On Wed, 2015-10-21 at 21:45 +0200, Fabiano Fidêncio wrote:
> On Wed, Oct 21, 2015 at 2:00 PM, Frediano Ziglio
> wrote:
> > From: Marc-André Lureau
> >
> > Clean up, more extensible.
> >
>
> Not exactly sure what it means.
>
> > Avoid server hanging when no client are connected.
>
> How does
Hi,
On 22-10-15 17:41, Victor Toso wrote:
Hi,
On Thu, Oct 22, 2015 at 05:28:44PM +0200, Hans de Goede wrote:
Hi,
On 22-10-15 16:07, Victor Toso wrote:
For streaming devices it might be necessary from application to drop
data for different reasons. This patch provides a new callback that it
i
On Wed, 2015-10-21 at 13:00 +0100, Frediano Ziglio wrote:
> From: Marc-André Lureau
>
> Move code around to declare and place it where it fits better.
> ---
> server/red_dispatcher.c | 106 +-
> --
> server/red_dispatcher.h | 7
> server/red_wor
Ah, now I see it's this patch which needs to disable zero copy. ACK.
On Wed, Oct 14, 2015 at 05:33:26PM +0200, Francois Gouget wrote:
> configure will use GStreamer 1.0 if present and fall back to
> GStreamer 0.10 otherwise.
>
> Signed-off-by: Francois Gouget
> ---
> configure.ac
ACK.
On Wed, Oct 14, 2015 at 05:33:19PM +0200, Francois Gouget wrote:
> This prevents the bit rate from going insanely high on fast networks
> for no visible video quality improvement.
>
> Signed-off-by: Francois Gouget
> ---
> server/gstreamer_encoder.c | 2 ++
> 1 file changed, 2 insertions(+
On Wed, Oct 14, 2015 at 05:33:55PM +0200, Francois Gouget wrote:
> Otherwise it may get wrapped to a much lower value than intended.
Quite complicated, but looks good, ACK.
>
> Signed-off-by: Francois Gouget
> ---
> server/gstreamer_encoder.c | 65
> +++
On Tue, Oct 20, 2015 at 12:21:58PM +0200, Francois Gouget wrote:
>
> This iteration got some review on the spice-gtk side for which I
> provided updated patches. Since last friday there have been some commits
> in the spice repository that cause some conflicts so I'll resubvmit the
> patch series.
On Wed, Oct 14, 2015 at 05:33:07PM +0200, Francois Gouget wrote:
> This typically happens when sending very small frames (less than
> 16 pixels in one dimension) to the x264enc encoder.
> This avoids repeatedly wasting time rebuilding the pipeline.
>
> Signed-off-by: Francois Gouget
> ---
> serv
Looks good, ACK.
On Wed, Oct 14, 2015 at 05:32:46PM +0200, Francois Gouget wrote:
> Signed-off-by: Francois Gouget
> ---
> server/gstreamer_encoder.c | 17 -
> server/red_dispatcher.c| 2 ++
> 2 files changed, 18 insertions(+), 1 deletion(-)
>
> diff --git a/server/gstream
On Wed, 2015-10-21 at 08:21 -0400, Frediano Ziglio wrote:
> >
> > From: Marc-André Lureau
> >
> > ---
> > server/dispatcher.c | 13 +
> > server/dispatcher.h | 2 ++
> > server/red_dispatcher.c | 21 +++--
> > server/red_dispatcher.h | 6 +-
> > server/
On Thu, 2015-10-22 at 13:10 +0200, Fabiano Fidêncio wrote:
> On Thu, Oct 22, 2015 at 1:03 PM, Victor Toso
> wrote:
> > Hi,
> >
> > On Thu, Oct 22, 2015 at 10:06:04AM +0200, Fabiano Fidêncio wrote:
> > > On Thu, Oct 22, 2015 at 9:13 AM, Victor Toso <
> > > li...@victortoso.com> wrote:
> > > > Hi,
Hi,
On Thu, Oct 22, 2015 at 05:28:44PM +0200, Hans de Goede wrote:
> Hi,
>
> On 22-10-15 16:07, Victor Toso wrote:
> >For streaming devices it might be necessary from application to drop
> >data for different reasons. This patch provides a new callback that it
> >is called before queueing the mos
Hi,
On 22-10-15 16:07, Victor Toso wrote:
For streaming devices it might be necessary from application to drop
data for different reasons. This patch provides a new callback that it
is called before queueing the most recent iso packages.
Related: https://bugzilla.redhat.com/show_bug.cgi?id=1264
Fixes color order on PowerPC when using LZ image compression.
---
Changes since v2:
- No changes
Changes since v1:
- New commit
---
common/canvas_base.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/common/canvas_base.c b/common/canvas_base.c
index 31f3cef..57b80f6 10
Fixes color order on PowerPC when using jpeg compression.
---
Changes since v2:
- New commit
---
common/canvas_base.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/common/canvas_base.c b/common/canvas_base.c
index 4c1bd2c..65b1c3f 100644
--- a/common/canvas_base.c
+++ b/
When checking, whether we use alpha channel, we need to compare color
format with constant with the same byte order.
When clearing alpha channel, we must clear in proper byte order.
---
Changes since v2:
- Added macro for proper alpha clearing
- Improved commit msg
Changes since v1:
- New comm
Fixes color order on PowerPC when using LZ4 image compression.
---
Changes since v2:
- No changes
Changes since v1:
- New commit
---
common/canvas_base.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/common/canvas_base.c b/common/canvas_base.c
index 57b80f6..4c1bd2c 1
Fixes color order on PowerPC when using QUIC image compression.
---
Changes since v2:
Changes since v1:
- No changes
---
common/canvas_base.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/common/canvas_base.c b/common/canvas_base.c
index 6f48340..31f3cef 100644
--- a/c
On LE machine, color order will always be ARGB.
On BE machines color order will sometimes be ARGB and sometimes BGRA.
(This is because surface is sometimes created using native byte order,
and sometimes using the network byte order, which is always LE)
This change introduces cases for BGRA color b
After the previous commit, spice_bitmap_try_as_pixman() can be
simplified as its #ifdef WORDS_BIGENDIAN exactly match what the
PIXMAN_LE_ constants do.
---
Changes since v2:
- Improved commit msg
Changes since v1:
- New commit
---
common/pixman_utils.c | 18 +++---
1 file changed, 3
When using image compression on PowerPC architecture, colors are in
wrong order ARGB -> BGRA.
This commit introduces macros, that will change the color order
according to machine endianness.
Theese macros are similar to QEMU macros in qemu-pixman.h
---
Changes since v2:
- None
Changes since v1:
When using image compression on BE client, the color order was wrong.
(BGRA instead of ARGB)
Theese patches try to fix this issue in most of the image compressions.
Only those compressions I could test are changed.
For most of the fixes, new macros for color order were implemented.
Theese macros
On Wed, Oct 14, 2015 at 05:33:45PM +0200, Francois Gouget wrote:
> This relies on the GStreamer buffer's lifetime being short enough which
> it is because:
> - We encode frames one by one.
> - For all encoders but MJPEG, the first element of the pipeline will
>convert the bitmap to another im
Hi,
there is a leftover (mentioned bellow). Fixed locally.
On Thu, Oct 22, 2015 at 04:07:18PM +0200, Victor Toso wrote:
> For streaming devices it might be necessary from application to drop
> data for different reasons. This patch provides a new callback that it
> is called before queueing the m
When channel wants to send much more data then the wire can handle, the
queue grows fast. This patch does not limit the queue growth but
introduces an internal API to check if queue size is too big.
This internal API is used with usbredirhost_can_write_iso callback from
usbredir. This way, we gara
Hi,
Let me know if the values of threshold makes sense (or any other issue).
changes from last version:
- threshold set by usbredirhost instead of client;
- helper function to set the threshold
- using bool for dropping packages;
- making the amount of packages sent be *at least* the difference
For streaming devices it might be necessary from application to drop
data for different reasons. This patch provides a new callback that it
is called before queueing the most recent iso packages.
Related: https://bugzilla.redhat.com/show_bug.cgi?id=1264156
---
usbredirhost/usbredirhost.c | 63 +++
>
> > On Thu, Oct 22, 2015 at 11:48 AM, Frediano Ziglio
> > wrote:
> > >>
> > >> On Wed, Oct 21, 2015 at 3:53 PM, Frediano Ziglio
> > >> wrote:
> > >> >>
> > >> >> On Wed, Oct 21, 2015 at 08:37:25AM -0400, Frediano Ziglio wrote:
> > >> >> >
> > >> >> > >
> > >> >> > > From: Marc-André Lureau
>
On Wed, Oct 14, 2015 at 05:33:38PM +0200, Francois Gouget wrote:
> This way the video encoder is not forced to use malloc()/free().
> This also allows more flexibility in how the video encoder manages the
> buffer which allows for a zero-copy implementation in both video
> encoders.
> The current i
Hi,
Almost all installers provide this kind of info, and the idea is to show it so
everyone, including ones who don't really know how and where to search will get
a pointer.
So I think it's a good idea to tell everyone who runs the installer "it's
produced with NSIS", so that anyone who may nee
On Thu, Oct 22, 2015 at 02:22:19PM +0200, Lukas Venhoda wrote:
> Only check for address lenght, when connecting trough IP address.
Fwiw, 'length', not 'lenght', 'through' rather than 'trough'
> It is not used, when connecting trough DNS hostname.
> ---
> Changes since v3:
> - Nothing
>
> Change
Change inet_aton function to glib functions.
inet_aton only supported IPv4 addresses, and wasn't available on windows
machines. GInetAddress functions support IPv6 natively, and requires less
boilerplate code then IPv6 gettaddrinfo().
---
Changes since v2:
- Changed printf to spice_debug
Changes
Only check for address lenght, when connecting trough IP address.
It is not used, when connecting trough DNS hostname.
---
Changes since v3:
- Nothing
Changes since v2:
- Also moved variable declarations
Changes since v1:
- New patch
---
common/ssl_verify.c | 15 ---
1 file change
This is required by the new GInetAddress functions.
---
Changes since v3:
- Changed commit message
- Now cleanly applies after f7ec855af3deff18aeb43ddbf2467c565f4bf346
Changes since v2:
- New patch
---
m4/spice-deps.m4 | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/m4/spic
On Thu, Oct 22, 2015 at 11:42 AM, Fabiano Fidêncio
wrote:
> Is there a specific reason for changing the g_debug() to printf()?
>
> >
> > -if ((addr_len == alt_ip_len)&&
> > -!memcmp(ASN1_STRING_data(name->d.iPAddress), &addr,
> addr_len)) {
> > -
On Wed, Oct 14, 2015 at 05:32:03PM +0200, Francois Gouget wrote:
> The Spice server administrator can specify the preferred encoder and
> codec preferences to optimize for CPU or bandwidth usage. Preferences
> are described in a semi-colon separated list of encoder:codec pairs.
> For instance 'gstr
Hi,
On Wed, Oct 21, 2015 at 03:52:50PM -0500, Jonathon Jongsma wrote:
> In certain circumstances we were printing an error message even though
> the file transfer had completed successfully. It didn't cause any
> problems, but it pointed out an issue in the handling of outgoing agent
> messages.
>
Hi,
On Thu, Oct 22, 2015 at 01:10:04PM +0200, Fabiano Fidêncio wrote:
> On Thu, Oct 22, 2015 at 1:03 PM, Victor Toso wrote:
> Sure it is.
> OTOH, g_clear_object() works basically as g_clear_pointer(&object,
> g_object_unref); and, in this case, it is not what we want, right?.
Indeed, my mistake.
On Thu, Oct 22, 2015 at 12:22:11PM +0300, Yedidyah Bar David wrote:
> On Thu, Oct 22, 2015 at 11:49 AM, Christophe Fergeau
> wrote:
> > On Thu, Oct 22, 2015 at 11:39:42AM +0300, Yedidyah Bar David wrote:
> >> On Thu, Oct 22, 2015 at 11:28 AM, Christophe Fergeau
> >> wrote:
> >> > Having smaller/m
On Thu, Oct 22, 2015 at 06:39:56AM -0400, Lev Veyde wrote:
> Hi Christophe,
>
> Yes, that's the reason.
>
> I don't see any reason to hide the fact that we use another open
> source project (NSIS) to produce the installer, even if it provides
> that option.
> Unless of course, we absolutely must
On Thu, Oct 22, 2015 at 1:03 PM, Victor Toso wrote:
> Hi,
>
> On Thu, Oct 22, 2015 at 10:06:04AM +0200, Fabiano Fidêncio wrote:
>> On Thu, Oct 22, 2015 at 9:13 AM, Victor Toso wrote:
>> > Hi,
>> >
>> > On Wed, Oct 21, 2015 at 03:52:49PM -0500, Jonathon Jongsma wrote:
>> >> +if (c->file_xfer_t
Hi,
On Thu, Oct 22, 2015 at 10:06:04AM +0200, Fabiano Fidêncio wrote:
> On Thu, Oct 22, 2015 at 9:13 AM, Victor Toso wrote:
> > Hi,
> >
> > On Wed, Oct 21, 2015 at 03:52:49PM -0500, Jonathon Jongsma wrote:
> >> +if (c->file_xfer_tasks) {
> >> +g_hash_table_unref(c->file_xfer_tasks);
>
Hi Christophe,
Yes, that's the reason.
I don't see any reason to hide the fact that we use another open source project
(NSIS) to produce the installer, even if it provides that option.
Unless of course, we absolutely must use the window space real-estate for
something else, and note that fact s
> On Thu, Oct 22, 2015 at 11:48 AM, Frediano Ziglio wrote:
> >>
> >> On Wed, Oct 21, 2015 at 3:53 PM, Frediano Ziglio
> >> wrote:
> >> >>
> >> >> On Wed, Oct 21, 2015 at 08:37:25AM -0400, Frediano Ziglio wrote:
> >> >> >
> >> >> > >
> >> >> > > From: Marc-André Lureau
> >> >> > >
> >> >> > > --
On Thu, Oct 22, 2015 at 11:48 AM, Frediano Ziglio wrote:
>>
>> On Wed, Oct 21, 2015 at 3:53 PM, Frediano Ziglio wrote:
>> >>
>> >> On Wed, Oct 21, 2015 at 08:37:25AM -0400, Frediano Ziglio wrote:
>> >> >
>> >> > >
>> >> > > From: Marc-André Lureau
>> >> > >
>> >> > > ---
>> >> > > server/red_di
On Thu, Oct 22, 2015 at 11:17 AM, Frediano Ziglio wrote:
>
>>
>> On Wed, Oct 21, 2015 at 2:15 PM, Frediano Ziglio wrote:
>> >
>> >>
>> >> From: Alon Levy
>> >>
>> >> ---
>> >> server/red_worker.c | 5 +
>> >> 1 file changed, 5 insertions(+)
>> >>
>> >> diff --git a/server/red_worker.c b/ser
>
> On Wed, Oct 21, 2015 at 3:53 PM, Frediano Ziglio wrote:
> >>
> >> On Wed, Oct 21, 2015 at 08:37:25AM -0400, Frediano Ziglio wrote:
> >> >
> >> > >
> >> > > From: Marc-André Lureau
> >> > >
> >> > > ---
> >> > > server/red_dispatcher.c | 6 --
> >> > > server/red_dispatcher.h | 2 +-
> >>
On Wed, Oct 14, 2015 at 4:57 PM, Lukas Venhoda wrote:
> Change inet_aton function to glib functions.
>
> inet_aton only supported IPv4 addresses, and wasn't available on windows
> machines. GInetAddress functions support IPv6 natively, and requires less
> boilerplate code then IPv6 gettaddrinfo().
On Wed, Oct 14, 2015 at 05:32:14PM +0200, Francois Gouget wrote:
> This introduces a pared down GStreamer-based video encoder to serve as
> the basis for later enhancements.
> In this form the new encoder supports both regular and sized streams
> but lacks any rate control. It should still work fin
On Thu, Oct 22, 2015 at 11:26 AM, Fabiano Fidêncio wrote:
> On Wed, Oct 14, 2015 at 4:57 PM, Lukas Venhoda wrote:
>> This is required by the new IPv6 verify_hostname.
>
> I would like to have a mention that it's required because of
> GInetAddress functions instead of "new IPv6 verify_hostanem"
>
On Wed, Oct 14, 2015 at 4:57 PM, Lukas Venhoda wrote:
> This is required by the new IPv6 verify_hostname.
I would like to have a mention that it's required because of
GInetAddress functions instead of "new IPv6 verify_hostanem"
> ---
> Changes since v2:
> - New patch
> ---
> m4/spice-deps.m4
On Wed, Oct 14, 2015 at 4:57 PM, Lukas Venhoda wrote:
> Only check for address lenght, when connecting trough IP address.
> It is not used, when connecting trough DNS hostname.
> ---
> Changes since v2:
> - Also moved variable declarations
>
> Changes since v1:
> - New patch
> ---
> common/ssl_
On Thu, Oct 22, 2015 at 11:49 AM, Christophe Fergeau
wrote:
> On Thu, Oct 22, 2015 at 11:39:42AM +0300, Yedidyah Bar David wrote:
>> On Thu, Oct 22, 2015 at 11:28 AM, Christophe Fergeau
>> wrote:
>> > Having smaller/more focused commits will allow to have
>> > more specific commit logs :)
>>
>> P
>
> On Wed, Oct 21, 2015 at 2:15 PM, Frediano Ziglio wrote:
> >
> >>
> >> From: Alon Levy
> >>
> >> ---
> >> server/red_worker.c | 5 +
> >> 1 file changed, 5 insertions(+)
> >>
> >> diff --git a/server/red_worker.c b/server/red_worker.c
> >> index ef529f1..225c272 100644
> >> --- a/server
> On Wed, Oct 21, 2015 at 2:26 PM, Frediano Ziglio wrote:
> >
> >>
> >> From: Marc-André Lureau
> >>
> >> Group enums with their respective struct location.
> >> ---
> >> server/red_dispatcher.h | 97
> >> -
> >> server/red_worker.h | 93
> >>
>
> On Wed, 2015-10-21 at 13:00 +0100, Frediano Ziglio wrote:
> > These patches are extracted from a branch intended to refactory
> > spice-server.
> >
> > This patchset supersed last patchset and apply to updated master.
>
>
> Can you provide a little bit of explanation for why this new set
>
On Thu, Oct 22, 2015 at 11:39:42AM +0300, Yedidyah Bar David wrote:
> On Thu, Oct 22, 2015 at 11:28 AM, Christophe Fergeau
> wrote:
> > Having smaller/more focused commits will allow to have
> > more specific commit logs :)
>
> Perhaps I should instead provide more details in the commit message,
On Thu, Oct 22, 2015 at 10:16 AM, Pavel Grunt wrote:
> On Thu, 2015-10-22 at 10:06 +0200, Fabiano Fidêncio wrote:
>> On Thu, Oct 22, 2015 at 9:13 AM, Victor Toso wrote:
>> > Hi,
>> >
>> > On Wed, Oct 21, 2015 at 03:52:49PM -0500, Jonathon Jongsma wrote:
>> > > In order to avoid reference cycles,
On Thu, Oct 22, 2015 at 08:42:08AM +0300, Yedidyah Bar David wrote:
> On Wed, Oct 21, 2015 at 7:55 PM, Christophe Fergeau
> wrote:
> > On Tue, Oct 20, 2015 at 02:35:23PM +0300, Yedidyah Bar David wrote:
> >> From: Christophe Fergeau
> >>
> >> Now that we are able to generate different installers
On Thu, Oct 22, 2015 at 11:28 AM, Christophe Fergeau
wrote:
> On Thu, Oct 22, 2015 at 08:36:49AM +0300, Yedidyah Bar David wrote:
>> >
>> >> Makefile| 100
>> >>
>> >
>> > one adding the Makefile
>> >
>> >> automation/README.md
On Thu, Oct 22, 2015 at 08:36:49AM +0300, Yedidyah Bar David wrote:
> >
> >> Makefile| 100
> >>
> >
> > one adding the Makefile
> >
> >> automation/README.md| 8 +++
> >> automation/build-artifacts.packages | 10
On Thu, 2015-10-22 at 10:06 +0200, Fabiano Fidêncio wrote:
> On Thu, Oct 22, 2015 at 9:13 AM, Victor Toso wrote:
> > Hi,
> >
> > On Wed, Oct 21, 2015 at 03:52:49PM -0500, Jonathon Jongsma wrote:
> > > In order to avoid reference cycles, you're supposed to release
> > > references in dispose, espe
On Thu, Oct 22, 2015 at 10:45 AM, Christophe Fergeau
wrote:
> On Thu, Oct 22, 2015 at 08:31:19AM +0300, Yedidyah Bar David wrote:
>> On Wed, Oct 21, 2015 at 7:52 PM, Christophe Fergeau
>> wrote:
>> > I suspect these scripts are no longer needed? I don't see them being
>> > used after applying yo
On Thu, Oct 22, 2015 at 9:13 AM, Victor Toso wrote:
> Hi,
>
> On Wed, Oct 21, 2015 at 03:52:49PM -0500, Jonathon Jongsma wrote:
>> In order to avoid reference cycles, you're supposed to release
>> references in dispose, especially to those objects that can hold
>> references to yourself. This prob
On Thu, Oct 22, 2015 at 08:28:15AM +0300, Yedidyah Bar David wrote:
> On Wed, Oct 21, 2015 at 7:50 PM, Christophe Fergeau
> wrote:
> > Sorry if I missed it/forgot the answer
>
> You didn't :-)
>
> You already asked and I replied asking Lev.
>
> > , but why do we need/want this
> > change?
>
>
On Thu, Oct 22, 2015 at 08:31:19AM +0300, Yedidyah Bar David wrote:
> On Wed, Oct 21, 2015 at 7:52 PM, Christophe Fergeau
> wrote:
> > I suspect these scripts are no longer needed? I don't see them being
> > used after applying your automation patch.
>
> Again, just used your ovirt branch.
>
>
Hi,
On Wed, Oct 21, 2015 at 03:52:51PM -0500, Jonathon Jongsma wrote:
> This isn't terribly useful for general debugging and makes the log
> pretty noisy when transferring large files.
Definitely :)
Ack!
> ---
> src/channel-main.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/src/chan
Hi,
On Wed, Oct 21, 2015 at 03:52:49PM -0500, Jonathon Jongsma wrote:
> In order to avoid reference cycles, you're supposed to release
> references in dispose, especially to those objects that can hold
> references to yourself. This probably wasn't causing any leaks, since
> the file transfer task
68 matches
Mail list logo
