[Git][xorg-team/app/mesa-demos][upstream-unstable] 167 commits: glxinfo/wglinfo: remove print_shader_limits() code

["Timo Aaltonen (@tjaalton)" <tjaalton@debian.org>]

Title: GitLab

Timo Aaltonen pushed to branch upstream-unstable at X Strike Force / app / mesa-demos

Commits:

• aa9e66f2
  by Brian Paul at 2018-03-07T08:42:50-07:00

  glxinfo/wglinfo: remove print_shader_limits() code
  
  And add queries of GL 2.0 limits.  All the VS, FS, GS limits which were
  reported with print_shader_limits() are also covered by the GL 2.0 and
  3.2 queries.
  
  Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105285
  Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
  Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
  

• 9804a083
  by Brian Paul at 2018-03-07T08:42:50-07:00

  glxinfo/wglinfo: fix VP/FP strings, indentation
  
  Print extensions string instead of program target.
  Use less indentation to match other sections.
  
  Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105285
  Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
  Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
  

• 6a7cee63
  by Brian Paul at 2018-03-07T08:42:50-07:00

  glxinfo/wglinfo: print (110) instead of empty line
  
  The GL_SHADING_LANGUAGE_VERSION query returns an empty string for
  GLSL 1.10.  Instead of printing an empty line, print (110).
  
  Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105285
  Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
  Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
  

• 27a0e5af
  by Brian Paul at 2018-03-07T08:42:50-07:00

  glxinfo/wglinfo: improve GL_ARB_imaging query code
  
  Print GL_MAX_CONVOLUTION_WIDTH, HEIGHT on separate lines.
  Print GL_MAX_COLOR_MATRIX_STACK_DEPTH in same section.
  
  Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
  Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
  

• 1830dcb6
  by Jose Fonseca at 2018-03-12T20:27:30+00:00

  cmake: Warn that MSVC is unmaintained.
  
  It has been broken for a while, and not much interest from our side
  in keeping it going, as most of us are using MinGW cross-compilers
  anyway.
  
  I'm not sure if there are other people interested on MSVC, so just
  warn for the time being.
  
  Also tone down the MSVC warnings, so they don't overwhelm the error
  messages.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  Reviewed-by: Roland Scheidegger <sroland@vmware.com>
  

• ecccd4f1
  by Adam Jackson at 2019-08-12T09:46:54-04:00

  auxbuffer: Don't override $DISPLAY from the environment
  
  Previously we would always try to connect to :0 unless you passed
  -display, which is just gratuitously unlike how every other X app works.
  
  Reviewed-by: Eric Engestrom <eric@engestrom.ch>
  

• 56690689
  by Adam Jackson at 2019-09-04T15:55:09-04:00

  eglinfo: Add support for EGL_MESA_query_driver
  
  Since we don't actually create a context it can be difficult to know
  which driver's capabilities you think you're looking at.
  

• 93b9b19b
  by Yurii Kolesnykov at 2019-09-11T20:49:13+00:00

  fix texdown on macOS
  
  Signed-off-by: Yurii Kolesnykov <root@yurikoles.com>
  

• 99f95b38
  by Matwey V. Kornilov at 2019-09-16T11:03:48-04:00

  eglinfo: Add support for surfaceless EGL platform
  
  Signed-off-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
  

• d905d051
  by Eric Engestrom at 2019-12-29T14:49:37+01:00

  gitlab-ci: add cmake & autotools builds
  
  Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
  Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
  

• cbfb39b9
  by Christian GMEINER at 2019-12-29T14:59:36+01:00

  fix -Wunused-variable compile warings
  
  Signed-off-by: Christian GMEINER <Christian.GMEINER@bachmann.info>
  Acked-by: Eric Engestrom <eric.engestrom@intel.com>
  

• 95c1a57c
  by Christian Gmeiner at 2019-12-31T06:28:35+01:00

  fix -Wunused-const-variable compile warnings
  
  Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
  Acked-by: Eric Engestrom <eric.engestrom@intel.com>
  

• 1973b27f
  by Mathias Fröhlich at 2020-02-19T07:15:43+01:00

  eglinfo: Extend for EGL_EXT_platform_device.
  
  Acked-by: Marek Olšák <marek.olsak@amd.com>
  Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
  

• 4d169621
  by Erik Faye-Lund at 2020-04-16T11:06:18+02:00

  wglgears: do not include unistd.h on MSVC
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• b1b19e43
  by Erik Faye-Lund at 2020-04-16T11:08:25+02:00

  wglgears: change window-title
  
  While this application has its roots from glxgears, it's name is
  wglgears, and we should reflect that in the title.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• b09bf9e3
  by Erik Faye-Lund at 2020-04-16T11:08:30+02:00

  wglgears: make usage and error-handling similar to glxgears
  
  This makes the code look a bit more like the similar code for glxgears.
  There's no good reason to do this differently for these two
  applications.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 78130637
  by Erik Faye-Lund at 2020-04-16T11:08:34+02:00

  wglgears: respect x and y arguments to make_window
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 16c9a130
  by Erik Faye-Lund at 2020-04-16T11:08:37+02:00

  wglgears: add geometry option
  
  The parser isn't exactly identical with XParseGeometry, but should be
  close enough for general use.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 7fa3a433
  by Erik Faye-Lund at 2020-04-16T11:10:47+02:00

  wglgears: mark application as dpi-aware
  
  This makes rendering nice and crisp on high-dpi systems. Since we're not
  doing any layouting or anything, we can just claim to be aware without
  doing much.
  
  We keep the current window-size when the DPI changes, because the size
  is specified in pixels, so it seems more useful to keep that as-is.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• d24d852f
  by Erik Faye-Lund at 2020-04-16T22:11:29+02:00

  wglgears: set whole process as dpi-aware
  
  It seems different GPU drivers behave differently in the case of
  per-thread DPI-awareness, causing incorrect rendering on drivers like
  the NVIDIA-driver. This is probably related to some internal threading
  in the drivers, and is probably a subtle driver-bug.
  
  In either case, setting per-process DPI awareness seems to do the trick.
  It doesn't matter to us if we're doing per process or per thread, so
  let's just go with the more robust solution.
  
  Fixes: 7fa3a433 ("wglgears: mark application as dpi-aware")
  

• 82b6c7a9
  by Erik Faye-Lund at 2020-04-17T10:27:36+00:00

  wglgears: drop random printf in sRGB code-path
  
  This seems a bit arbitrary, we don't generally print messages about
  internals like this. So let's drop it.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 8209ea79
  by Erik Faye-Lund at 2020-04-17T10:27:36+00:00

  wglgears: implement support for multi-sampling
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• d6e7b4fc
  by Erik Faye-Lund at 2020-04-17T10:27:58+00:00

  wglgears: fix rotation speed to 70 degrees per second
  
  This copies the logic from glxgears, omitting the rotating-state,
  because we have no such option (yet).
  
  In order to do this, we also need to swap out our timer. In order to
  keep things nice and easy, we can use timeGetTime(), but set the minumum
  period before the event-loop.
  
  And for the timing-calculation, this is a lot better with proper
  resolution anyway.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 223e9561
  by Erik Faye-Lund at 2020-04-20T11:48:09+02:00

  wglgears: remove unused variable
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 8044dac8
  by Erik Faye-Lund at 2020-04-20T11:48:28+02:00

  wglgears: match usage format more closely to glxgears
  
  Two things:
  - We spell "Usage" with an upper-case 'u'.
  - We print usage to stdout, not stderr.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 9a88b8e1
  by Erik Faye-Lund at 2020-04-20T11:48:33+02:00

  wglgears: omit paranoid error-checking
  
  This code contains a lot of code that can't realistically fail; these
  things can only fail on incorrect input, or under memory pressure. We're
  passing correct input, and we're not yet at a point where memory
  pressure is really possible.
  
  So for the sake of readability, let's remove some overly paranoid
  error-checking. It's not like these error-messages are all that useful
  to most users, compared to running this in a debugger anyway.
  
  This makes the code reasier to read, and makes it easier to add
  meaningful error-checking where it actually matters in the future.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• b571364a
  by Erik Faye-Lund at 2020-04-20T11:48:40+02:00

  wglgears: exit with error-code on error
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 27813895
  by Erik Faye-Lund at 2020-04-20T11:48:44+02:00

  wglgears: do not re-create window after re-creating context
  
  When we re-created the window, we accidentally leaked the old window.
  It's not so easy to notice, because it's invisible, but it's still there
  and taking up some memory.
  
  But there's more problems; when I recently added the -geometry switch,
  I missed the extra CreateWindowEx call, meaning this call didn't respect
  the x and y offset.
  
  But instead of fixing these things indevidually, let's take a step back
  here. There's no reason to re-create the window here in the first place.
  We can just set a new pixel-format on hDC and re-create hRC instead.
  
  This makes the code both easier to maintain, and squashes a couple of
  bugs.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 12a2b4cc
  by Erik Faye-Lund at 2020-04-20T11:48:51+02:00

  wglgears: error properly when no pixel-format is found
  
  If we're too strict with our requirements, we should properly report the
  issue. We shouldn't assert, and we should list the requirements so it's
  easy to understand why.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• a394b83f
  by Erik Faye-Lund at 2020-04-20T11:50:56+02:00

  wglgears: implement animate-state
  
  This allows the animation to be paused and resumed, just like with
  glxgears.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• a81375c8
  by Erik Faye-Lund at 2020-04-20T11:53:23+02:00

  wglgears: implemenmt wgl-extension helper
  
  This helper allows us to check if a WGL-extension is supported or not.
  It will be used in the next commit.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 35ce140e
  by Erik Faye-Lund at 2020-04-20T11:53:23+02:00

  wglgears: check what the current vsync setting is
  
  This seems to work fine on NVIDIA, but seems to report incorrect results
  on Intel drivers. Reading the WGL_EXT_swap_control specification, this
  seems like subtle driver-bug, and not an application problem, so let's
  do this anyway.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 17d95904
  by Erik Faye-Lund at 2020-04-20T12:02:22+02:00

  wglgears: add support for -fullscreen switch
  
  In order to get the correct screen-resolution on hidpi displays, we need
  to mark the process as dpi-aware first.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 5b7b7975
  by Erik Faye-Lund at 2020-05-07T21:56:35+02:00

  Revert "wglgears: do not re-create window after re-creating context"
  
  This fix wasn't a great idea; even if it seems to work fine on modern
  Windows versions, the documentation says that SetPixelFormat can only be
  set once per window. And worse, Mesa's OpenGL32.DLL emulation code seems
  to depend on this, ignoring any consecutive calls. This means that with
  Mesa, we get no multisampling here.
  
  So, let's revert this, and instead follow what the docs say.
  
  This reverts commit 27813895adef5866e9aee62975c7f5965cd9c601.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• d36ca86b
  by Erik Faye-Lund at 2020-05-07T21:56:59+02:00

  wglgears: respect x and y positions
  
  When adding support for the "-geometry" option, I forgot to update the
  second call to CreateWindowEx, meaning it only respected this option
  when not using srgb or multisampling. Let's fix that.
  
  Fixes: 16c9a130 ("wglgears: add geometry option")
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 8a28836d
  by Erik Faye-Lund at 2020-05-07T21:57:03+02:00

  wglgears: destroy old window when recreating
  
  Even if the window is not yet displayed, the object still exists and
  takes up memory. So let's destroy it before recreating.
  
  Fixes: 14688d84 ("wglgears.c: add -srgb option")
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 3260f665
  by Erik Faye-Lund at 2020-07-06T14:04:38+02:00

  glxgears: add sRGB mode support
  
  wglgears already have this, so let's also add it for glxgears!
  
  Reviewed-by: Marek Olšák <marek.olsak@amd.com>
  

• 8c73555b
  by Erik Faye-Lund at 2020-07-06T14:04:41+02:00

  glxgears / wglgears: adjust colors for sRGB
  
  Without this, the colors have a very different look, making the red more
  orange. This adjust the colors to give the same appearance.
  
  Reviewed-by: Marek Olšák <marek.olsak@amd.com>
  

• da28ca32
  by Erik Faye-Lund at 2020-07-10T09:27:33+02:00

  wglgears: match glxgears formatting
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 60044b34
  by Erik Faye-Lund at 2020-07-10T09:27:59+02:00

  wglgears: style-fixup
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 8bf5a21b
  by Erik Faye-Lund at 2020-07-10T09:28:02+02:00

  wglgears: show stereo-option in usage
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 29802678
  by Erik Faye-Lund at 2020-07-10T09:28:06+02:00

  wglgears: add srgb-mode to usage
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  

• 0de7436b
  by Erik Faye-Lund at 2020-08-04T12:19:43+02:00

  Revert "wglgears: show stereo-option in usage"
  
  I don't know what I was thinking; we don't have support for
  stereo-rendering in wglgears.
  
  This reverts commit 8bf5a21becc274d94a790a7e7e1ff460f9b82f02.
  
  Reviewed-by: Brian Paul <brianp@vmware.com>
  Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
  

• 67db4a25
  by Adam Jackson at 2020-09-11T14:06:30-04:00

  glxgears: Make the swap interval configurable
  
  As a bonus feature you can set negative values if your driver supports
  GLX_EXT_swap_control_tear, though heaven help you if your renderer is so
  slow that that improves things.
  
  Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
  

• 9c095ff0
  by Erik Faye-Lund at 2020-09-14T11:05:31+00:00

  egl: match geometry position of glxgears
  
  There's two things that differs about the geometry position of this port
  of glxgears:
  
  1. The camera is moved back by the wrong amount.
  2. The projection matrix is computed the wrong way.
  
  So let's remedy this, and make the rendering closer to the original.
  
  Reviewed-by: Adam Jackson <ajax@redhat.com>
  

• 8c1bc349
  by Erik Faye-Lund at 2020-09-14T11:05:31+00:00

  egl: match lighting of wglgears
  
  glxgears uses the default value for ambient light, which is 0.2. We
  should do the same. This makes these two demos look the same.
  
  Reviewed-by: Adam Jackson <ajax@redhat.com>
  

• 846ec9bf
  by Erik Faye-Lund at 2020-09-14T11:05:31+00:00

  egl: use smooth normals for inner face
  
  glxgears smooth-shades the inner face, so let's do that here as well for
  visual parity.
  
  Reviewed-by: Adam Jackson <ajax@redhat.com>
  

• db5ad06a
  by Jan Beich at 2020-09-15T18:12:46+00:00

  glxgears / wglgears: avoid C99 "for" without explicit -std=
  
  GCC < 5 defaults to -std=gnu89, failing to build:
  
  glxgears.c: In function 'init':
  glxgears.c:413: error: 'for' loop initial declaration used outside C99 mode
  
  Fixes: 8c73555b4dd5 ("glxgears / wglgears: adjust colors for sRGB")
  Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
  

• 98c93dd1
  by Jose Fonseca at 2020-09-16T10:01:51+01:00

  cmake: Build glslstateschange too.
  

• cd45113d
  by Jose Fonseca at 2020-09-16T10:02:17+01:00

  wgl/rtotex: Remove.
  
  This sample is broken, it's not easy to fix, and it's too
  complicated/niche to be worth the hassle.
  

• 13e0b926
  by Jose Fonseca at 2020-09-16T10:02:41+01:00

  wgl: Prevent GL usage before wglMakeCurrent.
  

• 13cf8beb
  by Jose Fonseca at 2020-09-16T10:02:41+01:00

  glad: Initial import.
  

• 5435fc7f
  by Jose Fonseca at 2020-09-16T10:03:35+01:00

  Use glad instead of GLEW.
  
  glad is a more modern, cleaner, and flexible alternative to GLEW.  In
  particular it has first-class citizen support for OpenGL ES end EGL.
  
  Glad can't be used just for defines.  For old demos which only need
  defines, we could bundle Khronos glext, but for now place the necessary
  defines on gl_wrap.h
  

• faaa319d
  by Erik Faye-Lund at 2020-10-08T11:24:02+00:00

  egl: pack all strips per gear into a single draw
  
  We can pack multiple triangle-strips into a single draw, by inserting
  degenerate triangles between them. This should be a big win in terms of
  CPU usage.
  
  To avoid inserting needless degenerate triangles to invert face-winding,
  we can manually invert the order of the back-face as well as emit the
  inner face last. This ensures every strip starts with the right
  face-winding.
  
  Reviewed-by: Alexandros Frantzis <alexandros.frantzis@collbora.com>
  

• 0f9e7d99
  by Jose Fonseca at 2021-03-24T06:51:46+00:00

  gitlab-ci: Add MinGW build job.
  
  A rudimentary job, just to ensure Windows builds don't go south.
  
  The FreeGLUT dependencies are derived from the Linux headers plus a .def
  file, to avoid any further downloads whatsoever.
  
  v2: Quote all bash variable expansions and treat unset vars as errors,
  per Eric Engestrom.
  

• bd4d268d
  by Jordan Justen at 2021-08-02T15:31:38-07:00

  mesa-demos: NOTE! Default branch is now main
  
  To update your local repository to use the new default branch, these
  commands may help:
  
  $ git fetch origin
  $ git checkout master
  $ git branch -m main
  $ git branch --set-upstream-to=origin/main
  $ git remote set-head origin --auto
  
  Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
  Reviewed-by: Dave Airlie <airlied@redhat.com>
  

• 5fff4a18
  by Yann Dirson at 2021-09-01T20:35:59+02:00

  egltri: don't abort if unused glMapBufferOES is not available (fixes #21).
  
  If a demo would like to make use of this function, it should first
  checking that the GL_OES_mapbuffer extension is available.  When it is
  not, there is no reason to abort here.
  

• f0bff6db
  by Dave Airlie at 2021-11-22T09:23:25+10:00

  glad: regenerate glad files from glad upstream
  

• ef6f1927
  by Kevin Locke at 2021-11-21T23:41:01+00:00

  Distribute glad/README.md, not nonexistent README
  
  EXTRA_DIST in src/glad/Makefile.am included README, which doesn't exist.
  Distribute README.md instead.  This fixes the `make dist` error:
  
      make[6]: *** No rule to make target 'README', needed by 'distdir-am'.  Stop.
      ...
      make: *** [Makefile:648: dist] Error 2
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• a212bf4d
  by Kevin Locke at 2021-11-21T23:47:30+00:00

  CMake: Build eglinfo and peglgears
  
  For parity with Autotools.
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• e55bd633
  by Erik Faye-Lund at 2022-02-10T16:33:02+01:00

  eglut_wayland: only mark opaque region if no alpha
  
  If we get a config with alpha, we don't know if the application will use
  translucent pixels or not. So let's not mark the area as opaque in that
  case.
  
  This fixes an issue with es2gears_wayland, which clears the background
  with an transparent black clear-color.
  

• 80b3c13e
  by Erik Faye-Lund at 2022-02-10T16:38:30+01:00

  egl: do not modulate alpha with light
  
  We don't want to modify the alpha with the lightness, otherwise we make
  fragments translucent when they should get darker.
  

• b4657303
  by Kevin Locke at 2022-04-27T11:00:22-06:00

  Add .editorconfig shared editor configuration
  
  [EditorConfig] is a format for specifying configuration information,
  such as indentation, supported by many common editors and IDEs.  A
  benefit is encouraging consistency across editors from a single shared
  configuration file, rather than editor-specific modelines in each file
  or requiring each developer to configure their own project-specific
  settings.
  
  This commit adds an .editorconfig file with settings that match the
  most common current indent styles.  It is not intended to proscribe any
  changes to current style, but to encourage consistency going forward.
  
  [EditorConfig]: https://editorconfig.org
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• 43a1cb23
  by Erik Faye-Lund at 2022-04-29T09:08:06+00:00

  remove slang tests
  
  These tests have been failing for a long time, and are largely
  unmaintained.
  
  But they also predate Piglit, which covers pretty much the same thing.
  So there's little point in fixing these up. They are unused, and nobody
  seems to care about it.
  

• 7e0445a0
  by Erik Faye-Lund at 2022-04-29T11:17:14+02:00

  remove rbug-client
  
  This doesn't build, and seemingly never got wired up correctly since the
  move to autotools. That's a good indication that nobody is using it, so
  let's get rid of it.
  

• 7c6e04e9
  by Erik Faye-Lund at 2022-05-04T09:48:47+02:00

  include windows.h in gl_wrap.h
  
  Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
  

• 01be0383
  by Erik Faye-Lund at 2022-05-04T09:49:03+02:00

  cmake: do not force inclusion of windows.h
  
  While it's true that nobody loves to include windows.h, we now have
  gl_wrap.h where we can do this automatically. This saves us from
  including windows.h in util sources etc, which makes things a tiny bit
  more sane.
  
  Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
  

• e6d0d9a8
  by Erik Faye-Lund at 2022-05-04T11:04:42+00:00

  remove openvg support
  
  The OpenVG state tracker was removed from Mesa back in 2015, over 7
  years ago to date. The last release of an OpenVG version was back in
  2008. It's pretty obvious that OpenVG is more or less dead.
  
  So let's get rid of the OpenVG demos, as it seems they're not really
  buildable or runnable on modern systems, at least not without a lot of
  work.
  
  Since the OpenVG code is the only code that uses Freetype2, this means
  we can also yank out the code for that.
  
  Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
  

• be8c929b
  by Alan Coopersmith at 2022-05-04T12:42:26+00:00

  Update URLs to reflect migration to gitlab
  
  Also update download URL from ftp to https since browsers are
  dropping ftp support.
  
  Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
  

• 986df97d
  by Kevin Locke at 2022-05-04T12:46:27+00:00

  egl/opengl: remove duplicate LDFLAGS
  
  Fix autoreconf warning due to duplicate LDFLAGS:
  
      src/egl/opengl/Makefile.am:82: warning: eglgears_x11_LDFLAGS multiply defined in condition TRUE ...
      src/egl/opengl/Makefile.am:71: ... 'eglgears_x11_LDFLAGS' previously defined here
      src/egl/opengl/Makefile.am:83: warning: egltri_x11_LDFLAGS multiply defined in condition TRUE ...
      src/egl/opengl/Makefile.am:72: ... 'egltri_x11_LDFLAGS' previously defined here
  
  Both LDFLAGS variables are defined to `$(AM_LDFLAGS) $(X11_LIBS)` in
  both definitions, so either should be safe to remove.
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• e47015ed
  by Erik Faye-Lund at 2022-05-04T18:06:41+02:00

  gitlab-ci: make sure we have the packages we need
  
  We want a few packages that we currently get implicitly through other
  dependencies. But when upgrading the CI images, it's much less
  error-prone if we explicitly list the ones we need.
  
  So let's add a couple of missing packags.
  
  Reviewed-by: Martin Roukala <martin.roukala@mupuf.org>
  

• e9f65a9c
  by Erik Faye-Lund at 2022-05-04T18:06:45+02:00

  gitlab-ci: switch to freedesktop/ci-templates
  
  Reviewed-by: Martin Roukala <martin.roukala@mupuf.org>
  

• d892161d
  by Erik Faye-Lund at 2022-05-04T18:06:48+02:00

  gitlab-ci: move to debian bullseye
  
  This gets us newer GCC versions etc.
  
  A couple of packages has been renamed, so let's update the names for
  better forward compatibility.
  
  Reviewed-by: Martin Roukala <martin.roukala@mupuf.org>
  

• 74d529da
  by Erik Faye-Lund at 2022-05-04T18:13:13+02:00

  cmake: make glut a soft dependency
  
  There's parts of the tree that is useful even without GLUT, so let's
  allow building those without requiring GLUT. But let's emit a warning so
  users have a chance to understand why they get such a stripped-down
  build.
  
  This is especially useful on Windows, where GLUT is a bit of an exotic
  dependency.
  
  This gets us a bit closer to what we're doing with autotools.
  
  Acked-by: Martin Roukala <martin.roukala@mupuf.org>
  

• 7abb1c4a
  by Erik Faye-Lund at 2022-05-04T16:20:46+00:00

  tests: remove getprocaddress test
  
  While this test does *kinda* work, there's a few issues with it:
  - It uses Python 2 rather than Python 3
  - It doesn't build out-of-tree, because of some hard-coded paths
  - It doesn't currently link, due to a missing X11 dependency
  
  That last point especially seems to indicate that nobody has been using
  this since the switch to autotools, which is a long time ago.
  
  While all of the above can be fixed, we have since gotten Piglit, which
  already extensively tests glXGetProcAddress. Let's get rid of it.
  
  This also removes our last dependency on the mesa source code, so we can
  now remove the logic for that as well.
  

• c0a6366f
  by Erik Faye-Lund at 2022-05-05T08:51:02+02:00

  eglut_wayland: give error when wl-shell is not supported
  
  Instead of crashing and burning far down in the wayland code due to
  a null-pointer being passed, let's give the user a graceful error
  message about what went wrong.
  

• 3a1af638
  by Erik Faye-Lund at 2022-05-05T11:25:30+00:00

  remove stale makefile
  
  This file seems to be a leftover from before the Autotools rewrite, and
  is no longer in use or maintained. Let's just get rid of it!
  
  Reviewed-by: Martin Roukala <martin.roukala@mupuf.org>
  

• ddc35ca0
  by Erik Faye-Lund at 2022-05-05T12:28:19+00:00

  cmake: move GLAPIENTRY definition to wrapper-headers
  
  We have a neat place to patch missing GLAPIENTRY definitions now, let's
  use that for Windows as well.
  
  Acked-by: Martin Roukala <martin.roukala@mupuf.org>
  

• 786aee82
  by Martin Roukala (né Peres) at 2022-05-06T07:34:00+00:00

  glxinfo: show the swap method when listing visuals
  
  Signed-off-by: Martin Roukala <martin.roukala@mupuf.org>
  Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
  

• 084d65ae
  by Erik Faye-Lund at 2022-05-06T11:53:30+00:00

  eglut_wayland: set opaque region every frame
  
  It's going to be a lot easier to deal with this if we do it right before
  we swap buffers once we support resizing.
  

• 0cdaa670
  by Kevin Locke at 2022-05-06T11:53:30+00:00

  cmake: build Wayland binaries
  
  For parity with Autotools, as added in b12c2995d, 66136b954, and
  5a466668d.
  
  Depend on [Extra CMake Modules] from the KDE project for
  FindWayland.cmake (and additional modules added in subsequent commits).
  
  Increase the required CMake version from 2.8 to 2.8.12, as required by
  ECMFindModuleHelpers.cmake used by FindWayland.cmake.
  
  [Extra CMake Modules]: https://api.kde.org/ecm/
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• 707e542f
  by Kevin Locke at 2022-05-06T11:53:30+00:00

  eglut_wayland: exit event loop on POLLNVAL
  
  If the file descriptor becomes invalid (e.g. because
  wl_display_disconnect has been called) exit the event loop.  There's no
  value from continuing to poll.
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• 7392c383
  by Kevin Locke at 2022-05-06T11:53:30+00:00

  eglut_wayland: destroy surface and compositor
  
  Although the compositor should destroy any orphaned resources, as a
  matter of good style, prefer to destroy the resources that we create.
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• 757a6ba2
  by Kevin Locke at 2022-05-06T11:53:30+00:00

  eglut_wayland: switch from wl_shell to xdg_shell
  
  The [wl_shell protocol is deprecated in favor of xdg_shell] and some
  compositors (e.g. wlroots-based compositors like Sway) have already
  dropped support for wl_shell.
  
  Add build dependencies on wayland-protocols and wayland-scanner, along
  with code to run wayland-scanner to produce the requisite code.
  
  Define BUILD_WAYLAND CMake variable to simplify conditionals in the
  build files so that every directory doesn't need to check the presence
  of the individual eglut Wayland dependencies.
  
  Fixes: #16
  
  [wl_shell protocol is deprecated in favor of xdg_shell]: https://wayland.freedesktop.org/docs/html/apa.html#protocol-spec-wl_shell
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• ce6dcb84
  by Kevin Locke at 2022-05-06T11:53:30+00:00

  eglut_wayland: add xdg_toplevel listener
  
  To handle close and resize events.
  
  Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
  

• e65ddf1a
  by Witold Baryluk at 2022-05-08T19:45:29+00:00

  gltestperf: Don't overflow int, use float as intended
  
  Convert to consistently use float for time and per-sec measurements.
  
  On fast hardware the `numelem * bmark->size[0]` can easily overflow 32-bit int.
  

• e9600c47
  by Erik Faye-Lund at 2022-05-09T12:22:20+00:00

  cmake: fix out-of-tree build
  
  When we started using ecm_add_wayland_client_protocol, we missed the
  fact that for out-of-tree builds, we also need to add the current build
  directory to the include-path, otherwise we can't include the generated
  header.
  

• 6122c4a6
  by Erik Faye-Lund at 2022-05-09T12:22:20+00:00

  gitlab-ci: bump ci-tag
  
  This was forgotten in when adding the ECM and waylands-protocols stuff,
  so we accidentally built without wayland support. Whoops!
  

• f2e942a9
  by Erik Faye-Lund at 2022-05-10T07:49:52+00:00

  glut_wrapper: fix macos glut/glad compatibility
  
  When a source-file includes the glad-header, the glad header defines
  APIENTRY to nothing, and defines GLAPIENTRY to APIENTRY. So far so good.
  
  But when they then include macOS' glut.h afterwads, glut.h ends up
  undefining APIENTRY, meaning GLAPIENTRY now expands to the identifier
  APIENTRY, not the variable as it used to.
  
  In order to patch over that problem, let's define APIENTRY back to
  nothing, so things expand the way we want it. This should be safe,
  because this only happend on macOS, and only when not using FreeGLUT,
  which doesn't have the same problem.
  
  Closes: https://gitlab.freedesktop.org/mesa/demos/-/issues/24
  Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
  

• 2e94e033
  by Erik Faye-Lund at 2022-05-10T09:53:20+02:00

  pbutil: dynamically load GLX extension functions
  
  These extension functions were used directly from libGLX, which isn't
  supported on all platforms. Let's load them dynamically instead.
  
  Closes: https://gitlab.freedesktop.org/mesa/demos/-/issues/14
  Reviewed-by: Adam Jackson <ajax@redhat.com>
  

• b876c51b
  by Erik Faye-Lund at 2022-05-10T08:50:09+00:00

  progs/fp: avoid undefined result in scs-test
  
  According to the ARB_fragment_program spec, section 3.11.5.22 "SCS:
  Sine/Cosine":
  
  > The z and w components of the result vector are undefined.
  
  So let's just write the x and y components, and do a separate MOV for
  the z and w components.
  
  Closes: https://gitlab.freedesktop.org/mesa/demos/-/issues/23
  Reviewed-by: Emma Anholt <emma@anholt.net>
  

• edb0f8d0
  by Yann Dirson at 2022-05-10T10:28:46+00:00

  es[12]_info.c: query GL_VENDOR and GL_SHADING_LANGUAGE_VERSION as well
  
  glxinfo already shows those but es2info did not.
  
  Signed-off-by: Yann Dirson <ydirson@free.fr>
  

• c106b674
  by Erik Faye-Lund at 2022-05-10T10:41:36+00:00

  cmake: set gl-preference to legacy
  
  Otherwise, we might fail to link some demos, due to missing GL symbols.
  

• d6530425
  by Erik Faye-Lund at 2022-05-10T10:41:36+00:00

  cmake: bump minimum version
  
  Using ECM requires a more recent CMake version, let's update the root
  CMakeList.txt to reflect this. This silences a CMake warning.
  

• 4d6ed539
  by Ethan Hsieh at 2022-05-12T05:51:21+00:00

  opengles1/opengles2: Exit cleanly by releasing all EGL resources on termination
  

• fd75b63d
  by Patrik Plihal at 2022-05-12T11:33:12+00:00

  trispd: fix double formatting in output
  
  Fixes the display formatting of the demo output regarding pixelRate:
  
    from : ... tri/s  -2147483648 pixels/s
       to: ... tri/s  5.50052e+09 pixels/s
  

• 2f5750ad
  by Erik Faye-Lund at 2022-05-13T08:12:33+00:00

  wgl: remove stray semicolon
  
  There were a stray semicolon here, making this always take the hDC !=
  NULL code-path, which clearly isn't what was intended.
  
  This was noticed by a compiler-warning on CI.
  

• 1b0801fa
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  es2tri.c: remove write-only variable
  

• f4b0fe4d
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  bindtex.c: remove write-only variable
  
  While we're at it, let's reformat the wrapped line here as well.
  

• df621204
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  xeglgears.c: remove write-only variable
  

• 1b5c266d
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  msaa.c: remove write-only variable
  

• d5998ae6
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  pbuffer.c: remove write-only variable
  

• df362dcb
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  render_tex.c: remove write-only variable
  

• efdc3a47
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  two_win.c: remove write-only variable
  

• f46ad507
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  yuvrect_client.c: fixup indent
  
  This bit had a mixture of spaces and tabs. Let's fix that before we make
  further changes.
  

• 4824b40c
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  yuvrect_client.c: remove write-only variables
  

• 0cb39b58
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  blendeq.c: remove write-only variable
  

• e2ebed45
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  shtest.c: remove write-only variable
  

• 0c9a0ee5
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  objview.c: remove write-only variable
  

• 97921bfa
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  auxbuffer.c: remove write-only variable
  

• 6d10462c
  by Erik Faye-Lund at 2022-05-13T08:17:38+00:00

  olympic.c: remove write-only variables
  

• 85978907
  by Erik Faye-Lund at 2022-05-13T10:27:11+02:00

  dinoshade.c: cast glu tess callbacks
  
  The gluTessCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  
  Let's use a typedef to keep this short and to the point, so we don't
  need to repeat the same function prototype over and over again.
  

• 96a74877
  by Erik Faye-Lund at 2022-05-13T10:27:11+02:00

  tessdemo.c: fixup indent
  

• 67728ed9
  by Erik Faye-Lund at 2022-05-13T10:27:12+02:00

  tessdemo.c: cast glu tess callbacks
  
  The gluTessCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  
  Let's use a typedef to keep this short and to the point, so we don't
  need to repeat the same function prototype over and over again.
  

• e46a03e5
  by Erik Faye-Lund at 2022-05-13T10:27:12+02:00

  quadric.c: cast glu quadric callbacks
  
  The gluQuadricCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  

• f5d34cb5
  by Erik Faye-Lund at 2022-05-13T10:27:12+02:00

  surfpoints.c: cast glu nurbs callbacks
  
  The gluNurbsCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  
  Let's use a typedef to keep this short and to the point, so we don't
  need to repeat the same function prototype over and over again.
  

• c807297b
  by Erik Faye-Lund at 2022-05-13T10:27:12+02:00

  tess.c: cast glu tess callbacks
  
  The gluTessCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  
  Let's use a typedef to keep this short and to the point, so we don't
  need to repeat the same function prototype over and over again.
  
  While we're at it, let's remove the pointless & operators that takes the
  address of the function; the address and the value of a function is the
  same thing.
  

• 875ec6ab
  by Erik Faye-Lund at 2022-05-13T10:27:12+02:00

  tesswind.c: cast glu tess callbacks
  
  The gluTessCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  
  Let's use a typedef to keep this short and to the point, so we don't
  need to repeat the same function prototype over and over again.
  
  While we're at it, let's remove the pointless & operators that takes the
  address of the function; the address and the value of a function is the
  same thing.
  

• bc646bd5
  by Erik Faye-Lund at 2022-05-13T10:27:12+02:00

  trim.c: cast glu nurbs callbacks
  
  The gluNurbsCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  

• 56a408f4
  by Erik Faye-Lund at 2022-05-13T10:27:12+02:00

  nurb.c: cast glu nurbs callbacks
  
  The gluNurbsCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  

• 960cfe0b
  by Erik Faye-Lund at 2022-05-13T10:27:12+02:00

  quad.c: cast glu quadric callbacks
  
  The gluQuadricCallback interface is kinda horrible, because it takes
  pointers without really knowing what the correct signature is. Sadly,
  this leads to compiler warnings, so let's cast the functions to quiet
  these.
  

• 6d242fc0
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  es1_info.c: remove excessive argument to printf
  

• 38b09d16
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  opengles1/tri.c: remove unused variable
  

• 6ab2900a
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  sphere.c: remove unused function
  

• 563713c2
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  glslstateschange.c: fix constness of pointers
  

• 4440759f
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  es2tri.c: remove unused variable
  

• 2893166b
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  texaaline.c: clean up reused calculations
  
  We calculated both sets of deltas for corordinates 2/4 and 3/5, even
  though they were the same. But we only used the 2/3 set twice, ignoring
  the 3/4 sets.
  
  Let's instead rename these variables to show that they are valid for
  both positions. That way we get less code, but it still makes sense.
  

• 1cd42146
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  pbuffer.c: make sure vertex-arrays are initialized
  
  It's really hard for the compiler to understand that glVertexPointer et
  al only stores the pointer itself, and doesn't read the data until
  glDrawArrays, where it's actually initialized.
  
  So let's zero-initialize these arrays, so the compiler doesn't have to
  know.
  
  Yeah, it has a minor performance impact, but it seems better than adding
  compiler specific warning disables.
  

• 418bef1a
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  render_tex.c: make sure vertex-arrays are initialized
  
  It's really hard for the compiler to understand that glVertexPointer et
  al only stores the pointer itself, and doesn't read the data until
  glDrawArrays, where it's actually initialized.
  
  So let's zero-initialize these arrays, so the compiler doesn't have to
  know.
  
  Yeah, it has a minor performance impact, but it seems better than adding
  compiler specific warning disables.
  

• 074e53ba
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  opengles1/torus.c: make sure vertex-arrays are initialized
  
  It's really hard for the compiler to understand that glVertexPointer et
  al only stores the pointer itself, and doesn't read the data until
  glDrawArrays, where it's actually initialized.
  
  So let's zero-initialize these arrays, so the compiler doesn't have to
  know.
  
  Yeah, it has a minor performance impact, but it seems better than adding
  compiler specific warning disables.
  

• bda44fe1
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  xeglgears.c: cast through uintptr_t
  
  GCC doesn't like casting integers to pointers of different sizes
  directly. Let's fix this, by casting to uintptr_t first, showing that we
  intend to go to a pointer-size.
  

• 1dde2d92
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  eglfbdev.c: cast through uintptr_t
  
  GCC doesn't like casting integers to pointers of different sizes
  directly. Let's fix this, by casting to uintptr_t first, showing that we
  intend to go to a pointer-size.
  

• e67dac7e
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  shaderutil.c: cast glShaderSourceARB to correct signature
  
  The signatures for glShaderSourceARB and glShaderSource differs
  slightly, but luckily only in constness, which isn't going to cause any
  real issues.
  
  So let's cast this to avoid compiler warnings; we're aware of the
  difference, and it's OK.
  

• 6a4071a0
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  msctest.c: use PRId64 instead of lld
  
  These values are int64_t values, so using %lld isn't really appropriate
  on all platforms.
  
  Let's use PRId64 instead, which is both of the right size and sign.
  

• 8510fa5f
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  msctest.c: drop unused argument parsing
  
  The only argument we parsed was the verbose-flag, which we don't
  respect. Just get rid of the entire argument-parsing, it's not doing
  anything useful.
  

• 4e4c6d4a
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  omlsync.c: remove unused verbose flag
  
  This flag has clearly been carelessly copied from glsync.c, but we never
  respected it. Let's remove the argument.
  

• 89a2cb64
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  samples/tkmap.c: mark functions as inline
  
  This source file gets included by other source files, and not all
  sources that include it call all functions in it.
  
  Marking these as inline helps the compiler to understand that this is
  expected, avoiding a bunch of warnings.
  

• 723df973
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  do not enable -Wbad-function-cast
  
  This compiler flag adds some pretty bonkers warnings, where stuff like
  (float)rand() / RAND_MAX produce errors unless the result of rand() is
  stored in a temporary variable.
  
  Let's not enable that switch. According to GCCs own issue tracker, this
  warning is a "strange, legacy warning that is more about GCC's internal
  implementation details ([...]) than anything in user code":
  
  https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92117
  

• 410ac86d
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  eglkms.c: add missing initialization of ret
  

• 2acd7853
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  eglkms.c: initialize connector and encoder
  
  It's a bit hard for the compiler to see that it's impossible for
  connector to be uninitialized. But it's actually possible for encoder to
  be uninitialize, in the case where count_encoders is zero.
  
  In any case, it's probably best to just initialize these to NULL anyway.
  

• 457a6efa
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  samplers.c: show compiler limit on integer
  
  These strings always fits inside the buffer, but it's not quite obvious
  to the compiler why. So let's explicitly show the limit on the size of
  s, effectively eliminating the warning.
  

• b2992ffe
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  glsync.c: do not invoke undefined behavior on integer overflow
  

• 59539104
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  texobjshare.c: remove hard-coded variable
  
  This variable is always NULL, which makes the compiler complain. Let's
  remove it.
  

• 4640d312
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  osdemos: ifdef-out unused materials
  

• 574244f0
  by Erik Faye-Lund at 2022-05-13T08:34:03+00:00

  osdemos: ifdef-out unused functions
  

• de01c5b2
  by Erik Faye-Lund at 2022-05-13T08:39:11+00:00

  es2tri.c: fix assertion
  
  >From the EGL spec:
  
  > Querying EGL_CONTEXT_CLIENT_VERSION returns the version of the client
  > API this context actually supports (which may differ from the version
  > specified at context creation time).
  
  So, we need to allow this value to be larger than two here.
  
  Closes: https://gitlab.freedesktop.org/mesa/demos/-/issues/19
  

• 2c4a83eb
  by Erik Faye-Lund at 2022-05-16T14:57:00+02:00

  gitlab-ci: rename mingw glut-lib
  
  This matches better what we're searching for in the automake
  build-system.
  
  Acked-by: Jose Fonseca <jfonseca@vmware.com>
  Acked-by: Eric Engestrom <eric@engestrom.ch>
  

• 4b1ceff0
  by Erik Faye-Lund at 2022-05-16T14:57:04+02:00

  gitlab-ci: move glut to /tmp
  
  Due to some boring details, Meson doesn't like if we specify library and
  include paths to compiler.find_library inside the source dir like we did
  with CMake. So let's prepare the GLUT stuff in /tmp instead.
  
  Acked-by: Jose Fonseca <jfonseca@vmware.com>
  Acked-by: Eric Engestrom <eric@engestrom.ch>
  

• 79d18fbf
  by Erik Faye-Lund at 2022-05-16T14:57:08+02:00

  gitlab-ci: refactor mingw-build
  
  This makes things more similar to what we're going to need for the Meson
  build, let's refactor things a bit.
  
  Acked-by: Jose Fonseca <jfonseca@vmware.com>
  Acked-by: Eric Engestrom <eric@engestrom.ch>
  

• 4973b62b
  by Erik Faye-Lund at 2022-05-16T14:57:12+02:00

  add a meson build system
  
  We're currently maintaining two separate build-systems, one based on
  autotools, and one based on cmake. These are quite similar, but some
  details are missing from each of them, most notably:
  
  - The autotools build-system can't build the src/wgl sources
  - The cmake build-system doesn't install things properly
  
  There's a bunch of other smaller differences, but they're not that
  important.
  
  In theory we could have massaged the cmake build system to work well
  enough for all use-cases, but there's a big elephan in the room here:
  
  Mesa's main repository has moved to meson. So if we keep doing cmake,
  the Mesa development community needs to learn two different build
  systems that fills the same needs. That's not really ideal.
  
  So instead, let's add a Meson build system, and later on remove the
  other ones.
  
  I think this is all; this should be a union of the feature-sets of the
  autotools and cmake build systems.
  
  Acked-by: Jose Fonseca <jfonseca@vmware.com>
  Acked-by: Eric Engestrom <eric@engestrom.ch>
  

• 21c4a918
  by Erik Faye-Lund at 2022-05-16T14:57:16+02:00

  index.html: document how to build using meson
  
  Acked-by: Jose Fonseca <jfonseca@vmware.com>
  Acked-by: Eric Engestrom <eric@engestrom.ch>
  

• 70ea768b
  by Erik Faye-Lund at 2022-05-16T14:57:20+02:00

  gitlab-ci: build with meson
  
  This adds a CI build for Meson in addition to the CMake and Autotools
  builds we already have.
  
  Acked-by: Jose Fonseca <jfonseca@vmware.com>
  Acked-by: Eric Engestrom <eric@engestrom.ch>
  

• 2df8e971
  by Erik Faye-Lund at 2022-05-16T14:57:23+02:00

  configure.ac: mark autotools and cmake as deprecated
  
  This follows what we did when switching mesa over to autotools, except
  it applies to two build-systems instead of one.
  
  Acked-by: Jose Fonseca <jfonseca@vmware.com>
  Acked-by: Eric Engestrom <eric@engestrom.ch>
  

• 12ff8b7b
  by Erik Faye-Lund at 2022-05-18T09:51:48+00:00

  eglut_wayland: do not use frame-listener
  
  Even the simple-egl.c example client from Weston hasn't been using a
  frame-listener since some time in 2013, and the way we used it causes
  issues with the FPS counter, due to not keeping multiple frames in
  flight.
  
  Let's drop using the frame-listener, and simplify the code a bit.
  
  Closes: https://gitlab.freedesktop.org/mesa/demos/-/issues/12
  

• 2472f475
  by Erik Faye-Lund at 2022-05-18T17:38:55+00:00

  meson: fix separate build of x11 and wayland egl demos
  
  This fixes building with either the x11 or wayland features disabled.
  
  Closes: https://gitlab.freedesktop.org/mesa/demos/-/issues/25
  Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
  

• c0f9927d
  by Erik Faye-Lund at 2022-05-19T13:08:38+00:00

  meson: make glut optional
  
  We have sufficiently guarded our glut-usage, so glut is already
  practically speaking optional. Let's make it fully optional, to match
  the behavior we have on CMake for Windows.
  

• d5ae5d58
  by Erik Faye-Lund at 2022-05-20T10:55:39+00:00

  gitlab-ci: add osmesa dev package
  

• f03a89aa
  by Erik Faye-Lund at 2022-05-20T10:55:39+00:00

  gitlab-ci: enable all auto-features on linux
  

• dc3053e7
  by Erik Faye-Lund at 2022-05-30T10:11:13+00:00

  autotools: add glad-headers to distribution
  
  Without this, make distcheck fails.
  

• d0144eda
  by Erik Faye-Lund at 2022-05-30T10:11:13+00:00

  autotools: add missing windows-sources
  
  This adds a few missing sources to glad, wgl and xdemos directories, which
  are needed for the Windows build.
  
  While we're at it, sort the list of sources.
  

• fdc37106
  by Erik Faye-Lund at 2022-05-30T10:11:13+00:00

  autotools: fix distcheck after deprecation
  

• 5d6dbbbf
  by Erik Faye-Lund at 2022-05-30T10:11:13+00:00

  autotools: distribute documentation
  
  index.html is the closest thing we have to proper documentation. Let's
  distribute it as well!
  

• 379f6a14
  by Erik Faye-Lund at 2022-05-30T10:11:13+00:00

  autotools: distribute meson build-files
  
  Without this, we can't really build meson from the autotools
  distribution. So let's add these files!
  
  In the case of the tools-folder, we also need to add dds.c to the
  distrubution, because we actually build it in the meson build. In the
  past, we only built it in the cmake-build, which we never added to the
  distribution.
  

• 47826086
  by Erik Faye-Lund at 2022-05-30T12:14:24+02:00

  convert index.html to README.rst
  
  This is a bit more discoverable, and easier to read for humans.
  

• 20a9adf6
  by Erik Faye-Lund at 2022-05-30T10:16:16+00:00

  meson: disable gl-deprecation warnings
  
  OpenGL has been deprecated for a good while on macOS. But since this
  repo is mostly about OpenGL, let's deprecate that warning; we're not
  going to port these demos to macOS.
  
  This does more or less the same as we do in the CMake build system, but
  it uses a bit more targeted approach.
  

• 2feaa0b0
  by Erik Faye-Lund at 2022-05-30T10:16:16+00:00

  xdemos: fix bulding of pbutil
  
  When building _libpbutil, we might need to add some include directories
  for some of the dependencies. Make sure we pass the dependencies as we
  should.
  

• cda7fa14
  by Erik Faye-Lund at 2022-05-30T14:01:02+02:00

  bump version to 8.5.0 for release
  

30 changed files:

• + .editorconfig
• .gitignore
• + .gitlab-ci.yml
• + .gitlab-ci/freeglut.def
• + .gitlab-ci/mingw32.cmake
• + .gitlab-ci/prepare-glut-mingw.sh
• + .gitlab-ci/x86_64-w64-mingw32
• CMakeLists.txt
• Makefile.am
• + README.rst
• configure.ac
• − index.html
• + meson.build
• + meson_options.txt
• src/CMakeLists.txt
• src/Makefile.am
• src/data/Makefile.am
• + src/data/meson.build
• src/demos/CMakeLists.txt
• src/demos/Makefile.am
• − src/demos/Makefile.cygnus
• src/demos/arbfplight.c
• src/demos/arbfslight.c
• src/demos/arbocclude.c
• src/demos/arbocclude2.c
• src/demos/copypix.c
• src/demos/cubemap.c
• src/demos/cuberender.c
• src/demos/dinoshade.c
• src/demos/engine.c

The diff was not included because it is too large.

—
View it on GitLab.
You're receiving this email because of your account on salsa.debian.org. If you'd like to receive fewer emails, you can adjust your notification settings.