Re: [OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Burton, Ross]

On 27 February 2017 at 11:40, Burton, Ross  wrote:

> Typical. :/  Thanks for trying, I must have something locally that
> triggers it.
>

So I grabbed scanelf to look into this quickly:

scanelf: scanelf_file_textrels(): ELF pulseaudio has TEXTREL markings but
doesnt appear to have any real TEXTREL's !?

Sigh.  Tooling bug.

Ross
-- 
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: [OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Burton, Ross]

On 25 February 2017 at 15:30, Tanu Kaskinen  wrote:

> I couldn't reproduce the problem. I don't get the warning, and "readelf
> -d tmp/work/corei7-64-poky-linux/pulseaudio/10.0-r0/packages-
> split/pulseaudio-server/usr/bin/pulseaudio | grep TEXT" returns
> nothing.
>
> My test was performed on a fresh poky clone with default configuration,
> using command "bitbake pulseaudio". In addition to the default qemux86,
> I tried with genericx86-64 and also intel-corei7-64 from meta-intel,
> since your error message seemed to be from a corei7 build.
>

Typical. :/  Thanks for trying, I must have something locally that triggers
it.

Ross
-- 
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: [OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Tanu Kaskinen]

On Wed, 2017-02-22 at 14:48 +, Burton, Ross wrote:
> On 3 February 2017 at 07:06, Tanu Kaskinen  wrote:
> 
> > Relase notes:
> > https://www.freedesktop.org/wiki/Software/PulseAudio/Notes/10.0/
> > 
> 
> I'm seeing a new warning now:
> 
> WARNING: pulseaudio-10.0-r0 do_package_qa: QA Issue: ELF binary
> '/data/poky-master/tmp-glibc/work/corei7-64-poky-linux/pulseaudio/10.0-r0/packages-split/pulseaudio-server/usr/bin/pulseaudio'
> has relocations in .text [textrel]
> 
> Would you be able to identify where this comes from?

I couldn't reproduce the problem. I don't get the warning, and "readelf
-d tmp/work/corei7-64-poky-linux/pulseaudio/10.0-r0/packages-
split/pulseaudio-server/usr/bin/pulseaudio | grep TEXT" returns
nothing.

My test was performed on a fresh poky clone with default configuration,
using command "bitbake pulseaudio". In addition to the default qemux86,
I tried with genericx86-64 and also intel-corei7-64 from meta-intel,
since your error message seemed to be from a corei7 build.

-- 
Tanu

https://www.patreon.com/tanuk
-- 
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: [OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Burton, Ross]

On 22 February 2017 at 16:24, Tanu Kaskinen  wrote:

> I'll investigate it. I'm not familiar with the topic, though, so I'll
> first have to figure out some basic stuff: What does it mean to have
> "relocations in .text"? Why is it a bad thing to have those? How can I
> check if a binary has relocations in .text? If you have quick answers
> to any of the questions, that would be helpful.
>

https://wiki.gentoo.org/wiki/Hardened/Textrels_Guide is useful.

Ross
-- 
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: [OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Tanu Kaskinen]

On Wed, 2017-02-22 at 14:48 +, Burton, Ross wrote:
> On 3 February 2017 at 07:06, Tanu Kaskinen  wrote:
> 
> > Relase notes:
> > https://www.freedesktop.org/wiki/Software/PulseAudio/Notes/10.0/
> > 
> 
> I'm seeing a new warning now:
> 
> WARNING: pulseaudio-10.0-r0 do_package_qa: QA Issue: ELF binary
> '/data/poky-master/tmp-glibc/work/corei7-64-poky-linux/pulseaudio/10.0-r0/packages-split/pulseaudio-server/usr/bin/pulseaudio'
> has relocations in .text [textrel]
> 
> Would you be able to identify where this comes from?

I'll investigate it. I'm not familiar with the topic, though, so I'll
first have to figure out some basic stuff: What does it mean to have
"relocations in .text"? Why is it a bad thing to have those? How can I
check if a binary has relocations in .text? If you have quick answers
to any of the questions, that would be helpful.

-- 
Tanu

https://www.patreon.com/tanuk
-- 
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: [OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Burton, Ross]

On 3 February 2017 at 07:06, Tanu Kaskinen  wrote:

> Relase notes:
> https://www.freedesktop.org/wiki/Software/PulseAudio/Notes/10.0/
>

I'm seeing a new warning now:

WARNING: pulseaudio-10.0-r0 do_package_qa: QA Issue: ELF binary
'/data/poky-master/tmp-glibc/work/corei7-64-poky-linux/pulseaudio/10.0-r0/packages-split/pulseaudio-server/usr/bin/pulseaudio'
has relocations in .text [textrel]

Would you be able to identify where this comes from?

Ross
-- 
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: [OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Tanu Kaskinen]

On Fri, 2017-02-03 at 13:42 +0200, Alexander Kanavin wrote:
> On 02/03/2017 09:06 AM, Tanu Kaskinen wrote:
> > 
> > Dropped json-c and gdbm from DEPENDS. The new release doesn't use json-c
> > any more. gdbm isn't used when --with-database=simple is passed to
> > configure, so it should have been removed from DEPENDS a long time ago.
> 
> Does anything else in oe-core still use json-c? We're generally trying 
> to trim such dangling recipes.

According to "git grep json-c", the only thing still using it is
systemtap.

-- 
Tanu

https://www.patreon.com/tanuk
-- 
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: [OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Alexander Kanavin]

On 02/03/2017 09:06 AM, Tanu Kaskinen wrote:


Dropped json-c and gdbm from DEPENDS. The new release doesn't use json-c
any more. gdbm isn't used when --with-database=simple is passed to
configure, so it should have been removed from DEPENDS a long time ago.


Does anything else in oe-core still use json-c? We're generally trying 
to trim such dangling recipes.



Alex

--
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

[OE-core] [PATCH] pulseaudio: 9.0 -> 10.0

[Tanu Kaskinen]

Relase notes:
https://www.freedesktop.org/wiki/Software/PulseAudio/Notes/10.0/

The checksum of the LICENSE file changed due to some clarifications.
There were no changes to the actual licensing terms.

The LICENSE variable was not accurate, so I made changes to it.
Specifically:
 * there's no GPL code in PulseAudio so I dropped GPL from the list
 * the LGPL code allows using later versions of the license rather than
   limiting to just 2.1
 * there are some MIT and BSD licensed bits

I added more files to LIC_FILES_CHKSUM to have better coverage of all
the differently licensed code.

Dropped json-c and gdbm from DEPENDS. The new release doesn't use json-c
any more. gdbm isn't used when --with-database=simple is passed to
configure, so it should have been removed from DEPENDS a long time ago.

The new release dropped the Xen module, so the --without-xen configure
option isn't needed any more.

Added a comment for why --without-fftw is used.

Disabled the adrian echo canceller, because it has an unusual license,
and disabling the code was simpler than adding a new license to OE-Core.

Dropped upstreamed patches.
---
 meta/recipes-multimedia/pulseaudio/pulseaudio.inc  |  80 +++-
 ...oth-fail-if-user-requested-profile-doesn-.patch |  61 ---
 ...ard-don-t-allow-the-CARD_NEW-hook-to-fail.patch |  37 --
 ...-move-profile-selection-after-pa_card_new.patch | 429 -
 ...rd-remove-pa_card_new_data.active_profile.patch |  72 
 ...vailability-for-some-unavailable-profiles.patch |  79 
 .../pulseaudio/pulseaudio_10.0.bb  |  14 +
 .../pulseaudio/pulseaudio_9.0.bb   |  19 -
 8 files changed, 88 insertions(+), 703 deletions(-)
 delete mode 100644 
meta/recipes-multimedia/pulseaudio/pulseaudio/0001-alsa-bluetooth-fail-if-user-requested-profile-doesn-.patch
 delete mode 100644 
meta/recipes-multimedia/pulseaudio/pulseaudio/0002-card-don-t-allow-the-CARD_NEW-hook-to-fail.patch
 delete mode 100644 
meta/recipes-multimedia/pulseaudio/pulseaudio/0003-card-move-profile-selection-after-pa_card_new.patch
 delete mode 100644 
meta/recipes-multimedia/pulseaudio/pulseaudio/0004-card-remove-pa_card_new_data.active_profile.patch
 delete mode 100644 
meta/recipes-multimedia/pulseaudio/pulseaudio/0005-alsa-set-availability-for-some-unavailable-profiles.patch
 create mode 100644 meta/recipes-multimedia/pulseaudio/pulseaudio_10.0.bb
 delete mode 100644 meta/recipes-multimedia/pulseaudio/pulseaudio_9.0.bb

diff --git a/meta/recipes-multimedia/pulseaudio/pulseaudio.inc 
b/meta/recipes-multimedia/pulseaudio/pulseaudio.inc
index f5c5ed29c9..818ff560cc 100644
--- a/meta/recipes-multimedia/pulseaudio/pulseaudio.inc
+++ b/meta/recipes-multimedia/pulseaudio/pulseaudio.inc
@@ -2,16 +2,69 @@ SUMMARY = "Sound server for Linux and Unix-like operating 
systems"
 HOMEPAGE = "http://www.pulseaudio.org;
 AUTHOR = "Lennart Poettering"
 SECTION = "libs/multimedia"
-LICENSE = "GPLv2+ & LGPLv2.1"
-LIC_FILES_CHKSUM = "file://LICENSE;md5=d9ae089c8dc5339f8ac9d8563038a29f \
+
+# Most of PulseAudio code is under LGPLv2.1+. There are a few exceptions:
+#
+# The "adrian" echo canceller variant has code under a non-standard permissive
+# license. See src/modules/echo-cancel/adrian-license.txt for details. This
+# recipe disables the adrian echo canceller to avoid hassle with the unusual
+# license.
+#
+# The src/modules/reserve* and src/pulsecore/rtkit* files are under the MIT
+# license.
+#
+# The src/pulsecore/filter/ directory contains code under the 3-clause BSD
+# license.
+#
+# src/utils/qpaeq is licensed under AGPL. qpaeq is not installed by this
+# recipe, however, which is why AGPL is not mentioned in LICENSE.
+#
+# People who distribute PulseAudio binaries need to also consider that there
+# are some dependencies to GPL libraries. LGPL code that depends on GPL
+# libraries probably becomes effectively GPL-licensed (at compile-time? or at
+# at link-time?). I'm not a lawyer, though, so I'm not sure of the exact
+# implications. The GPL dependencies only affect the server, not the client
+# library, with the exception of libdbus that affects both. These are the GPL
+# library dependencies:
+#
+# One of the resampler implementations uses libsamplerate. This recipe doesn't
+# enable that resampler, however.
+#
+# One of the database implementations uses gdbm. This recipe doesn't enable
+# that database implementation, however.
+#
+# module-lirc (enabled by PACKAGECONFIG[lirc]) uses LIRC.
+#
+# module-equalizer-sink uses FFTW. This recipe disables that, however.
+#
+# The dependency with the most complicated licensing considerations is libdbus.
+# When PACKAGECONFIG[dbus] is enabled (like it is by default), libdbus will be
+# used by both the server and the client library (libpulse). Does this affect
+# applications that use libpulse? It should be also noted that libdbus is
+# dual-licensed: either GPLv2+ or AFL-2 terms apply. Whose decision is it which
+# of the licenses apply? What a mess. Some people