Author: Daniel van Vugt
Revision Date: 2014-04-30 02:49:30 UTC

Merge latest development-branch

Author: Daniel van Vugt
Revision Date: 2016-11-09 05:29:01 UTC

Try again

Author: Daniel van Vugt
Revision Date: 2015-01-16 07:55:24 UTC

Update tests

Author: Daniel van Vugt
Revision Date: 2013-03-13 03:29:34 UTC

Simplified, and removed the type safety problem.

Author: Daniel van Vugt
Revision Date: 2015-04-23 10:08:15 UTC

Merge latest trunk and fix conflicts

Author: Daniel van Vugt
Revision Date: 2014-11-17 01:53:56 UTC

Merge latest paren branch (and trunk) and fix a conflict.

Author: Daniel van Vugt
Revision Date: 2014-11-17 01:49:19 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2014-02-20 07:02:12 UTC

Remember to return true from handle(event)

Author: Daniel van Vugt
Revision Date: 2014-02-18 10:29:06 UTC

Reimplement depth initialization

Author: Daniel van Vugt
Revision Date: 2015-03-27 10:14:43 UTC

Prototype: Plasma without cosine.

It was meant to be faster, but on my real modern GPU seems slower right now.

Author: Daniel van Vugt
Revision Date: 2015-07-16 07:01:46 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2015-02-04 10:06:26 UTC

Merge latest parent branch.

Author: Daniel van Vugt
Revision Date: 2014-11-18 08:12:28 UTC

Remove the wait_for_page_flip experiment

Author: Daniel van Vugt
Revision Date: 2014-07-28 04:00:51 UTC

Merge latest 0.5 branch and fix conflicts.

Author: Daniel van Vugt
Revision Date: 2015-05-08 06:21:22 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2013-10-07 04:06:28 UTC

Fix Android build failure.

Author: Daniel van Vugt
Revision Date: 2013-11-05 03:52:40 UTC

Merge latest privatize-surfaces-Surface

Author: Daniel van Vugt
Revision Date: 2013-11-05 03:52:04 UTC

Merge latest development-branch

Author: Daniel van Vugt
Revision Date: 2014-12-29 02:36:46 UTC

Merge latest trunk and fix conflicts.

Author: Daniel van Vugt
Revision Date: 2015-07-17 10:17:04 UTC

Prototype

Author: Daniel van Vugt
Revision Date: 2014-04-28 06:50:33 UTC

Reintroduce the radial rounding experiment from London

Author: Daniel van Vugt
Revision Date: 2015-02-23 09:29:26 UTC

Revert the 55Hz input rate kludge (r1937).

We don't need seem to need it any more, and my testing is showing me that
Ubuntu Touch on newer devices benefits in animation smoothness from the
full 60Hz (which is more noticeable I think).

Author: Daniel van Vugt
Revision Date: 2014-12-15 08:42:22 UTC

Much cleaner get_lock() to abstract the locking type.

Author: Daniel van Vugt
Revision Date: 2016-08-25 04:22:18 UTC

Merge latest trunk

I want to see CI pass this a second time...

Author: Daniel van Vugt
Revision Date: 2013-02-07 10:02:24 UTC

Remove mir_connect callback from the public API.

First pass at simplifying the client API by removing callbacks. This is part
of a greater effort to resolve LP: #1112195, with more proposals to come.

Author: Daniel van Vugt
Revision Date: 2017-04-06 09:50:01 UTC

Try again

Author: Daniel van Vugt
Revision Date: 2017-01-06 08:18:34 UTC

Remove more dead code

Author: Daniel van Vugt
Revision Date: 2014-02-26 05:27:35 UTC

Merge latest development-branch.

Author: Daniel van Vugt
Revision Date: 2014-01-29 08:54:23 UTC

Undo get_front_buffer_object rename. I think it's actually correct. Just have
to remember that the front buffer according to GBM isn't actually at the
front and visible until it's also flipped.

Author: Daniel van Vugt
Revision Date: 2014-05-05 07:39:07 UTC

Tweak focussed/unfocussed colour differences to be less dramatic

Author: Daniel van Vugt
Revision Date: 2016-11-14 10:18:14 UTC

Enable the fix and the test now passes

Author: Daniel van Vugt
Revision Date: 2013-12-06 12:08:18 UTC

Prototype translated_event

Author: Daniel van Vugt
Revision Date: 2014-07-23 09:56:02 UTC

Add comments.

Author: Daniel van Vugt
Revision Date: 2013-11-07 10:05:20 UTC

Add resize support to GBMClientBuffer. This just means getting the width and
height from MirBufferPackage each frame.

Author: Daniel van Vugt
Revision Date: 2013-11-12 11:16:17 UTC

Remove the (now redundant) size parameter from
*ClientBufferFactory::create_buffer()

Author: Daniel van Vugt
Revision Date: 2015-02-17 04:46:03 UTC

test_client_library.cpp: Fix weird type/state mixup (which apparently was
a preexisting problem).

Author: Daniel van Vugt
Revision Date: 2014-09-24 03:48:52 UTC

Really don't revert ABI bumps.

Author: Daniel van Vugt
Revision Date: 2015-02-06 13:17:42 UTC

Revert r2288.

Author: Daniel van Vugt
Revision Date: 2015-02-18 03:29:26 UTC

Don't mention "static".
Also restore support for testing multiple identical rectangles.

Author: Daniel van Vugt
Revision Date: 2014-01-13 08:25:01 UTC

Rotation angle should be a float by the time it reaches Renderer... so we
can support smooth rotation animations in future.

Author: Daniel van Vugt
Revision Date: 2015-10-30 10:26:10 UTC

Some improvement but still not ideal.

Author: Daniel van Vugt
Revision Date: 2015-10-30 11:01:23 UTC

Another (better) approach to corner rounding

Author: Daniel van Vugt
Revision Date: 2016-01-07 05:31:59 UTC

Cherry pick from the other branch

Author: Daniel van Vugt
Revision Date: 2014-01-28 08:23:22 UTC

Initialize on construction

Author: Daniel van Vugt
Revision Date: 2017-05-26 02:12:41 UTC

Series 0.28

Author: Daniel van Vugt
Revision Date: 2014-09-18 04:29:12 UTC

Bump the server ABI to 26. I don't expect this to pass CI as it's
missing the prerequisite fix for bug 1293944.

Author: Daniel van Vugt
Revision Date: 2014-08-08 04:05:24 UTC

Remember the contents of debian/libmirserver25.install

Author: Daniel van Vugt
Revision Date: 2016-08-24 05:42:18 UTC

Again please

Author: Daniel van Vugt
Revision Date: 2017-04-12 09:29:57 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2016-10-07 07:26:05 UTC

Move it

Author: Daniel van Vugt
Revision Date: 2013-02-08 06:48:15 UTC

Demonstrate two ways of avoiding callbacks in the client API:
  1. conn = mir_wait_for_connection(mir_connect(...));
  2. conn = mir_wait_connect(...);

I was trying to avoid the redundancy of having multiple functions per
operation in my previous proposal. However since that was disapproved, this
is the next best thing.

Arguably, we don't need mir_wait_connect() because it's just shorthand for:
  conn = mir_wait_for_connection(mir_connect(server, app, nullptr, nullptr));

And even then, we don't need mir_wait_for_connection(). I just added it to
keep the purists happy and avoid the weakly typed alternative:
  conn = (MirConnection*)mir_wait_for(mir_connect(...));

Author: Daniel van Vugt
Revision Date: 2015-03-12 03:51:05 UTC

Fix FTBFS from upstream changes

Author: Daniel van Vugt
Revision Date: 2017-03-03 11:21:41 UTC

An experiment

Author: Daniel van Vugt
Revision Date: 2014-05-09 03:15:40 UTC

Merge latest development-branch

Author: Daniel van Vugt
Revision Date: 2013-12-19 08:42:43 UTC

Port Android DisplayReports

Author: Daniel van Vugt
Revision Date: 2014-05-05 02:55:33 UTC

Merge latest development-branch

Author: Daniel van Vugt
Revision Date: 2015-06-18 10:01:14 UTC

Add a comment

Author: Daniel van Vugt
Revision Date: 2015-04-17 16:07:20 UTC

Set changelog for PPA

Author: Daniel van Vugt
Revision Date: 2014-12-11 07:36:26 UTC

Merge latest trunk and fix a conflict

Author: Daniel van Vugt
Revision Date: 2016-04-07 03:37:32 UTC

Reapply the strengthening diff

Author: Daniel van Vugt
Revision Date: 2015-03-30 06:54:15 UTC

Migrate to MirEvent 2.1 syntax (fixes build failure)

Author: Daniel van Vugt
Revision Date: 2017-01-31 08:15:36 UTC

Prototype

Author: Daniel van Vugt
Revision Date: 2016-10-20 13:15:50 UTC

Merge branch 'quit-eglapp'

Author: Daniel van Vugt
Revision Date: 2014-03-30 03:10:32 UTC

A small experiment to see what it would look like to taper/bleed the surface
colour into the title bar, eliminating the hard edge between the two.

Author: Daniel van Vugt
Revision Date: 2014-01-21 07:40:26 UTC

Add a regression test for LP: #1270964 (disabled till it's fixed)

Author: Daniel van Vugt
Revision Date: 2015-05-05 02:49:03 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2015-05-05 05:52:15 UTC

Another intentional ABI break for testing

Author: Daniel van Vugt
Revision Date: 2014-08-08 06:07:35 UTC

Add missing mirplatform.pc, which is now a requirement again.

Author: Daniel van Vugt
Revision Date: 2017-03-09 09:34:39 UTC

Rewrite the load balancing against latest trunk

Author: Daniel van Vugt
Revision Date: 2017-02-27 08:16:27 UTC

Much better test

Author: Daniel van Vugt
Revision Date: 2014-08-12 03:21:44 UTC

Merge latest development-branch

Author: Daniel van Vugt
Revision Date: 2015-03-11 03:49:17 UTC

Revert r2381. It created too many DSOs.

Author: Daniel van Vugt
Revision Date: 2016-04-07 02:15:58 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2015-02-11 07:34:46 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2017-02-01 05:17:29 UTC

Documentation

Author: Daniel van Vugt
Revision Date: 2015-07-20 03:20:46 UTC

integration-tests: Fix failing queue_size_scales_with_client_performance
due to implicit coupling to the old thresholds. Now it detects the
scaling delay and tests according to that.

Author: Daniel van Vugt
Revision Date: 2016-09-26 05:41:03 UTC

Re-enable the test

Author: Daniel van Vugt
Revision Date: 2016-11-03 07:45:48 UTC

Implement typo::Renderer::unicode_from_utf8

Author: Daniel van Vugt
Revision Date: 2014-11-28 02:38:51 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2013-04-16 08:51:38 UTC

Remove deprecated invalid/error ID feature

Author: Daniel van Vugt
Revision Date: 2014-05-06 02:57:38 UTC

Merge latest development-branch and fix conflicts

Author: Daniel van Vugt
Revision Date: 2014-09-17 08:21:22 UTC

Prototype avoiding resampling in nested servers (again)

Author: Daniel van Vugt
Revision Date: 2015-08-03 08:39:39 UTC

Bump the client platform ABI, since it's now broken.

Author: Daniel van Vugt
Revision Date: 2013-04-03 07:45:55 UTC

Delegate construction C++11-style.

Author: Daniel van Vugt
Revision Date: 2016-11-07 09:23:28 UTC

Some ideas, not working yet

Author: Daniel van Vugt
Revision Date: 2016-04-01 08:13:49 UTC

Try an RAII Observer approach

Author: Daniel van Vugt
Revision Date: 2014-10-10 07:51:17 UTC

Actually, yes you do need swap interval 0. I was accidentally testing
the wrong build.

Author: Daniel van Vugt
Revision Date: 2015-03-10 07:42:36 UTC

Fix style.

Author: Daniel van Vugt
Revision Date: 2017-01-19 10:59:16 UTC

Tidy up expectations and improve documentation

Author: Daniel van Vugt
Revision Date: 2015-06-19 06:30:23 UTC

Merge latest trunk and fix conflict

Author: Daniel van Vugt
Revision Date: 2014-12-24 10:45:24 UTC

Prototype wakelocks

Author: Daniel van Vugt
Revision Date: 2015-01-15 08:07:30 UTC

Merge latest trunk

Author: Daniel van Vugt
Revision Date: 2014-11-24 02:42:55 UTC

And fix a typo

Author: Daniel van Vugt
Revision Date: 2015-12-14 06:39:12 UTC

Merge latest trunk again; adds new suppressions from Cemil.

Author: Daniel van Vugt
Revision Date: 2015-12-11 09:38:29 UTC

Be more forgiving of a very slow test environment (valgrind on armhf).
This is another stab at LP: #1499229.

Author: Daniel van Vugt
Revision Date: 2017-03-09 10:30:57 UTC

Better comments

Author: Daniel van Vugt
Revision Date: 2015-08-17 03:08:50 UTC

Use the wily build environment for cross compiling, not vivid.

Author: Alexandros Frantzis
Revision Date: 2014-07-12 06:17:41 UTC

android,tests: Fix g++ 4.9 build

It seems that g++ 4.9 has become more aggressive about constant folding of
expressions that use sizeof(), thus triggering warnings/errors related to
constant overflows that didn't occur with g++ 4.8.

For example, g++-4.8 warns about overflowing i3, but not i4, whereas g++-4.9
warns about both i3 and i4:

#define VALUE1 0x70000000UL
#define VALUE2 (VALUE1 >> sizeof(unsigned char))
#define VALUE3 0x80000000UL
#define VALUE4 (VALUE1 << sizeof(unsigned char))

int main()
{
    int i1 = VALUE1;
    int i2 = VALUE2;
    int i3 = VALUE3;
    int i4 = VALUE4;
}

This affects our ioctl() wrapper calls because we use 'int' for the request
parameter, whereas the request numbers produced by the kernel macros (using
sizeof() in the calculation) are 'unsigned long'. This MP fixes the problem by
casting the request numbers to 'int', which is a safe conversion since the
request numbers are constrained to 32 bits by design.

Note that we don't have a problem with all ioctl request numbers, although they
are produced by the same set of macros, because almost all of them don't have
the high bit set (like VALUE2 above) and don't trigger the overflow error. The
one we have a problem with, SYNC_IOC_MERGE, has the high bit set (like VALUE4
above).

Also note that the glibc ioctl() function signature uses 'unsigned long'
instead of 'int' for the request number. So, alternatively, we could fix the
problem by changing our wrappers to use 'unsigned long', too. However, the
glibc signature goes against POSIX and there have been discussions about
switching it back. See https://sourceware.org/bugzilla/show_bug.cgi?id=14362 .

Author: Steve Langasek
Revision Date: 2014-07-12 05:53:13 UTC

Fix prototype of ioctl() so that it's compatible on amd64, as detected by g++ 4.9.

Author: Steve Langasek
Revision Date: 2014-02-03 21:46:23 UTC

releasing package mir version 0.1.3+14.04.20140108-0ubuntu3