Re: [Spice-devel] [PATCH v6 5/5] Do endian swapping.

2017-01-23 Thread Christophe de Dinechin
On 23/01/2017 14:53, Michal Suchanek wrote: This allows running big endian and little endian guest side by side using cut & paste between them. There is a general design idea that swapping should come as close to virtio_read/virtio_write as possible. In particular, the protocol between

Re: [Spice-devel] [PATCH v2 spice-gtk 1/2] authentication: Handle failed SASL authentication separately

2017-02-22 Thread Christophe de Dinechin
> On 21 Feb 2017, at 16:28, Snir Sheriber <ssher...@redhat.com> wrote: > > Hi, > > > On 02/20/2017 07:00 PM, Christophe de Dinechin wrote: >>> On 19 Feb 2017, at 15:47, Snir Sheriber <ssher...@redhat.com> wrote: >>> >>> Remove hand

Re: [Spice-devel] [spice v8] dcc: handle preferred video codec message

2017-02-20 Thread Christophe de Dinechin
> On 10 Feb 2017, at 14:14, Victor Toso wrote: > > Hi, > > On Fri, Feb 10, 2017 at 07:51:05AM -0500, Frediano Ziglio wrote: >>> [0] SPICE_MSGC_DISPLAY_PREFERRED_VIDEO_CODEC_TYPE >>> >>> This message provides a list of video codecs based on client's order >>> of

Re: [Spice-devel] [spice v8] dcc: handle preferred video codec message

2017-02-20 Thread Christophe de Dinechin
> On 9 Feb 2017, at 09:21, Victor Toso wrote: > > [0] SPICE_MSGC_DISPLAY_PREFERRED_VIDEO_CODEC_TYPE > > This message provides a list of video codecs based on client's order > of preference. > > We duplicate the video codecs array from reds.c and sort it using the > order

Re: [Spice-devel] [PATCH] Don't close all but one display during reboot.

2017-02-20 Thread Christophe de Dinechin
Is it possible to make the max number of monitors something persistent (e.g. set / get that using libvirt), so that the behavior would be consistent between a first boot and a reboot? Christophe > On 20 Feb 2017, at 15:49, Jonathon Jongsma wrote: > > On Thu, 2017-02-02

Re: [Spice-devel] [PATCH v2 spice-gtk 1/2] authentication: Handle failed SASL authentication separately

2017-02-20 Thread Christophe de Dinechin
> On 19 Feb 2017, at 15:47, Snir Sheriber wrote: > > Remove handling with failures in the SASL authentication > process to separate function > --- > src/spice-channel.c | 44 +++- > 1 file changed, 27 insertions(+), 17 deletions(-) >

Re: [Spice-devel] [spice-gtk v1] gtk-session: reply agent's clipboard request on failure

2017-02-24 Thread Christophe de Dinechin
> On 22 Feb 2017, at 13:29, Victor Toso wrote: > > From: Victor Toso > > We need to reply back to the agent all clipboard requests even in case > of failure otherwise it will have a pending request. The following > error message can be seen in

Re: [Spice-devel] [PATCH spice-gtk] main: Avoid sqrt in monitor compare

2017-02-24 Thread Christophe de Dinechin
> On 24 Feb 2017, at 13:26, Pavel Grunt wrote: > > qsort cares about the sign of the difference and that should stay > the same > --- > src/channel-main.c | 5 ++--- > 1 file changed, 2 insertions(+), 3 deletions(-) > > diff --git a/src/channel-main.c b/src/channel-main.c >

Re: [Spice-devel] GStreamer's zero-copy code is broken

2017-02-24 Thread Christophe de Dinechin
> On 24 Feb 2017, at 11:58, Francois Gouget wrote: > > > The patch below breaks the zero-copy code in the GStreamer video > encoder: > > commit c3d237075b994fe67e58f2b3164cb579e6f4 > Author: Frediano Ziglio > Date: Mon Dec 12 17:22:50 2016

Re: [Spice-devel] [PATCH] Don't close all but one display during reboot.

2017-02-23 Thread Christophe de Dinechin
--client have corresponding windows enabled (and placed correctly, e.g. > onto different monitors) > > Everything looks fine above, but why do we have the bug? The symptom > looks as if the guest "forget" to enable the monitor or someone change it. > Let's dig it o

Re: [Spice-devel] Remote 3d support - streaming and lag

2017-02-14 Thread Christophe de Dinechin
One thing I started to discuss last Wed is how video encoding may also require / introduce a lag on its own. The reason is that some video encodings use predictive (P) or bidirectional (B) frames that are relative to surrounding frames. Only a small fraction of the frames, the intracoded frames

Re: [Spice-devel] [PATCH 1/7] qxl-wddm-dod: Return EDID data to the OS

2017-02-13 Thread Christophe de Dinechin
> On 12 Feb 2017, at 14:09, Yuri Benditovich > wrote: > > Solves failure of HLK "Test for EDID requirements" > EDID contains capabilities and manufacturer data of > the emulated display device. Main parameters are: > Manufacturer code: QXL > Product ID: 0001 >

Re: [Spice-devel] [PATCH 0/5] Documentation and cleanup of Drawable Tree

2017-02-10 Thread Christophe de Dinechin
Overall, I like the patches. But I wonder if it would make sense to switch to a doxygen format for large comments like this? Christophe > On 9 Feb 2017, at 20:46, Jonathon Jongsma wrote: > > On Fri, 2017-02-03 at 16:55 -0600, Jonathon Jongsma wrote: >> Here's a slightly

Re: [Spice-devel] bool or gboolean

2017-02-16 Thread Christophe de Dinechin
> On 16 Feb 2017, at 13:17, Pavel Grunt wrote: > > On Wed, 2017-02-15 at 11:39 -0500, Frediano Ziglio wrote: >> Hi, >> question was raised recently on the ML and IRC. >> >> Some time ago we decided to use gboolean but some of us would like >> to discuss again. >> >> As

Re: [Spice-devel] [spice-server v4 5/5] Change some spice_printerr() to spice_debug()

2017-02-15 Thread Christophe de Dinechin
> On 15 Feb 2017, at 12:45, Frediano Ziglio wrote: > >> @@ -134,7 +134,7 @@ static void red_qxl_display_migrate(RedChannelClient >> *rcc) >> } >> g_object_get(channel, "channel-type", , "id", , NULL); >> dispatcher = (Dispatcher

[Spice-devel] [PATCH] Remove X == TRUE tests for non-boolean values

2017-02-27 Thread Christophe de Dinechin
Using X == TRUE is fragile, since the input value is a uint8_t. It would be surprising if the value was set to 2 or 0xFF and the test failed. Signed-off-by: Christophe de Dinechin <dinec...@redhat.com> --- server/dcc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff

Re: [Spice-devel] Multi-head Spice

2017-02-28 Thread Christophe de Dinechin
> On 28 Feb 2017, at 08:58, Christophe Fergeau <cferg...@redhat.com> wrote: > > On Mon, Feb 27, 2017 at 04:18:12PM -0600, Jonathon Jongsma wrote: >> On Mon, 2017-02-27 at 17:53 +0100, Christophe de Dinechin wrote: >>> I tried to setup a multi-head guest,

Re: [Spice-devel] Multi-head Spice

2017-02-28 Thread Christophe de Dinechin
Jonathon Jongsma wrote: >>>> On Mon, 2017-02-27 at 17:53 +0100, Christophe de Dinechin wrote: >>>>> I tried to setup a multi-head guest, but I’m having trouble >>>>> getting >>>>> multiple displays to work correctly >>>>> >&

Re: [Spice-devel] Multi-head Spice

2017-02-28 Thread Christophe de Dinechin
> On 28 Feb 2017, at 10:01, Snir Sheriber <ssher...@redhat.com> wrote: > > Hi, > > On 02/28/2017 09:58 AM, Christophe Fergeau wrote: >> On Mon, Feb 27, 2017 at 04:18:12PM -0600, Jonathon Jongsma wrote: >>> On Mon, 2017-02-27 at 17:53 +0100, Christophe de D

Re: [Spice-devel] Multi-head Spice

2017-02-28 Thread Christophe de Dinechin
> On 27 Feb 2017, at 23:18, Jonathon Jongsma <jjong...@redhat.com> wrote: > > On Mon, 2017-02-27 at 17:53 +0100, Christophe de Dinechin wrote: >> I tried to setup a multi-head guest, but I’m having trouble getting >> multiple displays to work correctly >> >

Re: [Spice-devel] [PATCH] Remove X == TRUE tests for non-boolean values

2017-02-27 Thread Christophe de Dinechin
he value was set to 2 or 0xFF and the test failed. >>> >>> Signed-off-by: Christophe de Dinechin <dinec...@redhat.com >>> <mailto:dinec...@redhat.com>> >> >> Acked-by: Frediano Ziglio <fzig...@redhat.com <mailto:fzig...@redhat.com>> >

Re: [Spice-devel] [PATCH] Remove X == TRUE tests for non-boolean values

2017-02-27 Thread Christophe de Dinechin
t; > Using X == TRUE is fragile, since the input value is a uint8_t. It would be > surprising if the value was set to 2 or 0xFF and the test failed. > > Signed-off-by: Christophe de Dinechin <dinec...@redhat.com > <mailto:dinec...@redhat.com>> > >

[Spice-devel] Multi-head Spice

2017-02-27 Thread Christophe de Dinechin
I tried to setup a multi-head guest, but I’m having trouble getting multiple displays to work correctly For Jonathon, the part that I had trouble with was adding “heads = ‘4’" to the video / QXL configuration. I don’t see it documented in

Re: [Spice-devel] [PATCH 02/12] qxl-wddm-dod: Use rendering offload thread

2017-03-21 Thread Christophe de Dinechin
> On 20 Mar 2017, at 17:37, Christophe de Dinechin <cdupo...@redhat.com > <mailto:cdupo...@redhat.com>> wrote: > > Side note: the .gitattribute in that project seems all wrong. It makes source > files a pain to edit under Linux or macOS. Why not mark

[Spice-devel] Allocation style in QXLDOD

2017-03-21 Thread Christophe de Dinechin
Hi Vadim, Looking at the code for QxlDod.cpp, I see allocations that look like this: 5c52e50b (Vadim Rozenfeld 2014-09-02 17:36:27 +1000 3169) m_ModeNumbers = reinterpret_cast (new (PagedPool) BYTE [sizeof (USHORT) * ModeCount]); and corresponding deallocations that look

[Spice-devel] Gnome vino-server not listening to IPv4

2017-03-31 Thread Christophe de Dinechin
Hello, [If you don’t use VNC to access a Linux host, you can safely skip this e-mail] I recently lost access to three Linux hosts over VNC, and it took me long enough to figure out what was going on that I thought I’d share and ask for advice. This is recorded as

Re: [Spice-devel] [spice-common] log: Add missing stdio.h include

2017-03-31 Thread Christophe de Dinechin
BTW, putting local headers first makes it possible to detect missing / implicit dependencies in the header, i.e.: #include “macros.h” #include #include #include #include Christophe > On 31 Mar 2017, at 12:03, Christophe Fergeau > wrote:

Re: [Spice-devel] [spice-gtk v2 1/2] Fix possible multimedia time overflow

2017-03-17 Thread Christophe de Dinechin
> On 16 Mar 2017, at 11:49, Marc-André Lureau > wrote: > > Hi > > - Original Message - >> The multimedia time can easily overflow as is encoded in an >> unsigned 32 bit and have a unit of milliseconds so it wrap >> up every 49 days. Use

Re: [Spice-devel] [PATCH spice-server] build: Fix some indentation issue on Makefile.am

2017-03-16 Thread Christophe de Dinechin
On 16 Mar 2017, at 17:08, Christophe Fergeau wrote: > > On Thu, Mar 16, 2017 at 10:26:24AM -0500, Jonathon Jongsma wrote: >> ACK for consistency (although I agree with you: I don't like this style >> of lining up the slashes. I'd rather just place them directly after the >>

Re: [Spice-devel] [PATCH 02/12] qxl-wddm-dod: Use rendering offload thread

2017-03-21 Thread Christophe de Dinechin
> On 20 Mar 2017, at 19:03, Frediano Ziglio > wrote: > > > Side note: the .gitattribute in that project seems all wrong. It makes source > files a pain to edit under Linux or macOS. Why not mark the source files as > text? > To avoid cr/lf

Re: [Spice-devel] [PATCH 02/12] qxl-wddm-dod: Use rendering offload thread

2017-03-21 Thread Christophe de Dinechin
> On 21 Mar 2017, at 13:59, Yuri Benditovich > wrote: > > > > On Mon, Mar 20, 2017 at 2:08 PM, Frediano Ziglio > wrote: > > > > Instead of sending drawable commands down from presentation > > callback, collect

Re: [Spice-devel] [PATCH 02/12] qxl-wddm-dod: Use rendering offload thread

2017-03-21 Thread Christophe de Dinechin
> On 21 Mar 2017, at 11:51, Frediano Ziglio > wrote: > So does Emacs, but it shows additional ^M within the file if CR/LF is > inconsistent throughout the file, as it the case here, presumably because the > .gitattributes was added in the middle

Re: [Spice-devel] [PATCH 02/12] qxl-wddm-dod: Use rendering offload thread

2017-03-20 Thread Christophe de Dinechin
Side note: the .gitattribute in that project seems all wrong. It makes source files a pain to edit under Linux or macOS. Why not mark the source files as text? > On 20 Mar 2017, at 13:08, Frediano Ziglio > wrote: > >> >> Instead of sending

Re: [Spice-devel] [spice-space-pages PATCH v2] Spice Proxy documentation

2017-03-20 Thread Christophe de Dinechin
> On 19 Mar 2017, at 13:54, Uri Lublin wrote: > > Signed-off-by: Uri Lublin > > > Changes since v1: > 1. :modified: was removed from the top > 2. 2 ways to set up proxy -- are numbered > 3. Comment added to not use http_proxy > 4. "Installation (Fedora)"

Re: [Spice-devel] [spice-space-pages PATCH] home.rst: s/and and/and/

2017-03-20 Thread Christophe de Dinechin
> On 19 Mar 2017, at 15:22, Uri Lublin wrote: > > Reported-by: romulous75 (on IRC) > > Signed-off-by: Uri Lublin > --- > home.rst | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/home.rst b/home.rst > index 7bef988..36fe95e 100644 >

Re: [Spice-devel] [spice-gtk 1/2] Fix possible multimedia time overflow

2017-03-17 Thread Christophe de Dinechin
> On 16 Mar 2017, at 11:56, Frediano Ziglio > wrote: > >> >> - Original Message - Hi - Original Message - > The multimedia time can easily overflow as is encoded in an > unsigned 32 bit and have a

Re: [Spice-devel] [PATCH 02/12] qxl-wddm-dod: Use rendering offload thread

2017-03-21 Thread Christophe de Dinechin
> On 21 Mar 2017, at 14:52, Christophe de Dinechin <cdupo...@redhat.com> wrote: > >> >> On 21 Mar 2017, at 13:59, Yuri Benditovich <yuri.benditov...@daynix.com >> <mailto:yuri.benditov...@daynix.com>> wrote: >> >> >> >> On

Re: [Spice-devel] [PATCH v3 0/2] RHEL7 improvements

2017-03-22 Thread Christophe de Dinechin
> On 6 Mar 2017, at 11:42, Frediano Ziglio wrote: > > On 3 Mar 2017, at 13:10, Marc-André Lureau > wrote: > > Hi > > - Original Message - > > Hi > > - Original Message - > These 2 patches attempt to join

Re: [Spice-devel] [PATCH v2 12/12] qxl-wddm-dod: Non-forced memory allocations with VSync

2017-04-04 Thread Christophe de Dinechin
> On 1 Apr 2017, at 18:40, Yuri Benditovich wrote: > > From: "yuri.benditov...@daynix.com" > > In case of VSync is active allocate bitmaps for drawable objects > using non-forced requests. The relation with VSync being active is not

Re: [Spice-devel] [PATCH spice-server] test-leaks: Checks some leaks using TLS

2017-03-09 Thread Christophe de Dinechin
[Resend, apparently stayed in my outbox for a few days] > On 7 Mar 2017, at 17:56, Frediano Ziglio wrote: > > Verify stuff are freed correctly (like TLS context). > The different PKI file required are generated with > base values (localhost and rsa 1024). I went back into

Re: [Spice-devel] [RFC] Spice channels compression

2017-03-09 Thread Christophe de Dinechin
> On 7 Mar 2017, at 16:37, Frediano Ziglio > wrote: > >>> >>> >>> Hi Snir, >>> >>> Regarding compression and another topic recently evoked, endianness, I >>> wonder >>> if encoding using LEB128 had been discussed? >>> >>> LEB128 encodes any

[Spice-devel] Is it possible to build without vala?

2017-03-08 Thread Christophe de Dinechin
In the configure.ac, the default for vala is “no”. But if I build without it, I get: make[4]: Entering directory `/Users/ddd/Work/spice/spice-gtk/src/controller' *** Error: missing valac! *** You must run autogen.sh or configure --enable-vala Is it possible to build spice-gtk without

Re: [Spice-devel] Is it possible to build without vala?

2017-03-08 Thread Christophe de Dinechin
> On 8 Mar 2017, at 19:17, Marc-André Lureau wrote: > > Hi > > - Original Message - >> In the configure.ac, the default for vala is “no”. But if I build without it, >> I get: >> >> make[4]: Entering directory `/Users/ddd/Work/spice/spice-gtk/src/controller' >> >>

[Spice-devel] Alignment of data within spice packets

2017-03-08 Thread Christophe de Dinechin
I have several errors like this: spice-channel.c:1923:12: error: cast from 'uint8_t *' (aka 'unsigned char *') to 'uint32_t *' (aka 'unsigned int *') increases required alignment from 1 to 4 [-Werror,-Wcast-align] ...(uint32_t *)((uint8_t *)c->peer_msg +

Re: [Spice-devel] Alignment of data within spice packets

2017-03-09 Thread Christophe de Dinechin
> On 8 Mar 2017, at 22:16, Victor Toso <li...@victortoso.com> wrote: > > Hi, > > On Wed, Mar 08, 2017 at 07:21:15PM +0100, Christophe de Dinechin wrote: >> I have several errors like this: >> >> spice-channel.c:1923:12: error: ca

Re: [Spice-devel] [RFC] Spice channels compression

2017-03-03 Thread Christophe de Dinechin
Hi Snir, Regarding compression and another topic recently evoked, endianness, I wonder if encoding using LEB128 had been discussed? LEB128 encodes any value below 128 as 1 bytes, any value below ~16000 as 2 bytes, and so on. It is size and endianness independent. This means that we can extend

[Spice-devel] Use of sized ints

2017-03-03 Thread Christophe de Dinechin
We have recently discussed the use of the bool type. What about sized int types? What is the policy here? Notably, on a part of the code I’m presently working on, I saw that surface_id could be either an int or an uint32_t. There is apparently no clear winner: $ git grep "int.*surface_id" | wc

Re: [Spice-devel] [PATCH v3 0/2] RHEL7 improvements

2017-03-03 Thread Christophe de Dinechin
> On 3 Mar 2017, at 13:10, Marc-André Lureau wrote: > > Hi > > - Original Message - >>> >>> Hi >>> >>> - Original Message - These 2 patches attempt to join images split by RHEL7 graphic stack (Mesa) decreasing commands handled by spice-server.

Re: [Spice-devel] Use of sized ints

2017-03-03 Thread Christophe de Dinechin
> On 3 Mar 2017, at 13:17, Frediano Ziglio wrote: > >> >> We have recently discussed the use of the bool type. What about sized int >> types? What is the policy here? >> >> Notably, on a part of the code I’m presently working on, I saw that >> surface_id could be either an

Re: [Spice-devel] [PATCH v3 1/2] display-channel: Join drawables to improve rhel7 behaviour

2017-03-03 Thread Christophe de Dinechin
> On 3 Mar 2017, at 12:46, Frediano Ziglio wrote: > > Due to the way RHEL7 works the images came out from guest using multiple > commands. This increase the commands to the client and cause the > video code to create and handle multiple streams creating some > visual

Re: [Spice-devel] [PATCH spice-protocol 2/2] agent: Add macro for clearing capability

2017-03-01 Thread Christophe de Dinechin
> On 28 Feb 2017, at 18:26, Frediano Ziglio > wrote: > >> >> On Tue, 2017-02-28 at 09:29 -0500, Frediano Ziglio wrote: Related: https://bugzilla.redhat.com/show_bug.cgi?id=1373725

Re: [Spice-devel] [PATCH spice] gl: fix client mouse mode

2017-03-01 Thread Christophe de Dinechin
> On 30 Jan 2017, at 13:50, marcandre.lur...@redhat.com wrote: > > From: Marc-André Lureau > > Since 2.8, QEMU now longer creates QXL primary surfaces when using Typo, “no longer" > GL. This change broke client-side mouse mode, because Spice server > relies on

Re: [Spice-devel] Survey of repository preferences

2017-07-31 Thread Christophe de Dinechin
> On 28 Jul 2017, at 12:10, Pavel Grunt <pgr...@redhat.com> wrote: > > Hi, > > On Fri, 2017-07-28 at 09:15 +0200, Christophe de Dinechin wrote: >>> On 27 Jul 2017, at 17:07, Marc-André Lureau <marcandre.lur...@redhat.com> >>> wrote: >>>

Re: [Spice-devel] [PATCH spice-gtk] Update submodules using the --merge option

2017-07-31 Thread Christophe de Dinechin
> On 28 Jul 2017, at 12:19, Marc-André Lureau <marcandre.lur...@redhat.com> > wrote: > > Hi > > - Original Message ----- >> From: Christophe de Dinechin <dinec...@redhat.com >> <mailto:dinec...@redhat.com>> >> >> B

Re: [Spice-devel] [PATCH spice-gtk] Update submodules using the --merge option

2017-07-31 Thread Christophe de Dinechin
> On 31 Jul 2017, at 10:58, Frediano Ziglio <fzig...@redhat.com> wrote: > >> >> Hi >> >> ----- Original Message ----- >>> From: Christophe de Dinechin <dinec...@redhat.com >>> <mailto:dinec...@redhat.com>> >>> >&g

[Spice-devel] Getting patchwork to acknowledge acks?

2017-08-04 Thread Christophe de Dinechin
During the discussion on PRs and stuff, several people pointed to patchwork. Does anyone know why this tool does not acknowledge acks? For example https://patchwork.freedesktop.org/series/27298/ has a "Acked-by: Christophe de Dinechin <dinec...@redhat.com>” in the fifth comment, a

Re: [Spice-devel] [PATCH spice-gtk v2] spice-gtk: Use time comparisons that still work after wraparound

2017-08-04 Thread Christophe de Dinechin
> On 18 Jul 2017, at 16:48, Christophe Fergeau <cferg...@redhat.com> wrote: > > Hey, > > On Tue, Jul 18, 2017 at 04:37:29PM +0200, Christophe de Dinechin wrote: >> OK. Since you seem to feel more strongly about this than me, I changed it. >> Pushed to freedes

Re: [Spice-devel] Getting patchwork to acknowledge acks?

2017-08-04 Thread Christophe de Dinechin
> On 4 Aug 2017, at 12:48, Victor Toso <victort...@redhat.com> wrote: > > Hi, > > On Fri, Aug 04, 2017 at 12:42:39PM +0200, Christophe de Dinechin wrote: >> During the discussion on PRs and stuff, several people pointed to >> patchwork. >> >> Does a

[Spice-devel] [PATCH spice-gtk] Fix occasional black screen at startup

2017-07-13 Thread Christophe de Dinechin
From: Christophe de Dinechin <dinec...@redhat.com> The problem occurs when we call spice_playback_channel_set_delay before the channel had received any data setting c->last_time, but after session initialization had set mm_time in the session. The result was that the (good) value i

[Spice-devel] [PATCH spice-gtk] spice-gtk: Use time comparisons that still work after wraparound

2017-07-13 Thread Christophe de Dinechin
From: Christophe de Dinechin <dinec...@redhat.com> The mm timer is a millisecond timer that wraps around after ~49 days. All comparisons that look like a --- src/channel-display-gst.c | 4 ++-- src/channel-display-mjpeg.c | 4 ++-- src/channel-display.c | 2 +- src/channel-play

[Spice-devel] [PATCH spice-gtk 1/3] Add SPICE_TWEAK_DEFINE and SPICE_TWEAK macros

2017-07-13 Thread Christophe de Dinechin
From: Christophe de Dinechin <dinec...@redhat.com> These two macros are intended to identify places in the code where there are ad-hoc values that can be tweaked to optimize behavior. They may be connected later to a mechanism that allows such "tweaks" to be made without

[Spice-devel] [PATCH spice-gtk 2/3] Drop frames if the backlog is above some limit

2017-07-13 Thread Christophe de Dinechin
From: Christophe de Dinechin <dinec...@redhat.com> Experience has shown that if the machine running the guest is overloaded, it may pile up a lot of backlog in the frames queue. This patch clears the queue if it exceeds 100 entries. This value is arbitrary. It corresponds to a few s

[Spice-devel] [PATCH spice-gtk 0/3] spice-gtk: Drop frames if they pile up because client is overloaded

2017-07-13 Thread Christophe de Dinechin
From: Christophe de Dinechin <dinec...@redhat.com> This patch series is a crude attempt at getting rid of the frames backlog when the spice client is overloaded. Another parameter I have been playing wiht is the max-bytes parameter in the GStreamer pipeline, which is currently set to 0 (unl

[Spice-devel] [PATCH spice-gtk 3/3] Add configurable value for gst_max_bytes

2017-07-13 Thread Christophe de Dinechin
From: Christophe de Dinechin <dinec...@redhat.com> The GStreamer configuration currently sets max-bytes=0 in the pipeline. The rationale is that we need to be able to keep frames in case later frames require them for decoding. However, having experienced scenarios where spice would accu

Re: [Spice-devel] [PATCH spice-gtk v2] spice-gtk: Use time comparisons that still work after wraparound

2017-07-17 Thread Christophe de Dinechin
> On 17 Jul 2017, at 13:49, Christophe Fergeau <cferg...@redhat.com > <mailto:cferg...@redhat.com>> wrote: > > On Mon, Jul 17, 2017 at 11:44:04AM +0200, Christophe de Dinechin wrote: >> From: Christophe de Dinechin <dinec...@redhat.com >> <mailto

Re: [Spice-devel] [PATCH spice-gtk] spice-gtk: Use time comparisons that still work after wraparound

2017-07-17 Thread Christophe de Dinechin
> On 17 Jul 2017, at 10:33, Christophe Fergeau <cferg...@redhat.com> wrote: > > On Thu, Jul 13, 2017 at 04:21:36PM +0200, Christophe de Dinechin wrote: >> From: Christophe de Dinechin <dinec...@redhat.com> >> >> The mm timer is a millisecond timer t

[Spice-devel] [PATCH spice-gtk v2] spice-gtk: Use time comparisons that still work after wraparound

2017-07-17 Thread Christophe de Dinechin
From: Christophe de Dinechin <dinec...@redhat.com> The mm timer is a millisecond timer that wraps around after ~49 days. All comparisons that look like a --- src/channel-display-gst.c | 4 ++-- src/channel-display-mjpeg.c | 4 ++-- src/channel-display.c | 2 +- src/channel-play

Re: [Spice-devel] [PATCH spice-gtk] gst-audio: Do not update mmtime without real audio channel

2017-07-14 Thread Christophe de Dinechin
t; >> Because the fake channel is only needed to get the volume info, make >> sure to not update the mm-time with it. >> >> Reported-by: Christophe de Dinechin <dinec...@redhat.com> >> --- >> src/spice-gstaudio.c | 11 +++ >> 1 file changed, 7 insertio

Re: [Spice-devel] [PATCH spice-gtk v2] spice-gtk: Use time comparisons that still work after wraparound

2017-07-18 Thread Christophe de Dinechin
> On 18 Jul 2017, at 16:48, Christophe Fergeau <cferg...@redhat.com> wrote: > > Hey, > > On Tue, Jul 18, 2017 at 04:37:29PM +0200, Christophe de Dinechin wrote: >> OK. Since you seem to feel more strongly about this than me, I changed it. >> Pushed to freedes

Re: [Spice-devel] [PATCH spice-gtk] gst-audio: Do not update mmtime without real audio channel

2017-07-18 Thread Christophe de Dinechin
On 14 Jul 2017, at 14:27, Christophe de Dinechin <dinec...@redhat.com> wrote: > > This looks better than my fix. Can’t ack right now (I did not build and test > it yet), but that looks like a reasonable approach > >> On 14 Jul 2017, at 14:24, Pavel Grunt &

Re: [Spice-devel] [PATCH spice-gtk] gst-audio: Do not update mmtime without real audio channel

2017-07-18 Thread Christophe de Dinechin
o… > > Uri. > >> may cause a (big) delay in displaying a video stream. >> Because the fake channel is only needed to get the volume info, make >> sure to not update the mm-time with it. >> Reported-by: Christophe de Dinechin <dinec...@redhat.com> >

Re: [Spice-devel] [PATCH spice-gtk v2] spice-gtk: Use time comparisons that still work after wraparound

2017-07-18 Thread Christophe de Dinechin
> > On 17 Jul 2017, at 14:20, Christophe Fergeau <cferg...@redhat.com> wrote: > > On Mon, Jul 17, 2017 at 02:00:01PM +0200, Christophe de Dinechin wrote: >> >>> On 17 Jul 2017, at 13:49, Christophe Fergeau <cferg...@redhat.com >>> <mailto:cfer

Re: [Spice-devel] [PATCH spice-gtk] gst-audio: Do not update mmtime without real audio channel

2017-07-18 Thread Christophe de Dinechin
> On 18 Jul 2017, at 17:59, Christophe de Dinechin <dinec...@redhat.com> wrote: > >> >> On 18 Jul 2017, at 17:54, Pavel Grunt <pgr...@redhat.com> wrote: >> >> On Tue, 2017-07-18 at 17:04 +0200, Christophe de Dinechin wrote: >>> On 1

Re: [Spice-devel] [PATCH spice-gtk] gst-audio: Do not update mmtime without real audio channel

2017-07-18 Thread Christophe de Dinechin
> On 18 Jul 2017, at 17:54, Pavel Grunt <pgr...@redhat.com> wrote: > > On Tue, 2017-07-18 at 17:04 +0200, Christophe de Dinechin wrote: >> On 14 Jul 2017, at 14:27, Christophe de Dinechin <dinec...@redhat.com> wrote: >>> >>> This looks better than

Re: [Spice-devel] [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code

2017-07-20 Thread Christophe de Dinechin
> On 19 Jul 2017, at 19:21, Frediano Ziglio <fzig...@redhat.com> wrote: > >> >> On Wed, Jul 19, 2017 at 08:03:49AM -0400, Frediano Ziglio wrote: >>>> >>>> On Wed, Jul 19, 2017 at 12:09:23PM +0200, Christophe de Dinechin wrote: >>>>&

Re: [Spice-devel] [vdagent-win PATCH v4 3/5] Write code to decode PNG format

2017-07-19 Thread Christophe de Dinechin
> On 18 Jul 2017, at 17:54, Christophe Fergeau wrote: > > On Mon, Jul 17, 2017 at 09:54:04AM +0100, Frediano Ziglio wrote: >> Signed-off-by: Frediano Ziglio >> --- >> Makefile.am | 6 +- >> configure.ac | 3 + >> vdagent/image.cpp

Re: [Spice-devel] [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code

2017-07-19 Thread Christophe de Dinechin
> On 18 Jul 2017, at 17:28, Christophe Fergeau wrote: > > On Mon, Jul 17, 2017 at 11:01:22AM +0100, Frediano Ziglio wrote: >> Remove CxImage linking. >> Support Windows BMP format. > > Too bad there is no small/maintained library which would do that for us > :-/ From a

Re: [Spice-devel] [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code

2017-07-21 Thread Christophe de Dinechin
> On 20 Jul 2017, at 11:23, Christophe Fergeau <cferg...@redhat.com> wrote: > > On Thu, Jul 20, 2017 at 09:42:26AM +0200, Christophe de Dinechin wrote: >>> I moved to "unsigned int" 2 versions ago. >> >> It was courteous of you. > > Yup, tha

Re: [Spice-devel] Proposal: review branches (was Re: [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code)

2017-07-25 Thread Christophe de Dinechin
> On 24 Jul 2017, at 16:06, Marc-André Lureau > wrote: > > Hi > > - Original Message - >>> On 21 Jul 2017, at 12:41, Frediano Ziglio wrote: > > On Fri, Jul 21, 2017 at 06:18:49AM -0400, Frediano Ziglio wrote:

Re: [Spice-devel] Proposal: review branches (was Re: [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code)

2017-07-25 Thread Christophe de Dinechin
> On 25 Jul 2017, at 12:36, Christophe Fergeau <cferg...@redhat.com> wrote: > > On Tue, Jul 25, 2017 at 12:23:34PM +0200, Christophe de Dinechin wrote: >>> and there are plenty of public places you can push your work. >> >> So plenty of places, but

Re: [Spice-devel] Proposal: review branches (was Re: [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code)

2017-07-27 Thread Christophe de Dinechin
> On 26 Jul 2017, at 12:19, Frediano Ziglio <fzig...@redhat.com> wrote: > >>> >>> On 25 Jul 2017, at 19:37, Christophe Fergeau <cferg...@redhat.com> wrote: >>> >>> On Tue, Jul 25, 2017 at 02:26:36PM +0200, Christophe de Dinechin wrote: &g

Re: [Spice-devel] Proposal: review branches (was Re: [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code)

2017-07-27 Thread Christophe de Dinechin
> On 26 Jul 2017, at 11:23, Marc-André Lureau > wrote: > > Hi > > - Original Message - >> Now, any objection to >> >> 1. Recommending that we use git URLs in patches? > > If that may help, but as Christophe said, this may be overkill for small > series.

Re: [Spice-devel] [vdagent-win PATCH] Do not use dash in rpm version

2017-07-27 Thread Christophe de Dinechin
> On 26 Jul 2017, at 17:03, Frediano Ziglio wrote: > > RPM does not allow dash in version string. > Remove everything after the dash (which should be the "dirty" > git version). > This make easier to run "make dist" followed by rpmbuild. > > Signed-off-by: Frediano Ziglio

Re: [Spice-devel] [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 15:27, Frediano Ziglio wrote: > > On 27 Jul 2017, at 15:07, Frediano Ziglio wrote: > > On 27 Jul 2017, at 12:39, Christophe Fergeau wrote: > > On Thu, Jul 27, 2017 at 11:54:08AM +0200, Victor Toso wrote: > On

Re: [Spice-devel] [PATCH spice-server] README: Fix typo (dependancies -> dependencies)

2017-07-27 Thread Christophe de Dinechin
> build SPICE > > Spice protocol >= 0.9.0 > @@ -34,7 +34,7 @@ build SPICE > zlib > Cyrus-SASL > > -The following optional dependancies increase the available > +The following optional dependencies increase the available > functionality > > GE

Re: [Spice-devel] [PATCH spice-server] README: Update required protocol version

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 15:25, Frediano Ziglio wrote: > > Signed-off-by: Frediano Ziglio > --- > README | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/README b/README > index 45fbe89c..0fd6f071 100644 > --- a/README > +++ b/README

Re: [Spice-devel] [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 15:07, Frediano Ziglio wrote: > > On 27 Jul 2017, at 12:39, Christophe Fergeau > wrote: > > On Thu, Jul 27, 2017 at 11:54:08AM +0200, Victor Toso wrote: > On Mon, Jul 24, 2017 at 10:47:34AM -0400,

Re: [Spice-devel] [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 12:53, Daniel P. Berrange wrote: > > On Thu, Jul 27, 2017 at 12:39:32PM +0200, Christophe Fergeau wrote: >> On Thu, Jul 27, 2017 at 11:54:08AM +0200, Victor Toso wrote: >>> On Mon, Jul 24, 2017 at 10:47:34AM -0400, Frediano Ziglio wrote: Not really

Re: [Spice-devel] Proposal: review branches (was Re: [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code)

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 14:28, Marc-André Lureau > wrote: > > Hi > > - Original Message - >> >>> On 26 Jul 2017, at 11:23, Marc-André Lureau >>> wrote: >>> >>> Hi >>> >>> - Original Message - Now, any objection to

Re: [Spice-devel] [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 12:39, Christophe Fergeau wrote: > > On Thu, Jul 27, 2017 at 11:54:08AM +0200, Victor Toso wrote: >> On Mon, Jul 24, 2017 at 10:47:34AM -0400, Frediano Ziglio wrote: >>> Not really familiar with GitLab merge requests but on GitHub they >>> remain open

Re: [Spice-devel] [PATCH spice-server] README: Update required protocol version

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 16:04, Christophe Fergeau <cferg...@redhat.com> wrote: > > On Thu, Jul 27, 2017 at 03:29:11PM +0200, Christophe de Dinechin wrote: >> >>> On 27 Jul 2017, at 15:25, Frediano Ziglio <fzig...@redhat.com> wrote: >>> >>&

[Spice-devel] Survey of repository preferences

2017-07-27 Thread Christophe de Dinechin
Following the recent discussions about pull requests, merge requests and marsupilamis, I’ve put together a small Google form at https://goo.gl/forms/bE3UON0zH30sGrg13 to capture the team preferences. I’ve used multiple-choice answers on purpose, so that we end up with actionable numbers. For a

Re: [Spice-devel] Survey of repository preferences

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 16:40, Frediano Ziglio wrote: > >> >> Following the recent discussions about pull requests, merge requests and >> marsupilamis, I’ve put together a small Google form at >> https://goo.gl/forms/bE3UON0zH30sGrg13 to capture the team preferences. I’ve >>

Re: [Spice-devel] Survey of repository preferences

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 16:49, Marc-André Lureau > wrote: > > Hi > > - Original Message - >> Following the recent discussions about pull requests, merge requests and >> marsupilamis, I’ve put together a small Google form at >>

Re: [Spice-devel] [PATCH spice-server] README: Update required protocol version

2017-07-27 Thread Christophe de Dinechin
> On 27 Jul 2017, at 17:00, Victor Toso <victort...@redhat.com> wrote: > > Hi, > > On Thu, Jul 27, 2017 at 04:28:59PM +0200, Christophe de Dinechin wrote: >> >>> On 27 Jul 2017, at 16:04, Christophe Fergeau <cferg...@redhat.com> wrote: >>

[Spice-devel] Form regarding preferences for repo layout

2017-07-27 Thread christophe . de . dinechin
I've invited you to fill out the following form: Reorganizing Spice repositories To fill it out, visit: https://docs.google.com/forms/d/e/1FAIpQLSelX_VxzoO64ruoBN1rAbkVQWcyLrJNvBS_vVF_fK09p43x3w/viewform?c=0w=1includes_info_params=trueusp=mail_form_link Following the recent discussion on

Re: [Spice-devel] Survey of repository preferences

2017-07-28 Thread Christophe de Dinechin
> On 27 Jul 2017, at 17:07, Marc-André Lureau > wrote: > > Hi > > - Original Message - >>> I think we should rather find a consensus on the mailing list rather than >>> avoiding the discussion. >> >> “Avoiding the discussion" sounds like a cheap and

Re: [Spice-devel] [PATCH spice-server] README: Update required protocol version

2017-07-28 Thread Christophe de Dinechin
> On 27 Jul 2017, at 17:06, Christophe Fergeau <cferg...@redhat.com> wrote: > > On Thu, Jul 27, 2017 at 04:28:59PM +0200, Christophe de Dinechin wrote: >> >>> On 27 Jul 2017, at 16:04, Christophe Fergeau <cferg...@redhat.com> wrote: >>> >>>

Re: [Spice-devel] [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code

2017-07-28 Thread Christophe de Dinechin
> On 27 Jul 2017, at 15:53, Christophe Fergeau <cferg...@redhat.com> wrote: > > On Thu, Jul 27, 2017 at 03:00:27PM +0200, Christophe de Dinechin wrote: >> No, that’s not correct (at least for me). The review itself can happen over >> mail, >> what I find

Re: [Spice-devel] Proposal: review branches (was Re: [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code)

2017-07-28 Thread Christophe de Dinechin
> On 27 Jul 2017, at 17:08, Frediano Ziglio wrote: > > Try to sum up the initial problem was patches/series tracking > > So far there are 3 proposal > 1) PR/MR (GitLab/GitHub style) > 2) patchew > 3a) shared git repository > 3b) links to external git repositories > > 1) PR

  1   2   3   4   5   6   7   8   9   >