[PATCH 1/6] runqemu: add options that enable virgl with the SDL frontend

[PATCH 1/6] runqemu: add options that enable virgl with the SDL frontend

[Alexander Kanavin]

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
---
 scripts/runqemu | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/scripts/runqemu b/scripts/runqemu
index a05facd0db6..5c56c3fe6c1 100755
--- a/scripts/runqemu
+++ b/scripts/runqemu
@@ -67,9 +67,9 @@ of the following environment variables (in any order):
     nographic - disable video console
     sdl - choose the SDL UI frontend
     gtk - choose the Gtk UI frontend
-    gl - enable virgl-based GL acceleration (also needs gtk option)
-    gl-es - enable virgl-based GL acceleration, using OpenGL ES (also needs gtk option)
-    egl-headless - enable headless EGL output; use vnc or spice to see it
+    gl - enable virgl-based GL acceleration (also needs gtk or sdl options)
+    gl-es - enable virgl-based GL acceleration, using OpenGL ES (also needs gtk or sdl options)
+    egl-headless - enable headless EGL output; use vnc (via publicvnc option) or spice to see it
     serial - enable a serial console on /dev/ttyS0
     serialstdio - enable a serial console on the console (regardless of graphics mode)
     slirp - enable user networking, no root privileges is required
@@ -437,7 +437,12 @@ class BaseConfig(object):
                 self.qemu_opt_script += ' -nographic'
                 self.kernel_cmdline_script += ' console=ttyS0'
             elif arg == 'sdl':
-                self.qemu_opt_script += ' -display sdl'
+                if 'gl' in sys.argv[1:]:
+                    self.qemu_opt_script += ' -vga virtio -display sdl,gl=on'
+                elif 'gl-es' in sys.argv[1:]:
+                    self.qemu_opt_script += ' -vga virtio -display sdl,gl=es'
+                else:
+                    self.qemu_opt_script += ' -display sdl'
             elif arg == 'gtk':
                 if 'gl' in sys.argv[1:]:
                     self.qemu_opt_script += ' -vga virtio -display gtk,gl=on'
-- 
2.17.1

[PATCH 2/6] selftest: skip virgl test on centos 7 entirely

[Alexander Kanavin]

With the sdl frontend, qemu isn't able to even boot fully,
so let's skip the test early.

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
---
 meta-selftest/lib/oeqa/runtime/cases/virgl.py | 5 -----
 meta/lib/oeqa/selftest/cases/runtime_test.py  | 2 ++
 2 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/meta-selftest/lib/oeqa/runtime/cases/virgl.py b/meta-selftest/lib/oeqa/runtime/cases/virgl.py
index d301a19fa42..c0abfd1b161 100644
--- a/meta-selftest/lib/oeqa/runtime/cases/virgl.py
+++ b/meta-selftest/lib/oeqa/runtime/cases/virgl.py
@@ -13,11 +13,6 @@ class VirglTest(OERuntimeTestCase):
 
     @OETestDepends(['virgl.VirglTest.test_kernel_driver'])
     def test_kmscube(self):
-
-        distro = oe.lsb.distro_identifier()
-        if distro and distro == 'centos-7':
-            self.skipTest('kmscube is not working when centos 7 is the host OS')
-
         status, output = self.target.run('kmscube', timeout=30)
         self.assertEqual(status, 0, "kmscube exited with non-zero status %d and output:\n%s" %(status, output))
         self.assertIn('renderer: "virgl"', output, "kmscube does not seem to use virgl:\n%s" %(output))
diff --git a/meta/lib/oeqa/selftest/cases/runtime_test.py b/meta/lib/oeqa/selftest/cases/runtime_test.py
index 3f212bd0eac..7d3922ce448 100644
--- a/meta/lib/oeqa/selftest/cases/runtime_test.py
+++ b/meta/lib/oeqa/selftest/cases/runtime_test.py
@@ -179,6 +179,8 @@ class TestImage(OESelftestTestCase):
         distro = oe.lsb.distro_identifier()
         if distro and distro == 'debian-8':
             self.skipTest('virgl isn\'t working with Debian 8')
+        if distro and distro == 'centos-7':
+            self.skipTest('virgl isn\'t working with Centos 7')
 
         qemu_packageconfig = get_bb_var('PACKAGECONFIG', 'qemu-system-native')
         features = 'INHERIT += "testimage"\n'
-- 
2.17.1

[PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Alexander Kanavin]

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
---
 meta/lib/oeqa/selftest/cases/runtime_test.py | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/meta/lib/oeqa/selftest/cases/runtime_test.py b/meta/lib/oeqa/selftest/cases/runtime_test.py
index 7d3922ce448..3632ae81ede 100644
--- a/meta/lib/oeqa/selftest/cases/runtime_test.py
+++ b/meta/lib/oeqa/selftest/cases/runtime_test.py
@@ -166,9 +166,9 @@ class TestImage(OESelftestTestCase):
         bitbake('core-image-full-cmdline socat')
         bitbake('-c testimage core-image-full-cmdline')
 
-    def test_testimage_virgl_gtk(self):
+    def test_testimage_virgl_gtk_sdl(self):
         """
-        Summary: Check host-assisted accelerate OpenGL functionality in qemu with gtk frontend
+        Summary: Check host-assisted accelerate OpenGL functionality in qemu with gtk and SDL frontends
         Expected: 1. Check that virgl kernel driver is loaded and 3d acceleration is enabled
                   2. Check that kmscube demo runs without crashing.
         Product: oe-core
@@ -183,18 +183,27 @@ class TestImage(OESelftestTestCase):
             self.skipTest('virgl isn\'t working with Centos 7')
 
         qemu_packageconfig = get_bb_var('PACKAGECONFIG', 'qemu-system-native')
+        sdl_packageconfig = get_bb_var('PACKAGECONFIG', 'libsdl2-native')
         features = 'INHERIT += "testimage"\n'
         if 'gtk+' not in qemu_packageconfig:
             features += 'PACKAGECONFIG_append_pn-qemu-system-native = " gtk+"\n'
+        if 'sdl' not in qemu_packageconfig:
+            features += 'PACKAGECONFIG_append_pn-qemu-system-native = " sdl"\n'
         if 'virglrenderer' not in qemu_packageconfig:
             features += 'PACKAGECONFIG_append_pn-qemu-system-native = " virglrenderer"\n'
         if 'glx' not in qemu_packageconfig:
             features += 'PACKAGECONFIG_append_pn-qemu-system-native = " glx"\n'
+        if 'opengl' not in sdl_packageconfig:
+            features += 'PACKAGECONFIG_append_pn-libsdl2-native = " opengl"\n'
         features += 'TEST_SUITES = "ping ssh virgl"\n'
         features += 'IMAGE_FEATURES_append = " ssh-server-dropbear"\n'
         features += 'IMAGE_INSTALL_append = " kmscube"\n'
-        features += 'TEST_RUNQEMUPARAMS = "gtk gl"\n'
-        self.write_config(features)
+        features_gtk = features + 'TEST_RUNQEMUPARAMS = "gtk gl"\n'
+        self.write_config(features_gtk)
+        bitbake('core-image-minimal')
+        bitbake('-c testimage core-image-minimal')
+        features_sdl = features + 'TEST_RUNQEMUPARAMS = "sdl gl"\n'
+        self.write_config(features_sdl)
         bitbake('core-image-minimal')
         bitbake('-c testimage core-image-minimal')
 
-- 
2.17.1

[PATCH 4/6] gcr: update to 3.34.0

[Alexander Kanavin]

inherit gettext, as gcr is now gettextized.

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
---
 meta/recipes-gnome/gcr/{gcr_3.28.1.bb => gcr_3.34.0.bb} | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
 rename meta/recipes-gnome/gcr/{gcr_3.28.1.bb => gcr_3.34.0.bb} (83%)

diff --git a/meta/recipes-gnome/gcr/gcr_3.28.1.bb b/meta/recipes-gnome/gcr/gcr_3.34.0.bb
similarity index 83%
rename from meta/recipes-gnome/gcr/gcr_3.28.1.bb
rename to meta/recipes-gnome/gcr/gcr_3.34.0.bb
index 0a9867451ee..f13bfecbe19 100644
--- a/meta/recipes-gnome/gcr/gcr_3.28.1.bb
+++ b/meta/recipes-gnome/gcr/gcr_3.34.0.bb
@@ -11,12 +11,12 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=55ca817ccb7d5b5b66355690e9abc605"
 DEPENDS = "intltool-native gtk+3 p11-kit glib-2.0 libgcrypt \
            ${@bb.utils.contains('GI_DATA_ENABLED', 'True', 'libxslt-native', '', d)}"
 
-inherit gnomebase gtk-icon-cache gtk-doc distro_features_check upstream-version-is-even vala gobject-introspection
+inherit gnomebase gtk-icon-cache gtk-doc distro_features_check upstream-version-is-even vala gobject-introspection gettext
 # depends on gtk+3, but also x11 through gtk+-x11
 REQUIRED_DISTRO_FEATURES = "x11"
 
-SRC_URI[archive.md5sum] = "afd88cacfd54c1ac5b3e0eb35e3aa674"
-SRC_URI[archive.sha256sum] = "95204aa2111c301778ebfbe60975ce3ed698c958430ffcc2a785ac5e593d168b"
+SRC_URI[archive.md5sum] = "4af28919fb1dd36d93603e8230283b6f"
+SRC_URI[archive.sha256sum] = "29df50974a90987af694c0fb8926a6b366e68cacd8abd813817cfe1eb5d54524"
 
 FILES_${PN} += " \
     ${datadir}/dbus-1 \
-- 
2.17.1

[PATCH 5/6] btrfs-tools: update to 5.3

[Alexander Kanavin]

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
---
 .../btrfs-tools/{btrfs-tools_5.2.2.bb => btrfs-tools_5.3.bb}    | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
 rename meta/recipes-devtools/btrfs-tools/{btrfs-tools_5.2.2.bb => btrfs-tools_5.3.bb} (97%)

diff --git a/meta/recipes-devtools/btrfs-tools/btrfs-tools_5.2.2.bb b/meta/recipes-devtools/btrfs-tools/btrfs-tools_5.3.bb
similarity index 97%
rename from meta/recipes-devtools/btrfs-tools/btrfs-tools_5.2.2.bb
rename to meta/recipes-devtools/btrfs-tools/btrfs-tools_5.3.bb
index 6b73c01dccb..12b5c4562dc 100644
--- a/meta/recipes-devtools/btrfs-tools/btrfs-tools_5.2.2.bb
+++ b/meta/recipes-devtools/btrfs-tools/btrfs-tools_5.3.bb
@@ -14,7 +14,7 @@ DEPENDS = "util-linux attr e2fsprogs lzo acl"
 DEPENDS_append_class-target = " udev"
 RDEPENDS_${PN} = "libgcc"
 
-SRCREV = "55a8c9626fb906c20c3206f8fd39b9a8fb259b79"
+SRCREV = "de7856cee5907938441f765ebab7cc106b7faf70"
 SRC_URI = "git://git.kernel.org/pub/scm/linux/kernel/git/kdave/btrfs-progs.git \
            file://0001-Add-a-possibility-to-specify-where-python-modules-ar.patch \
            "
-- 
2.17.1

[PATCH 6/6] libmodulemd-v1: update to 1.8.16

[Alexander Kanavin]

Upstream added support for optional docs, so
0001-Do-not-generate-gtkdoc-or-python-bindings.patch is replaced
with an option to disable gtk-doc (as the modulemd feature is not used
in oe-core anyway).

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
---
 ...t-generate-gtkdoc-or-python-bindings.patch | 60 -------------------
 .../libmodulemd/libmodulemd-v1_git.bb         |  7 +--
 2 files changed, 3 insertions(+), 64 deletions(-)
 delete mode 100644 meta/recipes-devtools/libmodulemd/libmodulemd-v1/0001-Do-not-generate-gtkdoc-or-python-bindings.patch

diff --git a/meta/recipes-devtools/libmodulemd/libmodulemd-v1/0001-Do-not-generate-gtkdoc-or-python-bindings.patch b/meta/recipes-devtools/libmodulemd/libmodulemd-v1/0001-Do-not-generate-gtkdoc-or-python-bindings.patch
deleted file mode 100644
index d950ad5867e..00000000000
--- a/meta/recipes-devtools/libmodulemd/libmodulemd-v1/0001-Do-not-generate-gtkdoc-or-python-bindings.patch
+++ /dev/null
@@ -1,60 +0,0 @@
-From 71c51206e037c0bb5759e01b307b7ce1d5934703 Mon Sep 17 00:00:00 2001
-From: Alexander Kanavin <alex.kanavin@gmail.com>
-Date: Fri, 6 Sep 2019 17:07:00 +0200
-Subject: [PATCH] Do not generate gtkdoc or python bindings
-
-All of these really need a configuration option.
-
-Upstream-Status: Inappropriate [oe-core specific]
-Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
----
- meson.build          | 12 ------------
- modulemd/meson.build |  8 --------
- 2 files changed, 20 deletions(-)
-
-diff --git a/meson.build b/meson.build
-index 155c9e7..fe35d5e 100644
---- a/meson.build
-+++ b/meson.build
-@@ -51,25 +51,13 @@ gnome = import('gnome')
- pkg = import('pkgconfig')
- gobject = dependency('gobject-2.0')
- yaml = dependency('yaml-0.1')
--gtkdoc = dependency('gtk-doc')
- 
- glib_prefix = dependency('glib-2.0').get_pkgconfig_variable('prefix')
--glib_docpath = join_paths(glib_prefix, 'share', 'gtk-doc', 'html')
- 
- sh = find_program('sh')
- sed = find_program('sed')
- test = find_program('test')
- 
--ret = run_command ([test, '-e', join_paths(glib_docpath, 'glib/index.html')])
--if ret.returncode() != 0
--  error('Missing documentation for GLib.')
--endif
--
--ret = run_command ([test, '-e', join_paths(glib_docpath, 'gobject/index.html')])
--if ret.returncode() != 0
--  error('Missing documentation for GObject.')
--endif
--
- python_name = get_option('python_name')
- 
- if python_name != ''
-diff --git a/modulemd/meson.build b/modulemd/meson.build
-index 9a164b5..349c982 100644
---- a/modulemd/meson.build
-+++ b/modulemd/meson.build
-@@ -523,11 +523,3 @@ configure_file(
-   configuration : xcdata
- )
- 
--gnome.gtkdoc(
--    'modulemd-1.0',
--    install_dir: 'modulemd-1.0',
--    src_dir : './modulemd',
--    main_xml : 'modulemd-docs.xml',
--    install : true,
--)
--
diff --git a/meta/recipes-devtools/libmodulemd/libmodulemd-v1_git.bb b/meta/recipes-devtools/libmodulemd/libmodulemd-v1_git.bb
index 9790470f4fe..5409051d79f 100644
--- a/meta/recipes-devtools/libmodulemd/libmodulemd-v1_git.bb
+++ b/meta/recipes-devtools/libmodulemd/libmodulemd-v1_git.bb
@@ -4,18 +4,17 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=25a3927bff3ee4f5b21bcb0ed3fcd6bb"
 
 SRC_URI = "git://github.com/fedora-modularity/libmodulemd;protocol=https;branch=1.x-maint \
            file://0001-spec_tmpl.sh-use-bin-sh-not-usr-bin-sh.patch \
-           file://0001-Do-not-generate-gtkdoc-or-python-bindings.patch \
            "
 
-PV = "1.8.15"
-SRCREV = "2d461725f781c6fdcf32893d8dcfa40bcef8dda5"
+PV = "1.8.16"
+SRCREV = "d0dcf7b373b3cf85cd39eb3bc23d31e06195a75a"
 UPSTREAM_CHECK_GITTAGREGEX = "libmodulemd-(?P<pver>1.*\d)"
 
 S = "${WORKDIR}/git"
 
 inherit meson gobject-introspection
 
-EXTRA_OEMESON = "-Ddeveloper_build=false"
+EXTRA_OEMESON = "-Ddeveloper_build=false -Dwith_docs=false"
 
 DEPENDS += "glib-2.0 libyaml glib-2.0-native python3"
 
-- 
2.17.1

Re: [PATCH 4/6] gcr: update to 3.34.0

[Andreas Müller]

On Thu, Oct 31, 2019 at 10:38 AM Alexander Kanavin
<alex.kanavin@gmail.com> wrote:
>
> inherit gettext, as gcr is now gettextized.
>
> Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
> ---
>  meta/recipes-gnome/gcr/{gcr_3.28.1.bb => gcr_3.34.0.bb} | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>  rename meta/recipes-gnome/gcr/{gcr_3.28.1.bb => gcr_3.34.0.bb} (83%)
>
> diff --git a/meta/recipes-gnome/gcr/gcr_3.28.1.bb b/meta/recipes-gnome/gcr/gcr_3.34.0.bb
> similarity index 83%
> rename from meta/recipes-gnome/gcr/gcr_3.28.1.bb
> rename to meta/recipes-gnome/gcr/gcr_3.34.0.bb
> index 0a9867451ee..f13bfecbe19 100644
> --- a/meta/recipes-gnome/gcr/gcr_3.28.1.bb
> +++ b/meta/recipes-gnome/gcr/gcr_3.34.0.bb
> @@ -11,12 +11,12 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=55ca817ccb7d5b5b66355690e9abc605"
>  DEPENDS = "intltool-native gtk+3 p11-kit glib-2.0 libgcrypt \
>             ${@bb.utils.contains('GI_DATA_ENABLED', 'True', 'libxslt-native', '', d)}"
Don't know if this is a complete V2. If so intltool-native dependency
can go (as Ross suggested).

Andreas

Re: [PATCH 4/6] gcr: update to 3.34.0

[Alexander Kanavin]

[-- Attachment #1: Type: text/plain, Size: 1332 bytes --]

I think Ross has a separate patch queued to address that.

Alex

On Thu, 31 Oct 2019 at 10:56, Andreas Müller <schnitzeltony@gmail.com>
wrote:

> On Thu, Oct 31, 2019 at 10:38 AM Alexander Kanavin
> <alex.kanavin@gmail.com> wrote:
> >
> > inherit gettext, as gcr is now gettextized.
> >
> > Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
> > ---
> >  meta/recipes-gnome/gcr/{gcr_3.28.1.bb => gcr_3.34.0.bb} | 6 +++---
> >  1 file changed, 3 insertions(+), 3 deletions(-)
> >  rename meta/recipes-gnome/gcr/{gcr_3.28.1.bb => gcr_3.34.0.bb} (83%)
> >
> > diff --git a/meta/recipes-gnome/gcr/gcr_3.28.1.bb
> b/meta/recipes-gnome/gcr/gcr_3.34.0.bb
> > similarity index 83%
> > rename from meta/recipes-gnome/gcr/gcr_3.28.1.bb
> > rename to meta/recipes-gnome/gcr/gcr_3.34.0.bb
> > index 0a9867451ee..f13bfecbe19 100644
> > --- a/meta/recipes-gnome/gcr/gcr_3.28.1.bb
> > +++ b/meta/recipes-gnome/gcr/gcr_3.34.0.bb
> > @@ -11,12 +11,12 @@ LIC_FILES_CHKSUM =
> "file://COPYING;md5=55ca817ccb7d5b5b66355690e9abc605"
> >  DEPENDS = "intltool-native gtk+3 p11-kit glib-2.0 libgcrypt \
> >             ${@bb.utils.contains('GI_DATA_ENABLED', 'True',
> 'libxslt-native', '', d)}"
> Don't know if this is a complete V2. If so intltool-native dependency
> can go (as Ross suggested).
>
> Andreas
>

[-- Attachment #2: Type: text/html, Size: 2628 bytes --]

Re: [PATCH 4/6] gcr: update to 3.34.0

[Andreas Müller]

On Thu, Oct 31, 2019 at 10:58 AM Alexander Kanavin
<alex.kanavin@gmail.com> wrote:
>
> I think Ross has a separate patch queued to address that.
>
> Alex
Yes - just thought is is a complete V2 -> everything fine :)

Andreas

Re: [PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Ross Burton]

Same failures on the Debian 10 worker:

https://autobuilder.yoctoproject.org/typhoon/#/builders/56/builds/778

Ross

Re: [PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Alexander Kanavin]

[-- Attachment #1: Type: text/plain, Size: 831 bytes --]

On Fri, 1 Nov 2019 at 18:46, Ross Burton <ross.burton@intel.com> wrote:

> Same failures on the Debian 10 worker:
>
> https://autobuilder.yoctoproject.org/typhoon/#/builders/56/builds/778
>

runqemu - ERROR - Failed to run qemu: Xlib:  extension "RANDR" missing on
display ":1".
qemu-system-x86_64: ../libepoxy-1.5.3/src/dispatch_common.c:863:
epoxy_get_proc_address: Assertion `0 && "Couldn't find current GLX or EGL
context.\n"' failed.

Thanks - the Gtk part is passing fine, so it's something SDL does that
upsets the X/GL stack on the Debian 10 host.
I am reluctant to disable the SDL part of the virgl tests on Debian 10, as
it is a new distro (unlike centos 7).

I'll try to see if I can run Debian 10 in a VM here and try to reproduce.
Or is it possible to debug directly on a Debian 10 worker?

Alex

[-- Attachment #2: Type: text/html, Size: 1443 bytes --]

Re: [PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Alexander Kanavin]

[-- Attachment #1: Type: text/plain, Size: 1384 bytes --]

On Sat, 2 Nov 2019 at 13:07, Alexander Kanavin <alex.kanavin@gmail.com>
wrote:

> Same failures on the Debian 10 worker:
>
>>
>> https://autobuilder.yoctoproject.org/typhoon/#/builders/56/builds/778
>>
>
> runqemu - ERROR - Failed to run qemu: Xlib:  extension "RANDR" missing on
> display ":1".
> qemu-system-x86_64: ../libepoxy-1.5.3/src/dispatch_common.c:863:
> epoxy_get_proc_address: Assertion `0 && "Couldn't find current GLX or EGL
> context.\n"' failed.
>
> Thanks - the Gtk part is passing fine, so it's something SDL does that
> upsets the X/GL stack on the Debian 10 host.
> I am reluctant to disable the SDL part of the virgl tests on Debian 10, as
> it is a new distro (unlike centos 7).
>
> I'll try to see if I can run Debian 10 in a VM here and try to reproduce.
> Or is it possible to debug directly on a Debian 10 worker?
>

So I actually went ahead, and installed Debian 10 into a qemu image, then
transferred a pre-populated build directory into it, and ran runqemu there
against tigervnc (hurray for the nested kvm feature!).

Both 'runqemu kvm sdl gl' and 'runqemu kvm gtk gl' work fine, including
running kmscube!

So I'd like to see what packages are installed on the Debian 10 worker vs.
my Debian 10 installation.

Can you issue 'dpkg -l' on the worker, and send me the output, please?
Maybe something is missing?

Alex

[-- Attachment #2: Type: text/html, Size: 2227 bytes --]

Re: [PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Alexander Kanavin]

[-- Attachment #1: Type: text/plain, Size: 2179 bytes --]

On Sat, 2 Nov 2019 at 23:29, Alexander Kanavin <alex.kanavin@gmail.com>
wrote:

> Same failures on the Debian 10 worker:
>
>>
>>> https://autobuilder.yoctoproject.org/typhoon/#/builders/56/builds/778
>>>
>>
>> runqemu - ERROR - Failed to run qemu: Xlib:  extension "RANDR" missing on
>> display ":1".
>> qemu-system-x86_64: ../libepoxy-1.5.3/src/dispatch_common.c:863:
>> epoxy_get_proc_address: Assertion `0 && "Couldn't find current GLX or EGL
>> context.\n"' failed.
>>
>> Thanks - the Gtk part is passing fine, so it's something SDL does that
>> upsets the X/GL stack on the Debian 10 host.
>> I am reluctant to disable the SDL part of the virgl tests on Debian 10,
>> as it is a new distro (unlike centos 7).
>>
>> I'll try to see if I can run Debian 10 in a VM here and try to reproduce.
>> Or is it possible to debug directly on a Debian 10 worker?
>>
>
> So I actually went ahead, and installed Debian 10 into a qemu image, then
> transferred a pre-populated build directory into it, and ran runqemu there
> against tigervnc (hurray for the nested kvm feature!).
>
> Both 'runqemu kvm sdl gl' and 'runqemu kvm gtk gl' work fine, including
> running kmscube!
>
> So I'd like to see what packages are installed on the Debian 10 worker vs.
> my Debian 10 installation.
>
> Can you issue 'dpkg -l' on the worker, and send me the output, please?
> Maybe something is missing?
>

After additional digging I reproduced this. The culprit is the outdated VNC
server implementation that runs on the Debian 10 autobuilder (and maybe
others as well).

Specifically, it's tightvncserver, where all Linux development has ceased
10 years ago (!).
https://www.tightvnc.com/

If you replace that with tigervnc (a modern, supported fork of tightvnc),
then the tests pass fine.
https://tigervnc.org/

As Fedora has already obsoleted tightvnc in favor of tigervnc, I think we
should do the same on all debian machines (debian provides both tightvnc
and tigervnc, but treats them as equal).
https://src.fedoraproject.org/rpms/tightvnc/blob/master/f/dead.package
I also checked that tightvnc is not available for opensuse either.

Alex

[-- Attachment #2: Type: text/html, Size: 3473 bytes --]

Re: [PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Michael Halstead]

[-- Attachment #1: Type: text/plain, Size: 2912 bytes --]

On Mon, Nov 4, 2019 at 1:11 PM Alexander Kanavin <alex.kanavin@gmail.com>
wrote:
>
> On Sat, 2 Nov 2019 at 23:29, Alexander Kanavin <alex.kanavin@gmail.com>
wrote:
>>
>> Same failures on the Debian 10 worker:
>>>>
>>>>
>>>> https://autobuilder.yoctoproject.org/typhoon/#/builders/56/builds/778
>>>
>>>
>>> runqemu - ERROR - Failed to run qemu: Xlib:  extension "RANDR" missing
on display ":1".
>>> qemu-system-x86_64: ../libepoxy-1.5.3/src/dispatch_common.c:863:
epoxy_get_proc_address: Assertion `0 && "Couldn't find current GLX or EGL
context.\n"' failed.
>>>
>>> Thanks - the Gtk part is passing fine, so it's something SDL does that
upsets the X/GL stack on the Debian 10 host.
>>> I am reluctant to disable the SDL part of the virgl tests on Debian 10,
as it is a new distro (unlike centos 7).
>>>
>>> I'll try to see if I can run Debian 10 in a VM here and try to
reproduce. Or is it possible to debug directly on a Debian 10 worker?
>>
>>
>> So I actually went ahead, and installed Debian 10 into a qemu image,
then transferred a pre-populated build directory into it, and ran runqemu
there against tigervnc (hurray for the nested kvm feature!).
>>
>> Both 'runqemu kvm sdl gl' and 'runqemu kvm gtk gl' work fine, including
running kmscube!
>>
>> So I'd like to see what packages are installed on the Debian 10 worker
vs. my Debian 10 installation.
>>
>> Can you issue 'dpkg -l' on the worker, and send me the output, please?
Maybe something is missing?
>
>
> After additional digging I reproduced this. The culprit is the outdated
VNC server implementation that runs on the Debian 10 autobuilder (and maybe
others as well).
>
> Specifically, it's tightvncserver, where all Linux development has ceased
10 years ago (!).
> https://www.tightvnc.com/
>
> If you replace that with tigervnc (a modern, supported fork of tightvnc),
then the tests pass fine.
> https://tigervnc.org/
>

TigerVNC is now installed from the standard repos where available.

>
> As Fedora has already obsoleted tightvnc in favor of tigervnc, I think we
should do the same on all debian machines (debian provides both tightvnc
and tigervnc, but treats them as equal).
> https://src.fedoraproject.org/rpms/tightvnc/blob/master/f/dead.package
> I also checked that tightvnc is not available for opensuse either.


Server components of TigerVNC are in xorg-x11-Xvnc package which is
installed on all the openSUSE workers so I expect the VNC server there to
work. Please let me know if they do not.


On Ubuntu 16.04 and Debian 8 workers I've installed from
https://bintray.com/tigervnc/stable/tigervnc/1.9.0. This should allow tests
to complete. Installing packages outside of the distribution's package
manager might invalidate these as sanity tested distros. I'd appreciate
feedback about this.

--
Michael Halstead
Yocto Project / SysAdmin


>
> Alex

[-- Attachment #2: Type: text/html, Size: 3777 bytes --]

Re: [PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Alexander Kanavin]

[-- Attachment #1: Type: text/plain, Size: 1578 bytes --]

On Fri, 8 Nov 2019 at 18:20, Michael Halstead <mhalstead@linuxfoundation.org>
wrote:

> > If you replace that with tigervnc (a modern, supported fork of
> tightvnc), then the tests pass fine.
> > https://tigervnc.org/
> >
>
> TigerVNC is now installed from the standard repos where available.
>
> >
> > As Fedora has already obsoleted tightvnc in favor of tigervnc, I think
> we should do the same on all debian machines (debian provides both tightvnc
> and tigervnc, but treats them as equal).
> > https://src.fedoraproject.org/rpms/tightvnc/blob/master/f/dead.package
> > I also checked that tightvnc is not available for opensuse either.
>
>
> Server components of TigerVNC are in xorg-x11-Xvnc package which is
> installed on all the openSUSE workers so I expect the VNC server there to
> work. Please let me know if they do not.
>
>
> On Ubuntu 16.04 and Debian 8 workers I've installed from
> https://bintray.com/tigervnc/stable/tigervnc/1.9.0. This should allow
> tests to complete. Installing packages outside of the distribution's
> package manager might invalidate these as sanity tested distros. I'd
> appreciate feedback about this.
>

Thanks a lot! There are two things to check:
1. Old instances of vnc may still be running and should be shut down (via
'vncserver -kill :1')
2. vncserver executable should point to tigervnc (if both tigervnc and
tightvnc are installed - I don't know if you removed the tightvnc packages).

Otherwise everything should be ready for re-testing the patches. I can
resend them if needed.

Alex

[-- Attachment #2: Type: text/html, Size: 2206 bytes --]

Re: [PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Michael Halstead]

[-- Attachment #1: Type: text/plain, Size: 1886 bytes --]

On Fri, Nov 8, 2019 at 9:56 AM Alexander Kanavin <alex.kanavin@gmail.com>
wrote:

> On Fri, 8 Nov 2019 at 18:20, Michael Halstead <
> mhalstead@linuxfoundation.org> wrote:
>
>> > If you replace that with tigervnc (a modern, supported fork of
>> tightvnc), then the tests pass fine.
>> > https://tigervnc.org/
>> >
>>
>> TigerVNC is now installed from the standard repos where available.
>>
>> >
>> > As Fedora has already obsoleted tightvnc in favor of tigervnc, I think
>> we should do the same on all debian machines (debian provides both tightvnc
>> and tigervnc, but treats them as equal).
>> > https://src.fedoraproject.org/rpms/tightvnc/blob/master/f/dead.package
>> > I also checked that tightvnc is not available for opensuse either.
>>
>>
>> Server components of TigerVNC are in xorg-x11-Xvnc package which is
>> installed on all the openSUSE workers so I expect the VNC server there to
>> work. Please let me know if they do not.
>>
>>
>> On Ubuntu 16.04 and Debian 8 workers I've installed from
>> https://bintray.com/tigervnc/stable/tigervnc/1.9.0. This should allow
>> tests to complete. Installing packages outside of the distribution's
>> package manager might invalidate these as sanity tested distros. I'd
>> appreciate feedback about this.
>>
>
> Thanks a lot! There are two things to check:
> 1. Old instances of vnc may still be running and should be shut down (via
> 'vncserver -kill :1')
> 2. vncserver executable should point to tigervnc (if both tigervnc and
> tightvnc are installed - I don't know if you removed the tightvnc packages).
>

I removed tightvnc and all workers have been rebooted so we are good to go.

Any comment on using packages from outside of the distro's official
packaging?


>
> Otherwise everything should be ready for re-testing the patches. I can
> resend them if needed.
>
> Alex
>

[-- Attachment #2: Type: text/html, Size: 3021 bytes --]

Re: [PATCH 3/6] oe-selftest: extend virgl gtk test to also check the SDL option

[Alexander Kanavin]

[-- Attachment #1: Type: text/plain, Size: 878 bytes --]

On Fri, 8 Nov 2019 at 19:01, Michael Halstead <mhalstead@linuxfoundation.org>
wrote:

>
>> Thanks a lot! There are two things to check:
>> 1. Old instances of vnc may still be running and should be shut down (via
>> 'vncserver -kill :1')
>> 2. vncserver executable should point to tigervnc (if both tigervnc and
>> tightvnc are installed - I don't know if you removed the tightvnc packages).
>>
>
> I removed tightvnc and all workers have been rebooted so we are good to go.
>
> Any comment on using packages from outside of the distro's official
> packaging?
>

A vnc server is entirely optional when working with yocto, it's used on the
autobuilder (machines without a physical screen) for the purpose of running
qemu with graphical frontends, but is not mentioned in the manuals as
something that users have to install or use. I'd say it's fine.

Alex

[-- Attachment #2: Type: text/html, Size: 1497 bytes --]