[PATCH 0/8] gobject introspection fixes

[PATCH 0/8] gobject introspection fixes

[Andre McCurdy]

Main change to avoid use of DISTRO/MACHINE_FEATURES_BACKFILL for
gobject-introspection-data (it's not appropriate, since gobject
introspection was not previously enabled by default).

  http://www.yoctoproject.org/docs/current/mega-manual/mega-manual.html#ref-features-backfill

Andre McCurdy (8):
  bitbake.conf: remove 'gobject-introspection-data' from DISTRO/MACHINE_FEATURES_BACKFILL
  default-distrovars.inc: add gobject-introspection-data to DISTRO_FEATURES_DEFAULT
  qemu.inc: add gobject-introspection-data to MACHINE_FEATURES
  arch-x86.inc: disable gobject introspection for x86-64 x32 builds
  qemuarm64.conf: don't clear MACHINE_FEATURES
  gobject-introspection.bbclass: wrap comments at 80 columns
  gobject-introspection.bbclass: make additional DEPENDS conditional
  gobject-introspection.bbclass: mark as machine specific due to COMBINED_FEATURES

 meta/classes/gobject-introspection.bbclass      | 29 +++++++++++++------------
 meta/conf/bitbake.conf                          |  4 ++--
 meta/conf/distro/include/default-distrovars.inc |  2 +-
 meta/conf/machine/include/qemu.inc              |  2 +-
 meta/conf/machine/include/x86/arch-x86.inc      |  2 +-
 meta/conf/machine/qemuarm64.conf                |  2 --
 6 files changed, 20 insertions(+), 21 deletions(-)

-- 
1.9.1

[PATCH 1/8] bitbake.conf: remove 'gobject-introspection-data' from DISTRO/MACHINE_FEATURES_BACKFILL

[Andre McCurdy]

Support for gobject introspection was not previously enabled by
default, therefore it's not appropriate to backfill into existing
distro or machine configurations.

  http://www.yoctoproject.org/docs/current/mega-manual/mega-manual.html#ref-features-backfill

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
 meta/conf/bitbake.conf | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/meta/conf/bitbake.conf b/meta/conf/bitbake.conf
index 7ed5ffb..9cec7dd 100644
--- a/meta/conf/bitbake.conf
+++ b/meta/conf/bitbake.conf
@@ -761,8 +761,8 @@ MACHINE_ESSENTIAL_EXTRA_RRECOMMENDS ?= ""
 EXTRA_IMAGE_FEATURES ??= ""
 IMAGE_FEATURES += "${EXTRA_IMAGE_FEATURES}"
 
-DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5 gobject-introspection-data"
-MACHINE_FEATURES_BACKFILL = "rtc gobject-introspection-data"
+DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5"
+MACHINE_FEATURES_BACKFILL = "rtc"
 
 COMBINED_FEATURES = "${@oe.utils.set_intersect('DISTRO_FEATURES', 'MACHINE_FEATURES', d)}"
 COMBINED_FEATURES[vardeps] += "DISTRO_FEATURES MACHINE_FEATURES"
-- 
1.9.1

[PATCH 2/8] default-distrovars.inc: add gobject-introspection-data to DISTRO_FEATURES_DEFAULT

[Andre McCurdy]

Enable gobject introspection by default for all distros which respect
DISTRO_FEATURES_DEFAULT.

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
 meta/conf/distro/include/default-distrovars.inc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/conf/distro/include/default-distrovars.inc b/meta/conf/distro/include/default-distrovars.inc
index ce42bde..8c9c5c1 100644
--- a/meta/conf/distro/include/default-distrovars.inc
+++ b/meta/conf/distro/include/default-distrovars.inc
@@ -9,7 +9,7 @@ ENABLE_BINARY_LOCALE_GENERATION ?= "1"
 LOCALE_UTF8_ONLY ?= "0"
 LOCALE_UTF8_IS_DEFAULT ?= "1"
 
-DISTRO_FEATURES_DEFAULT ?= "alsa argp bluetooth ext2 irda largefile pcmcia usbgadget usbhost wifi xattr nfs zeroconf pci 3g nfc x11"
+DISTRO_FEATURES_DEFAULT ?= "alsa argp bluetooth ext2 irda largefile pcmcia usbgadget usbhost wifi xattr nfs zeroconf pci 3g nfc x11 gobject-introspection-data"
 DISTRO_FEATURES_LIBC_DEFAULT ?= "ipv4 ipv6 libc-backtrace libc-big-macros libc-bsd libc-cxx-tests libc-catgets libc-charsets libc-crypt \
 					libc-crypt-ufc libc-db-aliases libc-envz libc-fcvt libc-fmtmsg libc-fstab libc-ftraverse \
 					libc-getlogin libc-idn libc-inet-anl libc-libm libc-locales libc-locale-code \
-- 
1.9.1

[PATCH 3/8] qemu.inc: add gobject-introspection-data to MACHINE_FEATURES

[Andre McCurdy]

Enable gobject-introspection-data by default for all qemu machines.

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
 meta/conf/machine/include/qemu.inc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/conf/machine/include/qemu.inc b/meta/conf/machine/include/qemu.inc
index f44dbe6..70879bb 100644
--- a/meta/conf/machine/include/qemu.inc
+++ b/meta/conf/machine/include/qemu.inc
@@ -11,7 +11,7 @@ XSERVER ?= "xserver-xorg \
             xf86-video-fbdev \
             xf86-input-keyboard"
 
-MACHINE_FEATURES = "alsa bluetooth usbgadget screen"
+MACHINE_FEATURES = "alsa bluetooth gobject-introspection-data screen usbgadget"
 
 MACHINEOVERRIDES =. "qemuall:"
 
-- 
1.9.1

[PATCH 4/8] arch-x86.inc: disable gobject introspection for x86-64 x32 builds

[Andre McCurdy]

Since gobject introspection is no longer backfilled into existing
machine configurations, it needs to be removed from MACHINE_FEATURES
directly when building for x86-64 x32.

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
 meta/conf/machine/include/x86/arch-x86.inc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/conf/machine/include/x86/arch-x86.inc b/meta/conf/machine/include/x86/arch-x86.inc
index ba7be8a..665da89 100644
--- a/meta/conf/machine/include/x86/arch-x86.inc
+++ b/meta/conf/machine/include/x86/arch-x86.inc
@@ -25,7 +25,7 @@ TUNE_CCARGS .= "${@bb.utils.contains('TUNE_FEATURES', 'mx32', ' -mx32', '', d)}"
 TUNE_LDARGS += "${@bb.utils.contains('TUNE_FEATURES', 'mx32', '-m elf32_x86_64', '', d)}"
 TUNE_ASARGS += "${@bb.utils.contains('TUNE_FEATURES', 'mx32', '-x32', '', d)}"
 # user mode qemu doesn't support x32
-MACHINE_FEATURES_BACKFILL_CONSIDERED_append = " ${@bb.utils.contains('TUNE_FEATURES', 'mx32', 'gobject-introspection-data', '', d)}"
+MACHINE_FEATURES_remove = "${@bb.utils.contains('TUNE_FEATURES', 'mx32', 'gobject-introspection-data', '', d)}"
 
 # ELF64 ABI
 TUNEVALID[m64] = "IA32e (x86_64) ELF64 standard ABI"
-- 
1.9.1

[PATCH 5/8] qemuarm64.conf: don't clear MACHINE_FEATURES

[Andre McCurdy]

Accept the default MACHINE_FEATURES from qemu.inc (qemuarm64
shouldn't need to be a special case).

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
 meta/conf/machine/qemuarm64.conf | 2 --
 1 file changed, 2 deletions(-)

diff --git a/meta/conf/machine/qemuarm64.conf b/meta/conf/machine/qemuarm64.conf
index 8459d0f..f59fb15 100644
--- a/meta/conf/machine/qemuarm64.conf
+++ b/meta/conf/machine/qemuarm64.conf
@@ -5,8 +5,6 @@
 require conf/machine/include/arm/arch-armv8.inc
 require conf/machine/include/qemu.inc
 
-MACHINE_FEATURES = ""
-
 KERNEL_IMAGETYPE = "Image"
 
 SERIAL_CONSOLES = "38400;ttyAMA0 38400;hvc0"
-- 
1.9.1

[PATCH 6/8] gobject-introspection.bbclass: wrap comments at 80 columns

[Andre McCurdy]

Whitespace and comment updates only, no functional changes.

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
 meta/classes/gobject-introspection.bbclass | 26 ++++++++++++--------------
 1 file changed, 12 insertions(+), 14 deletions(-)

diff --git a/meta/classes/gobject-introspection.bbclass b/meta/classes/gobject-introspection.bbclass
index ef51629..d8b8b3e 100644
--- a/meta/classes/gobject-introspection.bbclass
+++ b/meta/classes/gobject-introspection.bbclass
@@ -1,22 +1,22 @@
 # Inherit this class in recipes to enable building their introspection files
 
-# This allows disabling introspection support in recipes
-# (and therefore avoiding the use of qemu)
-# if gobject-introspection-data is omitted from DISTRO_FEATURES and MACHINE_FEATURES.
+# This allows disabling introspection support (and therefore avoiding the use
+# of qemu) in recipes if gobject-introspection-data is omitted from
+# DISTRO_FEATURES or MACHINE_FEATURES.
 EXTRA_OECONF_prepend = "${@bb.utils.contains('COMBINED_FEATURES', 'gobject-introspection-data', '--enable-introspection', '--disable-introspection', d)} "
 
 UNKNOWN_CONFIGURE_WHITELIST_append = " --enable-introspection --disable-introspection"
 
-# Generating introspection data depends on a combination of native and target introspection tools,
-# and qemu to run the target tools.
+# Generating introspection data depends on a combination of native and target
+# introspection tools, and qemu to run the target tools.
 DEPENDS_append = " gobject-introspection gobject-introspection-native qemu-native"
 
-# This is necessary for python scripts to succeed - distutils
-# failes if these are not set
+# This is necessary for python scripts to succeed - distutils fails if these
+# are not set
 export BUILD_SYS
 export HOST_SYS
-export STAGING_LIBDIR
 export STAGING_INCDIR
+export STAGING_LIBDIR
 
 # This is used by introspection tools to find .gir includes
 export XDG_DATA_DIRS = "${STAGING_DATADIR}"
@@ -28,12 +28,10 @@ do_configure_prepend_class-target () {
     cp ${STAGING_DIR_TARGET}/${datadir}/aclocal/introspection.m4 ${S}/m4
 }
 
-
-# .typelib files are needed at runtime and so they go to the main package
-# (so they'll be together with libraries they support).
+# .typelib files are needed at runtime and so they go to the main package (so
+# they'll be together with libraries they support).
 FILES_${PN}_append = " ${libdir}/girepository-*/*.typelib" 
     
-# .gir files go to dev package, as they're needed for developing (but not for running)
-# things that depends on introspection.
+# .gir files go to dev package, as they're needed for developing (but not for
+# running) things that depends on introspection.
 FILES_${PN}-dev_append = " ${datadir}/gir-*/*.gir"
-
-- 
1.9.1

[PATCH 7/8] gobject-introspection.bbclass: make additional DEPENDS conditional

[Andre McCurdy]

Make gobject introspection specific dependencies conditional on
gobject introspection being enabled.

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
 meta/classes/gobject-introspection.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/gobject-introspection.bbclass b/meta/classes/gobject-introspection.bbclass
index d8b8b3e..eb0e7ad 100644
--- a/meta/classes/gobject-introspection.bbclass
+++ b/meta/classes/gobject-introspection.bbclass
@@ -9,7 +9,7 @@ UNKNOWN_CONFIGURE_WHITELIST_append = " --enable-introspection --disable-introspe
 
 # Generating introspection data depends on a combination of native and target
 # introspection tools, and qemu to run the target tools.
-DEPENDS_append = " gobject-introspection gobject-introspection-native qemu-native"
+DEPENDS_append = " ${@bb.utils.contains('COMBINED_FEATURES', 'gobject-introspection-data', 'gobject-introspection gobject-introspection-native qemu-native', '', d)}"
 
 # This is necessary for python scripts to succeed - distutils fails if these
 # are not set
-- 
1.9.1

[PATCH 8/8] gobject-introspection.bbclass: mark as machine specific due to COMBINED_FEATURES

[Andre McCurdy]

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
 meta/classes/gobject-introspection.bbclass | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/meta/classes/gobject-introspection.bbclass b/meta/classes/gobject-introspection.bbclass
index eb0e7ad..8a09e34 100644
--- a/meta/classes/gobject-introspection.bbclass
+++ b/meta/classes/gobject-introspection.bbclass
@@ -35,3 +35,6 @@ FILES_${PN}_append = " ${libdir}/girepository-*/*.typelib"
 # .gir files go to dev package, as they're needed for developing (but not for
 # running) things that depends on introspection.
 FILES_${PN}-dev_append = " ${datadir}/gir-*/*.gir"
+
+# Due to use of COMBINED_FEATURES
+PACKAGE_ARCH = "${MACHINE_ARCH}"
-- 
1.9.1

Re: [PATCH 8/8] gobject-introspection.bbclass: mark as machine specific due to COMBINED_FEATURES

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1177 bytes --]

On Thu, Mar 17, 2016 at 10:19:29AM -0700, Andre McCurdy wrote:
> Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
> ---
>  meta/classes/gobject-introspection.bbclass | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/meta/classes/gobject-introspection.bbclass b/meta/classes/gobject-introspection.bbclass
> index eb0e7ad..8a09e34 100644
> --- a/meta/classes/gobject-introspection.bbclass
> +++ b/meta/classes/gobject-introspection.bbclass
> @@ -35,3 +35,6 @@ FILES_${PN}_append = " ${libdir}/girepository-*/*.typelib"
>  # .gir files go to dev package, as they're needed for developing (but not for
>  # running) things that depends on introspection.
>  FILES_${PN}-dev_append = " ${datadir}/gir-*/*.gir"
> +
> +# Due to use of COMBINED_FEATURES
> +PACKAGE_ARCH = "${MACHINE_ARCH}"

This was already proposed in:
http://lists.openembedded.org/pipermail/openembedded-core/2016-March/118743.html
thread, but unfortunately rejected :/.

Also this isn't enough, you would need to change all recipes depending
on one recipes with this inherit to be MACHINE_ARCH as well.

Regards,

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 188 bytes --]

Re: [PATCH 7/8] gobject-introspection.bbclass: make additional DEPENDS conditional

[Burton, Ross]

[-- Attachment #1: Type: text/plain, Size: 654 bytes --]

On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com> wrote:

>  # Generating introspection data depends on a combination of native and
> target
>  # introspection tools, and qemu to run the target tools.
> -DEPENDS_append = " gobject-introspection gobject-introspection-native
> qemu-native"
> +DEPENDS_append = " ${@bb.utils.contains('COMBINED_FEATURES',
> 'gobject-introspection-data', 'gobject-introspection
> gobject-introspection-native qemu-native', '', d)}"
>

Have you verified this works with g-i disabled and a clean sysroot?  I'm
thinking builds will still need at least some of this to provide the m4
macros.

Ross

[-- Attachment #2: Type: text/html, Size: 1110 bytes --]

Re: [PATCH 1/8] bitbake.conf: remove 'gobject-introspection-data' from DISTRO/MACHINE_FEATURES_BACKFILL

[Burton, Ross]

[-- Attachment #1: Type: text/plain, Size: 523 bytes --]

On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com> wrote:

> -DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5
> gobject-introspection-data"
> -MACHINE_FEATURES_BACKFILL = "rtc gobject-introspection-data"
> +DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5"
> +MACHINE_FEATURES_BACKFILL = "rtc"
>

So every BSP (apart from the qemu ones) would need to add the feature to
MACHINE_FEATURES?

Maybe we should remove from DISTRO backfill but keep backfilling for
MACHINE features?

Ross

[-- Attachment #2: Type: text/html, Size: 1028 bytes --]

Re: [PATCH 7/8] gobject-introspection.bbclass: make additional DEPENDS conditional

[Andre McCurdy]

On Thu, Mar 17, 2016 at 10:29 AM, Burton, Ross <ross.burton@intel.com> wrote:
>
> On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com> wrote:
>>
>>  # Generating introspection data depends on a combination of native and
>> target
>>  # introspection tools, and qemu to run the target tools.
>> -DEPENDS_append = " gobject-introspection gobject-introspection-native
>> qemu-native"
>> +DEPENDS_append = " ${@bb.utils.contains('COMBINED_FEATURES',
>> 'gobject-introspection-data', 'gobject-introspection
>> gobject-introspection-native qemu-native', '', d)}"
>
>
> Have you verified this works with g-i disabled and a clean sysroot?  I'm
> thinking builds will still need at least some of this to provide the m4
> macros.

I'm travelling at the moment, without access to any kind of reasonable
build machine... so I have to confess that I didn't test the patch set
at all. Please treat everything as RFC.

> Ross

Re: [PATCH 1/8] bitbake.conf: remove 'gobject-introspection-data' from DISTRO/MACHINE_FEATURES_BACKFILL

[Andre McCurdy]

On Thu, Mar 17, 2016 at 10:31 AM, Burton, Ross <ross.burton@intel.com> wrote:
>
> On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com> wrote:
>>
>> -DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5
>> gobject-introspection-data"
>> -MACHINE_FEATURES_BACKFILL = "rtc gobject-introspection-data"
>> +DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5"
>> +MACHINE_FEATURES_BACKFILL = "rtc"
>
> So every BSP (apart from the qemu ones) would need to add the feature to
> MACHINE_FEATURES?
>
> Maybe we should remove from DISTRO backfill but keep backfilling for MACHINE
> features?

Or don't control via a MACHINE feature at all (which would also solve
the package PACKAGE_ARCH issue) ?

Each CPU tuning file would then instead need to somehow express "this
tuning target creates binaries which can / can't be run with qemu",
but maybe that's an improvement too - isn't it better to define and
maintain that information centrally in files controlled by oe-core
rather than leave it up to BSPs to get right?

> Ross

Re: [PATCH 1/8] bitbake.conf: remove 'gobject-introspection-data' from DISTRO/MACHINE_FEATURES_BACKFILL

[Richard Purdie]

On Thu, 2016-03-17 at 11:17 -0700, Andre McCurdy wrote:
> On Thu, Mar 17, 2016 at 10:31 AM, Burton, Ross <ross.burton@intel.com
> > wrote:
> > 
> > On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com>
> > wrote:
> > > 
> > > -DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5
> > > gobject-introspection-data"
> > > -MACHINE_FEATURES_BACKFILL = "rtc gobject-introspection-data"
> > > +DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5"
> > > +MACHINE_FEATURES_BACKFILL = "rtc"
> > 
> > So every BSP (apart from the qemu ones) would need to add the
> > feature to
> > MACHINE_FEATURES?
> > 
> > Maybe we should remove from DISTRO backfill but keep backfilling
> > for MACHINE
> > features?
> 
> Or don't control via a MACHINE feature at all (which would also solve
> the package PACKAGE_ARCH issue) ?
> 
> Each CPU tuning file would then instead need to somehow express "this
> tuning target creates binaries which can / can't be run with qemu",
> but maybe that's an improvement too - isn't it better to define and
> maintain that information centrally in files controlled by oe-core
> rather than leave it up to BSPs to get right?

MACHINE_FEATURES is a variable which represents the features the target
hardware has. There is nothing which says "this must be set in
MACHINE.conf". In fact there is significant precedent for setting up
common include files which define the hardware properties.

If you look at one of my patches, it alters a core common include file
to exclude introspection in a case where we can't use it (x32).

So I think we actually want the same thing which is for the common tune
files to setup the right data.

We have made a decision to defaulting to introspection, rightly or
wrongly. I'm also quite happy to disable it where we know qemu can't
work and tune files are likely the right place to that.

I don't think this patchset is right.

Cheers,

Richard

Re: [PATCH 7/8] gobject-introspection.bbclass: make additional DEPENDS conditional

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1211 bytes --]

On Thu, Mar 17, 2016 at 05:29:59PM +0000, Burton, Ross wrote:
> On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com> wrote:
> 
> >  # Generating introspection data depends on a combination of native and
> > target
> >  # introspection tools, and qemu to run the target tools.
> > -DEPENDS_append = " gobject-introspection gobject-introspection-native
> > qemu-native"
> > +DEPENDS_append = " ${@bb.utils.contains('COMBINED_FEATURES',
> > 'gobject-introspection-data', 'gobject-introspection
> > gobject-introspection-native qemu-native', '', d)}"
> >
> 
> Have you verified this works with g-i disabled and a clean sysroot?  I'm
> thinking builds will still need at least some of this to provide the m4
> macros.

Even with gobject-introspection-data removed from DISTRO_FEATURES I'm seeing
couple recipes (our version of gst-*) failing with:

| configure:22282: error: possibly undefined macro: PKG_CONFIG_SYSROOT_DIR
|       If this token and others are legitimate, please use m4_pattern_allow.
|       See the Autoconf documentation.

and PKG_CONFIG_SYSROOT_DIR is added by introspection.m4 from g-i patchset.

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 188 bytes --]

Re: [PATCH 7/8] gobject-introspection.bbclass: make additional DEPENDS conditional

[Richard Purdie]

On Thu, 2016-03-17 at 22:36 +0100, Martin Jansa wrote:
> On Thu, Mar 17, 2016 at 05:29:59PM +0000, Burton, Ross wrote:
> > On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com>
> > wrote:
> > 
> > >  # Generating introspection data depends on a combination of
> > > native and
> > > target
> > >  # introspection tools, and qemu to run the target tools.
> > > -DEPENDS_append = " gobject-introspection gobject-introspection
> > > -native
> > > qemu-native"
> > > +DEPENDS_append = " ${@bb.utils.contains('COMBINED_FEATURES',
> > > 'gobject-introspection-data', 'gobject-introspection
> > > gobject-introspection-native qemu-native', '', d)}"
> > > 
> > 
> > Have you verified this works with g-i disabled and a clean sysroot?
> >   I'm
> > thinking builds will still need at least some of this to provide
> > the m4
> > macros.
> 
> Even with gobject-introspection-data removed from DISTRO_FEATURES I'm
> seeing
> couple recipes (our version of gst-*) failing with:
> 
> > configure:22282: error: possibly undefined macro:
> > PKG_CONFIG_SYSROOT_DIR
> >       If this token and others are legitimate, please use
> > m4_pattern_allow.
> >       See the Autoconf documentation.
> 
> and PKG_CONFIG_SYSROOT_DIR is added by introspection.m4 from g-i
> patchset.

Sounds like a missing pkgconfig depends somewhere. Does the g-i class
inherit pkgconfig?

Cheers,

Richard

Re: [PATCH 1/8] bitbake.conf: remove 'gobject-introspection-data' from DISTRO/MACHINE_FEATURES_BACKFILL

[Andre McCurdy]

On Thu, Mar 17, 2016 at 2:02 PM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> On Thu, 2016-03-17 at 11:17 -0700, Andre McCurdy wrote:
>> On Thu, Mar 17, 2016 at 10:31 AM, Burton, Ross <ross.burton@intel.com
>> > wrote:
>> >
>> > On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com>
>> > wrote:
>> > >
>> > > -DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5
>> > > gobject-introspection-data"
>> > > -MACHINE_FEATURES_BACKFILL = "rtc gobject-introspection-data"
>> > > +DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5"
>> > > +MACHINE_FEATURES_BACKFILL = "rtc"
>> >
>> > So every BSP (apart from the qemu ones) would need to add the
>> > feature to
>> > MACHINE_FEATURES?
>> >
>> > Maybe we should remove from DISTRO backfill but keep backfilling
>> > for MACHINE
>> > features?
>>
>> Or don't control via a MACHINE feature at all (which would also solve
>> the package PACKAGE_ARCH issue) ?
>>
>> Each CPU tuning file would then instead need to somehow express "this
>> tuning target creates binaries which can / can't be run with qemu",
>> but maybe that's an improvement too - isn't it better to define and
>> maintain that information centrally in files controlled by oe-core
>> rather than leave it up to BSPs to get right?
>
> MACHINE_FEATURES is a variable which represents the features the target
> hardware has. There is nothing which says "this must be set in
> MACHINE.conf". In fact there is significant precedent for setting up
> common include files which define the hardware properties.
>
> If you look at one of my patches, it alters a core common include file
> to exclude introspection in a case where we can't use it (x32).
>
> So I think we actually want the same thing which is for the common tune
> files to setup the right data.

Yes, I think so.

Deciding if/how qemu can run binaries for each tuning target is
already partially addressed by the QEMU_EXTRAOPTIONS logic in
qemu.bbclass. My suggestion would be to somehow generalise that
existing logic, so (at least for machines using one of the standard
oe-core tuning files) the question "can qemu run binaries created for
${MACHINE}?" could be answered automatically.

> We have made a decision to defaulting to introspection, rightly or
> wrongly. I'm also quite happy to disable it where we know qemu can't
> work and tune files are likely the right place to that.
>
> I don't think this patchset is right.
>
> Cheers,
>
> Richard
>

Re: [PATCH 7/8] gobject-introspection.bbclass: make additional DEPENDS conditional

[alexander.kanavin]

> Even with gobject-introspection-data removed from DISTRO_FEATURES I'm
> seeing
> couple recipes (our version of gst-*) failing with:
>
> | configure:22282: error: possibly undefined macro: PKG_CONFIG_SYSROOT_DIR
> |       If this token and others are legitimate, please use
> m4_pattern_allow.
> |       See the Autoconf documentation.
>
> and PKG_CONFIG_SYSROOT_DIR is added by introspection.m4 from g-i patchset.

Disabling building of introspection data will not guarantee absence of
build failures, I'm afraid. Some recipes will work just fine unchanged,
but others, including gstreamer, do need custom fixing, usually because
they have non-standard locations for m4 macros, or those macros themselves
are non-standard. So you need to rebase your recipes on top of those
available now in oe-core, and particularly this snippet will fix the
immediate problem you have above:

delete_pkg_m4_file() {
        # This m4 file is out of date and is missing
PKG_CONFIG_SYSROOT_PATH tweaks which we need for introspection
        rm "${S}/common/m4/pkg.m4" || true
}

do_configure[prefuncs] += " delete_pkg_m4_file"


Alex

Re: [PATCH 7/8] gobject-introspection.bbclass: make additional DEPENDS conditional

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1370 bytes --]

On Fri, Mar 18, 2016 at 09:22:29PM +0200, alexander.kanavin@linux.intel.com wrote:
> > Even with gobject-introspection-data removed from DISTRO_FEATURES I'm
> > seeing
> > couple recipes (our version of gst-*) failing with:
> >
> > | configure:22282: error: possibly undefined macro: PKG_CONFIG_SYSROOT_DIR
> > |       If this token and others are legitimate, please use
> > m4_pattern_allow.
> > |       See the Autoconf documentation.
> >
> > and PKG_CONFIG_SYSROOT_DIR is added by introspection.m4 from g-i patchset.
> 
> Disabling building of introspection data will not guarantee absence of
> build failures, I'm afraid. Some recipes will work just fine unchanged,
> but others, including gstreamer, do need custom fixing, usually because
> they have non-standard locations for m4 macros, or those macros themselves
> are non-standard. So you need to rebase your recipes on top of those
> available now in oe-core, and particularly this snippet will fix the
> immediate problem you have above:
> 
> delete_pkg_m4_file() {
>         # This m4 file is out of date and is missing
> PKG_CONFIG_SYSROOT_PATH tweaks which we need for introspection
>         rm "${S}/common/m4/pkg.m4" || true
> }
> 
> do_configure[prefuncs] += " delete_pkg_m4_file"

Thanks, this did the trick.

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 188 bytes --]

Re: [PATCH 1/8] bitbake.conf: remove 'gobject-introspection-data' from DISTRO/MACHINE_FEATURES_BACKFILL

[Alexander Kanavin]

On 03/18/2016 01:08 PM, Andre McCurdy wrote:

> Deciding if/how qemu can run binaries for each tuning target is
> already partially addressed by the QEMU_EXTRAOPTIONS logic in
> qemu.bbclass. My suggestion would be to somehow generalise that
> existing logic, so (at least for machines using one of the standard
> oe-core tuning files) the question "can qemu run binaries created for
> ${MACHINE}?" could be answered automatically.

I agree: machine features should specify the capabilities of a machine, 
and 'generate of gobject introspection data' is only a specific 
application of 'machine has qemu usermode support' feature. We might 
have other uses for qemu userspace in the future, for example gtkdoc 
generation.

So I'm thinking of having a machine feature called 'qemu-usermode', 
which is turned on by default, and can be disabled in specific machine 
configurations. And then a distro feature called 'gobject introspection 
data', which is on or off depending on whether the machine supports it, 
but can also be turned off explicitly. And then the recipes and recipe 
classes do the right thing. How's that?


Alex

Re: [PATCH 1/8] bitbake.conf: remove 'gobject-introspection-data' from DISTRO/MACHINE_FEATURES_BACKFILL

[Richard Purdie]

On Fri, 2016-03-18 at 04:08 -0700, Andre McCurdy wrote:
> On Thu, Mar 17, 2016 at 2:02 PM, Richard Purdie
> <richard.purdie@linuxfoundation.org> wrote:
> > On Thu, 2016-03-17 at 11:17 -0700, Andre McCurdy wrote:
> > > On Thu, Mar 17, 2016 at 10:31 AM, Burton, Ross <
> > > ross.burton@intel.com
> > > > wrote:
> > > > 
> > > > On 17 March 2016 at 17:19, Andre McCurdy <armccurdy@gmail.com>
> > > > wrote:
> > > > > 
> > > > > -DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5
> > > > > gobject-introspection-data"
> > > > > -MACHINE_FEATURES_BACKFILL = "rtc gobject-introspection-data"
> > > > > +DISTRO_FEATURES_BACKFILL = "pulseaudio sysvinit bluez5"
> > > > > +MACHINE_FEATURES_BACKFILL = "rtc"
> > > > 
> > > > So every BSP (apart from the qemu ones) would need to add the
> > > > feature to
> > > > MACHINE_FEATURES?
> > > > 
> > > > Maybe we should remove from DISTRO backfill but keep
> > > > backfilling
> > > > for MACHINE
> > > > features?
> > > 
> > > Or don't control via a MACHINE feature at all (which would also
> > > solve
> > > the package PACKAGE_ARCH issue) ?
> > > 
> > > Each CPU tuning file would then instead need to somehow express
> > > "this
> > > tuning target creates binaries which can / can't be run with
> > > qemu",
> > > but maybe that's an improvement too - isn't it better to define
> > > and
> > > maintain that information centrally in files controlled by oe
> > > -core
> > > rather than leave it up to BSPs to get right?
> > 
> > MACHINE_FEATURES is a variable which represents the features the
> > target
> > hardware has. There is nothing which says "this must be set in
> > MACHINE.conf". In fact there is significant precedent for setting
> > up
> > common include files which define the hardware properties.
> > 
> > If you look at one of my patches, it alters a core common include
> > file
> > to exclude introspection in a case where we can't use it (x32).
> > 
> > So I think we actually want the same thing which is for the common
> > tune
> > files to setup the right data.
> 
> Yes, I think so.
> 
> Deciding if/how qemu can run binaries for each tuning target is
> already partially addressed by the QEMU_EXTRAOPTIONS logic in
> qemu.bbclass. My suggestion would be to somehow generalise that
> existing logic, so (at least for machines using one of the standard
> oe-core tuning files) the question "can qemu run binaries created for
> ${MACHINE}?" could be answered automatically.

I think the "correct" way to do this is through MACHINE_FEATURES with
QEMU_EXTRAOPTIONS being used at the class level to sort out the flags
for qemu. The naming of the option within MACHINE_FEATURES is obviously
suboptimal at the moment though as Alex mentions.

I guess we did it this way since it meant we could easily use
COMBINED_FEATURES but we could have differently named options in there
at the cost of a bit of extra code.

Cheers,

Richard