Hi All,
With F19 fully updated host and guest, with or without virt-preview, is
this supposed to work when configuring launching from virt-manager /
libvirtd?
I use remote-viewer to connect to vm.
I'd like to be able to have two "monitors", adjust the resolution to
match my physical (two 19
On 09/23/2013 11:23 AM, Christophe Fergeau wrote:
On Sun, Sep 22, 2013 at 02:39:36PM +0300, Uri Lublin wrote:
On 09/20/2013 06:07 PM, Christophe Fergeau wrote:
What is v->verifyop value when this problem occurs ?
When this occurs, v->verifyop would be SPICE_SSL_VERIFY_OP_HOSTNAME |
SPICE_SSL_VE
On 09/24/2013 02:42 PM, Marc-André Lureau wrote:
Hi,
Please find a few patches fixing gl_fbo usage and glitches.
spice-common needs to be updated too, with the opengl related fixes.
(the submodule update will be done before, so ignore updates during
review of these patches please)
Looks good
- Original Message -
> On 09/24/2013 02:42 PM, Marc-André Lureau wrote:
>
>
> Looks good, but can you explain why you disable diff check for fill, text?
Setting CHECKDIFF=1 environment variable will compare the rendering of the
selected canvas with a software canvas. This is useful fo
- Original Message -
> > It should only be called in enter or focus events, which I assume shouldn't
> > happen when the widget is not realized.
>
> it receives a focus event.
Ok, interesting.
>
> > > So this results is a call to GDK_WINDOW_HWND(NULL), which simply crash
> > > remote
- Original Message -
> On 09/24/2013 02:47 PM, Marc-André Lureau wrote:
> > I don't see why the fallback method should be the default.
> >
> > glCopyTexImage2D and glCopyPixels are equally broken with current Intel
> > mesa/drivers (I'll be filing a bug). But sw rendering is correct.
> >
On 09/24/2013 02:42 PM, Marc-André Lureau wrote:
This appears to be unrelated to the gl checks, same as the nograb fix
(excellent). Can you move both to the start of the patchset?
---
client/display_channel.cpp | 4
1 file changed, 4 insertions(+)
diff --git a/client/display_channel.c
On 09/24/2013 02:42 PM, Marc-André Lureau wrote:
Looks good, but can you explain why you disable diff check for fill, text?
---
client/display_channel.cpp | 88 +-
client/display_channel.h | 2 ++
client/red_pixmap.h| 19 ++
cl
On 09/24/2013 02:47 PM, Marc-André Lureau wrote:
I don't see why the fallback method should be the default.
glCopyTexImage2D and glCopyPixels are equally broken with current Intel
mesa/drivers (I'll be filing a bug). But sw rendering is correct.
---
common/glc.c | 2 +-
1 file changed, 1 inse
> It should only be called in enter or focus events, which I assume shouldn't
> happen when the widget is not realized.
it receives a focus event.
> > So this results is a call to GDK_WINDOW_HWND(NULL), which simply crash
> > remote-viewer.exe
>
> Ok that macro isn't safe. Why do you check (!di
- Original Message -
> I try to cross compile virt-viewer to win32 using Debian Jessie. So far it
> works,
> but only if I apply the following patch to spice-widget.c:
>
> Index: new/gtk/spice-widget.c
> ===
> --- new.orig/g
Hi,
> Someone can help me to find the problem that makes qxl unusable please?
#1 git cherry-pick c58c7b959b93b864a27fd6b3646ee1465ab8832b
#2 When using f19 try without X11 first. You should have a working
framebuffer console on qxldrmfb before trying to get X11 going.
#3 qxl has a bunch o
I try to cross compile virt-viewer to win32 using Debian Jessie. So far it
works,
but only if I apply the following patch to spice-widget.c:
Index: new/gtk/spice-widget.c
===
--- new.orig/gtk/spice-widget.c 2013-06-26 15:55:08.00
Updating commit message for rationale:
The primary buffer doesn't use stencil test. However, this should be
explicitely disabled, since the canvas might change stencil state, and
this will affect primary stencil buffer, making some of further update
operations clipped in unwanted ways.
On Tue, Se
I don't see why the fallback method should be the default.
glCopyTexImage2D and glCopyPixels are equally broken with current Intel
mesa/drivers (I'll be filing a bug). But sw rendering is correct.
---
common/glc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/common/glc.c b/
Keeping the warning, because I don't think this should happen anyway.
---
common/gl_canvas.c | 10 +++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/common/gl_canvas.c b/common/gl_canvas.c
index 176c3ef..fe152ef 100644
--- a/common/gl_canvas.c
+++ b/common/gl_canvas.c
@@ -75
Similar to sw canvas blit copy operation, and not a blend.
---
common/gl_canvas.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/common/gl_canvas.c b/common/gl_canvas.c
index 0736a08..176c3ef 100644
--- a/common/gl_canvas.c
+++ b/common/gl_canvas.c
@@ -373,6 +373,7 @@ static void gl_canvas_dr
Similar to how sw canvas ignore mask if !bitmap in canvas_mask_pixman().
---
common/gl_canvas.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/common/gl_canvas.c b/common/gl_canvas.c
index 2adc986..0736a08 100644
--- a/common/gl_canvas.c
+++ b/common/gl_canvas.c
@@ -207,7 +2
Hi,
Here is a few fixes for the gl canvas.
(the gl canvas still doesn't support offscreen surfaces)
Marc-André Lureau (4):
gl: ignore if mask has not bitmap
gl: ignore alpha channel in draw_copy
gl: cope with positive stride in put_image()
gl: use glCopyPixels()
common/gl_canvas.c | 14
---
client/display_channel.cpp | 4
1 file changed, 4 insertions(+)
diff --git a/client/display_channel.cpp b/client/display_channel.cpp
index c9ce3d8..47076ad 100644
--- a/client/display_channel.cpp
+++ b/client/display_channel.cpp
@@ -1278,6 +1278,10 @@ void DisplayChannel::create_canvas(i
---
client/x11/platform.cpp | 2 +-
client/x11/red_pixmap_gl.cpp | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/client/x11/platform.cpp b/client/x11/platform.cpp
index 475ee6a..b8563b3 100644
--- a/client/x11/platform.cpp
+++ b/client/x11/platform.cpp
@@ -3231,7 +3231,7
Ensure the GL state is appropriate for calling GCanvas::copy_pixels.
---
client/red_gl_canvas.cpp | 2 ++
1 file changed, 2 insertions(+)
diff --git a/client/red_gl_canvas.cpp b/client/red_gl_canvas.cpp
index 81203d9..e8c3f1e 100644
--- a/client/red_gl_canvas.cpp
+++ b/client/red_gl_canvas.cpp
@@
---
client/display_channel.cpp | 88 +-
client/display_channel.h | 2 ++
client/red_pixmap.h| 19 ++
client/utils.h | 3 ++
4 files changed, 88 insertions(+), 24 deletions(-)
diff --git a/client/display_channel.cpp b/clie
---
client/cache.hpp | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/client/cache.hpp b/client/cache.hpp
index e267f42..7dce671 100644
--- a/client/cache.hpp
+++ b/client/cache.hpp
@@ -45,7 +45,8 @@ public:
while (*item) {
if ((*item)->id == id) {
-
Similar to spice-gtk, disable mouse and keyboard grab (useful when
running under a debugger)
---
client/x11/red_window.cpp | 7 +++
1 file changed, 7 insertions(+)
diff --git a/client/x11/red_window.cpp b/client/x11/red_window.cpp
index d82f5b7..7e16fcd 100644
--- a/client/x11/red_window.cpp
---
client/x11/red_window.cpp | 31 ---
client/x11/red_window_p.h | 3 ---
2 files changed, 34 deletions(-)
diff --git a/client/x11/red_window.cpp b/client/x11/red_window.cpp
index b6e11e1..d82f5b7 100644
--- a/client/x11/red_window.cpp
+++ b/client/x11/red_window.cpp
This visually reduces glitches without noticeable speed difference.
It's also the traditionnal way of doing opengl.
---
client/red_window.h | 2 ++
client/screen.cpp| 1 +
client/x11/platform.cpp | 2 +-
client/x11/red_pixmap_gl.cpp | 2 --
client/x11/red_window.cpp
First, context must set it, then Draw/ReadBuffer must be set to FRONT,
and then explicit Flush is needed.
This patch is mostly for future reference, it is mostly discarded in
following patch using double-buffer.
---
client/x11/platform.cpp | 1 +
client/x11/red_drawable.cpp | 1 +
client/x1
---
client/x11/pixels_source_p.h | 1 +
client/x11/red_drawable.cpp | 12 +++
client/x11/red_pixmap_gl.cpp | 84 +++-
3 files changed, 50 insertions(+), 47 deletions(-)
diff --git a/client/x11/pixels_source_p.h b/client/x11/pixels_source_p.h
index 64b
This helps to identify the code location of exception
---
client/utils.h | 3 +++
1 file changed, 3 insertions(+)
diff --git a/client/utils.h b/client/utils.h
index c23e04f..4657279 100644
--- a/client/utils.h
+++ b/client/utils.h
@@ -19,6 +19,7 @@
#define _H_UTILS
#include "common.h"
+#inclu
Hi,
Please find a few patches fixing gl_fbo usage and glitches.
spice-common needs to be updated too, with the opengl related fixes.
(the submodule update will be done before, so ignore updates during
review of these patches please)
Marc-André Lureau (11):
spicec: warn when throwing exception
This property indicates whether to look into the system
CA database when validating certificates in a TLS connection.
This property defaults to TRUE, but is automatically set to
FALSE when SpiceSession:ca-file is set.
---
gtk/spice-option.c | 8
gtk/spice-session-priv.h | 3 +++
g
Hey, here is a v4 of my series.
Changes since v3:
- added a line at the end of configure indicating which CA store will be
used (if any), this was requested during review but I forgot to add it in v3
- fix compilation when using --without-ca-certificates
Currently, spice-gtk will look in $HOME/.spicec/spice_truststore.pem
by default for its trust certificate store (to verify the certificates
used during SPICE TLS connections). However, these days a system-wide
trust store can be found in /etc/pki or /etc/ssl.
This commit checks at compile time wher
34 matches
Mail list logo
