package/spice: disable tests
PATCH v2 1/4 package/spice: disable tests
By disabling tests, we'll remove the optional gdk-pixbuf dependency
Fixes:
- http://autobuild.buildroot.org/results/96c786f85d35f33508e9c71778043d16b87f72cd
---
Changes v1 -> v2 (after review of Yann E. Morin):
- Disable tests instead of adding and fixing gdk-pixbuf optional dependency
.../0001-configure.ac-add-enable-tests.patch | 54 +++++++++++++++++++ package/spice/spice.mk | 5 +- 2 files changed, 58 insertions(+), 1 deletion(-) create mode 100644 package/spice/0001-configure.ac-add-enable-tests.patch
package/spice/0001-configure.ac-add-enable-tests.patch new file +54 -0
@@ -0,0 +1,54 @@ +From 1b6eaf5589a14763452cbe53382cc699cdeca141 Mon Sep 17 00:00:00 2001 +From: Fabrice Fontaine <[email protected]> +Date: Fri, 15 Nov 2019 11:36:14 +0100 +Subject: [PATCH] configure.ac: add --enable-tests + +Allow the user to disable tests through --disable-tests, this is +especially useful for example to disable gdk-pixbuf dependency + +Signed-off-by: Fabrice Fontaine <[email protected]> +[Upstream status: +https://gitlab.freedesktop.org/spice/spice-common/merge_requests/6] +--- + Makefile.am | 6 +++++- + configure.ac | 7 +++++++ + 2 files changed, 12 insertions(+), 1 deletion(-) + +diff --git a/subprojects/spice-common/Makefile.am b/subprojects/spice-common/Makefile.am +index 5402499..00623a1 100644 +--- a/subprojects/spice-common/Makefile.am ++++ b/subprojects/spice-common/Makefile.am +@@ -1,7 +1,11 @@ + NULL = + ACLOCAL_AMFLAGS = -I m4 + +-SUBDIRS = python_modules common tests docs ++SUBDIRS = python_modules common docs ++ ++if ENABLE_TESTS ++SUBDIRS += tests ++endif + + EXTRA_DIST = \ + meson.build \ +diff --git a/subprojects/spice-common/configure.ac b/subprojects/spice-common/configure.ac +index 9d10287..2dba7c8 100644 +--- a/subprojects/spice-common/configure.ac ++++ b/subprojects/spice-common/configure.ac +@@ -33,6 +33,13 @@ AC_SEARCH_LIBS(regcomp, [regex rx]) + SPICE_CHECK_SYSDEPS + SPICE_EXTRA_CHECKS + ++AC_ARG_ENABLE([tests], ++ AS_HELP_STRING([--enable-tests], ++ [Enable tests @<:@default=yes@:>@]), ++ [], ++ enable_tests="yes") ++AM_CONDITIONAL(ENABLE_TESTS, test "x$enable_tests" = "xyes") ++ + AC_ARG_ENABLE([alignment-checks], + AS_HELP_STRING([--enable-alignment-checks], + [Enable runtime checks for cast alignment @<:@default=no@:>@]), +-- +2.23.0 +
package/spice/spice.mk +4 -1
@@ -17,6 +17,8 @@ SPICE_DEPENDENCIES = \ openssl \ pixman \ spice-protocol +# We're patching subprojects/spice-common/configure.ac +SPICE_AUTORECONF = YES # We disable everything for now, because the dependency tree can become # quite deep if we try to enable some features, and I have not tested that.
@@ -25,7 +27,8 @@ SPICE_CONF_OPTS = \ --disable-opengl \ --disable-smartcard \ --without-sasl \ - --disable-manual + --disable-manual \ + --disable-tests SPICE_DEPENDENCIES += host-pkgconf
Signature · 1 line
2.24.0
PATCH v2 2/4 package/celt051: needs dynamic library
celt051 can't be built statically with opus as opus embeds its own celt version. So add a dynamic library dependency to celt051 (and not to opus, as celt051 is only used optionally by spice whereas opus is selected by 9 packages)
This will fix static build of spice with opus and celt:
/home/fabrice/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/i686-buildroot-linux-uclibc/7.4.0/../../../../i686-buildroot-linux-uclibc/bin/ld: /home/fabrice/buildroot/output/host/i686-buildroot-linux-uclibc/sysroot/usr/lib/libopus.a(vq.o): in function `renormalise_vector':
vq.c:(.text+0x5ee): multiple definition of `renormalise_vector'; /home/fabrice/buildroot/output/host/i686-buildroot-linux-uclibc/sysroot/usr/lib/libcelt051.a(vq.o):vq.c:(.text+0x6e6): first defined here
Fixes:
- http://autobuild.buildroot.org/results/96c786f85d35f33508e9c71778043d16b87f72cd
--- package/celt051/Config.in | 4 ++++ 1 file changed, 4 insertions(+)
package/celt051/Config.in +4 -0
@@ -1,5 +1,6 @@ config BR2_PACKAGE_CELT051 bool "celt051" + depends on !BR2_STATIC_LIBS select BR2_PACKAGE_LIBOGG help The CELT ultra-low delay audio codec
@@ -13,3 +14,6 @@ config BR2_PACKAGE_CELT051 Note: this is version 0.5.1.3 of celt. http://www.celt-codec.org/ + +comment "celt051 needs a toolchain w/ dynamic library" + depends on BR2_STATIC_LIBS
Signature · 1 line
2.24.0
PATCH v2 2/4 package/celt051: needs dynamic library
Quoted 7 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
celt051 can't be built statically with opus as opus embeds its own celt version. So add a dynamic library dependency to celt051 (and not to opus, as celt051 is only used optionally by spice whereas opus is selected by 9 packages)
Quoted 7 lines
This will fix static build of spice with opus and celt:
/home/fabrice/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/i686-buildroot-linux-uclibc/7.4.0/../../../../i686-buildroot-linux-uclibc/bin/ld: /home/fabrice/buildroot/output/host/i686-buildroot-linux-uclibc/sysroot/usr/lib/libopus.a(vq.o): in function `renormalise_vector':
vq.c:(.text+0x5ee): multiple definition of `renormalise_vector'; /home/fabrice/buildroot/output/host/i686-buildroot-linux-uclibc/sysroot/usr/lib/libcelt051.a(vq.o):vq.c:(.text+0x6e6): first defined hereFixes:
- http://autobuild.buildroot.org/results/96c786f85d35f33508e9c71778043d16b87f72cd
So this is only really an issue for spice, if spice is built statically and both celt051 and opus are enabled?
If so, then I would prefer to handle it in spice.mk instead, E.G. blacklist the celt51 + opus + static combination with a sensible comment, something like:
ifeq ($(BR2_PACKAGE_CELT051),y)
# opus embeds a copy of celt, causing multiple definitions of a number # of symbols when linked statically together with celt051, so disable # celt051 support for that combination ifeq ($(BR2_PACKAGE_OPUS)$(BR2_STATIC_LIBS),yy) SPICE_CONF_OPTS += --disable-celt051 else SPICE_CONF_OPTS += --enable-celt051 SPICE_DEPENDENCIES += celt051 endif
elseSPICE_CONF_OPTS += --disable-celt051
endif
Care to send a patch doing something like that? Thanks.
Signature · 1 line
Bye, Peter Korsgaard
PATCH v2 2/4 package/celt051: needs dynamic library
Dear Peter,
Le sam. 23 nov. 2019 ? 14:38, Peter Korsgaard <[email protected]> a ?crit :
Quoted 35 lines
Quoted 8 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
celt051 can't be built statically with opus as opus embeds its own celt version. So add a dynamic library dependency to celt051 (and not to opus, as celt051 is only used optionally by spice whereas opus is selected by 9 packages)
Quoted 7 lines
This will fix static build of spice with opus and celt:
/home/fabrice/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/i686-buildroot-linux-uclibc/7.4.0/../../../../i686-buildroot-linux-uclibc/bin/ld: /home/fabrice/buildroot/output/host/i686-buildroot-linux-uclibc/sysroot/usr/lib/libopus.a(vq.o): in function `renormalise_vector':
vq.c:(.text+0x5ee): multiple definition of `renormalise_vector'; /home/fabrice/buildroot/output/host/i686-buildroot-linux-uclibc/sysroot/usr/lib/libcelt051.a(vq.o):vq.c:(.text+0x6e6): first defined hereFixes:
- http://autobuild.buildroot.org/results/96c786f85d35f33508e9c71778043d16b87f72cdSo this is only really an issue for spice, if spice is built statically and both celt051 and opus are enabled?
If so, then I would prefer to handle it in spice.mk instead, E.G. blacklist the celt51 + opus + static combination with a sensible comment, something like:
ifeq ($(BR2_PACKAGE_CELT051),y)
# opus embeds a copy of celt, causing multiple definitions of a number # of symbols when linked statically together with celt051, so disable # celt051 support for that combination ifeq ($(BR2_PACKAGE_OPUS)$(BR2_STATIC_LIBS),yy) SPICE_CONF_OPTS += --disable-celt051 else SPICE_CONF_OPTS += --enable-celt051 SPICE_DEPENDENCIES += celt051 endif
Quoted 5 lines
else
SPICE_CONF_OPTS+= --disable-celt051
endifCare to send a patch doing something like that? Thanks.
An other option would be to remove celt051 package indeed the root cause of this static build failure is that celt has been merged into the IETF Opus codec and is now obsolete (see http://celt-codec.org/).
Quoted 3 lines
--
Bye, Peter Korsgaard
Best Regards,
Fabrice
PATCH v2 2/4 package/celt051: needs dynamic library
Quoted 10 lines · Fabrice Fontaine
Dear Peter,
Le sam. 23 nov. 2019 ? 14:38, Peter Korsgaard <[email protected]> a ?crit :
Quoted 7 lines
Quoted 7 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
celt051 can't be built statically with opus as opus embeds its own celt version. So add a dynamic library dependency to celt051 (and not to opus, as celt051 is only used optionally by spice whereas opus is selected by 9 packages)
[snip]
Quoted 4 lines
Quoted 1 line
Care to send a patch doing something like that? Thanks.
An other option would be to remove celt051 package indeed the root cause of this static build failure is that celt has been merged into the IETF Opus codec and is now obsolete (see http://celt-codec.org/).
Indeed, that makes more sense. Care to remove it? Legacy handling should selectBR2_PACKAGE_OPUS then (with appropriate explanation).
Yann is developer for celt051, so please give a peep if it can't be removed.
Regards,
Arnout
PATCH v2 2/4 package/celt051: needs dynamic library
Arnout, All,
On 2020-01-23 23:23 +0100, Arnout Vandecappelle spake thusly:
Quoted 13 lines · Fabrice Fontaine
Quoted 10 lines
Dear Peter,
Le sam. 23 nov. 2019 ? 14:38, Peter Korsgaard <[email protected]> a ?crit :
Quoted 7 lines
Quoted 7 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
celt051 can't be built statically with opus as opus embeds its own celt version. So add a dynamic library dependency to celt051 (and not to opus, as celt051 is only used optionally by spice whereas opus is selected by 9 packages)
[snip]
Quoted 1 line
Quoted 1 line
Care to send a patch doing something like that? Thanks.
Quoted 8 lines
Quoted 3 lines
An other option would be to remove celt051 package indeed the root cause of this static build failure is that celt has been merged into the IETF Opus codec and is now obsolete (see http://celt-codec.org/).
Indeed, that makes more sense. Care to remove it? Legacy handling should select
BR2_PACKAGE_OPUSthen (with appropriate explanation).Yann is developer for celt051, so please give a peep if it can't be removed.
Well, celt-0.51 was introduced because at the time. spice on;y supported that, and not the later versions, which is still documented in the package itself.
But sicne then (7 years ago now), spice was updated, so maybe it can support newer versions, and maybe it works with Opus instead.
Regards,
Yann E. MORIN.
Signature · 6 lines
.-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 561 099 427 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------'
PATCH v2 2/4 package/celt051: needs dynamic library
Quoted 23 lines · Yann E. MORIN
Arnout, All,
On 2020-01-23 23:23 +0100, Arnout Vandecappelle spake thusly:
Quoted 20 lines · Fabrice Fontaine
Quoted 10 lines
Dear Peter,
Le sam. 23 nov. 2019 ? 14:38, Peter Korsgaard <[email protected]> a ?crit :
Quoted 7 lines
Quoted 7 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
celt051 can't be built statically with opus as opus embeds its own celt version. So add a dynamic library dependency to celt051 (and not to opus, as celt051 is only used optionally by spice whereas opus is selected by 9 packages)
[snip]
Quoted 4 lines
Quoted 1 line
Care to send a patch doing something like that? Thanks.
An other option would be to remove celt051 package indeed the root cause of this static build failure is that celt has been merged into the IETF Opus codec and is now obsolete (see http://celt-codec.org/).
Indeed, that makes more sense. Care to remove it? Legacy handling should select
BR2_PACKAGE_OPUSthen (with appropriate explanation).
Quoted 8 lines
Quoted 1 line
Yann is developer for celt051, so please give a peep if it can't be removed.
Well, celt-0.51 was introduced because at the time. spice on;y supported that, and not the later versions, which is still documented in the package itself.
But sicne then (7 years ago now), spice was updated, so maybe it can support newer versions, and maybe it works with Opus instead.
Opus support in spice has already been added with the bump to 0.14.1 inf33f7a4f6407f.
So if someone could remove celt051 please :-)
Regards,
Arnout
PATCH v2 2/4 package/celt051: needs dynamic library
Hi Arnout,
Le lun. 27 janv. 2020 ? 17:22, Arnout Vandecappelle <[email protected]> a ?crit :
Quoted 33 lines · Yann E. MORIN
Quoted 29 lines
Arnout, All,
On 2020-01-23 23:23 +0100, Arnout Vandecappelle spake thusly:
Quoted 13 lines · Fabrice Fontaine
Quoted 10 lines
Dear Peter,
Le sam. 23 nov. 2019 ? 14:38, Peter Korsgaard <[email protected]> a ?crit :
Quoted 7 lines
Quoted 7 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
celt051 can't be built statically with opus as opus embeds its own celt version. So add a dynamic library dependency to celt051 (and not to opus, as celt051 is only used optionally by spice whereas opus is selected by 9 packages)
[snip]
Quoted 1 line
Quoted 1 line
Care to send a patch doing something like that? Thanks.
Quoted 8 lines
Quoted 3 lines
An other option would be to remove celt051 package indeed the root cause of this static build failure is that celt has been merged into the IETF Opus codec and is now obsolete (see http://celt-codec.org/).
Indeed, that makes more sense. Care to remove it? Legacy handling should select
BR2_PACKAGE_OPUSthen (with appropriate explanation).Yann is developer for celt051, so please give a peep if it can't be removed.
Well, celt-0.51 was introduced because at the time. spice on;y supported that, and not the later versions, which is still documented in the package itself.
Quoted 7 lines
Quoted 2 lines
But sicne then (7 years ago now), spice was updated, so maybe it can support newer versions, and maybe it works with Opus instead.
Opus support in spice has already been added with the bump to 0.14.1 in
f33f7a4f6407f.So if someone could remove celt051 please :-)
It's done since 29 hours and commit b32efbdb0354557db48852478b99080705ba315b ;-)
Quoted 4 lines
Regards,
Arnout
Regards,
Fabrice
PATCH v2 3/4 package/spice-protocol: bump to version 0.14.0
In reply to Fabrice Fontaine- This bump is needed for spice 0.14.2
- Add license hash
--- package/spice-protocol/spice-protocol.hash | 5 ++++- package/spice-protocol/spice-protocol.mk | 2 +- 2 files changed, 5 insertions(+), 2 deletions(-)
package/spice-protocol/spice-protocol.hash +4 -1
@@ -1,2 +1,5 @@ # Locally calculated after checking pgp signature -sha256 20350bc4309039fdf0d29ee4fd0033cde27bccf33501e13b3c1befafde9d0c9c spice-protocol-0.12.14.tar.bz2 +sha256 b6a4aa1ca32668790b45a494bbd000e9d05797b391d5a5d4b91adf1118216eac spice-protocol-0.14.0.tar.bz2 + +# Hash for license file +sha256 eb55ea746be694c449224c7ddce6340648b075fd377281712a19094e7fb7ef92 COPYING
package/spice-protocol/spice-protocol.mk +1 -1
@@ -4,7 +4,7 @@ # ################################################################################ -SPICE_PROTOCOL_VERSION = 0.12.14 +SPICE_PROTOCOL_VERSION = 0.14.0 SPICE_PROTOCOL_SOURCE = spice-protocol-$(SPICE_PROTOCOL_VERSION).tar.bz2 SPICE_PROTOCOL_SITE = http://www.spice-space.org/download/releases SPICE_PROTOCOL_LICENSE = BSD-3-Clause
Signature · 1 line
2.24.0
PATCH v2 3/4 package/spice-protocol: bump to version 0.14.0
Quoted 6 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
- This bump is needed for spice 0.14.2
- Add license hashSigned-off-by Fabrice Fontaine <[email protected]>
Committed, thanks.
Signature · 1 line
Bye, Peter Korsgaard
PATCH v2 3/4 package/spice-protocol: bump to version 0.14.0
In reply to Fabrice FontaineQuoted 6 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
- This bump is needed for spice 0.14.2
- Add license hashSigned-off-by Fabrice Fontaine <[email protected]>
Committed to 2019.02.x and 2019.08.x, thanks.
Signature · 1 line
Bye, Peter Korsgaard
PATCH v2 3/4 package/spice-protocol: bump to version 0.14.0
Hi Peter,
Quoted 8 lines · Peter Korsgaard
Quoted 6 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
- This bump is needed for spice 0.14.2
- Add license hashSigned-off-by Fabrice Fontaine <[email protected]>Committed to 2019.02.x and 2019.08.x, thanks.
Not in 2019.02.x as of commit 7de477493571b9. This breaks spice, I presume.
baruch
Signature · 3 lines
http://baruch.siach.name/blog/ ~. .~ Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
- baruch at tkos.co.il - tel: +972.2.679.5364, http://www.tkos.co.il -PATCH v2 3/4 package/spice-protocol: bump to version 0.14.0
Quoted 6 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Baruch" == Baruch Siach <[email protected]> writes:
Hi Peter,
Quoted 2 lines · Peter Korsgaard
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
- This bump is needed for spice 0.14.2
- Add license hashSigned-off-by Fabrice Fontaine <[email protected]>Committed to 2019.02.x and 2019.08.x, thanks.
Not in 2019.02.x as of commit
7de477493571b9. This breaks spice, I presume.
Ups, indeed not - Fixed now, thanks!
Signature · 1 line
Bye, Peter Korsgaard
PATCH v2 4/4 package/spice: security bump to version 0.14.2
In reply to Fabrice Fontaine- Fix CVE-2019-3813: fix off-by-one error in group/slot boundary check
- Add license hash
--- package/spice/spice.hash | 3 ++- package/spice/spice.mk | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
package/spice/spice.hash +2 -1
@@ -1,2 +1,3 @@ # Locally calculated -sha256 1ead5de63d06eededed4017db37240f07bef0abffbaf621899647e7e685a1519 spice-0.14.1.tar.bz2 +sha256 b203b3882e06f4c7249a3150d90c84e1a90490d41ead255a3d2cede46f4a29a7 spice-0.14.2.tar.bz2 +sha256 dc626520dcd53a22f727af3ee42c770e56c97a64fe3adb063799d8ab032fe551 COPYING
package/spice/spice.mk +1 -1
@@ -4,7 +4,7 @@ # ################################################################################ -SPICE_VERSION = 0.14.1 +SPICE_VERSION = 0.14.2 SPICE_SOURCE = spice-$(SPICE_VERSION).tar.bz2 SPICE_SITE = http://www.spice-space.org/download/releases/spice-server SPICE_LICENSE = LGPL-2.1+
Signature · 1 line
2.24.0
PATCH v2 4/4 package/spice: security bump to version 0.14.2
Quoted 6 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
- Fix CVE-2019-3813: fix off-by-one error in group/slot boundary check
- Add license hashSigned-off-by Fabrice Fontaine <[email protected]>
Committed, thanks.
Signature · 1 line
Bye, Peter Korsgaard
PATCH v2 4/4 package/spice: security bump to version 0.14.2
In reply to Fabrice FontaineQuoted 6 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
- Fix CVE-2019-3813: fix off-by-one error in group/slot boundary check
- Add license hashSigned-off-by Fabrice Fontaine <[email protected]>
Committed to 2019.02.x and 2019.08.x, thanks.
Signature · 1 line
Bye, Peter Korsgaard
PATCH v2 1/4 package/spice: disable tests
In reply to Fabrice FontaineQuoted 10 lines · On Sun, 17 Nov 2019 17:44:49 +0100 Fabrice Fontaine…
By disabling tests, we'll remove the optional gdk-pixbuf dependency
Fixes:
- http://autobuild.buildroot.org/results/96c786f85d35f33508e9c71778043d16b87f72cdSigned-off-by Fabrice Fontaine <[email protected]>---
Changes v1 -> v2 (after review of Yann E. Morin):
- Disable tests instead of adding and fixing gdk-pixbuf optional dependency
I've applied to next. However, your patch to upstream is not really complete, or at least an additional patch is needed: when tests are enabled, their configure.ac script should check if gdk-pixbuf is available. And if it's not available, it should bail out, or disable tests, but not allow the configure process to complete, and then have a build failure when compiling.
Thanks!
Thomas
Signature · 3 lines
Thomas Petazzoni, CTO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
PATCH v2 1/4 package/spice: disable tests
Hi Thomas,
Le lun. 18 nov. 2019 ? 22:47, Thomas Petazzoni
<[email protected]> a ?crit :
Quoted 15 lines
Quoted 10 lines · On Sun, 17 Nov 2019 17:44:49 +0100 Fabrice Fontaine…
By disabling tests, we'll remove the optional gdk-pixbuf dependency
Fixes:
- http://autobuild.buildroot.org/results/96c786f85d35f33508e9c71778043d16b87f72cdSigned-off-by Fabrice Fontaine <[email protected]>---
Changes v1 -> v2 (after review of Yann E. Morin):
- Disable tests instead of adding and fixing gdk-pixbuf optional dependency
Quoted 6 lines
I've applied to next. However, your patch to upstream is not really complete, or at least an additional patch is needed: when tests are enabled, their configure.ac script should check if gdk-pixbuf is available. And if it's not available, it should bail out, or disable tests, but not allow the configure process to complete, and then have a build failure when compiling.
You applied only the first patch of the serie and you applied it on next even if it fixes a build failure on master. Was it intentional?
Quoted 8 lines
Thanks!
Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
Best Regards,
Fabrice
PATCH v2 1/4 package/spice: disable tests
Quoted 7 lines · On Tue, 19 Nov 2019 08:47:52 +0100 Fabrice Fontaine…
Quoted 6 lines
I've applied to next. However, your patch to upstream is not really complete, or at least an additional patch is needed: when tests are enabled, their configure.ac script should check if gdk-pixbuf is available. And if it's not available, it should bail out, or disable tests, but not allow the configure process to complete, and then have a build failure when compiling.
You applied only the first patch of the serie
Yes, I didn't get around to applying the rest of the series.
Quoted 1 line
and you applied it on next even if it fixes a build failure on master. Was it intentional?
I was not sure if it was a build failure on master, and the autobuilder failure linked in the commit log
(http://autobuild.buildroot.org/results/96c786f85d35f33508e9c71778043d16b87f72cd)
happened on the next branch. I'll cherry-pick on master.
In such cases (i.e autobuilder failures happening on next, but issue also present in master), it's always good to be explicit in the patch, by adding a small note like this:
"""
balblabla
---
The autobuilder failure linked in the commit log was for next, but the problem also exists on master.
"""
This would really help.
Thomas
Signature · 3 lines
Thomas Petazzoni, CTO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
PATCH v2 1/4 package/spice: disable tests
In reply to Fabrice FontaineQuoted 11 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Fabrice" == Fabrice Fontaine <[email protected]> writes:
By disabling tests, we'll remove the optional gdk-pixbuf dependency Fixes:
- http://autobuild.buildroot.org/results/96c786f85d35f33508e9c71778043d16b87f72cdSigned-off-by Fabrice Fontaine <[email protected]>---
Changes v1 -> v2 (after review of Yann E. Morin):
- Disable tests instead of adding and fixing gdk-pixbuf optional dependency
Committed to 2019.02.x and 2019.08.x, thanks.
Signature · 1 line
Bye, Peter Korsgaard