*- */
> -/*
> -   Copyright (C) 2012-2014 Red Hat, Inc.
> -
> -   This library is free software; you can redistribute it and/or
> -   modify it under the terms of the GNU Lesser General Public
> -   License as published by the Free Software Foundation; either
> -   version 2.1 of the License, or (at your option) any later
> version.
> -
> -   This library is distributed in the hope that it will be useful,
> -   but WITHOUT ANY WARRANTY; without even the implied warranty of
> -   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> -   Lesser General Public License for more details.
> -
> -   You should have received a copy of the GNU Lesser General Public
> -   License along with this library; if not, see <http://www.gnu.org/
> licenses/>.
> -*/
> -#ifndef GTK_COMPAT_H
> -#define GTK_COMPAT_H
> -
> -#include "config.h"
> -
> -#include 
> -
> -static inline void gdk_drawable_get_size(GdkWindow *w, gint *ww,
> gint *wh)
> -{
> -*ww = gdk_window_get_width(w);
> -*wh = gdk_window_get_height(w);
> -}
> -
> -#endif /* GTK_COMPAT_H */
> diff --git a/src/spice-gtk-session.c b/src/spice-gtk-session.c
> index 4201ee0..380b0bb 100644
> --- a/src/spice-gtk-session.c
> +++ b/src/spice-gtk-session.c
> @@ -38,7 +38,6 @@
>  #include 
>  #include 
>  #include "desktop-integration.h"
> -#include "gtk-compat.h"
>  #include "spice-common.h"
>  #include "spice-gtk-session.h"
>  #include "spice-gtk-session-priv.h"
> diff --git a/src/spice-widget-cairo.c b/src/spice-widget-cairo.c
> index c92d4ec..8c45524 100644
> --- a/src/spice-widget-cairo.c
> +++ b/src/spice-widget-cairo.c
> @@ -17,7 +17,6 @@
>  */
>  #include "config.h"
>  
> -#include "gtk-compat.h"
>  #include "spice-widget.h"
>  #include "spice-widget-priv.h"
>  #include "spice-gtk-session-priv.h"
> @@ -78,7 +77,8 @@ void spicex_draw_event(SpiceDisplay *display,
> cairo_t *cr)
>  
>  spice_display_get_scaling(display, , , , , );
>  
> -gdk_drawable_get_size(gtk_widget_get_window(GTK_WIDGET(display))
> , , );
> +ww = gtk_widget_get_allocated_width(GTK_WIDGET(display));
> +wh = gtk_widget_get_allocated_height(GTK_WIDGET(display));
>  
>  /* We need to paint the bg color around the image */
>  rect.x = 0;
> diff --git a/src/spice-widget-egl.c b/src/spice-widget-egl.c
> index 5d839a3..4254628 100644
> --- a/src/spice-widget-egl.c
> +++ b/src/spice-widget-egl.c
> @@ -22,7 +22,6 @@
>  #define EGL_EGLEXT_PROTOTYPES
>  #define GL_GLEXT_PROTOTYPES
>  
> -#include "gtk-compat.h"
>  #include "spice-widget.h"
>  #include "spice-widget-priv.h"
>  #include "spice-gtk-session-priv.h"
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index 12be7c2..a9387ad 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -42,8 +42,6 @@
>  #include "vncdisplaykeymap.h"
>  #include "spice-grabsequence-priv.h"
>  
> -#include "gtk-compat.h"
> -
>  /* Some compatibility defines to let us build on both Gtk2 and Gtk3
> */
>  
>  /**
> @@ -2411,9 +2409,10 @@ void spice_di

Re: [Spice-devel] [spice-server] doc: Add virgl documentation

[Pavel Grunt]

Hi Christophe,

On Wed, 2016-03-23 at 17:18 +0100, Christophe Fergeau wrote:
> On Mon, Mar 21, 2016 at 03:28:41PM +0100, Christophe Fergeau wrote:
> > 
> > On Mon, Mar 21, 2016 at 02:57:18PM +0100, Pavel Grunt wrote:
> > > 
> > > Thanks!
> > > 
> > > Acked-by: Pavel Grunt <pgr...@redhat.com>
> > > 
> > > On Fri, 2016-03-18 at 17:12 +0100, Christophe Fergeau wrote:
> > > > 
> > > > +GL acceleration (virgl)
> > > > +===
> > > > +
> > > > +OpenGL acceleration is currently local only (it has to go
> > > > through a
> > > > Unix socket)
> > > > +and it needs guest support. It's currently limited to recent
> > > > linux
> > > > +distributions (for example up to date Fedora 23).
> > > > +
> > After running more tests, I realized that on a f23 guest, you need
> > to
> > update mesa from Gerd's copr, I'll have to adjust the wording here.
> I"ll change it to
> 
> -distributions (for example up to date Fedora 23).
> +distributions (for example up to date Fedora 23 + updated mesa
> packages from
> +copr:kraxel).

I would keep up to date Fedora 23. Mesa should have virgl support -
see http://koji.fedoraproject.org/koji/buildinfo?buildID=739036 

Pavel
> 
> Christophe
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH v3 6/9] egl: check and update scanout after egl init

[Pavel Grunt]

On Thu, 2016-03-24 at 12:32 +0100, Marc-André Lureau wrote:
> Once the egl/GL context are initialized, check if there is a scanout
> to
> associate to display widget. This solves races when scanout update is
> happening before the egl/GL context is ready.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  src/spice-widget-egl.c | 6 ++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/src/spice-widget-egl.c b/src/spice-widget-egl.c
> index 151970c..b9bec68 100644
> --- a/src/spice-widget-egl.c
> +++ b/src/spice-widget-egl.c
> @@ -272,6 +272,12 @@ end:
>  
>  d->egl.context_ready = TRUE;
>  
> +if (spice_display_get_gl_scanout(SPICE_DISPLAY_CHANNEL(d-
> >display)) != NULL) {
> +SPICE_DEBUG("scanout present during egl init, updating
> widget");
> +spice_display_widget_gl_scanout(display);
> +spice_display_widget_update_monitor_area(display);
> +}
> +
>  return TRUE;
>  }
>  
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 8/8] widget: modify update-area to not require primary surface

[Pavel Grunt]

On Wed, 2016-03-23 at 19:32 -0400, Marc-André Lureau wrote:
> Hi
> 
> - Original Message -
> > 
> > Hi,
> > 
> > as we discussed, I prefer this patch to go in after 9/8 to avoid
> > some
> > incorrect resizes
> Ok
> 
> > 
> > 
> > On Tue, 2016-03-22 at 13:57 +0100, Marc-André Lureau wrote:
> > > 
> > > If egl display is enabled, use GL scanout geometry to check
> > > intersection
> > > with the monitor area. This solves displaying GL display without
> > > software canvas.
> > What are the issues?
> There is no interesection with monitor area & missing primary canvas,
> so d->area is not set and gl update display fails.
> 
Thanks, ack

Pavel

> > 
> > 
> > Pavel
> > > 
> > > 
> > > Signed-off-by: Marc-André Lureau 
> > > ---
> > >  src/spice-widget.c | 41 --
> > > ---
> > >  1 file changed, 28 insertions(+), 13 deletions(-)
> > > 
> > > diff --git a/src/spice-widget.c b/src/spice-widget.c
> > > index e16126a..4b7e202 100644
> > > --- a/src/spice-widget.c
> > > +++ b/src/spice-widget.c
> > > @@ -2190,32 +2190,47 @@ static void update_area(SpiceDisplay
> > > *display,
> > >  gint x, gint y, gint width, gint height)
> > >  {
> > >  SpiceDisplayPrivate *d = display->priv;
> > > -GdkRectangle primary = {
> > > -.x = 0,
> > > -.y = 0,
> > > -.width = d->width,
> > > -.height = d->height
> > > -};
> > > -GdkRectangle area = {
> > > +GdkRectangle primary;
> > > +
> > > +SPICE_DEBUG("update area +%d+%d %dx%d", x, y, width,
> > > height);
> > > +d->area = (GdkRectangle) {
> > >  .x = x,
> > >  .y = y,
> > >  .width = width,
> > >  .height = height
> > >  };
> > >  
> > > -SPICE_DEBUG("update area, primary: %dx%d, area: +%d+%d
> > > %dx%d",
> > > d->width, d->height, area.x, area.y, area.width, area.height);
> > > +#ifndef G_OS_WIN32
> > > +if (d->egl.enabled) {
> > > +primary = (GdkRectangle) {
> > > +.width = d->egl.scanout.width,
> > > +.height = d->egl.scanout.height
> > > +};
> > > +} else
> > > +#endif
> > > +{
> > > +primary = (GdkRectangle) {
> > > +.width = d->width,
> > > +.height = d->height
> > > +};
> > > +}
> > >  
> > > -if (!gdk_rectangle_intersect(, , )) {
> > > +SPICE_DEBUG("primary: %dx%d", primary.width,
> > > primary.height);
> > > +if (!gdk_rectangle_intersect(, >area, >area))
> > > {
> > >  SPICE_DEBUG("The monitor area is not intersecting
> > > primary
> > > surface");
> > >  memset(>area, '\0', sizeof(d->area));
> > >  set_monitor_ready(display, false);
> > >  return;
> > >  }
> > >  
> > > -spicex_image_destroy(display);
> > > -d->area = area;
> > > -if (gtk_widget_get_realized(GTK_WIDGET(display)))
> > > -update_image(display);
> > > +#ifndef G_OS_WIN32
> > > +if (!d->egl.enabled)
> > > +#endif
> > > +{
> > > +spicex_image_destroy(display);
> > > +if (gtk_widget_get_realized(GTK_WIDGET(display)))
> > > +update_image(display);
> > > +}
> > >  
> > >  update_size_request(display);
> > >  
> > ___
> > Spice-devel mailing list
> > Spice-devel@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/spice-devel
> > 
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-html5] Fix 2 typos in comments

[Pavel Grunt]

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Tue, 2016-03-08 at 11:39 +0100, Christophe Fergeau wrote:
> ---
>  utils.js | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/utils.js b/utils.js
> index 9eb42ff..f1a5748 100644
> --- a/utils.js
> +++ b/utils.js
> @@ -100,7 +100,7 @@ function hexdump_buffer(a)
>  /*
> 
>  ** Converting keycodes to AT scancodes is very hard.
>  ** luckly there are some resources on the web and in the Xorg driver
> that help
> -** us figure out what browser depenend keycodes match to what
> scancodes.
> +** us figure out what browser dependent keycodes match to what
> scancodes.
>  **
>  ** This will most likely not work for non US keyboard and browsers
> other than
>  ** modern Chrome and FireFox.
> @@ -155,7 +155,7 @@ common_scanmap[121]= KEY_F10;
>  common_scanmap[122]= KEY_F11;
>  common_scanmap[123]= KEY_F12;
>  
> -/* These externded scancodes do not line up with values from
> atKeynames */
> +/* These extended scancodes do not line up with values from
> atKeynames */
>  common_scanmap[42] = 99;
>  common_scanmap[19] = 101;// Break
>  common_scanmap[111]= 0xE035; // KP_Divide
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [spice v10 07/27] replay: Add an option to change video codec

[Pavel Grunt]

Signed-off-by: Francois Gouget 
---
 server/tests/replay.c | 11 ++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/server/tests/replay.c b/server/tests/replay.c
index f3b670f..5bbca5e 100644
--- a/server/tests/replay.c
+++ b/server/tests/replay.c
@@ -284,7 +284,7 @@ int main(int argc, char **argv)
 {
 GError *error = NULL;
 GOptionContext *context = NULL;
-gchar *client = NULL, **file = NULL;
+gchar *client = NULL, *codecs = NULL, **file = NULL;
 gint port = 5000, compression = SPICE_IMAGE_COMPRESSION_AUTO_GLZ;
 gint streaming = SPICE_STREAM_VIDEO_FILTER;
 gboolean wait = FALSE;
@@ -294,6 +294,7 @@ int main(int argc, char **argv)
 { "client", 'c', 0, G_OPTION_ARG_STRING, , "Client", "CMD" },
 { "compression", 'C', 0, G_OPTION_ARG_INT, , "Compression 
(default 2)", "INT" },
 { "streaming", 'S', 0, G_OPTION_ARG_INT, , "Streaming 
(default 3)", "INT" },
+{ "video-codecs", 'v', 0, G_OPTION_ARG_STRING, , "Video 
codecs", "STRING" },
 { "port", 'p', 0, G_OPTION_ARG_INT, , "Server port (default 
5000)", "PORT" },
 { "wait", 'w', 0, G_OPTION_ARG_NONE, , "Wait for client", NULL },
 { "slow", 's', 0, G_OPTION_ARG_INT, , "Slow down replay. Delays 
USEC microseconds before each command", "USEC" },
@@ -378,6 +379,14 @@ int main(int argc, char **argv)
 server = spice_server_new();
 spice_server_set_image_compression(server, compression);
 spice_server_set_streaming_video(server, streaming);
+
+if (codecs != NULL) {
+if (spice_server_set_video_codecs(server, codecs) != 0) {
+g_warning("could not set codecs: %s", codecs);
+}
+g_free(codecs);
+}
+
 spice_server_set_port(server, port);
 spice_server_set_noauth(server);
 
-- 
2.7.0

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk] widget-egl: Define label only if can be used

[Pavel Grunt]

The label can be used only with GDK_WINDOWING_WAYLAND

Silence -Wunused-label
---
 src/spice-widget-egl.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/spice-widget-egl.c b/src/spice-widget-egl.c
index 51bb5c5..c54d9f6 100644
--- a/src/spice-widget-egl.c
+++ b/src/spice-widget-egl.c
@@ -266,7 +266,10 @@ gboolean spice_egl_init(SpiceDisplay *display, GError 
**err)
 eglMakeCurrent(d->egl.display, EGL_NO_SURFACE, EGL_NO_SURFACE,
d->egl.ctx);
 
+#ifdef GDK_WINDOWING_WAYLAND
 end:
+#endif
+
 if (!spice_egl_init_shaders(display, err))
 return FALSE;
 
-- 
2.8.0

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] "undefined reference" errors while running make

[Pavel Grunt]

Hi,

On Mon, 2016-04-04 at 09:59 +0200, Christophe Fergeau wrote:
> On Sat, Apr 02, 2016 at 02:14:11PM +, Mahmood Naderan wrote:
> > 
> > Hi Eduardo,I have to say that the latest git revision doesn't
> > include your patch.  So, I applied your patch manually into the
> > latest revision.
> > That undefined error has been solved, but now I see a new
> > error.Please note that it is saying I have to include two compiler
> > options. Should I do that or you have a netter idea?
> > 
> > Please see the full log below
> > 
> > 
> > 
> > 
> > root@cluster:spice-src# git rev-parse HEAD
> > 77946c395ba9ff3e6e03386a43f7451ddf1e38a0
> This does not correspond to a sha1 in the upstream repository.
> Things should be good starting from
> https://cgit.freedesktop.org/spice/spice-common/commit/?id=6c7552d543
> 69ced08a9b2c61587cad44c04d4ab9

There hasn't been a submodule update which contains that commit, that
is a reason why making spice server fails.

About the 'g_get_monotonic_time' it is available since glib 2.28 and I
think it is reasonable to start requiring it (at least for the unstable
branch). It's been there for five years.

And for the redefinition of typedef's there is a Christophe's patch &
discussion in the mailing list:
https://lists.freedesktop.org/archives/spice-devel/2016-March/027808.ht
ml

Pavel

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-common 1/3] log: Clamp SPICE_DEBUG_LEVEL if it's too high

[Pavel Grunt]

Hi,

ack for the series.

Pavel

On Tue, 2016-04-05 at 12:56 +0200, Christophe Fergeau wrote:
> This matches how SPICE_DEBUG_LEVEL behaved before switching to glib
> logging.
> ---
>  common/log.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/common/log.c b/common/log.c
> index 8d47cb6..5f08470 100644
> --- a/common/log.c
> +++ b/common/log.c
> @@ -67,6 +67,9 @@ static void spice_log_set_debug_level(void)
>  /* FIXME: To be removed after enough deprecation time */
>  g_warning("Setting SPICE_DEBUG_LEVEL is deprecated, use
> G_MESSAGES_DEBUG instead");
>  debug_level = atoi(debug_str);
> +if (debug_level > SPICE_LOG_LEVEL_DEBUG) {
> +debug_level = SPICE_LOG_LEVEL_DEBUG;
> +}
>  glib_debug_level = spice_log_level_to_glib(debug_level);
>  
>  /* If the debug level is too high, make sure we don't
> try to enable
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-server 1/3] Remove duplicate typedef from headers

[Pavel Grunt]

Hi,

I agree that there should be the only one typedef. But I don't
understand why not to include the needed header directly. We should try
to "untagle" these header dependencies.

Pavel

On Wed, 2016-03-30 at 11:46 +0200, Christophe Fergeau wrote:
> This is causing compile errors with older gcc versions.
> 
> reds.h includes reds-stream.h through red-channel.h so it will get
> the
> RedsState typedef this way.
> 
> red-qxl.h includes spice-qxl.h through red-channel.h and then
> spice.h,
> so it will get the QXLState typedef this way.
> ---
>  server/red-qxl.h | 2 --
>  server/reds.h| 2 --
>  2 files changed, 4 deletions(-)
> 
> diff --git a/server/red-qxl.h b/server/red-qxl.h
> index b1ebbe1..ce80f97 100644
> --- a/server/red-qxl.h
> +++ b/server/red-qxl.h
> @@ -20,8 +20,6 @@
>  
>  #include "red-channel.h"
>  
> -typedef struct QXLState QXLState;
> -
>  typedef struct AsyncCommand AsyncCommand;
>  
>  void red_qxl_init(SpiceServer *reds, QXLInstance *qxl);
> diff --git a/server/reds.h b/server/reds.h
> index 5966dc3..c2968c2 100644
> --- a/server/reds.h
> +++ b/server/reds.h
> @@ -31,8 +31,6 @@
>  #include "main-dispatcher.h"
>  #include "migration-protocol.h"
>  
> -typedef struct RedsState RedsState;
> -
>  static inline QXLInterface * qxl_get_interface(QXLInstance *qxl)
>  {
>  return SPICE_CONTAINEROF(qxl->base.sif, QXLInterface, base);
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-server 2/3] Update NEWS for 0.13.1

[Pavel Grunt]

The series looks good, ack

Pavel

On Wed, 2016-03-30 at 11:46 +0200, Christophe Fergeau wrote:
> ---
>  NEWS | 14 +-
>  1 file changed, 13 insertions(+), 1 deletion(-)
> 
> diff --git a/NEWS b/NEWS
> index e906e88..ecedcd8 100644
> --- a/NEWS
> +++ b/NEWS
> @@ -1,4 +1,4 @@
> -Major changes in 0.13.0:
> +Major changes in 0.13.1:
>  
>  
>  *** IMPORTANT ***
> @@ -6,6 +6,18 @@ The 0.13.x release series is an unstable release
> series, if you are looking for
>  stability for daily use, please use the latest 0.12.x release. We
> are trying to
>  keep the code as functional as 0.12, but regressions or bugs could
> still happen.
>  
> +* add spice_qxl_gl_scanout() spice_qxl_gl_draw_async() for local
> virgl support
> +* spice_server_set_keepalive_timeout() has been removed in favour of
> +  unconditionally sending keepalive probes every 10 minutes
> +* the rework started in 0.13.0 has moved forward. The global
> RedsState *reds
> +  variable has now been removed, GLib mainloop is now used for the
> RedWorker thread
> +  rather than an adhoc implementation, some core classes have been
> ported to
> +  GObject, ...
> +* some crash/leak fixes in corner cases
> +
> +Major changes in 0.13.0:
> +
> +
>  * Start of a large code rework to attempt to make the codebase more
>    maintainable. With this first release, the huge red_worker.c file
> has been split
>    in multiple smaller files.
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-gtk] build-sys: Update symbol files

[Pavel Grunt]

On Wed, 2016-03-30 at 10:59 +0200, Christophe Fergeau wrote:
> The OSX-specific symbol files were not updated after the recent API
> additions/removals, and some removed symbols were still listed in
> map-file.

Acked-by: Pavel Grunt <pgr...@redhat.com>

> ---
>  src/map-file| 4 +---
>  src/spice-glib-sym-file | 3 +++
>  src/spice-gtk-sym-file  | 2 --
>  3 files changed, 4 insertions(+), 5 deletions(-)
> 
> diff --git a/src/map-file b/src/map-file
> index 9df9017..589da21 100644
> --- a/src/map-file
> +++ b/src/map-file
> @@ -22,7 +22,6 @@ spice_client_error_quark;
>  spice_cursor_channel_get_type;
>  spice_display_change_preferred_compression;
>  spice_display_channel_get_type;
> -spice_display_copy_to_guest;
>  spice_display_get_gl_scanout;
>  spice_display_get_grab_keys;
>  spice_display_get_pixbuf;
> @@ -33,7 +32,6 @@ spice_display_key_event_get_type;
>  spice_display_mouse_ungrab;
>  spice_display_new;
>  spice_display_new_with_monitor;
> -spice_display_paste_from_guest;
>  spice_display_send_keys;
>  spice_display_set_grab_keys;
>  spice_file_transfer_task_cancel;
> @@ -133,12 +131,12 @@ spice_usb_device_manager_connect_device_finish;
>  spice_usb_device_manager_disconnect_device;
>  spice_usb_device_manager_disconnect_device_async;
>  spice_usb_device_manager_disconnect_device_finish;
> -spice_usb_device_manager_is_redirecting;
>  spice_usb_device_manager_get;
>  spice_usb_device_manager_get_devices;
>  spice_usb_device_manager_get_devices_with_filter;
>  spice_usb_device_manager_get_type;
>  spice_usb_device_manager_is_device_connected;
> +spice_usb_device_manager_is_redirecting;
>  spice_usb_device_widget_get_type;
>  spice_usb_device_widget_new;
>  spice_usbredir_channel_get_type;
> diff --git a/src/spice-glib-sym-file b/src/spice-glib-sym-file
> index d49a713..0657dcb 100644
> --- a/src/spice-glib-sym-file
> +++ b/src/spice-glib-sym-file
> @@ -108,11 +108,14 @@ spice_usb_device_manager_can_redirect_device
>  spice_usb_device_manager_connect_device_async
>  spice_usb_device_manager_connect_device_finish
>  spice_usb_device_manager_disconnect_device
> +spice_usb_device_manager_disconnect_device_async
> +spice_usb_device_manager_disconnect_device_finish
>  spice_usb_device_manager_get
>  spice_usb_device_manager_get_devices
>  spice_usb_device_manager_get_devices_with_filter
>  spice_usb_device_manager_get_type
>  spice_usb_device_manager_is_device_connected
> +spice_usb_device_manager_is_redirecting
>  spice_usbredir_channel_get_type
>  spice_util_get_debug
>  spice_util_get_version_string
> diff --git a/src/spice-gtk-sym-file b/src/spice-gtk-sym-file
> index 1574e07..e52334b 100644
> --- a/src/spice-gtk-sym-file
> +++ b/src/spice-gtk-sym-file
> @@ -1,4 +1,3 @@
> -spice_display_copy_to_guest
>  spice_display_get_grab_keys
>  spice_display_get_pixbuf
>  spice_display_get_type
> @@ -6,7 +5,6 @@ spice_display_key_event_get_type
>  spice_display_mouse_ungrab
>  spice_display_new
>  spice_display_new_with_monitor
> -spice_display_paste_from_guest
>  spice_display_send_keys
>  spice_display_set_grab_keys
>  spice_grab_sequence_as_string
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk 0/6] Fixes for autogenerated documentation

[Pavel Grunt]

On Wed, 2016-03-30 at 10:49 +0200, Christophe Fergeau wrote:
> Acked-by: Christophe Fergeau <cferg...@redhat.com>
> 
Thank you, pushed

> On Wed, Mar 30, 2016 at 09:00:15AM +0200, Pavel Grunt wrote:
> > 
> > Hi,
> > 
> > Following patches adjust autogenerated docs (can be found at [0])
> > to recent
> > changes (some classes are final now, there are new functions).
> > They also silence some warnings when the docs are generated.
> > 
> > Pavel
> > 
> > [0] http://www.spice-space.org/spice-gtk.html
> > 
> > Pavel Grunt (6):
> >   usb-device-manager: Fix docstring
> >   doc: Remove private structs
> >   doc: channel stability should be Stable
> >   spice-file-transfer-task: Fix docstring
> >   doc: Ignore spicy-connect
> >   doc: Add description for spice_display_get_gl_scanout()
> > 
> >  doc/reference/Makefile.am|  1 +
> >  doc/reference/spice-gtk-sections.txt |  8 +---
> >  src/channel-display.c|  2 ++
> >  src/channel-smartcard.c  |  2 +-
> >  src/channel-usbredir.c   |  2 +-
> >  src/spice-file-transfer-task.h   |  2 +-
> >  src/usb-device-manager.c | 14 +-
> >  7 files changed, 16 insertions(+), 15 deletions(-)
> > 
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-server] doc: Add virgl documentation

[Pavel Grunt]

On Thu, 2016-03-24 at 13:58 +0100, Christophe Fergeau wrote:
> On Wed, Mar 23, 2016 at 05:59:03PM +0100, Pavel Grunt wrote:
> > 
> > On Wed, 2016-03-23 at 17:45 +0100, Christophe Fergeau wrote:
> > > 
> > > But it seems these are good enough, when I tested this, Xorg was
> > > picking
> > > the QXL driver, which gave me a blackscreen with the f23
> > > packages.
> > > With
> > > packages from kraxel's repository, the correct driver was picked
> > > up,
> > > and
> > > I could get a working guest display. Which is why I want to
> > > mention
> > > them
> > > now ;)
> > > 
> > ok maybe it deserves a bug report.
> Do you mean bug report against fedora? Or somewhere else?
> 
Yes, I meant a bug for Fedora.

Pavel

> > 
> > I already acked the patch, but would you mind adding a link to the
> > copr repo [0] ?
> I added this and pushed.
> 
> Christophe
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk] widget: Use parentheses around ternary conditional

[Pavel Grunt]

"&&" has higher precedency than "?:"

Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=94738
---
 src/spice-widget.c | 7 +++
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/src/spice-widget.c b/src/spice-widget.c
index 21b1c14..67a9cc1 100644
--- a/src/spice-widget.c
+++ b/src/spice-widget.c
@@ -232,12 +232,11 @@ static void update_ready(SpiceDisplay *display)
 SpiceDisplayPrivate *d = display->priv;
 gboolean ready;
 
-ready = d->monitor_ready &&
 #ifndef G_OS_WIN32
-d->egl.enabled ? d->egl.image != NULL :
+ready = d->monitor_ready && (d->egl.enabled ? d->egl.image != NULL : 
d->mark != 0);
+#else
+ready = d->monitor_ready && d->mark != 0;
 #endif
-d->mark;
-
 /* If the 'resize-guest' property is set, the application expects spice-gtk
  * to manage the size and state of the displays, so update the 'enabled'
  * state here. If 'resize-guest' is false, we can assume that the
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH 01/11] Rename some missing names related to RedCharDevice

[Pavel Grunt]

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Fri, 2016-04-01 at 13:37 +0100, Frediano Ziglio wrote:
> ---
>  server/char-device.c | 6 +++---
>  server/char-device.h | 2 +-
>  server/reds.c| 2 +-
>  3 files changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/server/char-device.c b/server/char-device.c
> index 2957a1b..22f2de2 100644
> --- a/server/char-device.c
> +++ b/server/char-device.c
> @@ -92,7 +92,7 @@ static void red_char_device_ref(RedCharDevice
> *char_dev);
>  static void red_char_device_unref(RedCharDevice *char_dev);
>  static void
> red_char_device_write_buffer_unref(RedCharDeviceWriteBuffer
> *write_buf);
>  
> -static void spice_char_dev_write_retry(void *opaque);
> +static void red_char_device_write_retry(void *opaque);
>  
>  typedef struct RedCharDeviceMsgToClientItem {
>  RingItem link;
> @@ -566,7 +566,7 @@ static int
> red_char_device_write_to_device(RedCharDevice *dev)
>  return total;
>  }
>  
> -static void spice_char_dev_write_retry(void *opaque)
> +static void red_char_device_write_retry(void *opaque)
>  {
>  RedCharDevice *dev = opaque;
>  
> @@ -750,7 +750,7 @@ RedCharDevice
> *red_char_device_create(SpiceCharDeviceInstance *sin,
>  sif = spice_char_device_get_interface(char_dev->priv->sin);
>  if (sif->base.minor_version <= 2 ||
>  !(sif->flags & SPICE_CHAR_DEVICE_NOTIFY_WRITABLE)) {
> -char_dev->priv->write_to_dev_timer =
> reds_core_timer_add(reds, spice_char_dev_write_retry, char_dev);
> +char_dev->priv->write_to_dev_timer =
> reds_core_timer_add(reds, red_char_device_write_retry, char_dev);
>  if (!char_dev->priv->write_to_dev_timer) {
>  spice_error("failed creating char dev write timer");
>  }
> diff --git a/server/char-device.h b/server/char-device.h
> index f3ae6cb..b0efa1f 100644
> --- a/server/char-device.h
> +++ b/server/char-device.h
> @@ -158,7 +158,7 @@ int red_char_device_restore(RedCharDevice *dev,
>   * protocol:
>   *  The client tokens' are set only once, when the main channel is
> initialized.
>   *  Instead, it would have been more appropriate to reset them upon
> AGEN_CONNECT.
> - *  The client tokens are tracked as part of the
> SpiceCharDeviceClientState. Thus,
> + *  The client tokens are tracked as part of the
> RedCharDeviceClient. Thus,
>   *  in order to be backwartd compatible with the client, we need to
> track the tokens
>   *  event when the agent is detached. We don't destroy the
> char_device state, and
>   *  instead we just reset it.
> diff --git a/server/reds.c b/server/reds.c
> index 12de608..d913539 100644
> --- a/server/reds.c
> +++ b/server/reds.c
> @@ -425,7 +425,7 @@ static void reds_reset_vdp(RedsState *reds)
>   * tokens management in the vdagent protocol:
>   *  The client tokens' are set only once, when the main channel
> is initialized.
>   *  Instead, it would have been more appropriate to reset them
> upon AGEN_CONNECT.
> - *  The client tokens are tracked as part of the
> SpiceCharDeviceClientState. Thus,
> + *  The client tokens are tracked as part of the
> RedCharDeviceClient. Thus,
>   *  in order to be backward compatible with the client, we need
> to track the tokens
>   *  even if the agent is detached. We don't destroy the
> char_device state, and
>   *  instead we just reset it.
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk v2] widget: Avoid combining ternary with another operators

[Pavel Grunt]

Logical operators like "&&" has higher precedency than "?:".
Readiness of SpiceDisplay was wrongly determined due to missing
parentheses around the ternary operator.

Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=94738
---
 src/spice-widget.c | 11 ++-
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/src/spice-widget.c b/src/spice-widget.c
index 21b1c14..f605439 100644
--- a/src/spice-widget.c
+++ b/src/spice-widget.c
@@ -230,14 +230,15 @@ static gint get_display_id(SpiceDisplay *display)
 static void update_ready(SpiceDisplay *display)
 {
 SpiceDisplayPrivate *d = display->priv;
-gboolean ready;
+gboolean ready = FALSE;
 
-ready = d->monitor_ready &&
+if (d->monitor_ready) {
 #ifndef G_OS_WIN32
-d->egl.enabled ? d->egl.image != NULL :
+ready = d->egl.enabled ? d->egl.image != NULL : d->mark != 0;
+#else
+ready = d->mark != 0;
 #endif
-d->mark;
-
+}
 /* If the 'resize-guest' property is set, the application expects spice-gtk
  * to manage the size and state of the displays, so update the 'enabled'
  * state here. If 'resize-guest' is false, we can assume that the
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk] widget: Use parentheses around ternary conditional

[Pavel Grunt]

On Tue, 2016-03-29 at 15:39 +0200, Victor Toso wrote:
> Hey,
> 
> Works fine! Thanks!
> 
> On Tue, Mar 29, 2016 at 03:03:03PM +0200, Pavel Grunt wrote:
> > 
> > "&&" has higher precedency than "?:"
> Maybe a small summary of the issue that his patch fixes?
> 
> > 
> > 
> > Fixes:
> > https://bugs.freedesktop.org/show_bug.cgi?id=94738
> > ---
> >  src/spice-widget.c | 7 +++
> >  1 file changed, 3 insertions(+), 4 deletions(-)
> > 
> > diff --git a/src/spice-widget.c b/src/spice-widget.c
> > index 21b1c14..67a9cc1 100644
> > --- a/src/spice-widget.c
> > +++ b/src/spice-widget.c
> > @@ -232,12 +232,11 @@ static void update_ready(SpiceDisplay
> > *display)
> >  SpiceDisplayPrivate *d = display->priv;
> >  gboolean ready;
> > 
> > -ready = d->monitor_ready &&
> >  #ifndef G_OS_WIN32
> > -d->egl.enabled ? d->egl.image != NULL :
> > +ready = d->monitor_ready && (d->egl.enabled ? d->egl.image !=
> > NULL : d->mark != 0);
> > +#else
> > +ready = d->monitor_ready && d->mark != 0;
> >  #endif
> > -d->mark;
> I would go with something like
> 
> gboolean ready = FALSE;
> ...
> if (d->monitor_ready) {
> #ifndef G_OS_WIN32
>   ready = d->egl.enabled ? d->egl.image != NULL : d->mark != 0;
> #else
>   ready = d->mark != 0;
> #endif
> }
> 
> But either way is fine (I'm not really picky with this things)
> 
> Acked-by: Victor Toso <victort...@redhat.com>

I will go for your suggestion

Pavel

> 
> > 
> > -
> >  /* If the 'resize-guest' property is set, the application
> > expects spice-gtk
> >   * to manage the size and state of the displays, so update the
> > 'enabled'
> >   * state here. If 'resize-guest' is false, we can assume that
> > the
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 2/6] doc: Remove private structs

[Pavel Grunt]

---
 doc/reference/spice-gtk-sections.txt | 7 ---
 1 file changed, 7 deletions(-)

diff --git a/doc/reference/spice-gtk-sections.txt 
b/doc/reference/spice-gtk-sections.txt
index 0c05168..a6fcfbf 100644
--- a/doc/reference/spice-gtk-sections.txt
+++ b/doc/reference/spice-gtk-sections.txt
@@ -337,8 +337,6 @@ SpiceUsbDeviceManagerPrivate
 
 spice-gtk-session
 SpiceGtkSession
-SpiceGtkSession
-SpiceGtkSessionClass
 spice_gtk_session_get
 spice_gtk_session_copy_to_guest
 spice_gtk_session_paste_from_guest
@@ -357,8 +355,6 @@ SpiceGtkSessionPrivate
 
 spice-widget
 SpiceDisplay
-SpiceDisplay
-SpiceDisplayClass
 SpiceDisplayKeyEvent
 spice_display_new
 spice_display_new_with_monitor
@@ -368,7 +364,6 @@ spice_display_get_grab_keys
 spice_display_send_keys
 spice_display_get_pixbuf
 
-SpiceGrabSequence
 spice_grab_sequence_new
 spice_grab_sequence_new_from_string
 spice_grab_sequence_copy
@@ -393,8 +388,6 @@ SpiceDisplayPrivate
 
 usb-device-widget
 SpiceUsbDeviceWidget
-SpiceUsbDeviceWidget
-SpiceUsbDeviceWidgetClass
 
 spice_usb_device_widget_new
 
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 4/6] spice-file-transfer-task: Fix docstring

[Pavel Grunt]

---
 src/spice-file-transfer-task.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/spice-file-transfer-task.h b/src/spice-file-transfer-task.h
index b97d107..18dc509 100644
--- a/src/spice-file-transfer-task.h
+++ b/src/spice-file-transfer-task.h
@@ -37,7 +37,7 @@ typedef struct _SpiceFileTransferTaskPrivate 
SpiceFileTransferTaskPrivate;
 /**
  * SpiceFileTransferTask:
  *
- * The #FileTransferTask struct is opaque and should not be accessed directly.
+ * The #SpiceFileTransferTask struct is opaque and should not be accessed 
directly.
  */
 struct _SpiceFileTransferTask
 {
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 5/6] doc: Ignore spicy-connect

[Pavel Grunt]

---
 doc/reference/Makefile.am | 1 +
 1 file changed, 1 insertion(+)

diff --git a/doc/reference/Makefile.am b/doc/reference/Makefile.am
index cb01f97..5aa8557 100644
--- a/doc/reference/Makefile.am
+++ b/doc/reference/Makefile.am
@@ -46,6 +46,7 @@ IGNORE_HFILES=\
spice-uri-priv.h\
spice-util-priv.h   \
spice-widget-priv.h \
+   spicy-connect.h \
usb-acl-helper.h\
usb-device-manager-priv.h   \
usbutil.h   \
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 1/6] usb-device-manager: Fix docstring

[Pavel Grunt]

Add description, "Since" and change the parameter name in the docstring
to match parameter name in the function definition.
---
 doc/reference/spice-gtk-sections.txt |  1 +
 src/usb-device-manager.c | 14 +-
 2 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/doc/reference/spice-gtk-sections.txt 
b/doc/reference/spice-gtk-sections.txt
index 8947893..0c05168 100644
--- a/doc/reference/spice-gtk-sections.txt
+++ b/doc/reference/spice-gtk-sections.txt
@@ -311,6 +311,7 @@ spice_usb_device_manager_get
 spice_usb_device_manager_get_devices
 spice_usb_device_manager_get_devices_with_filter
 spice_usb_device_manager_is_device_connected
+spice_usb_device_manager_is_redirecting
 spice_usb_device_manager_disconnect_device
 spice_usb_device_manager_can_redirect_device
 spice_usb_device_manager_connect_device_async
diff --git a/src/usb-device-manager.c b/src/usb-device-manager.c
index c36a9a8..da7e596 100644
--- a/src/usb-device-manager.c
+++ b/src/usb-device-manager.c
@@ -232,11 +232,15 @@ G_DEFINE_BOXED_TYPE(SpiceUsbDevice, spice_usb_device, 
g_object_ref, g_object_unr
 #endif
 
 /**
-* spice_usb_device_manager_is_redirecting:
-* @manager: the #SpiceUsbDeviceManager manager
-*
-* Returns: %TRUE if device redirection negotiation flow is in progress
-*/
+ * spice_usb_device_manager_is_redirecting:
+ * @self: the #SpiceUsbDeviceManager manager
+ *
+ * Checks whether a device is being redirected
+ *
+ * Returns: %TRUE if device redirection negotiation flow is in progress
+ *
+ * Since: 0.32
+ */
 gboolean spice_usb_device_manager_is_redirecting(SpiceUsbDeviceManager *self)
 {
 #ifdef USE_USBREDIR
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 6/6] doc: Add description for spice_display_get_gl_scanout()

[Pavel Grunt]

---
 src/channel-display.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/channel-display.c b/src/channel-display.c
index 90d155e..431c2e7 100644
--- a/src/channel-display.c
+++ b/src/channel-display.c
@@ -532,6 +532,8 @@ void 
spice_display_change_preferred_compression(SpiceChannel *channel, gint comp
  * spice_display_get_gl_scanout:
  * @channel: a #SpiceDisplayChannel
  *
+ * Retrieves the GL scanout if available
+ *
  * Returns: the current GL scanout, or %NULL if none or not valid
  *
  * Since: 0.31
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 3/6] doc: channel stability should be Stable

[Pavel Grunt]

---
 src/channel-smartcard.c | 2 +-
 src/channel-usbredir.c  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/channel-smartcard.c b/src/channel-smartcard.c
index 7f3306e..21df98a 100644
--- a/src/channel-smartcard.c
+++ b/src/channel-smartcard.c
@@ -35,7 +35,7 @@
  * @title: Smartcard Channel
  * @section_id:
  * @see_also: #SpiceSmartcardManager, #SpiceSession
- * @stability: API Stable (channel in development)
+ * @stability: Stable
  * @include: spice-client.h
  *
  * The Spice protocol defines a set of messages to forward smartcard
diff --git a/src/channel-usbredir.c b/src/channel-usbredir.c
index a251855..9066b4a 100644
--- a/src/channel-usbredir.c
+++ b/src/channel-usbredir.c
@@ -42,7 +42,7 @@
  * @short_description: usb redirection
  * @title: USB Redirection Channel
  * @section_id:
- * @stability: API Stable (channel in development)
+ * @stability: Stable
  * @include: spice-client.h
  *
  * The Spice protocol defines a set of messages to redirect USB devices
-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 0/6] Fixes for autogenerated documentation

[Pavel Grunt]

Hi,

Following patches adjust autogenerated docs (can be found at [0]) to recent
changes (some classes are final now, there are new functions).
They also silence some warnings when the docs are generated.

Pavel

[0] http://www.spice-space.org/spice-gtk.html

Pavel Grunt (6):
  usb-device-manager: Fix docstring
  doc: Remove private structs
  doc: channel stability should be Stable
  spice-file-transfer-task: Fix docstring
  doc: Ignore spicy-connect
  doc: Add description for spice_display_get_gl_scanout()

 doc/reference/Makefile.am|  1 +
 doc/reference/spice-gtk-sections.txt |  8 +---
 src/channel-display.c|  2 ++
 src/channel-smartcard.c  |  2 +-
 src/channel-usbredir.c   |  2 +-
 src/spice-file-transfer-task.h   |  2 +-
 src/usb-device-manager.c | 14 +-
 7 files changed, 16 insertions(+), 15 deletions(-)

-- 
2.7.4

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk 1/2] widget: offset draw updates by widget allocation if necessary

[Pavel Grunt]

Hi Marc-André,

On Tue, 2016-03-22 at 19:27 +0100, Marc-André Lureau wrote:
> According to gtk_widget_queue_draw_area() documentation:
> 
> "The region here is specified in widget coordinates. Widget
> coordinates
> are a bit odd; for historical reasons, they are defined as
> widget->window coordinates for widgets that return TRUE for
> gtk_widget_get_has_window(), and are relative to widget->allocation.x 
> ,
> widget->allocation.y otherwise."
> 
> Since spice-gtk 57df040cc, the SpiceDisplay no longer forces its own
> window. During gtk+ 3.19.8, gtk stack also no longer use its own
> window (commit 9d0e8401). In order to make drawing update resilient
> to
> such changes, add a function applying offset if necessary.
> 
> This solves drawing glitches with spice-gtk git and gtk+ >= 3.19.8.
> 
This also fixes slowness of spicy for me

> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
> ---
>  src/spice-widget.c | 30 ++
>  1 file changed, 22 insertions(+), 8 deletions(-)
> 
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index 4b7e202..e45513e 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -2267,6 +2267,21 @@ static void primary_destroy(SpiceChannel
> *channel, gpointer data)
>  set_monitor_ready(display, false);
>  }
>  
> +static void queue_draw_area(SpiceDisplay *display, gint x, gint y,
> +gint width, gint height)
> +{
> +if (!gtk_widget_get_has_window(GTK_WIDGET(display))) {
> +GtkAllocation allocation;
> +
> +gtk_widget_get_allocation(GTK_WIDGET(display), );
> +x += allocation.x;
> +y += allocation.y;
> +}
> +
> +gtk_widget_queue_draw_area(GTK_WIDGET(display),
> +   x, y, width, height);
> +}
> +
>  static void invalidate(SpiceChannel *channel,
> gint x, gint y, gint w, gint h, gpointer
> data)
>  {
> @@ -2304,9 +2319,8 @@ static void invalidate(SpiceChannel *channel,
>  x2 = ceil ((rect.x - d->area.x + rect.width) * s);
>  y2 = ceil ((rect.y - d->area.y + rect.height) * s);
>  
> -gtk_widget_queue_draw_area(GTK_WIDGET(display),
> -   display_x + x1, display_y + y1,
> -   x2 - x1, y2-y1);
> +queue_draw_area(display, display_x + x1, display_y + y1,
> +x2 - x1, y2 - y1);

I would move 'display_x + x1, display_y + y1,' to separate line to keep
the same style.

>  }
>  
>  static void mark(SpiceDisplay *display, gint mark)
> @@ -2457,11 +2471,11 @@ static void cursor_invalidate(SpiceDisplay
> *display)
>  
>  spice_display_get_scaling(display, , , , NULL, NULL);
>  
> -gtk_widget_queue_draw_area(GTK_WIDGET(display),
> -   floor ((d->mouse_guest_x - d-
> >mouse_hotspot.x - d->area.x) * s) + x,
> -   floor ((d->mouse_guest_y - d-
> >mouse_hotspot.y - d->area.y) * s) + y,
> -   ceil (gdk_pixbuf_get_width(d-
> >mouse_pixbuf) * s),
> -   ceil (gdk_pixbuf_get_height(d-
> >mouse_pixbuf) * s));
> +queue_draw_area(display,
> +floor ((d->mouse_guest_x - d->mouse_hotspot.x -
> d->area.x) * s) + x,
> +floor ((d->mouse_guest_y - d->mouse_hotspot.y -
> d->area.y) * s) + y,
> +ceil (gdk_pixbuf_get_width(d->mouse_pixbuf) *
> s),
> +ceil (gdk_pixbuf_get_height(d->mouse_pixbuf) *
> s));
>  }
>  
>  static void cursor_move(SpiceCursorChannel *channel, gint x, gint y,
> gpointer data)

Acked-by: Pavel Grunt <pgr...@redhat.com>

and thanks for fixing it!

Pavel


___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk 2/2] widget: use allocation instead of window size

[Pavel Grunt]

Hi,

This fixes the "top border" in spicy.

considering how the allocation is used I would prefer some helper
function to get the width and height. 

Thanks,
Pavel

On Tue, 2016-03-22 at 19:27 +0100, Marc-André Lureau wrote:
> Since spice-gtk 57df040cc, the SpiceDisplay no longer forces its own
> window. Get allocation instead of window size.
> 
> Signed-off-by: Marc-André Lureau 
> ---
>  src/Makefile.am  |  1 -
>  src/gtk-compat.h | 31 ---
>  src/spice-gtk-session.c  |  1 -
>  src/spice-widget-cairo.c |  6 --
>  src/spice-widget-egl.c   |  1 -
>  src/spice-widget.c   | 11 ++-
>  src/vncdisplaykeymap.c   |  1 -
>  7 files changed, 10 insertions(+), 42 deletions(-)
>  delete mode 100644 src/gtk-compat.h
> 
> diff --git a/src/Makefile.am b/src/Makefile.am
> index 66ba58b..330c85d 100644
> --- a/src/Makefile.am
> +++ b/src/Makefile.am
> @@ -122,7 +122,6 @@ SPICE_GTK_LIBADD_COMMON = \
>   $(NULL)
>  
>  SPICE_GTK_SOURCES_COMMON =   \
> - gtk-compat.h\
>   spice-util.c\
>   spice-util-priv.h   \
>   spice-gtk-session.c \
> diff --git a/src/gtk-compat.h b/src/gtk-compat.h
> deleted file mode 100644
> index 39438f9..000
> --- a/src/gtk-compat.h
> +++ /dev/null
> @@ -1,31 +0,0 @@
> -/* -*- Mode: C; c-basic-offset: 4; indent-tabs-mode: nil -*- */
> -/*
> -   Copyright (C) 2012-2014 Red Hat, Inc.
> -
> -   This library is free software; you can redistribute it and/or
> -   modify it under the terms of the GNU Lesser General Public
> -   License as published by the Free Software Foundation; either
> -   version 2.1 of the License, or (at your option) any later
> version.
> -
> -   This library is distributed in the hope that it will be useful,
> -   but WITHOUT ANY WARRANTY; without even the implied warranty of
> -   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> -   Lesser General Public License for more details.
> -
> -   You should have received a copy of the GNU Lesser General Public
> -   License along with this library; if not, see  licenses/>.
> -*/
> -#ifndef GTK_COMPAT_H
> -#define GTK_COMPAT_H
> -
> -#include "config.h"
> -
> -#include 
> -
> -static inline void gdk_drawable_get_size(GdkWindow *w, gint *ww,
> gint *wh)
> -{
> -*ww = gdk_window_get_width(w);
> -*wh = gdk_window_get_height(w);
> -}
> -
> -#endif /* GTK_COMPAT_H */
> diff --git a/src/spice-gtk-session.c b/src/spice-gtk-session.c
> index 4201ee0..380b0bb 100644
> --- a/src/spice-gtk-session.c
> +++ b/src/spice-gtk-session.c
> @@ -38,7 +38,6 @@
>  #include 
>  #include 
>  #include "desktop-integration.h"
> -#include "gtk-compat.h"
>  #include "spice-common.h"
>  #include "spice-gtk-session.h"
>  #include "spice-gtk-session-priv.h"
> diff --git a/src/spice-widget-cairo.c b/src/spice-widget-cairo.c
> index c92d4ec..6a8afda 100644
> --- a/src/spice-widget-cairo.c
> +++ b/src/spice-widget-cairo.c
> @@ -17,7 +17,6 @@
>  */
>  #include "config.h"
>  
> -#include "gtk-compat.h"
>  #include "spice-widget.h"
>  #include "spice-widget-priv.h"
>  #include "spice-gtk-session-priv.h"
> @@ -78,7 +77,10 @@ void spicex_draw_event(SpiceDisplay *display,
> cairo_t *cr)
>  
>  spice_display_get_scaling(display, , , , , );
>  
> -gdk_drawable_get_size(gtk_widget_get_window(GTK_WIDGET(display))
> , , );
> +GtkAllocation allocation;
> +gtk_widget_get_allocation(GTK_WIDGET(display), );
> +ww = allocation.width;
> +wh = allocation.height;
>  
>  /* We need to paint the bg color around the image */
>  rect.x = 0;
> diff --git a/src/spice-widget-egl.c b/src/spice-widget-egl.c
> index 5fcdcaa..7a9ba73 100644
> --- a/src/spice-widget-egl.c
> +++ b/src/spice-widget-egl.c
> @@ -22,7 +22,6 @@
>  #define EGL_EGLEXT_PROTOTYPES
>  #define GL_GLEXT_PROTOTYPES
>  
> -#include "gtk-compat.h"
>  #include "spice-widget.h"
>  #include "spice-widget-priv.h"
>  #include "spice-gtk-session-priv.h"
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index e45513e..b409063 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -42,8 +42,6 @@
>  #include "vncdisplaykeymap.h"
>  #include "spice-grabsequence-priv.h"
>  
> -#include "gtk-compat.h"
> -
>  /* Some compatibility defines to let us build on both Gtk2 and Gtk3
> */
>  
>  /**
> @@ -2410,9 +2408,12 @@ void spice_display_get_scaling(SpiceDisplay
> *display,
>  int x, y, w, h;
>  double s;
>  
> -if (gtk_widget_get_realized (GTK_WIDGET(display)))
> -gdk_drawable_get_size(gtk_widget_get_window(GTK_WIDGET(displ
> ay)), , );
> -else {
> +if (gtk_widget_get_realized (GTK_WIDGET(display))) {
> +GtkAllocation allocation;
> +gtk_widget_get_allocation(GTK_WIDGET(display), );
> +ww = allocation.width;
> +wh = allocation.height;
> +} else {
>  ww = fbw;
>  wh = fbh;
>  }
> 

Re: [Spice-devel] [spice-common] tests: Fix glib version check

[Pavel Grunt]

Ack,
Pavel

On Thu, 2016-03-31 at 13:35 +0200, Christophe Fergeau wrote:
> The glib version checks were using #ifdef GLIB_CHECK_VERSION() rather
> than  #if GLIB_CHECK_VERSION()
> ---
>  tests/test-logging.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/test-logging.c b/tests/test-logging.c
> index c02905f..cb929f9 100644
> --- a/tests/test-logging.c
> +++ b/tests/test-logging.c
> @@ -43,7 +43,7 @@ LOG_OTHER_HELPER(message, MESSAGE)
>  LOG_OTHER_HELPER(warning, WARNING)
>  LOG_OTHER_HELPER(critical, CRITICAL)
>  
> -#ifdef GLIB_CHECK_VERSION(2,38,0)
> +#if GLIB_CHECK_VERSION(2,38,0)
>  /* Checks that spice_warning() aborts after changing
> SPICE_ABORT_LEVEL */
>  static void test_spice_abort_level(void)
>  {
> @@ -401,7 +401,7 @@ int main(int argc, char **argv)
>   * test cases are going to test */
>  g_log_set_always_fatal(fatal_mask & G_LOG_LEVEL_MASK);
>  
> -#ifdef GLIB_CHECK_VERSION(2,38,0)
> +#if GLIB_CHECK_VERSION(2,38,0)
>  g_test_add_func("/spice-common/spice-abort-level",
> test_spice_abort_level);
>  g_test_add_func("/spice-common/spice-abort-level-gwarning",
> test_spice_abort_level_g_warning);
>  g_test_add_func("/spice-common/spice-debug-level",
> test_spice_debug_level);
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH 01/14] Store display in Drawable struct

[Pavel Grunt]

Hi Jonathon,

On Thu, 2016-04-21 at 16:43 -0500, Jonathon Jongsma wrote:
> If the Drawable keeps a pointer to the Display channel that it is
> associated with, we can unref it directly and not need to pass the
> 'display' parameter separately to the unref function

So drawable needs 'display' just for unref'ing surface(s) - detach_stream()
doesn't need DisplayChannel, the same applies for
drawable_remove_dependencies(). Looking at the DisplayChannel struct it still
has an array of Drawables and free_drawables. (It is not clear to me - I need to
read previous reviews)

> ---
> NEW PATCH. This prepares for the UpgradeItem change as discussed on the
> previous review.
> 
>  server/dcc.c | 14 +-
>  server/display-channel.c | 17 +++--
>  server/display-channel.h |  4 +++-
>  3 files changed, 19 insertions(+), 16 deletions(-)
> 
> diff --git a/server/dcc.c b/server/dcc.c
> index 91c3f82..4f3f227 100644
> --- a/server/dcc.c
> +++ b/server/dcc.c
> @@ -277,13 +277,11 @@ static void
> add_drawable_surface_images(DisplayChannelClient *dcc, Drawable *dra
>  void drawable_pipe_item_free(PipeItem *item)
>  {
>  DrawablePipeItem *dpi = SPICE_CONTAINEROF(item, DrawablePipeItem,
> dpi_pipe_item);
> -DisplayChannel *display = DCC_TO_DC(dpi->dcc);
> -
>  spice_assert(item->refcount == 0);
>  
>  spice_warn_if_fail(!ring_item_is_linked(>link));
>  spice_warn_if_fail(!ring_item_is_linked(>base));
> -display_channel_drawable_unref(display, dpi->drawable);
> +drawable_unref(dpi->drawable);
>  free(dpi);
>  }
>  
> @@ -1591,20 +1589,18 @@ int dcc_handle_migrate_data(DisplayChannelClient *dcc,
> uint32_t size, void *mess
>  return TRUE;
>  }
>  
> -static void upgrade_item_unref(DisplayChannel *display, UpgradeItem *item)
> +static void upgrade_item_unref(UpgradeItem *item)
>  {
>  if (--item->refs != 0)
>  return;
>  
> -display_channel_drawable_unref(display, item->drawable);
> +drawable_unref(item->drawable);
>  free(item->rects);
>  free(item);
>  }
>  
>  static void release_item_after_push(DisplayChannelClient *dcc, PipeItem
> *item)
dcc is unused in the function, it should be removed

The patch makes a sense to me, in my opinion more cleanups should be done.

Pavel

>  {
> -DisplayChannel *display = DCC_TO_DC(dcc);
> -
>  switch (item->type) {
>  case PIPE_ITEM_TYPE_DRAW:
>  case PIPE_ITEM_TYPE_IMAGE:
> @@ -1613,7 +1609,7 @@ static void release_item_after_push(DisplayChannelClient
> *dcc, PipeItem *item)
>  pipe_item_unref(item);
>  break;
>  case PIPE_ITEM_TYPE_UPGRADE:
> -upgrade_item_unref(display, (UpgradeItem *)item);
> +upgrade_item_unref((UpgradeItem *)item);
>  break;
>  case PIPE_ITEM_TYPE_GL_SCANOUT:
>  case PIPE_ITEM_TYPE_GL_DRAW:
> @@ -1651,7 +1647,7 @@ static void
> release_item_before_push(DisplayChannelClient *dcc, PipeItem *item)
>  }
>  case PIPE_ITEM_TYPE_STREAM_CLIP:
>  case PIPE_ITEM_TYPE_UPGRADE:
> -upgrade_item_unref(display, (UpgradeItem *)item);
> +upgrade_item_unref((UpgradeItem *)item);
>  break;
>  case PIPE_ITEM_TYPE_IMAGE:
>  case PIPE_ITEM_TYPE_MONITORS_CONFIG:
> diff --git a/server/display-channel.c b/server/display-channel.c
> index 88dbc74..f2bdc1d 100644
> --- a/server/display-channel.c
> +++ b/server/display-channel.c
> @@ -397,7 +397,7 @@ static void current_remove_drawable(DisplayChannel
> *display, Drawable *item)
>  ring_remove(>tree_item.base.siblings_link);
>  ring_remove(>list_link);
>  ring_remove(>surface_list_link);
> -display_channel_drawable_unref(display, item);
> +drawable_unref(item);
>  display->current_size--;
>  }
>  
> @@ -495,7 +495,7 @@ static int current_add_equal(DisplayChannel *display,
> DrawItem *item, TreeItem *
>  pipes_add_drawable(display, drawable);
>  }
>  drawable_remove_from_pipes(other_drawable);
> -display_channel_drawable_unref(display, other_drawable);
> +drawable_unref(other_drawable);
>  return TRUE;
>  }
>  
> @@ -538,7 +538,7 @@ static int current_add_equal(DisplayChannel *display,
> DrawItem *item, TreeItem *
>  /* not sending other_drawable where possible */
>  drawable_remove_from_pipes(other_drawable);
>  
> -display_channel_drawable_unref(display, other_drawable);
> +drawable_unref(other_drawable);
>  return TRUE;
>  }
>  break;
> @@ -1192,7 +1192,7 @@ void display_channel_process_draw(DisplayChannel
> *display, RedDrawable *red_draw
>  
>  display_channel_add_drawable(display, drawable);
>  
> -display_channel_drawable_unref(display, drawable);
> +drawable_unref(drawable);
>  }
>  
>  int display_channel_wait_for_migrate_data(DisplayChannel *display)
> @@ -1380,6 +1380,10 @@ Drawable
> *display_channel_drawable_try_new(DisplayChannel *display,
>  }
>  
>  

Re: [Spice-devel] [spice-gtk] Spice-widget: Allow smaller widget with scaling enabled

[Pavel Grunt]

On Fri, 2016-04-22 at 09:57 +0200, Javier Celaya wrote:
> Hi
> 
> El 22/04/16 a las 09:19, Pavel Grunt escribió:
> > Hi Javier,
> > 
> > sorry, I must have missed the patch. Can you please explain the problem? How
> > to
> > reproduce it? Is it about spicy?
>  Sorry, the commit message may be a bit short. I'll send a second version with
> a longer explanation, along with the the declaration of update_size_request().
> I found the problem with spicy, but I suppose it affects other clients too.
> You can reproduce it checking the "scale display" option and unchecking the
> "resize guest to match window size" option. Then, the spice widget takes a
> minimum size of the current guest resolution. The window can be made bigger,
> but never smaller.

I see, so steps to reproduce are:
1) connect with spicy to a guest
2) resize the guest to be bigger than minimal spicy resolution 640x480
3) uncheck "resize guest", check "scale display"
4) Try to scale down

Thanks,
Pavel

> > About the patch - you should declare the update_size_request() function
> > instead
> > of moving scaling_updated() down - it would be more clear what has changed.
> > 
> > Thanks,
> > Pavel 
> > 
> > On Fri, 2016-04-22 at 08:06 +0200, Javier Celaya wrote:
> > > Can someone review this patch, please?
> > > Thank you
> > > 
> > > El 13/04/16 a las 13:42, Javier Celaya escribió:
> > > > With scaling enabled, the size request of the widget must be 640x480
> > > > (the same as with resize_guest_enable) so that the widget can be
> > > > actually scalled down.
> > > > ---
> > > >  src/spice-widget.c | 25 +
> > > >  1 file changed, 13 insertions(+), 12 deletions(-)
> > > > 
> > > > diff --git a/src/spice-widget.c b/src/spice-widget.c
> > > > index 6f638fb..f445f3f 100644
> > > > --- a/src/spice-widget.c
> > > > +++ b/src/spice-widget.c
> > > > @@ -169,23 +169,12 @@ static void
> > > > spice_display_get_property(GObject*object,
> > > >  }
> > > >  }
> > > >  
> > > > -static void scaling_updated(SpiceDisplay *display)
> > > > -{
> > > > -SpiceDisplayPrivate *d = display->priv;
> > > > -GdkWindow *window = gtk_widget_get_window(GTK_WIDGET(display));
> > > > -
> > > > -recalc_geometry(GTK_WIDGET(display));
> > > > -if (d->ximage && window) { /* if not yet shown */
> > > > -gtk_widget_queue_draw(GTK_WIDGET(display));
> > > > -}
> > > > -}
> > > > -
> > > >  static void update_size_request(SpiceDisplay *display)
> > > >  {
> > > >  SpiceDisplayPrivate *d = display->priv;
> > > >  gint reqwidth, reqheight;
> > > >  
> > > > -if (d->resize_guest_enable) {
> > > > +if (d->resize_guest_enable || d->allow_scaling) {
> > > >  reqwidth = 640;
> > > >  reqheight = 480;
> > > >  } else {
> > > > @@ -197,6 +186,18 @@ static void update_size_request(SpiceDisplay
> > > > *display)
> > > >  recalc_geometry(GTK_WIDGET(display));
> > > >  }
> > > >  
> > > > +static void scaling_updated(SpiceDisplay *display)
> > > > +{
> > > > +SpiceDisplayPrivate *d = display->priv;
> > > > +GdkWindow *window = gtk_widget_get_window(GTK_WIDGET(display));
> > > > +
> > > > +recalc_geometry(GTK_WIDGET(display));
> > > > +if (d->ximage && window) { /* if not yet shown */
> > > > +gtk_widget_queue_draw(GTK_WIDGET(display));
> > > > +}
> > > > +update_size_request(display);
> > > > +}
> > > > +
> > > >  static void update_keyboard_focus(SpiceDisplay *display, gboolean
> > > > state)
> > > >  {
> > > >  SpiceDisplayPrivate *d = display->priv;
> > >  
> > > -- 
> > > 
> > > 
> > >   Javier Celaya Software Engineer  javier.cel...@flexvdi.com  
> > > 
> > >   +34 876 60 00 73 @j_celaya 
> > > ___
> > > Spice-devel mailing list
> > > Spice-devel@lists.freedesktop.org
> > > https://lists.freedesktop.org/mailman/listinfo/spice-devel
>  
> -- 
> 
> 
>   Javier Celaya Software Engineer  javier.cel...@flexvdi.com  
>   +34 876 60 00 73 @j_celaya 
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-gtk] Spice-widget: Allow smaller widget with scaling enabled

[Pavel Grunt]

Hi Javier,

sorry, I must have missed the patch. Can you please explain the problem? How to
reproduce it? Is it about spicy?

About the patch - you should declare the update_size_request() function instead
of moving scaling_updated() down - it would be more clear what has changed.

Thanks,
Pavel 

On Fri, 2016-04-22 at 08:06 +0200, Javier Celaya wrote:
> Can someone review this patch, please?
> Thank you
> 
> El 13/04/16 a las 13:42, Javier Celaya escribió:
> > With scaling enabled, the size request of the widget must be 640x480
> > (the same as with resize_guest_enable) so that the widget can be
> > actually scalled down.
> > ---
> >  src/spice-widget.c | 25 +
> >  1 file changed, 13 insertions(+), 12 deletions(-)
> > 
> > diff --git a/src/spice-widget.c b/src/spice-widget.c
> > index 6f638fb..f445f3f 100644
> > --- a/src/spice-widget.c
> > +++ b/src/spice-widget.c
> > @@ -169,23 +169,12 @@ static void
> > spice_display_get_property(GObject*object,
> >  }
> >  }
> >  
> > -static void scaling_updated(SpiceDisplay *display)
> > -{
> > -SpiceDisplayPrivate *d = display->priv;
> > -GdkWindow *window = gtk_widget_get_window(GTK_WIDGET(display));
> > -
> > -recalc_geometry(GTK_WIDGET(display));
> > -if (d->ximage && window) { /* if not yet shown */
> > -gtk_widget_queue_draw(GTK_WIDGET(display));
> > -}
> > -}
> > -
> >  static void update_size_request(SpiceDisplay *display)
> >  {
> >  SpiceDisplayPrivate *d = display->priv;
> >  gint reqwidth, reqheight;
> >  
> > -if (d->resize_guest_enable) {
> > +if (d->resize_guest_enable || d->allow_scaling) {
> >  reqwidth = 640;
> >  reqheight = 480;
> >  } else {
> > @@ -197,6 +186,18 @@ static void update_size_request(SpiceDisplay *display)
> >  recalc_geometry(GTK_WIDGET(display));
> >  }
> >  
> > +static void scaling_updated(SpiceDisplay *display)
> > +{
> > +SpiceDisplayPrivate *d = display->priv;
> > +GdkWindow *window = gtk_widget_get_window(GTK_WIDGET(display));
> > +
> > +recalc_geometry(GTK_WIDGET(display));
> > +if (d->ximage && window) { /* if not yet shown */
> > +gtk_widget_queue_draw(GTK_WIDGET(display));
> > +}
> > +update_size_request(display);
> > +}
> > +
> >  static void update_keyboard_focus(SpiceDisplay *display, gboolean state)
> >  {
> >  SpiceDisplayPrivate *d = display->priv;
>  
> -- 
> 
> 
>   Javier Celaya Software Engineer  javier.cel...@flexvdi.com  
>   +34 876 60 00 73 @j_celaya 
> ___
> Spice-devel mailing list
> Spice-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/spice-devel
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] How to look information of the debugging of spice-server?

[Pavel Grunt]

Hello,

On Fri, 2016-04-22 at 15:43 +0800, #嚣张的狼# wrote:
>        Thank you for taking time out of your busy schedule to see this
> letter.I am student who has been studying Spice and have some questions about
> using Spice-server.
> 
>         Recently, I am studying some pathes released in "https://patchwork.fre
> edesktop.org" , But I don't know  where I can find information of
> "spice_debug". I have set SPICE_DEBUG_LEVEL=5, but there have no changes in 
> /var/log/libvirt/qemu/XX.log.
> 
Where did you set the variable ? If you use libvirt you need to edit the domain
(virsh edit domain) and do some changes.
Please see http://www.spice-space.org/page/SpiceLogging for more info about it.

Regards,
Pavel


>        Could you give me some help?
> ___
> Spice-devel mailing list
> Spice-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/spice-devel
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice-gtk v2] Spice-widget: Allow smaller widget with scaling enabled

[Pavel Grunt]

Hi Javier,

Acked-by: Pavel Grunt <pgr...@redhat.com>

On Fri, 2016-04-22 at 10:08 +0200, Javier Celaya wrote:
> When resize-guest-to-match-window-size is disabled, the size request
> (minimum size) of the spice widget is set to the current guest
> resolution. The client window cannot be made smaller, even with
> scale-display enabled.
> 
> This patch sets a size request of 640x480 when either
> resize-guest-to-match-window-size OR scale-display are enabled, making
> it posible to shrink the window and scale down the display.
> ---
>  src/spice-widget.c | 5 -
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/src/spice-widget.c b/src/spice-widget.c
> index 6f638fb..a7f31c4 100644
> --- a/src/spice-widget.c
> +++ b/src/spice-widget.c
> @@ -169,6 +169,8 @@ static void spice_display_get_property(GObject*object,
>  }
>  }
>  
> +static void update_size_request(SpiceDisplay *display);
> +
I will move the declaration up so it is groupped with other declaration and push
the patch.

Thanks,
Pavel


>  static void scaling_updated(SpiceDisplay *display)
>  {
>  SpiceDisplayPrivate *d = display->priv;
> @@ -178,6 +180,7 @@ static void scaling_updated(SpiceDisplay *display)
>  if (d->ximage && window) { /* if not yet shown */
>  gtk_widget_queue_draw(GTK_WIDGET(display));
>  }
> +update_size_request(display);
>  }
>  
>  static void update_size_request(SpiceDisplay *display)
> @@ -185,7 +188,7 @@ static void update_size_request(SpiceDisplay *display)
>  SpiceDisplayPrivate *d = display->priv;
>  gint reqwidth, reqheight;
>  
> -if (d->resize_guest_enable) {
> +if (d->resize_guest_enable || d->allow_scaling) {
>  reqwidth = 640;
>  reqheight = 480;
>  } else {
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice 4/4] manual: Remove spicec screenshot

[Pavel Grunt]

---
 docs/manual/images/spicec01.png | Bin 10244 -> 0 bytes
 docs/manual/manual.txt  |   4 
 2 files changed, 4 deletions(-)
 delete mode 100644 docs/manual/images/spicec01.png

diff --git a/docs/manual/images/spicec01.png b/docs/manual/images/spicec01.png
deleted file mode 100644
index 
e2cf8c5dd86e18516e36b343fdefde3bd0c709eb..
GIT binary patch
literal 0
HcmV?d1

literal 10244
zcmcI~WmFq&*KR5lEyW5gMG8e4NO6ioi@SS)prN?Cq)?m|DH^mCD;6lhwYWpEASqs;
zSaFBs3(xbs@A=L-?>g)JI44)!ji_C0HI)L%H1cRCIA2c_Z8%
zaoC?Q-firWTEhf#?Bn+H{Z|Ag$dXr_|EbMZU6uw)|HU_
z79cI-5%wUiyMl@gF6s^;Av2-gnKAaYi;k@ojIRGn#lmpy@h6-NtF9YiydFC
zRh81)*OHR@~t^ynU*=z1|s*W@Tf4L3ZIyPFh
zi-w*TCw}TClly7pg^HqrvoRk0lVQ)mFuA9jJ~-E
z1?W+N^L^g%lKw(t15JLti4g+_AKFOi<)b3L_+P>Zvy~X3q4>R+?u`8c)r*9^?0S-A
z?OIj;!91tUjEnt`mnfg^4i+mdG)HEy39Z+3p1rL%G>cD>gwv6
zo0|(KrWJngrGY1$pSbRWE)TIXK(CHsEFt53`HAqY7<3smaH(U7Twu#53z-%H8W8@5
zbBoJ}=_A^MdN{kL=A}?Wsv=85))t3Ro%39SNA286=y(-1ZEDRdcnSo~+En%4+pl
ztZ1d%`tW)-be`*j9lRdW6GPkRx#PQ$pF+uD0#XCVvlOsf;ZOKt4};2k9qM!kcJOh
zwTM|$XjIup4((A{VB==rjrMw~k#q^Dje(7Uf-y_+6_Ey
ztgfz(l|Y{nayB_%?$m55PsxXnyv_Jz29cYIk;XPxN`q^P}h1h@jXttM>X8W
z2@|E6wSf3OFA7b=)9mLA*ib6cwmn=M!X_<1l-t;N*{}saR64@SHC$eQad}zZd@{Qm
zuT1F%b0_}}e{7y7YvMs(
z;@A0+TA-PfRL8>A(1J3I$7-P^09E7Ac(xpU9hsxqkHOJY1m^!~+->26o)2VSOcWB+
z2s$4Q3HW85KEUifSUBqzp!45r!=>|HOkcQKD^N*CUt`%_YB{ew{A5v_VzpcRtc5%&3M>iF#7e)}Fw38fE@C7Y
z?XmNNO8Y1C*260(d3BF5uxm(jIEgAQe+r6zdmc;z1fe@pSDo^C-J1CX4{5XpX
zV)^;eVR_pn9<^IIdC!^J=9ym5bcjMJBr+m0GF6|1`x0}2ar)i9*hAiP?Y4KYNGto&
zqqs5FNAQcr*4o>NuvC4bITSs|Ri&?BdTJ_6R$?XxmzF;&-086|6WwKAci?7wZqEmw
zhaJrIn*CJ{=^@-?`kj$|Abc@*gh~EjasNe^vojGf-POWUt}&5S
z`cq1&2fv#j4|Z1d=y&|>;Orte3!sgtsbcJ}hJ3OfQKzj5x=*~Oq1zuH|5(ue?X
zsg+e-I;3tbu6OK(7ou-TxOVmOvl2ZLKL`}!s8`xV;!s1b+T+sS;4k1ud@P&|cDqByHShFy6
zN~1RY75O$Lx=+0U-!#1TCP!Nou5|moSfHk16TfcF*5m#TJIr9MUVang42SOC~WV
z1km(cAI;T{9C<2PsVPG683yq)X`$p{s?yH%5bh}A0ga`%U=Huv0=>JYm
zp{SD`B#&<@In44D9bS{F8Os>)J7+4~j?>oRwv%yi$@}W-R9@Z4Z2hzxFuXNL26JBv
zf?oEh{k+t>oG@~bzIr_UIE81;`;u_0+tS+?3?xE?d)+ctFk
z1c2dzi>T(7N~Ngc0;-pw%dK+%otTm}v3bUy2m?_rv@nPvLf+=j#>k`KNO_xl(X#F_
zMH>iYD6M*}X{WgD)!lc86n4h?yh`#NBv*?U0ohoV{7%ISn|USb=g$b`y42&72B*Uk
z83x!sQYl5*$|%}N6eW%s+R!P2)Hdcns7UL-$8k2%(>LEs?=$o{grBX<*zJB8F|gKO
z;x*9`A1z
z_9cgr5kUGI*>>xc3m`j!Hfv~Yma$Rt(?;?QuLt1T2$FHVs!v~dO~q*EX9lZw3kXY~
zYIkjpWA2)0k4f!S;2g-3GPE12DYW?=TWwBt=R1=u@^0E6$+PwMyBlh1B^>r3MH7+Q
z50V8>dFgUC&$ic+Ycg^zn7<`6HBZK?6pnX(!j+@+v1*Ox}7x~C=C*4gL2`2|?1
zqiGXwkm{^JmY~eMqs*P~tM=m4oo{E#MO?tHuPMHHfh82Irsu~;
z7TNtpQgNs>DN34goO%;x@M!6ea!h8G!#a}^V{*LU0Re!8yMo1?_z_e&=1bYm
zk4gt0Xq-=_BTkaUCV$(XV`>WG)x28SH#^=hgGjLhCtqF@Ch2+&2wlbtwAJ#hy
zeyQ$9dCJN4#COQ0rKi*MWQsVU6oa%Bil9BbZ~}F=GPaT@~oG!?$Pe<4LpRt(nW)bJ;flK)OL*2;n#A&;@kre6vR!)
zAX(4{Id2i4MQss~QsGeb)^SC=Fz2p1=;!Q>!}jt_kjEJQ;_unbkb7u-qO^_
z(5IL|0jbkt44in$LK2W#`d#?j#-Hr6Zmth;z{Em(wODHZPb1eMe2gsXG-6~y1!j*W
z*;j2eZTFP>@Hha1mti^tHZ?Is9M~HRsCBJYqQfI{Vz4SWv9g=ITwyMyqDU3G%$z7%KY%|CL
z4q$Wa5*2lUdS1ts-OM^`7gQ%b)CC+RAODCUQ#Q9h7Xv-*Frg(+e&(Ijc}+$cM6>_W
z8rj+T@df_VcP%tP51y`_&|9

[Spice-devel] [PATCH spice 3/4] manual: No need to use sudo to run qemu

[Pavel Grunt]

---
 docs/manual/manual.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/manual/manual.txt b/docs/manual/manual.txt
index 348e5e0..f172ab4 100644
--- a/docs/manual/manual.txt
+++ b/docs/manual/manual.txt
@@ -240,7 +240,7 @@ population. I assume that you have a locally stored ISO of 
your
 favourite operating system so you can use it for installation.
 
 [source,sh]
-host$ sudo qemu-kvm -boot order=dc -vga qxl \
+host$ qemu-kvm -boot order=dc -vga qxl \
 -spice port=3001,disable-ticketing -soundhw ac97 \
 -device virtio-serial -chardev 
spicevmc,id=vdagent,debug=0,name=vdagent \
 -device virtserialport,chardev=vdagent,name=com.redhat.spice.0 \
-- 
2.8.1

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice 2/4] manual: Recommend UsbDk instead of USB Clerk

[Pavel Grunt]

---
 docs/manual/manual.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/manual/manual.txt b/docs/manual/manual.txt
index 3f869bf..348e5e0 100644
--- a/docs/manual/manual.txt
+++ b/docs/manual/manual.txt
@@ -1026,8 +1026,8 @@ The recommended way of getting all the needed drivers 
installed is to
 use the all-in-one Spice guest tools installer which can be found on
 http://spice-space.org/download/windows/spice-guest-tools/[spice-space.org].
 
-To get USB redirection working on Windows, you need to install USB
-Clerk (TODO: add link)
+To get USB redirection working on Windows, you need to install
+http://www.spice-space.org/download/windows/usbdk/[UsbDk]
 
 If you want to manually install them, the QXL driver can be downloaded
 from http://spice-space.org/download/windows/qxl/[this location] ,
-- 
2.8.1

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 2/2] channel: Abort migration in delayed unref

[Pavel Grunt]

When channel is unref'ed during migration migrate_channel_event_cb
is called causing a crash by coroutine yielding to nonexistent channel.

As comment in spice_channel_coroutine says:
  Co-routine exits now - the SpiceChannel object may no longer exist,
  so don't do anything else now unless you like SEGVs

Related: rhbz#1318574
---
 src/spice-channel.c | 8 
 1 file changed, 8 insertions(+)

diff --git a/src/spice-channel.c b/src/spice-channel.c
index 19237b3..7b0a3dc 100644
--- a/src/spice-channel.c
+++ b/src/spice-channel.c
@@ -2296,6 +2296,7 @@ static gboolean spice_channel_delayed_unref(gpointer data)
 SpiceChannel *channel = SPICE_CHANNEL(data);
 SpiceChannelPrivate *c = channel->priv;
 gboolean was_ready = c->state == SPICE_CHANNEL_STATE_READY;
+SpiceSession *session;
 
 CHANNEL_DEBUG(channel, "Delayed unref channel %p", channel);
 
@@ -2303,6 +2304,13 @@ static gboolean spice_channel_delayed_unref(gpointer 
data)
 
 c->state = SPICE_CHANNEL_STATE_UNCONNECTED;
 
+session = spice_channel_get_session(channel);
+if (spice_session_is_for_migration(session)) {
+/* error during migration - abort migration */
+spice_session_abort_migration(session);
+return FALSE;
+}
+
 if (c->event != SPICE_CHANNEL_NONE) {
 g_coroutine_signal_emit(channel, signals[SPICE_CHANNEL_EVENT], 0, 
c->event);
 c->event = SPICE_CHANNEL_NONE;
-- 
2.8.1

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk 1/2] channel-main: Use CHANNEL_DEBUG for migration events

[Pavel Grunt]

To show which channel got an error or an unhandled event
---
 src/channel-main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/channel-main.c b/src/channel-main.c
index 604f333..3966e49 100644
--- a/src/channel-main.c
+++ b/src/channel-main.c
@@ -2286,7 +2286,7 @@ static void migrate_channel_event_cb(SpiceChannel 
*channel, SpiceChannelEvent ev
 coroutine_yieldto(mig->from, NULL);
 break;
 default:
-SPICE_DEBUG("error or unhandled channel event during migration: %d", 
event);
+CHANNEL_DEBUG(channel, "error or unhandled channel event during 
migration: %d", event);
 /* go back to main channel to report error */
 coroutine_yieldto(mig->from, NULL);
 }
-- 
2.8.1

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice 1/4] manual: Recommend enabling monitors using remote-viewer's menu

[Pavel Grunt]

On Fri, 2016-04-22 at 16:01 +0200, Christophe Fergeau wrote:
> ACK series.

Pushed, thanks
> 
> Christophe
> 
> On Fri, Apr 22, 2016 at 03:09:03PM +0200, Pavel Grunt wrote:
> > ---
> >  docs/manual/manual.txt | 5 ++---
> >  1 file changed, 2 insertions(+), 3 deletions(-)
> > 
> > diff --git a/docs/manual/manual.txt b/docs/manual/manual.txt
> > index 9e41c36..3f869bf 100644
> > --- a/docs/manual/manual.txt
> > +++ b/docs/manual/manual.txt
> > @@ -656,9 +656,8 @@ that is already there:
> >  Client
> >  --
> >  
> > -You can enable additional displays either from the "Display/Displays"
> > -menu in remote-viewer, or from your guest OS display configuration
> > -tool.
> > +You can enable additional displays from the "View -> Displays"
> > +menu in remote-viewer.
> >  
> >  TLS
> >  ===
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] How to modify spice version?

[Pavel Grunt]

Hi,

On Thu, 2016-04-28 at 14:09 +0800, Chen yu wrote:
> Thank you for taking time out of your busy schedule to see this letter.
> I am trying to compile spice downloaded from github, but the versions of these
> are 'UNKNOWN'. when I compiled QEMU with these spice, there have some errors
> in the process of 'make'. 
> Could I modify the version of spice (spice-server and spice-gtk) temporarily
> for testing and how should I do?


Please, use repos from freedesktop https://cgit.freedesktop.org/spice/
The github repos are mirrored, but they don't have tags, so they cannot generate
version.

Pavel

> ___
> Spice-devel mailing list
> Spice-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/spice-devel
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk 2/2] channel: Abort migration in delayed unref

[Pavel Grunt]

Hi,

On Tue, 2016-04-26 at 12:05 +0200, Victor Toso wrote:
> Hi,
> 
> On Fri, Apr 22, 2016 at 04:47:48PM +0200, Pavel Grunt wrote:
> > When channel is unref'ed during migration migrate_channel_event_cb
> > is called causing a crash by coroutine yielding to nonexistent channel.
> > 
> > As comment in spice_channel_coroutine says:
> >   Co-routine exits now - the SpiceChannel object may no longer exist,
> >   so don't do anything else now unless you like SEGVs
> > 
> > Related: rhbz#1318574
> > ---
> >  src/spice-channel.c | 8 
> >  1 file changed, 8 insertions(+)
> > 
> > diff --git a/src/spice-channel.c b/src/spice-channel.c
> > index 19237b3..7b0a3dc 100644
> > --- a/src/spice-channel.c
> > +++ b/src/spice-channel.c
> > @@ -2296,6 +2296,7 @@ static gboolean spice_channel_delayed_unref(gpointer
> > data)
> >  SpiceChannel *channel = SPICE_CHANNEL(data);
> >  SpiceChannelPrivate *c = channel->priv;
> >  gboolean was_ready = c->state == SPICE_CHANNEL_STATE_READY;
> > +SpiceSession *session;
> >  
> >  CHANNEL_DEBUG(channel, "Delayed unref channel %p", channel);
> >  
> > @@ -2303,6 +2304,13 @@ static gboolean spice_channel_delayed_unref(gpointer
> > data)
> >  
> >  c->state = SPICE_CHANNEL_STATE_UNCONNECTED;
> >  
> > +session = spice_channel_get_session(channel);
> > +if (spice_session_is_for_migration(session)) {
> > +/* error during migration - abort migration */
> > +spice_session_abort_migration(session);
> > +return FALSE;
> > +}
> > +
> 
> Just to be sure if I got it right. The migration fails for some reason which
> triggers and error in the migration process for spice-gtk. That will make the
> channel for the target host to be delayed_unref.
> 
> If we don't spice_session_abort_migration() now, migrate_channel_event_cb()
> will
> be called for the channel for the former host which will try to yield to a
> coroutine and cause the crash.
yes
> 
> Does this only happen under failure in the migration like [0] or it could
> happen
> if we do cancel the migration in the right moment?

if migration is canceled by the SPICE_MSG_MAIN_MIGRATE_CANCEL
message, spice_session_abort_migration() is called.

Looking at the backtrace of [0] the patch addresses the same issue.

Pavel

> 
> [0] https://bugs.freedesktop.org/show_bug.cgi?id=92019
> 
> Cheers,
>   toso
> 
> >  if (c->event != SPICE_CHANNEL_NONE) {
> >  g_coroutine_signal_emit(channel, signals[SPICE_CHANNEL_EVENT], 0,
> > c->event);
> >  c->event = SPICE_CHANNEL_NONE;
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH] Fix make dist

[Pavel Grunt]

On Mon, 2016-04-25 at 17:21 -0300, Eduardo Lima (Etrunko) wrote:
> spicec screenshot has been removed since commit
> b0186fe1267dfa2b8534c7c083dc301791279f08
> 
> Signed-off-by: Eduardo Lima (Etrunko) <etru...@redhat.com>
Acked-by: Pavel Grunt <pgr...@redhat.com>

Thanks,
Pavel
> ---
>  docs/manual/Makefile.am | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/docs/manual/Makefile.am b/docs/manual/Makefile.am
> index ead5f46..24a11fe 100644
> --- a/docs/manual/Makefile.am
> +++ b/docs/manual/Makefile.am
> @@ -5,7 +5,6 @@ ASCIIDOC_FLAGS = -a icons -a toc
>  EXTRA_DIST = \
>   images/icons/important.png  \
>   images/icons/note.png   \
> - images/spicec01.png \
>   manual.html \
>   manual.chunked  \
>   manual.txt  \
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v3 6/7] spice-uri: Check if port is in allowed range

[Pavel Grunt]

On Fri, 2016-05-20 at 14:50 +0200, Christophe Fergeau wrote:
> On Thu, May 19, 2016 at 06:38:08PM +0200, Pavel Grunt wrote:
> > Related: rhbz#1335239
> > ---
> >  src/spice-uri.c| 8 ++--
> >  tests/test-spice-uri.c | 2 ++
> >  2 files changed, 8 insertions(+), 2 deletions(-)
> > 
> > diff --git a/src/spice-uri.c b/src/spice-uri.c
> > index b483374..6a43461 100644
> > --- a/src/spice-uri.c
> > +++ b/src/spice-uri.c
> > @@ -165,8 +165,8 @@ gboolean spice_uri_parse(SpiceURI *self, const gchar
> > *_uri, GError **error)
> >  uri_port = uriv[1];
> >  
> >  if (uri_port != NULL) {
> > -char *endptr;
> > -guint port = strtoul(uri_port, , 10);
> > +gchar *endptr;
> > +gint64 port = g_ascii_strtoll(uri_port, , 10);
> 
> Not sure this is 100% related to this change? but why not

It helps to detect negative values (strtoul "converts" negative number to
positive/treats it as negative w/o the sign?) - I'll mention the reason for
change in the commit log.
> 
> >  if (*endptr != '\0') {
> >  g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED,
> >  "Invalid uri port: %s", uri_port);
> > @@ -175,6 +175,10 @@ gboolean spice_uri_parse(SpiceURI *self, const gchar
> > *_uri, GError **error)
> >  g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED, "Missing uri port");
> >  goto end;
> >  }
> > +if (port < 0 || port > 65535) {
> > +g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED, "Port out of range");
> > +goto end;
> > +}
> 
> I'd check for <= 0

Ok, I'll change it.

Thanks

Pavel

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [spice v14 00/29] Add GStreamer support for video streaming

[Pavel Grunt]

Hi Francois,

I pushed the spice-gtk part upstream.

Thanks!
Pavel

On Wed, 2016-05-04 at 11:40 +0200, Francois Gouget wrote:
> This patch series adds support for using GStreamer to encode and 
> decode the video streams, adding support for VP8 and h264 codecs.
> 
> As before the patches can also be grabbed from the gst branch of the
> repositories below:
> 
> spice:  https://github.com/fgouget/spice
> spice-gtk:  https://github.com/fgouget/spice-gtk
> xf86-video-qxl: https://github.com/fgouget/xf86-video-qxl
> spice-protocol: https://github.com/fgouget/spice-protocol
> 
> See also gst-sync for the old spice-gtk code. (there's also 'extras'
> branches with more experimental/future patches for the curious)
> 
> Note also that I added gst-v10, gst-v11, gst-v12 and gst-v13 branches to 
> help comparing these revisions.
> 
> Note that to test GStreamer support with QEMU you probably need to grab 
> the patch below and set the video-codecs option to something like 
> gstreamer:h264 or gstreamer:vp8.
> 
> https://lists.freedesktop.org/archives/spice-devel/2015-May/019771.html
> 
> 
> Changes from v13:
> 
> Rebased + some changes suggested by Christophe Fergeau.
> 
> 
> Changes from v12:
> 
> This round focused on performance and reliability in various special 
> cases such as low bandwidth operation.
> 
>  * The server no longer rebuilds the pipeline from scratch every time 
>    the frame size or bitrate changes. The trick to it is to transition 
>    to the NULL state (instead of the PAUSED one as before).
>  * I also unified configure_pipeline() and reconfigure_pipeline() by 
>    moving all the one-time configuration operations to 
>    create_pipeline() and tracking the configuration changes to apply.
>  * The server can now also apply bitrate changes directly if the codec 
>    advertises it can be changed in the PLAYING state. This is the case 
>    for the x264enc codec.
>  * The server can now detect when the GStreamer pipeline error occurs 
>    instead of getting stuck waiting for the compressed buffer.
>  * I fixed the value for the error-resilient vp8enc option.
>  * I removed the resize-allowed=true vp8enc setting as tests showed the 
>    decoder was sometimes not restoring a full-size frame.
>  * I removed the 'frame' SpiceGstVideoBuffer field as it was not used.
>  * push_raw_frame() now avoids allocating memory for chunk_copy() if 
>    zero_copy() was able to take care of the whole frame.
> 
>  * The client no longer rebuilds or even reconfigure the pipeline from 
>    scratch every time the frame size changes. It turns out that the 
>    decoders don't care (unlike the encoders).
>  * I have protected the PTS calculation against underflows. This fixes 
>    freezes that could occasionally happen with negative latencies near 
>    the start of a stream (i.e. in low bandwidth cases).
>  * I renamed SpiceFrameMeta to just SpiceFrame since it also contains 
>    the sample once the frame has been decoded, and is thus not just 
>    metadata. I also cleaned up the naming of the corresponding queues.
>  * The GStreamer pipeline sometimes returns multiple buffers with the 
>    same PTS. The first buffer would remove its SpiceFrame from the 
>    decoding_queue so that the next one would not find a match, thus 
>    clearing the decoding_queue in the process. As a result the next 
>    buffer would not find a match either, again clearing the 
>    decoding_queue in the process, etc. So buffers that have an unknown 
>    PTS no longer clear the decoding_queue.
>  * When frames are late for display the client now tries to display the 
>    least out of date one in the display_queue so there are at least 
>    some screen updates.
>  * I removed the qos=true appsink option. It's not clear that it was 
>    helping so I decided to remove it until there is more evidence that 
>    it is needed.
>  * I renamed stream_dropped_frame() to 
>    stream_dropped_frame_on_playback() to make it clearer that it's 
>    unrelated to the stream report data (which is based on arrival time).
> 
> 
> Besides rebases, the changes from v11 are limited to spice-gtk. 
> Specifically they fix two compatibility issues:
>  * v11 used a GAsyncQueue and relied on g_async_queue_push_front() which 
>    is only available in GLib >= 2.46 which is unfortunately not 
>    available in Debian 8 and presumably on many other Linux 
>    distributions.
>  * It also associated the original network message with the 
>    corresponding GStreamer buffer using the GstMeta API. This is 
>    necessary in order to know where to display the decoded frame as this 
>    could change from one frame to the next. However some GStreamer 
>    elements have (refcounting) bugs that caused the metadata to be lost 
>    in the pipeline. I had a workaround in place that worked on Debian 
>    Testing but did not on Debian 8.
>    https://bugzilla.gnome.org/show_bug.cgi?id=757254
> 
> So in the end I reworked both by putting the metadata in a 

Re: [Spice-devel] [PATCH 1/2] Remove unnecessary cursor_pipe_item_ref()

[Pavel Grunt]

Hi,

Am I right that your next step is to remove hold_item() ?

On Thu, 2016-05-19 at 14:05 -0500, Jonathon Jongsma wrote:
> Use the base red_pipe_item_ref() instead of adding an additional static
> wrapper function.

Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  server/cursor-channel.c | 14 +-
>  1 file changed, 1 insertion(+), 13 deletions(-)
> 
> diff --git a/server/cursor-channel.c b/server/cursor-channel.c
> index c0b2fda..fa462c5 100644
> --- a/server/cursor-channel.c
> +++ b/server/cursor-channel.c
> @@ -340,23 +340,11 @@ static void cursor_channel_send_item(RedChannelClient
> *rcc, RedPipeItem *pipe_it
>  red_channel_client_begin_send_message(rcc);
>  }
>  
> -static RedCursorPipeItem *cursor_pipe_item_ref(RedCursorPipeItem *item)
> -{
> -spice_return_val_if_fail(item, NULL);
> -
> -red_pipe_item_ref(item);
> -return item;
> -}
> -
> -
>  static void cursor_channel_hold_pipe_item(RedChannelClient *rcc, RedPipeItem
> *item)
>  {
> -RedCursorPipeItem *cursor_pipe_item;
> -
>  spice_return_if_fail(item);
>  
> -cursor_pipe_item = SPICE_CONTAINEROF(item, RedCursorPipeItem, base);
> -cursor_pipe_item_ref(cursor_pipe_item);
> +red_pipe_item_ref(item);
>  }
>  
>  CursorChannel* cursor_channel_new(RedWorker *worker)
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH 2/2] Simplify DisplayChannel hold_item()

[Pavel Grunt]

On Thu, 2016-05-19 at 14:05 -0500, Jonathon Jongsma wrote:
> Since all pipe items implement refcounting now, just call
> red_pipe_item_ref() and eliminate the switch statement.

Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  server/display-channel.c | 11 +--
>  1 file changed, 1 insertion(+), 10 deletions(-)
> 
> diff --git a/server/display-channel.c b/server/display-channel.c
> index 1487f10..3f414fd 100644
> --- a/server/display-channel.c
> +++ b/server/display-channel.c
> @@ -1934,16 +1934,7 @@ static void hold_item(RedChannelClient *rcc,
> RedPipeItem *item)
>  {
>  spice_return_if_fail(item);
>  
> -switch (item->type) {
> -case RED_PIPE_ITEM_TYPE_DRAW:
> -case RED_PIPE_ITEM_TYPE_IMAGE:
> -case RED_PIPE_ITEM_TYPE_STREAM_CLIP:
> -case RED_PIPE_ITEM_TYPE_UPGRADE:
> -red_pipe_item_ref(item);
> -break;
> -default:
> -spice_warn_if_reached();
> -}
> +red_pipe_item_ref(item);
>  }
>  
>  static int handle_migrate_flush_mark(RedChannelClient *rcc)
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v3 5/7] spice-uri: Do not allow empty port string

[Pavel Grunt]

On Fri, 2016-05-20 at 10:32 +0200, Victor Toso wrote:
> Hi,
> 
> On Thu, May 19, 2016 at 06:38:07PM +0200, Pavel Grunt wrote:
> > Related: rhbz#1335239
> > ---
> >  src/spice-uri.c| 3 +++
> >  tests/test-spice-uri.c | 1 +
> >  2 files changed, 4 insertions(+)
> > 
> > diff --git a/src/spice-uri.c b/src/spice-uri.c
> > index 04ea3cb..b483374 100644
> > --- a/src/spice-uri.c
> > +++ b/src/spice-uri.c
> > @@ -171,6 +171,9 @@ gboolean spice_uri_parse(SpiceURI *self, const gchar
> > *_uri, GError **error)
> >  g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED,
> >  "Invalid uri port: %s", uri_port);
> >  goto end;
> > +} else if (endptr == uri_port) {
> > +g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED, "Missing uri port");
> > +goto end;
> 
> You might want to use/include the return value of strtoul for zero in this
> case

I would not change the value, parsing failed so SpiceURI should not be consider
valid

Pavel

> 
> Acked-by: Victor Toso <victort...@redhat.com>
> 
> >  }
> >  spice_uri_set_port(self, port);
> >  }
> > diff --git a/tests/test-spice-uri.c b/tests/test-spice-uri.c
> > index 1b14dbb..42c9aad 100644
> > --- a/tests/test-spice-uri.c
> > +++ b/tests/test-spice-uri.c
> > @@ -34,6 +34,7 @@ static void test_spice_uri_ipv4(void)
> >  {"http://:80;, "http", NULL, 80, NULL, NULL}, /* missing hostname
> > */
> >  {"http://;, "http", NULL, 3128, NULL, NULL},
> >  {"http://127.0.0.1:port;, "http", "127.0.0.1", 3128, NULL, NULL},
> > /* invalid port */
> > +{"http://127.0.0.1:;, "http", "127.0.0.1", 3128, NULL, NULL},
> >  };
> >  const struct test_case valid_test_cases[] = {
> >  {"http://user:password@host:80;, "http", "host", 80, "user",
> > "password"},
> > -- 
> > 2.8.2
> > 
> > ___
> > Spice-devel mailing list
> > Spice-devel@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/spice-devel
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v3 3/7] spice-uri: Add missing include

[Pavel Grunt]

On Fri, 2016-05-20 at 10:14 +0200, Victor Toso wrote:
> Hi,
> 
> On Thu, May 19, 2016 at 06:38:05PM +0200, Pavel Grunt wrote:
> > Related: rhbz#1335239
> > ---
> >  src/spice-uri.c | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/src/spice-uri.c b/src/spice-uri.c
> > index 8cf870d..3bdb502 100644
> > --- a/src/spice-uri.c
> > +++ b/src/spice-uri.c
> > @@ -22,6 +22,7 @@
> > 
> >  #include "spice-client.h"
> >  #include "spice-uri.h"
> > +#include "spice-uri-priv.h"
> 
> The -priv is internal and it always includes the non -priv one.
> I would change the include above spice-uri.h to spice-uri-priv.h

I would include both, but can remove the other one if you prefer.

Pavel

> 
> Acked-by: Victor Toso <victort...@redhat.com>
> 
> >  
> >  /**
> >   * SECTION:spice-uri
> > -- 
> > 2.8.2
> > 
> > ___
> > Spice-devel mailing list
> > Spice-devel@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/spice-devel
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v3 7/7] spice-uri: Add ipv6 support

[Pavel Grunt]

On Fri, 2016-05-20 at 10:54 +0200, Victor Toso wrote:
> Hi,
> 
> On Thu, May 19, 2016 at 06:38:09PM +0200, Pavel Grunt wrote:
> > Just basic support -  http://user:password@[host]:port
> > 
> > Resolves: rhbz#1335239
> > ---
> >  src/spice-uri.c| 25 +
> >  tests/test-spice-uri.c | 42 ++
> >  2 files changed, 63 insertions(+), 4 deletions(-)
> > 
> > diff --git a/src/spice-uri.c b/src/spice-uri.c
> > index 6a43461..323fbf7 100644
> > --- a/src/spice-uri.c
> > +++ b/src/spice-uri.c
> > @@ -150,10 +150,29 @@ gboolean spice_uri_parse(SpiceURI *self, const gchar
> > *_uri, GError **error)
> >  uri = next;
> >  }
> >  
> > -/* max 2 parts, host:port */
> > -gchar **uriv = g_strsplit(uri, ":", 2);
> > +gchar **uriv;
> >  const gchar *uri_port = NULL;
> >  
> > +if (*uri == '[') { /* ipv6 address */
> > +uriv = g_strsplit(uri + 1, "]", 2);
> > +if (uriv[1] == NULL) {
> > +g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED,
> > +"Missing ']' in ipv6 uri");
> > +goto end;
> > +}
> > +if (*uriv[1] == ':') {
> > +uri_port = uriv[1] + 1;
> > +} else if (strlen(uriv[1]) > 0) { /* invalid string after the
> > hostname */
> > +g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED,
> > +"Invalid uri address");
> > +goto end;
> > +}
> > +} else {
> > +/* max 2 parts, host:port */
> > +uriv = g_strsplit(uri, ":", 2);
> > +uri_port = uriv[1];
> > +}
> > +
> 
> This seems to leak (I think Frediano said it before)
> 
This should not leak, uriv is initialized by g_strsplit() and in the
end g_strfreev(uriv) is called.

The leaks come from the patch "spice-uri: Reset SpiceUri before parsing" as you
noticed before

Thanks!
Pavel


> Full valgrind in the end.
> 
> 
> >  if (uriv[0] == NULL || strlen(uriv[0]) == 0) {
> >  g_set_error(error, SPICE_CLIENT_ERROR, SPICE_CLIENT_ERROR_FAILED,
> >  "Invalid hostname in uri address");
> > @@ -161,8 +180,6 @@ gboolean spice_uri_parse(SpiceURI *self, const gchar
> > *_uri, GError **error)
> >  }
> >  
> >  spice_uri_set_hostname(self, uriv[0]);
> > -if (uriv[0] != NULL)
> > -uri_port = uriv[1];
> >  
> >  if (uri_port != NULL) {
> >  gchar *endptr;
> > diff --git a/tests/test-spice-uri.c b/tests/test-spice-uri.c
> > index c32a343..9cdcb90 100644
> > --- a/tests/test-spice-uri.c
> > +++ b/tests/test-spice-uri.c
> > @@ -66,11 +66,53 @@ static void test_spice_uri_ipv4(void)
> >  g_object_unref(uri);
> >  }
> >  
> > +static void test_spice_uri_ipv6(void)
> > +{
> > +const struct test_case invalid_test_cases[] = {
> > +{"http://[]:80;, "http", NULL, 80, NULL, NULL},
> > +{"http://[::1;, "http", NULL, 3128, NULL, NULL},
> > +{"http://[host]1234;, "http", "host", 3128, NULL, NULL},
> > +{"http://[host]foo/;, "http", "host", 3128, NULL, NULL},
> > +{"http://[::1]:port;, "http", "::1", 3128, NULL, NULL},
> > +{"http://[::127.0.0.1]:;, "http", "::127.0.0.1", 3128, NULL, NULL},
> > +{"http://[::127.0.0.1]:-42;, "http", "::127.0.0.1", 3128, NULL,
> > NULL},
> > +{"[3ffe:2a00:100:7031::1]:4200", "http",
> > "3ffe:2a00:100:7031::1", 3128, NULL, NULL},
> > +};
> > +const struct test_case valid_test_cases[] = {
> > +{"http://user:password@[host]:80/;, "http", "host", 80, "user",
> > "password"},
> > +{"http://user@[1080:0:0:0:8:800:200C:4171]:100;, "http",
> > "1080:0:0:0:8:800:200C:4171", 100,
> > + "user", NULL},
> > +{"https://[1080::8:800:200C:417A];, "https",
> > "1080::8:800:200C:417A", 3129, NULL, NULL},
> > +{"[3ffe:2a00:100:7031::1]", "http", "3ffe:2a00:100:7031::1", 3128,
> > NULL, NULL},
> > +};
> > +
> >

Re: [Spice-devel] [PATCH spice-gtk v3 6/7] spice-uri: Check if port is in allowed range

[Pavel Grunt]

On Fri, 2016-05-20 at 10:36 +0200, Victor Toso wrote:
> Hi,
> 
> On Thu, May 19, 2016 at 06:38:08PM +0200, Pavel Grunt wrote:
> > Related: rhbz#1335239
> > ---
> >  src/spice-uri.c| 8 ++--
> >  tests/test-spice-uri.c | 2 ++
> >  2 files changed, 8 insertions(+), 2 deletions(-)
> > 
> > diff --git a/src/spice-uri.c b/src/spice-uri.c
> > index b483374..6a43461 100644
> > --- a/src/spice-uri.c
> > +++ b/src/spice-uri.c
> > @@ -165,8 +165,8 @@ gboolean spice_uri_parse(SpiceURI *self, const gchar
> > *_uri, GError **error)
> >  uri_port = uriv[1];
> >  
> >  if (uri_port != NULL) {
> > -char *endptr;
> > -guint port = strtoul(uri_port, , 10);
> > +gchar *endptr;
> > +gint64 port = g_ascii_strtoll(uri_port, , 10);
> >  if (*endptr != '\0') {
> >  g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED,
> >  "Invalid uri port: %s", uri_port);
> > @@ -175,6 +175,10 @@ gboolean spice_uri_parse(SpiceURI *self, const gchar
> > *_uri, GError **error)
> >  g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED, "Missing uri port");
> >  goto end;
> >  }
> > +if (port < 0 || port > 65535) {
> > +g_set_error(error, SPICE_CLIENT_ERROR,
> > SPICE_CLIENT_ERROR_FAILED, "Port out of range");
> > +goto end;
> > +}
> >  spice_uri_set_port(self, port);
> >  }
> >  
> > diff --git a/tests/test-spice-uri.c b/tests/test-spice-uri.c
> > index 42c9aad..c32a343 100644
> > --- a/tests/test-spice-uri.c
> > +++ b/tests/test-spice-uri.c
> > @@ -35,6 +35,8 @@ static void test_spice_uri_ipv4(void)
> >  {"http://;, "http", NULL, 3128, NULL, NULL},
> >  {"http://127.0.0.1:port;, "http", "127.0.0.1", 3128, NULL, NULL},
> > /* invalid port */
> >  {"http://127.0.0.1:;, "http", "127.0.0.1", 3128, NULL, NULL},
> > +{"http://127.0.0.1:-80;, "http", "127.0.0.1", 3128, NULL, NULL},
> > +{"http://127.0.0.1:800;, "http", "127.0.0.1", 3128, NULL,
> > NULL},
> >  };
> >  const struct test_case valid_test_cases[] = {
> >  {"http://user:password@host:80;, "http", "host", 80, "user",
> > "password"},
> 
> Looks good. You might want to include test to check the error you have
> included.

Check for the error message ? Code/domain is the same.

I will add it.

Thanks,
Pavel

> 
> Acked-by: Victor Toso <victort...@redhat.com>
> 
> > -- 
> > 2.8.2
> > 
> > ___
> > Spice-devel mailing list
> > Spice-devel@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/spice-devel
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCHv2 1/3] egl: don't destroy wayland egl context

[Pavel Grunt]

On Mon, 2016-05-23 at 13:01 +0200, Marc-André Lureau wrote:
> The egl context is from Gtk on Wayland. Destroy it only on X11.
> 
> Signed-off-by: Marc-André Lureau <marcandre.lur...@gmail.com>
Acked-by: Pavel Grunt <pgr...@redhat.com>
> ---
>  src/spice-widget-egl.c | 28 +---
>  1 file changed, 17 insertions(+), 11 deletions(-)
> 
> diff --git a/src/spice-widget-egl.c b/src/spice-widget-egl.c
> index 961dcfa..a3db856 100644
> --- a/src/spice-widget-egl.c
> +++ b/src/spice-widget-egl.c
> @@ -364,10 +364,6 @@ void spice_egl_unrealize_display(SpiceDisplay *display)
>  d->egl.tex_pointer_id = 0;
>  }
>  
> -if (d->egl.surface != EGL_NO_SURFACE) {
> -eglDestroySurface(d->egl.display, d->egl.surface);
> -d->egl.surface = EGL_NO_SURFACE;
> -}
>  if (d->egl.vbuf_id) {
>  glDeleteBuffers(1, >egl.vbuf_id);
>  d->egl.vbuf_id = 0;
> @@ -378,14 +374,24 @@ void spice_egl_unrealize_display(SpiceDisplay *display)
>  d->egl.prog = 0;
>  }
>  
> -if (d->egl.ctx) {
> -eglDestroyContext(d->egl.display, d->egl.ctx);
> -d->egl.ctx = 0;
> +if (GDK_IS_X11_DISPLAY(gdk_display_get_default())) {
> +/* egl.surface && egl.ctx are only created on x11, see
> +   spice_egl_init() */
> +
> +if (d->egl.surface != EGL_NO_SURFACE) {
> +eglDestroySurface(d->egl.display, d->egl.surface);
> +d->egl.surface = EGL_NO_SURFACE;
> +}
> +
> +if (d->egl.ctx) {
> +eglDestroyContext(d->egl.display, d->egl.ctx);
> +d->egl.ctx = 0;
> +}
> +
> +eglMakeCurrent(d->egl.display, EGL_NO_SURFACE, EGL_NO_SURFACE,
> +   EGL_NO_CONTEXT);
> +eglTerminate(d->egl.display);
>  }
> -
> -eglMakeCurrent(d->egl.display, EGL_NO_SURFACE, EGL_NO_SURFACE,
> -   EGL_NO_CONTEXT);
> -eglTerminate(d->egl.display);
>  }
>  
>  G_GNUC_INTERNAL
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

Re: [Spice-devel] [PATCH spice-gtk v2 3/5] spice-uri: Set user & password to NULL when not specified

[Pavel Grunt]

On Thu, 2016-05-19 at 06:24 -0400, Frediano Ziglio wrote:
> > 
> > 
> > Avoid using old values after parsing a new uri.
> > 
> > Related: rhbz#1335239
> > ---
> >  src/spice-uri.c| 3 +++
> >  tests/test-spice-uri.c | 2 +-
> >  2 files changed, 4 insertions(+), 1 deletion(-)
> > 
> > diff --git a/src/spice-uri.c b/src/spice-uri.c
> > index 0cc2195..35d773e 100644
> > --- a/src/spice-uri.c
> > +++ b/src/spice-uri.c
> > @@ -141,6 +141,9 @@ gboolean spice_uri_parse(SpiceURI *self, const gchar
> > *_uri, GError **error)
> >  spice_uri_set_user(self, user);
> >  spice_uri_set_password(self, pass);
> >  uri = next;
> > +} else {
> > +spice_uri_set_user(self, NULL);
> > +spice_uri_set_password(self, NULL);
> >  }
> >  
> 
> Why not setting all to NULL at the beginning ?

Sure
> Does the internal state has to be unchanged if the function returns FALSE ?

I don't think so
> 
> >  /* max 2 parts, host:port */
> > diff --git a/tests/test-spice-uri.c b/tests/test-spice-uri.c
> > index e8cfcc9..3d21fa7 100644
> > --- a/tests/test-spice-uri.c
> > +++ b/tests/test-spice-uri.c
> > @@ -36,10 +36,10 @@ static void test_spice_uri_ipv4(void)
> >  {"http://127.0.0.1:port;, "http", "127.0.0.1", 3128, NULL, NULL},
> > /*
> >  invalid port */
> >  };
> >  const struct test_case valid_test_cases[] = {
> > +{"http://user:password@host:80;, "http", "host", 80, "user",
> > "password"},
> >  {"http://127.0.0.1/;, "http", "127.0.0.1", 3128, NULL, NULL},
> >  {"https://127.0.0.1;, "https", "127.0.0.1", 3129, NULL, NULL},
> >  {"127.0.0.1", "http", "127.0.0.1", 3128, NULL, NULL},
> > -{"http://user:password@host:80;, "http", "host", 80, "user",
> > "password"},
> >  };
> >  
> 
> I think a comment stating why they are in this order would be good.

Right, it was in previous version.

Pavel
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Spice-devel] [PATCH spice-gtk v3 0/7] Allow ipv6 proxy url

[Pavel Grunt]

Hi,

these patches add ipv6 support to SpiceUri, so it can be used in SPICE_PROXY.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1335239

v3 per Frediano's review:
 - Added test for port out of range (0, 65535)
 - reset all values of SpiceUri before parsing

v2 contains changes suggested by Frediano and Victor:
 - more tests (missing port, missing ending bracket)
 - tests are defined in array, so it is easier to add new test cases
 - follow ipv6 address format http://user:password@[ipv6]:port

Thanks,*

Pavel Grunt (7):
  tests: Add test for SpiceUri
  spice-uri: Mark parameter as unused
  spice-uri: Add missing include
  spice-uri: Reset SpiceUri before parsing
  spice-uri: Do not allow empty port string
  spice-uri: Check if port is in allowed range
  spice-uri: Add ipv6 support

 src/spice-uri.c|  46 ---
 tests/Makefile.am  |   2 +
 tests/test-spice-uri.c | 118 +
 3 files changed, 159 insertions(+), 7 deletions(-)
 create mode 100644 tests/test-spice-uri.c

-- 
2.8.2

___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel