Add a function to look up an xrandr output for a given device display
id. This uses sysfs and the drm subsystem to lookup information about a
graphics device output. It then compares the drm output name to xrandr
output names to try to match that device output to an xrandr output.
This is necesary
When the agent gets a new device info message (from the daemon), we need
to re-calculate the guest output map and send that information back down
to the daemon so that it can handle mouse input events correctly.
---
src/vdagent/x11-randr.c | 3 +++
1 file changed, 3 insertions(+)
diff --git
There are basically three ways to refer to an output within vdagent:
- The index of the array of MonitorConfig message. This is essentially
a "spice display id"
- the index of the array of xrandr outputs. This is the "output index"
- the xrandr output id. This is the "output ID"
When sending the guest xorg resolution to the vdagentd daemon, we get
the current resolution with this function, which returns a
VDAgentMonitorsConfig structure that must then be converted to the
structure that we send to the daemon. Rather than re-using this function
that returns the wrong type,
---
src/vdagent/vdagent.c | 68 +
src/vdagent/x11-priv.h | 1 +
src/vdagent/x11-randr.c | 58 +++
src/vdagent/x11.c | 13
src/vdagent/x11.h | 1 +
5 files changed, 74 insertions(+), 67 deletions(-)
This is a patch set that handles the PCI address and device dispay ID
sent down to the agent by the server, and uses that to maintain a map
for looking up a particular xrandr output for a given spice display id.
This patch series builds on the patch from Lukas titled "Receive the
Instead of storing each device address and device display ID in the hash
table, simply use the lookup_xrandr_output_for_device_info() function to
look up the ID of the xrandr output and store that in the hash table.
---
src/vdagent/x11-priv.h | 2 +-
src/vdagent/x11-randr.c | 47
instead of using the spice display id directly as the xrandr output,
look it up using our new guest output map
---
src/vdagent/x11-randr.c | 78 -
1 file changed, 69 insertions(+), 9 deletions(-)
diff --git a/src/vdagent/x11-randr.c
Rather than getting the current guest resolution in a
VDAgentMonitorsConfig struct and then translating it to a new struct
type for sending down to the daemon, simply use the new function that
was factored out in a previous commit and populate the message struct
directly.
---
Add a display_id field to the structure that we use to send down the
list of guest display resolutions to the vdagentd daemon. This allows us
to map the spice display id to the proper X display for determining
mouse locations, etc.
---
src/vdagent/x11-randr.c | 35
On 12/27/18 8:51 AM, Niccolò Belli wrote:
> On mercoledì 26 dicembre 2018 13:38:28 CET, Frediano Ziglio wrote:
>> Yes, this looks like a format string error in the upper (not into
>> spice) layer.
>>
>> This potentially is a security problem.
>
> Considering the spice server is exposed to the
On Thu, 2019-01-03 at 16:46 +, Frediano Ziglio wrote:
> The SpiceStat structure can be 20 or 24 bytes depending on alignment.
> Being a memory mapped structure potentially used with lockless
> access, it
> is not good to have it unaligned.
> The current tool that reads this memory mapped file
Assuming that you've tested that it compiles
Acked-by: Jonathon Jongsma
On Thu, 2019-01-03 at 10:02 +, Frediano Ziglio wrote:
> Signed-off-by: Frediano Ziglio
> ---
> src/vdagent/x11-priv.h | 4
> src/vdagent/x11-randr.c | 1 +
> src/vdagent/x11.h | 4
> 3 files changed,
On Thu, Jan 3, 2019 at 9:22 PM Frediano Ziglio wrote:
>
> >
> > From: Marc-André Lureau
> >
> > Local caps are set during construction, no need to reset to the same
> > caps on channel_reset().
> >
> > This also solves calling spice_channel_reset_capabilities() without an
> > associated session
>
> From: Marc-André Lureau
>
> Local caps are set during construction, no need to reset to the same
> caps on channel_reset().
>
> This also solves calling spice_channel_reset_capabilities() without an
> associated session (after a channel disconnect).
>
> Also fix keeping CAP_AUTH_SASL on
>
> On Wed, Jan 02, 2019 at 02:58:45PM +, Frediano Ziglio wrote:
> > Windows needs some specific setting to use network.
> >
> > Signed-off-by: Frediano Ziglio
> > ---
> > configure.ac | 20 +++-
> > 1 file changed, 19 insertions(+), 1 deletion(-)
> >
> > Changes since v1:
From: Marc-André Lureau
Add a new "label" stack page. When EGL is unsupported and a GL scanout
message is received, switch to that page with an appropriate message
hint.
At this point, no further switch back to 2d drawing is implemented,
although it is theorically possible for a server to
From: Marc-André Lureau
Allow to set whether GL scanout capability is set, at the session level.
SPICE_DISABLE_GL_SCANOUT environment variable can be used for tweaking
and testing. Note: the server doesn't seem to handle this case very
nicely yet: the simplest solution could perhaps be to
From: Marc-André Lureau
In order to set capabilities based on session configuration, post-pone
display capabilities after all channel properties are set.
Signed-off-by: Marc-André Lureau
---
src/channel-display.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
From: Marc-André Lureau
Local caps are set during construction, no need to reset to the same
caps on channel_reset().
This also solves calling spice_channel_reset_capabilities() without an
associated session (after a channel disconnect).
Also fix keeping CAP_AUTH_SASL on reset.
Signed-off-by:
From: Marc-André Lureau
Hi,
This series addresses:
https://gitlab.freedesktop.org/spice/spice-gtk/issues/69
The first 2 patches are cleanups, the 3rd patch introduces a
SpiceSession:gl-scanout property, so a client can tweak the
capability. The last patch displays a message about EGL support
On Thu, Jan 03, 2019 at 11:54:15AM -0500, Frediano Ziglio wrote:
>
> What about:
>
> "interface" and "MAX_MONITORS" are defined in some Windows system
> headers causing garbage code to be fed to the compiler.
Yes this is the missing bit.
Christophe
signature.asc
Description: PGP signature
Acked-by: Jonathon Jongsma
On Mon, 2018-12-17 at 10:53 +, Frediano Ziglio wrote:
> Support for protocol version 1 was dropped in commit f77a1d50.
>
> Signed-off-by: Frediano Ziglio
> ---
> src/channel-base.c | 4 ---
> src/channel-main.c | 69 +++-
>
> From: Marc-André Lureau
>
> - "Since:" tag must have colon
> - s/gstreamer/GStreamer
> - s/regeister/register
>
> Signed-off-by: Marc-André Lureau
Acked-by: Frediano Ziglio
> ---
> src/channel-display.c | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git
> On Sat, Dec 22, 2018 at 09:27:02AM -0500, Frediano Ziglio wrote:
> >
> > > On Fri, Dec 21, 2018 at 4:04 PM Frediano Ziglio
> > > wrote:
> > > >
> > > > Signed-off-by: Frediano Ziglio
> > >
> > > I would suggest to wrap with #if _WIN32, otherwise it's really weird
> > > to see those undef
On Sat, Dec 22, 2018 at 09:27:02AM -0500, Frediano Ziglio wrote:
>
> > On Fri, Dec 21, 2018 at 4:04 PM Frediano Ziglio wrote:
> > >
> > > Signed-off-by: Frediano Ziglio
> >
> > I would suggest to wrap with #if _WIN32, otherwise it's really weird
> > to see those undef without context.
> >
>
From: Marc-André Lureau
- "Since:" tag must have colon
- s/gstreamer/GStreamer
- s/regeister/register
Signed-off-by: Marc-André Lureau
---
src/channel-display.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/src/channel-display.c b/src/channel-display.c
index
On Wed, Jan 02, 2019 at 02:56:43PM +, Frediano Ziglio wrote:
> Formatting string should be compatible with GLib.
> GLib uses formatting types compatible with GNU.
> For Linux this is not an issue as both systems (like a printf) and
> GLib one uses the same formatting type. However on Windows
Hi
On Thu, Jan 3, 2019 at 7:22 PM Frediano Ziglio wrote:
>
> >
> > On Thu, Jan 03, 2019 at 04:46:29PM +0200, Snir Sheriber wrote:
> > > GST_TYPE_PIPELINE is undeclared when gstvideo is disabled
> >
> > I'd rahter that we keep using the more specific type. Feel free to have
> > a
> > #ifndef
> On Thu, Jan 03, 2019 at 10:15:43AM -0500, Frediano Ziglio wrote:
> > >
> > > On Thu, Jan 03, 2019 at 02:14:24PM +, Frediano Ziglio wrote:
> > > > The SpiceStat structure can be 20 or 24 bytes depending on alignment.
> > > > Being a memory mapped structure potentially used with lockless
Acked-by: Jonathon Jongsma
On Thu, 2019-01-03 at 10:02 +, Frediano Ziglio wrote:
> Modify directly the new_wbuf->buf buffer instead of writing into
> a temporary buffer and then copy in the final one.
> This also fixes Coverity warning:
>
> | uninit_use_in_call: Using uninitialized value
On Wed, Jan 02, 2019 at 02:58:45PM +, Frediano Ziglio wrote:
> Windows needs some specific setting to use network.
>
> Signed-off-by: Frediano Ziglio
> ---
> configure.ac | 20 +++-
> 1 file changed, 19 insertions(+), 1 deletion(-)
>
> Changes since v1:
> - all changes in
On Thu, Jan 03, 2019 at 10:15:43AM -0500, Frediano Ziglio wrote:
> >
> > On Thu, Jan 03, 2019 at 02:14:24PM +, Frediano Ziglio wrote:
> > > The SpiceStat structure can be 20 or 24 bytes depending on alignment.
> > > Being a memory mapped structure potentially used with lockless access
> > >
On Thu, Jan 3, 2019 at 8:12 PM Christophe Fergeau wrote:
>
> Hey,
>
> On Thu, Jan 03, 2019 at 03:41:35PM +0400, marcandre.lur...@redhat.com wrote:
> > From: Marc-André Lureau
> >
> > Simple command line tools simply connect all channels, however
> > usbredir channels require that the manager was
On Thu, Jan 03, 2019 at 10:15:43AM -0500, Frediano Ziglio wrote:
> >
> > On Thu, Jan 03, 2019 at 02:14:24PM +, Frediano Ziglio wrote:
> > > The SpiceStat structure can be 20 or 24 bytes depending on alignment.
> > > Being a memory mapped structure potentially used with lockless access
> > >
Hey,
On Thu, Jan 03, 2019 at 03:41:35PM +0400, marcandre.lur...@redhat.com wrote:
> From: Marc-André Lureau
>
> Simple command line tools simply connect all channels, however
> usbredir channels require that the manager was previously
> initialized. Currently, running spicy-stats on a VM with
Can you explain the end goal here?
On Fri, 2018-12-21 at 12:02 +, Frediano Ziglio wrote:
> Mainly SPICE server uses lot of libraries to expose a TCP protocol.
> As TCP is implemented with socket library which is quite portable was
> not that hard to port.
> Beside some minor feature (see
On Fri, 2018-12-21 at 12:03 +, Frediano Ziglio wrote:
> Signed-off-by: Frediano Ziglio
> ---
> README.Windows | 18 ++
> 1 file changed, 18 insertions(+)
> create mode 100644 README.Windows
>
> diff --git a/README.Windows b/README.Windows
> new file mode 100644
> index
>
> On Thu, Jan 03, 2019 at 04:46:29PM +0200, Snir Sheriber wrote:
> > GST_TYPE_PIPELINE is undeclared when gstvideo is disabled
>
> I'd rahter that we keep using the more specific type. Feel free to have
> a
> #ifndef HAVE_GST_VIDEO
> #define GST_TYPE_PIPELINE G_TYPE_POINTER
> #endif
It sounds
Hi,
On Thu, Jan 03, 2019 at 03:41:35PM +0400, marcandre.lur...@redhat.com wrote:
> From: Marc-André Lureau
>
> Simple command line tools simply connect all channels, however
> usbredir channels require that the manager was previously
> initialized. Currently, running spicy-stats on a VM with
>
> On Thu, Jan 03, 2019 at 02:14:24PM +, Frediano Ziglio wrote:
> > The SpiceStat structure can be 20 or 24 bytes depending on alignment.
> > Being a memory mapped structure potentially used with lockless access
> > is not good to have it unaligned.
> > The current tool that reads this
Acked-by: Christophe Fergeau
On Thu, Jan 03, 2019 at 02:00:14PM +, Frediano Ziglio wrote:
> Due to previous commit "make syntax-check" command reports:
>
> prohibit_signal_without_use
> server/tests/test-display-base.c
> maint.mk: the above files include signal.h but don't use it
> make:
On Thu, Jan 03, 2019 at 02:14:24PM +, Frediano Ziglio wrote:
> The SpiceStat structure can be 20 or 24 bytes depending on alignment.
> Being a memory mapped structure potentially used with lockless access
> is not good to have it unaligned.
> The current tool that reads this memory mapped file
On Thu, Jan 03, 2019 at 04:46:29PM +0200, Snir Sheriber wrote:
> GST_TYPE_PIPELINE is undeclared when gstvideo is disabled
I'd rahter that we keep using the more specific type. Feel free to have
a
#ifndef HAVE_GST_VIDEO
#define GST_TYPE_PIPELINE G_TYPE_POINTER
#endif
but would that code be used
>
> GST_TYPE_PIPELINE is undeclared when gstvideo is disabled
>
> Signed-off-by: Snir Sheriber
This fixes current CI which is doing this test.
Also removes this warning (I just got it):
GISCAN SpiceClientGLib-2.0.gir
:: Warning: SpiceClientGLib: (Signal)gst-video-overlay: argument
GST_TYPE_PIPELINE is undeclared when gstvideo is disabled
Signed-off-by: Snir Sheriber
---
src/channel-display.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/channel-display.c b/src/channel-display.c
index 2a1d9d1..b0fe57e 100644
--- a/src/channel-display.c
+++
The SpiceStat structure can be 20 or 24 bytes depending on alignment.
Being a memory mapped structure potentially used with lockless access
is not good to have it unaligned.
The current tool that reads this memory mapped file (reds_stats) is
able to detect if the structure is either 20 or 24 bytes
Due to previous commit "make syntax-check" command reports:
prohibit_signal_without_use
server/tests/test-display-base.c
maint.mk: the above files include signal.h but don't use it
make: *** [maint.mk:639: sc_prohibit_signal_without_use] Error 1
Signed-off-by: Frediano Ziglio
---
>
> This patch is changing the way GstVideoOverlay is being set.
> Once pipeline is created a pointer is passed to the widget using
> GObject signal, so we can set there the overlay interface and call
> its functions from widget callbacks. This allows to solve issues
> like resizing the window.
>
This patch is changing the way GstVideoOverlay is being set.
Once pipeline is created a pointer is passed to the widget using
GObject signal, so we can set there the overlay interface and call
its functions from widget callbacks. This allows to solve issues
like resizing the window.
With previous patch changes it's now possible to call GstVideoOverlay
interface functions in response to matching gtk events. This patch fixes
the window resize and possibly other rendering issues.
Signed-off-by: Snir Sheriber
---
src/spice-widget.c | 42
On Thu, Jan 3, 2019 at 3:31 PM Frediano Ziglio wrote:
>
> If input contains the binary record we can't have it modified
> during read.
>
> Signed-off-by: Frediano Ziglio
Reviewed-by: Marc-André Lureau
> ---
> server/tests/replay.c | 5 -
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
From: Marc-André Lureau
Simple command line tools simply connect all channels, however
usbredir channels require that the manager was previously
initialized. Currently, running spicy-stats on a VM with usbredir
channels prints:
(lt-spicy-stats:25224): GSpice-CRITICAL **: 14:30:54.724:
If input contains the binary record we can't have it modified
during read.
Signed-off-by: Frediano Ziglio
---
server/tests/replay.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
Changes since v1:
- open file passed with name in binary mode too.
diff --git a/server/tests/replay.c
>
> On 1/3/19 10:43 AM, Frediano Ziglio wrote:
> >> If playback is not active and it's streaming mode set latency to 0
> >> so frames will not be synchronized with mm time.
> >>
> >> Signed-off-by: Snir Sheriber
> >> ---
> >>
> >> This patch is a suggestion to improve current state.
> >>
> >>
On 1/3/19 10:43 AM, Frediano Ziglio wrote:
If playback is not active and it's streaming mode set latency to 0
so frames will not be synchronized with mm time.
Signed-off-by: Snir Sheriber
---
This patch is a suggestion to improve current state.
Latency in display channel is the difference
Not supported, %m is a GNU extension of sscanf.
Signed-off-by: Frediano Ziglio
---
server/reds.c | 25 -
1 file changed, 12 insertions(+), 13 deletions(-)
Changes since v1:
- use a single code path.
diff --git a/server/reds.c b/server/reds.c
index 84e4bb8c..ec7fbc38
Signed-off-by: Frediano Ziglio
---
src/udscs.c| 2 ++
src/vdagent/x11.c | 2 ++
src/vdagentd/virtio-port.c | 1 +
3 files changed, 5 insertions(+)
diff --git a/src/udscs.c b/src/udscs.c
index 7fe74b9..fe2cf8c 100644
--- a/src/udscs.c
+++ b/src/udscs.c
@@ -343,12 +343,14
Signed-off-by: Frediano Ziglio
---
src/vdagent/vdagent.c | 10 ++
src/vdagentd/vdagentd.c | 10 ++
2 files changed, 12 insertions(+), 8 deletions(-)
diff --git a/src/vdagent/vdagent.c b/src/vdagent/vdagent.c
index 87dfa7c..eb741bc 100644
--- a/src/vdagent/vdagent.c
+++
We need to pass original argument list, however argv is modified by
g_option_context_parse so save a copy of the array to make possible
to restart with the same arguments.
Signed-off-by: Frediano Ziglio
---
src/vdagent/vdagent.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff
Different tools (compilers or analysers) recognise missing break
in switch as possible problems.
Many recognise "fall through" comment to tell is wanted.
Signed-off-by: Frediano Ziglio
---
src/vdagentd/vdagentd.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/vdagentd/vdagentd.c
Modify directly the new_wbuf->buf buffer instead of writing into
a temporary buffer and then copy in the final one.
This also fixes Coverity warning:
| uninit_use_in_call: Using uninitialized value "message_header". Field
| "message_header.data" is uninitialized when calling "memcpy".
Most removing some warnings related to Coverity.
Other just cleanups.
Frediano Ziglio (7):
virtio-port: Avoid temporary buffers
vdagentd: Tell different tools that we are falling through
Ignore some Coverity reports
Check errors setting standard file descriptors
Do not include config.h
Hi,
On Wed, Jan 2, 2019, 5:31 PM Victor Toso Hi,
>
> Thanks for taking a look!
>
> On Sun, Dec 30, 2018 at 10:23:02PM +0100, Jakub Janku wrote:
> > Hi,
> >
> > On Wed, Dec 19, 2018 at 3:30 PM Victor Toso
> wrote:
> > >
> > > From: Victor Toso
> > >
> > > If SpiceGtkSession is holding the
ping
>
> An array of WSABUF can be used with WSASend.
>
> Signed-off-by: Frediano Ziglio
> ---
> common/marshaller.c | 10 --
> common/marshaller.h | 4 ++--
> 2 files changed, 10 insertions(+), 4 deletions(-)
>
> This will allow to port spice-server to Windows,
ping
>
> > On Thu, Dec 13, 2018 at 02:20:25PM +, Frediano Ziglio wrote:
> > > Signed-off-by: Frediano Ziglio
> > > ---
> > > common/ssl_verify.c | 5 +-
> > > common/ssl_verify.h | 4 ++
> > > tests/Makefile.am | 20 ++
> > > tests/test-ssl-verify.c | 141
ping
>
> Support for protocol version 1 was dropped in commit f77a1d50.
>
> Signed-off-by: Frediano Ziglio
> ---
> src/channel-base.c | 4 ---
> src/channel-main.c | 69 +++---
> 2 files changed, 29 insertions(+), 44 deletions(-)
>
> Patch much
>
> If playback is not active and it's streaming mode set latency to 0
> so frames will not be synchronized with mm time.
>
> Signed-off-by: Snir Sheriber
> ---
>
> This patch is a suggestion to improve current state.
>
> Latency in display channel is the difference between multimedia time
>
> From: Marc-André Lureau
>
> The passed argument is already a SpiceChannel.
>
> Signed-off-by: Marc-André Lureau
Acked-by: Frediano Ziglio
> ---
> src/channel-display.c | 20 ++--
> 1 file changed, 10 insertions(+), 10 deletions(-)
>
> diff --git
>
> From: Marc-André Lureau
>
> Not strictly required, but slightly more clear and consistent with the
> rest of the code.
>
> Signed-off-by: Marc-André Lureau
Acked-by: Frediano Ziglio
> ---
> src/channel-display.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git
70 matches
Mail list logo
