* Re: [PATCH] u-boot: mkimage: Fix build of u-boot-mkimage
From: Khem Raj @ 2016-11-09 0:43 UTC (permalink / raw)
To: Burton, Ross, Marek Vasut; +Cc: OE-core
In-Reply-To: <CAJTo0LaSLN1eVRH1ju1w5Ba3PLh6d+E0ZwmRimxcN1DjKRk6iQ@mail.gmail.com>
[-- Attachment #1.1: Type: text/plain, Size: 1232 bytes --]
On 11/8/16 4:15 PM, Burton, Ross wrote:
>
> On 7 November 2016 at 18:20, Marek Vasut <marex@denx.de
> <mailto:marex@denx.de>> wrote:
>
> -EXTRA_OEMAKE = 'CROSS_COMPILE="${TARGET_PREFIX}" CC="${CC} ${CFLAGS}
> ${LDFLAGS}" STRIP=true V=1'
> +EXTRA_OEMAKE = 'HOSTCC="${BUILD_CC} ${BUILD_CFLAGS} ${BUILD_LDFLAGS}"'
> +EXTRA_OEMAKE_append_class-target = 'CROSS_COMPILE="${TARGET_PREFIX}"
> CC="${CC} ${CFLAGS} ${LDFLAGS}" STRIP=true V=1'
> +EXTRA_OEMAKE_append_class-native = 'CC="${BUILD_CC} ${BUILD_CFLAGS}
> ${BUILD_LDFLAGS}" STRIP=true V=1'
> +EXTRA_OEMAKE_append_class-nativesdk = 'CC="${BUILD_CC} ${BUILD_CFLAGS}
> ${BUILD_LDFLAGS}" STRIP=true V=1'
>
>
> You're going to start hating u-boot...
>
> Fails on the autobuilder if you're building for a 32-bit x86 target on a
> 64-bit x86 host:
>
> ERROR: QA Issue: Architecture did not match (x86-64, expected x86) on
> /work/i586-poky-linux-musl/u-boot-mkimage/1_2016.09.01-r0/packages-split/u-boot-mkimage/usr/bin/uboot-mkimage
> [arch]
>
> I'm guessing that it is using the host compiler instead of the cross compiler.
isnt mkimage a host tool ? why are we building it for target ?
>
> Ross
>
>
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 211 bytes --]
^ permalink raw reply
* Re: [PATCH] u-boot: mkimage: Fix build of u-boot-mkimage
From: Burton, Ross @ 2016-11-09 0:15 UTC (permalink / raw)
To: Marek Vasut; +Cc: OE-core
In-Reply-To: <20161107182006.4207-1-marex@denx.de>
[-- Attachment #1: Type: text/plain, Size: 1013 bytes --]
On 7 November 2016 at 18:20, Marek Vasut <marex@denx.de> wrote:
> -EXTRA_OEMAKE = 'CROSS_COMPILE="${TARGET_PREFIX}" CC="${CC} ${CFLAGS}
> ${LDFLAGS}" STRIP=true V=1'
> +EXTRA_OEMAKE = 'HOSTCC="${BUILD_CC} ${BUILD_CFLAGS} ${BUILD_LDFLAGS}"'
> +EXTRA_OEMAKE_append_class-target = 'CROSS_COMPILE="${TARGET_PREFIX}"
> CC="${CC} ${CFLAGS} ${LDFLAGS}" STRIP=true V=1'
> +EXTRA_OEMAKE_append_class-native = 'CC="${BUILD_CC} ${BUILD_CFLAGS}
> ${BUILD_LDFLAGS}" STRIP=true V=1'
> +EXTRA_OEMAKE_append_class-nativesdk = 'CC="${BUILD_CC} ${BUILD_CFLAGS}
> ${BUILD_LDFLAGS}" STRIP=true V=1'
>
You're going to start hating u-boot...
Fails on the autobuilder if you're building for a 32-bit x86 target on a
64-bit x86 host:
ERROR: QA Issue: Architecture did not match (x86-64, expected x86) on
/work/i586-poky-linux-musl/u-boot-mkimage/1_2016.09.01-r0/packages-split/u-boot-mkimage/usr/bin/uboot-mkimage
[arch]
I'm guessing that it is using the host compiler instead of the cross
compiler.
Ross
[-- Attachment #2: Type: text/html, Size: 1677 bytes --]
^ permalink raw reply
* Re: [Patch v2 12/14] ltp: Reduce duplication in MIPS variants.
From: Burton, Ross @ 2016-11-09 0:13 UTC (permalink / raw)
To: Zubair Lutfullah Kakakhel; +Cc: OE-core
In-Reply-To: <20161108171235.41837-13-Zubair.Kakakhel@imgtec.com>
[-- Attachment #1: Type: text/plain, Size: 324 bytes --]
On 8 November 2016 at 17:12, Zubair Lutfullah Kakakhel <
Zubair.Kakakhel@imgtec.com> wrote:
> Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
> variable is defined
>
I'm guessing that this is the cause of this error for qemumips64:
http://errors.yoctoproject.org/Errors/Details/99744/
Ross
[-- Attachment #2: Type: text/html, Size: 959 bytes --]
^ permalink raw reply
* [PATCH 2/2] systemtap: fix native linking on recent Ubuntu
From: Ross Burton @ 2016-11-09 0:10 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <1478650244-7002-1-git-send-email-ross.burton@intel.com>
The latest Ubuntu uses yet more aggressive hardening options, which causes the
unconventional build order used by systemtap to fail.
[ YOCTO #10521 ]
Signed-off-by: Ross Burton <ross.burton@intel.com>
---
.../systemtap/systemtap/fix-monitor-linking.patch | 41 ++++++++++++++++++++++
meta/recipes-kernel/systemtap/systemtap_git.inc | 1 +
2 files changed, 42 insertions(+)
create mode 100644 meta/recipes-kernel/systemtap/systemtap/fix-monitor-linking.patch
diff --git a/meta/recipes-kernel/systemtap/systemtap/fix-monitor-linking.patch b/meta/recipes-kernel/systemtap/systemtap/fix-monitor-linking.patch
new file mode 100644
index 0000000..2bf0742a1
--- /dev/null
+++ b/meta/recipes-kernel/systemtap/systemtap/fix-monitor-linking.patch
@@ -0,0 +1,41 @@
+Upstream-Status: Submitted
+Signed-off-by: Ross Burton <ross.burton@intel.com>
+
+From d940aa0079c253b958cf9158e9ec7922ecf464f9 Mon Sep 17 00:00:00 2001
+From: Ross Burton <ross.burton@intel.com>
+Date: Wed, 9 Nov 2016 00:00:48 +0000
+Subject: [PATCH] staprun: fix linking if monitor is enabled
+
+If the monitor is enabled, monitor.c needs to link against json-c and ncurses.
+Instead of adding these linker flags to AM_CFLAGS (global CC flags) add them to
+stapio_LDADD.
+
+Apart from being the "right" thing to do, this fixes build failures on systems
+such as current Ubuntu (which defaults to using --as-needed and similar) where
+link order is important.
+---
+ staprun/Makefile.am | 7 +++----
+ 1 file changed, 3 insertions(+), 4 deletions(-)
+
+diff --git a/staprun/Makefile.am b/staprun/Makefile.am
+index 63d4088..66b3e21 100644
+--- a/staprun/Makefile.am
++++ b/staprun/Makefile.am
+@@ -42,12 +42,11 @@ staprun_CXXFLAGS += $(nss_CFLAGS)
+ staprun_LDADD += $(nss_LIBS)
+ endif
+
+-if HAVE_MONITOR_LIBS
+-AM_CFLAGS += $(jsonc_LIBS) $(ncurses_LIBS)
+-endif
+-
+ stapio_SOURCES = stapio.c mainloop.c common.c ctl.c relay.c relay_old.c monitor.c
+ stapio_LDADD = libstrfloctime.a -lpthread
++if HAVE_MONITOR_LIBS
++stapio_LDADD += $(jsonc_LIBS) $(ncurses_LIBS)
++endif
+
+ man_MANS = staprun.8
+
+--
+2.8.1
diff --git a/meta/recipes-kernel/systemtap/systemtap_git.inc b/meta/recipes-kernel/systemtap/systemtap_git.inc
index 34cf227..a0f0e68 100644
--- a/meta/recipes-kernel/systemtap/systemtap_git.inc
+++ b/meta/recipes-kernel/systemtap/systemtap_git.inc
@@ -10,6 +10,7 @@ SRC_URI = "git://sourceware.org/git/systemtap.git \
file://x32_abi_time.patch \
file://monitor-option.patch \
file://no-msgfmt-check.patch \
+ file://fix-monitor-linking.patch \
"
# systemtap doesn't support mips
--
2.8.1
^ permalink raw reply related
* [PATCH 1/2] systemtap: remove explicit msgfmt check
From: Ross Burton @ 2016-11-09 0:10 UTC (permalink / raw)
To: openembedded-core
Passing --disable-nls should be enough to disable the requirement for a full
gettext to be present, but the upstream configure explicitly checks for msgfmt
even if it isn't going to be used. To avoid having to depend on gettext-native,
patch this check out.
Signed-off-by: Ross Burton <ross.burton@intel.com>
---
.../systemtap/systemtap/no-msgfmt-check.patch | 15 +++++++++++++++
meta/recipes-kernel/systemtap/systemtap_git.inc | 1 +
2 files changed, 16 insertions(+)
create mode 100644 meta/recipes-kernel/systemtap/systemtap/no-msgfmt-check.patch
diff --git a/meta/recipes-kernel/systemtap/systemtap/no-msgfmt-check.patch b/meta/recipes-kernel/systemtap/systemtap/no-msgfmt-check.patch
new file mode 100644
index 0000000..471e62c
--- /dev/null
+++ b/meta/recipes-kernel/systemtap/systemtap/no-msgfmt-check.patch
@@ -0,0 +1,15 @@
+There is no need to explicitly check that msgfmt was found as the gettext macros
+handle this for us if NLS is enabled.
+
+Upstream-Status: Pending
+Signed-off-by: Ross Burton <ross.burton@intel.com>
+
+diff --git a/configure.ac b/configure.ac
+index e56079a..284bbe0 100644
+--- a/configure.ac
++++ b/configure.ac
+@@ -39,4 +38,0 @@ AM_GNU_GETTEXT_VERSION([0.19.4])
+-if test "x$GMSGFMT" = "x:"; then
+- AC_MSG_ERROR([missing gnu /usr/bin/msgfmt])
+-fi
+-
diff --git a/meta/recipes-kernel/systemtap/systemtap_git.inc b/meta/recipes-kernel/systemtap/systemtap_git.inc
index 304eb99..34cf227 100644
--- a/meta/recipes-kernel/systemtap/systemtap_git.inc
+++ b/meta/recipes-kernel/systemtap/systemtap_git.inc
@@ -9,6 +9,7 @@ SRC_URI = "git://sourceware.org/git/systemtap.git \
file://configure-allow-to-disable-libvirt.patch \
file://x32_abi_time.patch \
file://monitor-option.patch \
+ file://no-msgfmt-check.patch \
"
# systemtap doesn't support mips
--
2.8.1
^ permalink raw reply related
* [PATCH] libyaml: Enable nativesdk bake
From: Brad Bishop @ 2016-11-08 20:45 UTC (permalink / raw)
To: openembedded-core
Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com>
---
meta/recipes-support/libyaml/libyaml_0.1.7.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/recipes-support/libyaml/libyaml_0.1.7.bb b/meta/recipes-support/libyaml/libyaml_0.1.7.bb
index 5c422ef..b167f03 100644
--- a/meta/recipes-support/libyaml/libyaml_0.1.7.bb
+++ b/meta/recipes-support/libyaml/libyaml_0.1.7.bb
@@ -17,4 +17,4 @@ S = "${WORKDIR}/yaml-${PV}"
inherit autotools
-BBCLASSEXTEND = "native"
+BBCLASSEXTEND = "native nativesdk"
--
1.8.3.1
^ permalink raw reply related
* Re: [PATCH] mkefidisk.sh: add deprecation warning to the output
From: Randy Witt @ 2016-11-08 20:44 UTC (permalink / raw)
To: Khem Raj, ed.bartosh; +Cc: Hawley, John, openembedded-core
In-Reply-To: <32b3605d-9c79-d1be-c6cc-80309cad7a1a@gmail.com>
>>> We should also document, the wic steps in wiki pages e.g. http://wiki.minnowboard.org/Yocto_Project
>> It's already documented in README.hardware: https://git.yoctoproject.org/cgit/cgit.cgi/poky/tree/README.hardware
>>
>> And in Yocto manual:
>> http://www.yoctoproject.org/docs/2.2/mega-manual/mega-manual.html#building-an-image-for-hardware
>
> Thanks, now can you also nudge the minnowboard.org to do the same ?
>
Pinging you John. :)
>>
>> --
>> Regards,
>> Ed
>>
^ permalink raw reply
* Re: compile error for handwritten "perl-RPM2", 'RPM_GNUC_NULL_TERMINATED' error
From: Robert P. J. Day @ 2016-11-08 19:16 UTC (permalink / raw)
To: Tim Orling; +Cc: OE Core mailing list
In-Reply-To: <CANx9H-Bfd6hKhcKhNbx8DVYBUTRPQbpxM7naw+aEW5dNc+vhKw@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 9049 bytes --]
On Tue, 8 Nov 2016, Tim Orling wrote:
> NOTE: by convention, OE perl recipes follow debian naming so this
> should be librpm2-perl_1.3.bb. Obviously, in your own layer you are
> free to do what you want, but it will require the name change if you
> want it to go into meta-perl.
i had noticed the naming convention after i started writing those
recipes, it won't be hard to go back and rename. i hadn't planned
submitting anything upstream, but maybe after all that work, i might
as well, so no one else has to go through the same effort.
> For dependencies, since you are looking at CentOS rpms, you can look
> at, e.g. [0] to see the DEPENDS (BuildRequires:) and RDEPENDS
> (Requires:). This would have shown you the need for rpm-devel.
true enough ... it didn't take long to realize i needed the rpm
devel package installed for this.
> I also use metacpan, e.g. [1], to look for dependencies in perl
> modules. It also has a very slick ElasticSearch JSON API that you
> can use to automatically query data [2].
i'd noticed that as well ... still playing with it.
> To pass in the extra variables, for 'inherit cpan' you probably need
> to set EXTRA_CPANFLAGS [3] or for 'inherit cpan_build'
> EXTRA_CPAN_BUILD_FLAGS [4], e.g. [5].
except i'm not sure what those variables are for ... it *appears*
they're for Perl-related settings, not simple toolchain-related
settings. i will have to read further to understand precisely what the
purpose of those variables is.
> You might want to file a documentation bug for the next release?
not sure what this means, i'm still trying to debug this issue
before i try to submit anything official. :-)
> [0] http://pkgs.fedoraproject.org/cgit/rpms/perl-RPM2.git/tree/perl-RPM2.spec?h=epel7
> [1] https://metacpan.org/pod/RPM2
> [2] https://github.com/metacpan/metacpan-api/blob/master/docs/API-docs.md
> [3] http://git.yoctoproject.org/cgit/cgit.cgi/poky/tree/meta/classes/cpan.bbclass#n20
> [4] http://git.yoctoproject.org/cgit/cgit.cgi/poky/tree/meta/classes/cpan_build.bbclass#n21
> [5] http://git.openembedded.org/meta-openembedded/tree/meta-perl/recipes-perl/libxml/libxml-libxml-perl_2.0121.bb#n37
>
> On Tue, Nov 8, 2016 at 3:17 AM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
>
> replying to my earlier post with additional info ... man, i want to
> debug this ...
>
> On Tue, 8 Nov 2016, Robert P. J. Day wrote:
>
> > following on my earlier post, here's the specific issue i'm
> > tackling. having found no existing OE recipe equivalent to a centos
> > "perl-RPM2" RPM, i wrote my own. here's the info about that RH RPM on
> > my fedora system, so i know what the final result should look like:
> >
> > $ rpm -ql perl-RPM2
> > /usr/lib64/perl5/vendor_perl/RPM2.pm
> > /usr/lib64/perl5/vendor_perl/auto/RPM2
> > /usr/lib64/perl5/vendor_perl/auto/RPM2/RPM2.so
> > /usr/share/doc/perl-RPM2
> > /usr/share/doc/perl-RPM2/Changes
> > /usr/share/doc/perl-RPM2/README
> > /usr/share/man/man3/RPM2.3pm.gz
> > $
> >
> > so, as before, track down the source at cpan:
> >
> > https://metacpan.org/pod/RPM2
> >
> > and write a boilerplate recipe, "perl-rpm2", containing the following
> > line since it's an older style recipe:
> >
> > inherit cpan_build
> >
> > fetching works fine:
> >
> > $ bitbake -c fetchall perl-rpm2
> >
> > but here's the problem:
> >
> > $ bitbake perl-rpm2
> >
> > ... snip ...
> >
> > | DEBUG: Executing shell function do_compile
> > | Copying lib/RPM2.pm -> blib/lib/RPM2.pm
> > | lib/RPM2.xs -> lib/RPM2.c
> > | powerpc-poky-linux-gcc -m32 -mhard-float -mcpu=7400 -mno-spe
> > --sysroot=/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc
> > -I/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/usr/lib/perl/5.22.1/CORE
> > -DVERSION="1.3" -DXS_VERSION="1.3" -fPIC -DRPM2_API=5004
> > -I/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/usr/include/rpm
> > -c -O2 -pipe -g -feliminate-unused-debug-types
> >
> -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/tmp/work/ppc7400-poky-linux/perl/5.22.1-r0=/usr/src/debug/perl/5.22.1-r0
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/x86_64-linux=
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc=
> > -DDEBIAN -D_REENTRANT -D_GNU_SOURCE -fwrapv -fno-strict-aliasing -pipe
> > -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -D_FORTIFY_SOURCE=2 -O2
> > -pipe -g -feliminate-unused-debug-types
> >
> -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/tmp/work/ppc7400-poky-linux/perl-rpm2/1.3-r0=/usr/src/debug/perl-rpm2/1.3-r0
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/x86_64-linux=
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc=
> > -O2 -o lib/RPM2.o lib/RPM2.c
> > | In file included from
> > /home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/usr/include/rpm/rpmcli.h:9:0,
> > | from lib/RPM2.xs:3:
> > |
> > /home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/usr/include/rpm/rpmmacro.h:
> > In function 'rpmExpand':
> > |
> > /home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/usr/include/rpm/rpmmacro.h:249:2:
> > error: expected declaration specifiers before
> > 'RPM_GNUC_NULL_TERMINATED'
> > | RPM_GNUC_NULL_TERMINATED
> > | ^~~~~~~~~~~~~~~~~~~~~~~~
> > |
> > /home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/usr/include/rpm/rpmmacro.h:260:2:
> > error: expected '=', ',', ';', 'asm' or '__attribute__' before
> > 'RPM_GNUC_NULL_TERMINATED'
> > | RPM_GNUC_NULL_TERMINATED
> > | ^~~~~~~~~~~~~~~~~~~~~~~~
> >
> > ... big snip, lots more errors ...
>
> ok, here's my wildly-uneducated record of what is going on:
>
> first, unpacking and configuring perl-RPM2 source (and running XS
> processor) gives me the source file "lib/RPM2.c" (which is what is
> generating the first build error above), which includes the header
> file:
>
> #include <rpm/rpmcli.h>
>
> next, wander over to the qemuppc sysroot, to usr/include/rpm/, to
> check out that header file, which contains (among other things):
>
> #include <popt.h>
> #include <rpmmacro.h> <----- there
> #include <rpmtypes.h>
> #include <rpmtag.h>
> #include <rpmps.h>
> #include <rpmrc.h>
> #include <rpmfi.h> /* XXX rpmfileAttrs */
> #include <rpmts.h> /* XXX rpmdepFlags *
>
> next, in the same directory, examine rpmmacro.h, which contains:
>
> #if defined(_MACRO_INTERNAL)
> #include <rpmiotypes.h>
> ... snip ...
>
> at this point, inclusion is now conditional on _MACRO_INTERNAL, so
> move on to rpmiotypes.h, which contains:
>
> #include <rpmutil.h>
>
> and finally visit that header file in the qemuppc sysroot to find:
>
> #if __GNUC__ >= 4
> #define RPM_GNUC_NULL_TERMINATED __attribute__((__sentinel__))
> #else
> #define RPM_GNUC_NULL_TERMINATED
> #endif
>
> where we find the definition of that macro.
>
> now, is it my imagination, or could a lot of this compile-time error
> be resolved by simply defining the macro "_MACRO_INTERNAL" for the
> build? and if so, what is the proper way to add that setting to the
> compile step for this recipe?
>
> would i use the standard EXTRA_* variables? or are there
> perl-specific variables one would use when build CPAN recipes? or am i
> totally off-track here?
>
> rday
>
> --
>
> ========================================================================
> Robert P. J. Day Ottawa, Ontario, CANADA
> http://crashcourse.ca
>
> Twitter: http://twitter.com/rpjday
> LinkedIn: http://ca.linkedin.com/in/rpjday
> ========================================================================
>
>
>
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core
>
>
>
>
--
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
^ permalink raw reply
* Re: compile error for handwritten "perl-RPM2", 'RPM_GNUC_NULL_TERMINATED' error
From: Tim Orling @ 2016-11-08 18:37 UTC (permalink / raw)
To: Robert P. J. Day; +Cc: OE Core mailing list
In-Reply-To: <alpine.LFD.2.20.1611080556150.4555@ca624034.mitel.com>
[-- Attachment #1: Type: text/plain, Size: 7328 bytes --]
NOTE: by convention, OE perl recipes follow debian naming so this should be
librpm2-perl_1.3.bb
Obviously, in your own layer you are free to do what you want, but it will
require the name change if you want it to go into meta-perl.
For dependencies, since you are looking at CentOS rpms, you can look at,
e.g. [0] to see the DEPENDS (BuildRequires:) and RDEPENDS (Requires:).
This would have shown you the need for rpm-devel.
I also use metacpan, e.g. [1], to look for dependencies in perl modules. It
also has a very slick ElasticSearch JSON API that you can use to
automatically query data [2].
To pass in the extra variables, for 'inherit cpan' you probably need to set
EXTRA_CPANFLAGS [3] or for 'inherit cpan_build' EXTRA_CPAN_BUILD_FLAGS [4],
e.g. [5].
You might want to file a documentation bug for the next release?
[0]
http://pkgs.fedoraproject.org/cgit/rpms/perl-RPM2.git/tree/perl-RPM2.spec?h=epel7
[1] https://metacpan.org/pod/RPM2
[2] https://github.com/metacpan/metacpan-api/blob/master/docs/API-docs.md
[3]
http://git.yoctoproject.org/cgit/cgit.cgi/poky/tree/meta/classes/cpan.bbclass#n20
[4]
http://git.yoctoproject.org/cgit/cgit.cgi/poky/tree/meta/classes/cpan_build.bbclass#n21
[5]
http://git.openembedded.org/meta-openembedded/tree/meta-perl/recipes-perl/libxml/libxml-libxml-perl_2.0121.bb#n37
On Tue, Nov 8, 2016 at 3:17 AM, Robert P. J. Day <rpjday@crashcourse.ca>
wrote:
>
> replying to my earlier post with additional info ... man, i want to
> debug this ...
>
> On Tue, 8 Nov 2016, Robert P. J. Day wrote:
>
> > following on my earlier post, here's the specific issue i'm
> > tackling. having found no existing OE recipe equivalent to a centos
> > "perl-RPM2" RPM, i wrote my own. here's the info about that RH RPM on
> > my fedora system, so i know what the final result should look like:
> >
> > $ rpm -ql perl-RPM2
> > /usr/lib64/perl5/vendor_perl/RPM2.pm
> > /usr/lib64/perl5/vendor_perl/auto/RPM2
> > /usr/lib64/perl5/vendor_perl/auto/RPM2/RPM2.so
> > /usr/share/doc/perl-RPM2
> > /usr/share/doc/perl-RPM2/Changes
> > /usr/share/doc/perl-RPM2/README
> > /usr/share/man/man3/RPM2.3pm.gz
> > $
> >
> > so, as before, track down the source at cpan:
> >
> > https://metacpan.org/pod/RPM2
> >
> > and write a boilerplate recipe, "perl-rpm2", containing the following
> > line since it's an older style recipe:
> >
> > inherit cpan_build
> >
> > fetching works fine:
> >
> > $ bitbake -c fetchall perl-rpm2
> >
> > but here's the problem:
> >
> > $ bitbake perl-rpm2
> >
> > ... snip ...
> >
> > | DEBUG: Executing shell function do_compile
> > | Copying lib/RPM2.pm -> blib/lib/RPM2.pm
> > | lib/RPM2.xs -> lib/RPM2.c
> > | powerpc-poky-linux-gcc -m32 -mhard-float -mcpu=7400 -mno-spe
> > --sysroot=/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc
> > -I/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/
> usr/lib/perl/5.22.1/CORE
> > -DVERSION="1.3" -DXS_VERSION="1.3" -fPIC -DRPM2_API=5004
> > -I/home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/
> usr/include/rpm
> > -c -O2 -pipe -g -feliminate-unused-debug-types
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/
> tmp/work/ppc7400-poky-linux/perl/5.22.1-r0=/usr/src/debug/perl/5.22.1-r0
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/
> tmp/sysroots/x86_64-linux=
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/
> tmp/sysroots/qemuppc=
> > -DDEBIAN -D_REENTRANT -D_GNU_SOURCE -fwrapv -fno-strict-aliasing -pipe
> > -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -D_FORTIFY_SOURCE=2 -O2
> > -pipe -g -feliminate-unused-debug-types
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/
> tmp/work/ppc7400-poky-linux/perl-rpm2/1.3-r0=/usr/src/
> debug/perl-rpm2/1.3-r0
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/
> tmp/sysroots/x86_64-linux=
> > -fdebug-prefix-map=/home/rpjday/oe/builds/msm_qemuppc/
> tmp/sysroots/qemuppc=
> > -O2 -o lib/RPM2.o lib/RPM2.c
> > | In file included from
> > /home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/
> usr/include/rpm/rpmcli.h:9:0,
> > | from lib/RPM2.xs:3:
> > |
> > /home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/
> usr/include/rpm/rpmmacro.h:
> > In function 'rpmExpand':
> > |
> > /home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/
> usr/include/rpm/rpmmacro.h:249:2:
> > error: expected declaration specifiers before
> > 'RPM_GNUC_NULL_TERMINATED'
> > | RPM_GNUC_NULL_TERMINATED
> > | ^~~~~~~~~~~~~~~~~~~~~~~~
> > |
> > /home/rpjday/oe/builds/msm_qemuppc/tmp/sysroots/qemuppc/
> usr/include/rpm/rpmmacro.h:260:2:
> > error: expected '=', ',', ';', 'asm' or '__attribute__' before
> > 'RPM_GNUC_NULL_TERMINATED'
> > | RPM_GNUC_NULL_TERMINATED
> > | ^~~~~~~~~~~~~~~~~~~~~~~~
> >
> > ... big snip, lots more errors ...
>
> ok, here's my wildly-uneducated record of what is going on:
>
> first, unpacking and configuring perl-RPM2 source (and running XS
> processor) gives me the source file "lib/RPM2.c" (which is what is
> generating the first build error above), which includes the header
> file:
>
> #include <rpm/rpmcli.h>
>
> next, wander over to the qemuppc sysroot, to usr/include/rpm/, to
> check out that header file, which contains (among other things):
>
> #include <popt.h>
> #include <rpmmacro.h> <----- there
> #include <rpmtypes.h>
> #include <rpmtag.h>
> #include <rpmps.h>
> #include <rpmrc.h>
> #include <rpmfi.h> /* XXX rpmfileAttrs */
> #include <rpmts.h> /* XXX rpmdepFlags *
>
> next, in the same directory, examine rpmmacro.h, which contains:
>
> #if defined(_MACRO_INTERNAL)
> #include <rpmiotypes.h>
> ... snip ...
>
> at this point, inclusion is now conditional on _MACRO_INTERNAL, so
> move on to rpmiotypes.h, which contains:
>
> #include <rpmutil.h>
>
> and finally visit that header file in the qemuppc sysroot to find:
>
> #if __GNUC__ >= 4
> #define RPM_GNUC_NULL_TERMINATED __attribute__((__sentinel__))
> #else
> #define RPM_GNUC_NULL_TERMINATED
> #endif
>
> where we find the definition of that macro.
>
> now, is it my imagination, or could a lot of this compile-time error
> be resolved by simply defining the macro "_MACRO_INTERNAL" for the
> build? and if so, what is the proper way to add that setting to the
> compile step for this recipe?
>
> would i use the standard EXTRA_* variables? or are there
> perl-specific variables one would use when build CPAN recipes? or am i
> totally off-track here?
>
> rday
>
> --
>
> ========================================================================
> Robert P. J. Day Ottawa, Ontario, CANADA
> http://crashcourse.ca
>
> Twitter: http://twitter.com/rpjday
> LinkedIn: http://ca.linkedin.com/in/rpjday
> ========================================================================
>
>
>
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core
>
[-- Attachment #2: Type: text/html, Size: 10037 bytes --]
^ permalink raw reply
* [PATCH] webkitgtk: drop patch 0001-WebKitMacros-Append-to-I-and-not-to-isystem.patch
From: Carlos Alberto Lopez Perez @ 2016-11-08 17:52 UTC (permalink / raw)
To: openembedded-core
* This patch is not longer needed. Upstream has fixed this issue in:
https://trac.webkit.org/changeset/205672 which is already included
in WebKitGTK+ >= 2.14.0
Signed-off-by: Carlos Alberto Lopez Perez <clopez@igalia.com>
---
...bKitMacros-Append-to-I-and-not-to-isystem.patch | 181 ---------------------
meta/recipes-sato/webkit/webkitgtk_2.14.1.bb | 1 -
2 files changed, 182 deletions(-)
delete mode 100644 meta/recipes-sato/webkit/files/0001-WebKitMacros-Append-to-I-and-not-to-isystem.patch
diff --git a/meta/recipes-sato/webkit/files/0001-WebKitMacros-Append-to-I-and-not-to-isystem.patch b/meta/recipes-sato/webkit/files/0001-WebKitMacros-Append-to-I-and-not-to-isystem.patch
deleted file mode 100644
index ef209c8..0000000
--- a/meta/recipes-sato/webkit/files/0001-WebKitMacros-Append-to-I-and-not-to-isystem.patch
+++ /dev/null
@@ -1,181 +0,0 @@
-From 20ee11dd188e1538f8cdd17a289dc6f9c63a011e Mon Sep 17 00:00:00 2001
-From: Khem Raj <raj.khem@gmail.com>
-Date: Sun, 17 Apr 2016 12:35:41 -0700
-Subject: [PATCH] WebKitMacros: Append to -I and not to -isystem
-
-gcc-6 has now introduced stdlib.h in libstdc++ for better
-compliance and its including the C library stdlib.h using
-include_next which is sensitive to order of system header
-include paths. Its infact better to not tinker with the
-system header include paths at all. Since adding /usr/include
-to -system is redundant and compiler knows about it moreover
-now with gcc6 it interferes with compiler's functioning
-and ends up with compile errors e.g.
-
-/usr/include/c++/6.0.0/cstdlib:75:25: fatal error: stdlib.h: No such file or directory
-
-Signed-off-by: Khem Raj <raj.khem@gmail.com>
-Upstream-Status: Pending
----
- Source/JavaScriptCore/shell/CMakeLists.txt | 2 +-
- Source/WebCore/PlatformGTK.cmake | 6 +++---
- Source/WebKit2/PlatformGTK.cmake | 2 +-
- Source/cmake/WebKitMacros.cmake | 2 +-
- Tools/DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt | 2 +-
- Tools/ImageDiff/CMakeLists.txt | 2 +-
- Tools/MiniBrowser/gtk/CMakeLists.txt | 2 +-
- Tools/TestWebKitAPI/PlatformGTK.cmake | 2 +-
- Tools/TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt | 2 +-
- Tools/WebKitTestRunner/CMakeLists.txt | 2 +-
- 10 files changed, 12 insertions(+), 12 deletions(-)
-
-diff --git a/Source/JavaScriptCore/shell/CMakeLists.txt b/Source/JavaScriptCore/shell/CMakeLists.txt
-index 155c797..80fe22b 100644
---- a/Source/JavaScriptCore/shell/CMakeLists.txt
-+++ b/Source/JavaScriptCore/shell/CMakeLists.txt
-@@ -20,7 +20,7 @@ WEBKIT_INCLUDE_CONFIG_FILES_IF_EXISTS()
-
- WEBKIT_WRAP_SOURCELIST(${JSC_SOURCES})
- include_directories(./ ${JavaScriptCore_INCLUDE_DIRECTORIES})
--include_directories(SYSTEM ${JavaScriptCore_SYSTEM_INCLUDE_DIRECTORIES})
-+include_directories(${JavaScriptCore_SYSTEM_INCLUDE_DIRECTORIES})
- add_executable(jsc ${JSC_SOURCES})
- target_link_libraries(jsc ${JSC_LIBRARIES})
-
-diff --git a/Source/WebCore/PlatformGTK.cmake b/Source/WebCore/PlatformGTK.cmake
-index 567bd79..1fabea8 100644
---- a/Source/WebCore/PlatformGTK.cmake
-+++ b/Source/WebCore/PlatformGTK.cmake
-@@ -340,7 +340,7 @@ if (ENABLE_PLUGIN_PROCESS_GTK2)
- ${GTK2_INCLUDE_DIRS}
- ${GDK2_INCLUDE_DIRS}
- )
-- target_include_directories(WebCorePlatformGTK2 SYSTEM PRIVATE
-+ target_include_directories(WebCorePlatformGTK2 PRIVATE
- ${WebCore_SYSTEM_INCLUDE_DIRECTORIES}
- )
- target_link_libraries(WebCorePlatformGTK2
-@@ -365,7 +365,7 @@ WEBKIT_SET_EXTRA_COMPILER_FLAGS(WebCorePlatformGTK)
- target_include_directories(WebCorePlatformGTK PRIVATE
- ${WebCore_INCLUDE_DIRECTORIES}
- )
--target_include_directories(WebCorePlatformGTK SYSTEM PRIVATE
-+target_include_directories(WebCorePlatformGTK PRIVATE
- ${WebCore_SYSTEM_INCLUDE_DIRECTORIES}
- ${GTK_INCLUDE_DIRS}
- ${GDK_INCLUDE_DIRS}
-@@ -383,7 +383,7 @@ include_directories(
- "${DERIVED_SOURCES_GOBJECT_DOM_BINDINGS_DIR}"
- )
-
--include_directories(SYSTEM
-+include_directories(
- ${WebCore_SYSTEM_INCLUDE_DIRECTORIES}
- )
-
-diff --git a/Source/WebKit2/PlatformGTK.cmake b/Source/WebKit2/PlatformGTK.cmake
-index e4805a4..c57df5d 100644
---- a/Source/WebKit2/PlatformGTK.cmake
-+++ b/Source/WebKit2/PlatformGTK.cmake
-@@ -822,7 +822,7 @@ if (ENABLE_PLUGIN_PROCESS_GTK2)
- target_include_directories(WebKitPluginProcess2 PRIVATE
- ${WebKit2CommonIncludeDirectories}
- )
-- target_include_directories(WebKitPluginProcess2 SYSTEM PRIVATE
-+ target_include_directories(WebKitPluginProcess2 PRIVATE
- ${WebKit2CommonSystemIncludeDirectories}
- ${GTK2_INCLUDE_DIRS}
- ${GDK2_INCLUDE_DIRS}
-diff --git a/Source/cmake/WebKitMacros.cmake b/Source/cmake/WebKitMacros.cmake
-index 043e78e..8b3b642 100644
---- a/Source/cmake/WebKitMacros.cmake
-+++ b/Source/cmake/WebKitMacros.cmake
-@@ -224,7 +224,7 @@ endmacro()
-
- macro(WEBKIT_FRAMEWORK _target)
- include_directories(${${_target}_INCLUDE_DIRECTORIES})
-- include_directories(SYSTEM ${${_target}_SYSTEM_INCLUDE_DIRECTORIES})
-+ include_directories(${${_target}_SYSTEM_INCLUDE_DIRECTORIES})
- add_library(${_target} ${${_target}_LIBRARY_TYPE}
- ${${_target}_HEADERS}
- ${${_target}_SOURCES}
-diff --git a/Tools/DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt b/Tools/DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt
-index c431667..6dff244 100644
---- a/Tools/DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt
-+++ b/Tools/DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt
-@@ -42,7 +42,7 @@ set(WebKitTestNetscapePlugin_SYSTEM_INCLUDE_DIRECTORIES
- )
-
- include_directories(${WebKitTestNetscapePlugin_INCLUDE_DIRECTORIES})
--include_directories(SYSTEM ${WebKitTestNetscapePlugin_SYSTEM_INCLUDE_DIRECTORIES})
-+include_directories(${WebKitTestNetscapePlugin_SYSTEM_INCLUDE_DIRECTORIES})
-
- set(WebKitTestNetscapePlugin_LIBRARIES
- ${X11_LIBRARIES}
-diff --git a/Tools/ImageDiff/CMakeLists.txt b/Tools/ImageDiff/CMakeLists.txt
-index b15443f..87e03bf 100644
---- a/Tools/ImageDiff/CMakeLists.txt
-+++ b/Tools/ImageDiff/CMakeLists.txt
-@@ -14,6 +14,6 @@ set(IMAGE_DIFF_LIBRARIES
- WEBKIT_INCLUDE_CONFIG_FILES_IF_EXISTS()
-
- include_directories(${IMAGE_DIFF_INCLUDE_DIRECTORIES})
--include_directories(SYSTEM ${IMAGE_DIFF_SYSTEM_INCLUDE_DIRECTORIES})
-+include_directories(${IMAGE_DIFF_SYSTEM_INCLUDE_DIRECTORIES})
- add_executable(ImageDiff ${IMAGE_DIFF_SOURCES})
- target_link_libraries(ImageDiff ${IMAGE_DIFF_LIBRARIES})
-diff --git a/Tools/MiniBrowser/gtk/CMakeLists.txt b/Tools/MiniBrowser/gtk/CMakeLists.txt
-index 0704bc6..619e5a5 100644
---- a/Tools/MiniBrowser/gtk/CMakeLists.txt
-+++ b/Tools/MiniBrowser/gtk/CMakeLists.txt
-@@ -58,7 +58,7 @@ endif ()
- add_definitions(-DGDK_VERSION_MIN_REQUIRED=GDK_VERSION_3_6)
-
- include_directories(${MiniBrowser_INCLUDE_DIRECTORIES})
--include_directories(SYSTEM ${MiniBrowser_SYSTEM_INCLUDE_DIRECTORIES})
-+include_directories(${MiniBrowser_SYSTEM_INCLUDE_DIRECTORIES})
- add_executable(MiniBrowser ${MiniBrowser_SOURCES})
- target_link_libraries(MiniBrowser ${MiniBrowser_LIBRARIES})
-
-diff --git a/Tools/TestWebKitAPI/PlatformGTK.cmake b/Tools/TestWebKitAPI/PlatformGTK.cmake
-index 7552cc2..2eb44d5 100644
---- a/Tools/TestWebKitAPI/PlatformGTK.cmake
-+++ b/Tools/TestWebKitAPI/PlatformGTK.cmake
-@@ -20,7 +20,7 @@ include_directories(
- ${WEBKIT2_DIR}/UIProcess/API/gtk
- )
-
--include_directories(SYSTEM
-+include_directories(
- ${GDK3_INCLUDE_DIRS}
- ${GLIB_INCLUDE_DIRS}
- ${GTK3_INCLUDE_DIRS}
-diff --git a/Tools/TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt b/Tools/TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt
-index b0b4739..434e4ca 100644
---- a/Tools/TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt
-+++ b/Tools/TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt
-@@ -23,7 +23,7 @@ include_directories(
- ${TOOLS_DIR}/TestWebKitAPI/gtk/WebKit2Gtk
- )
-
--include_directories(SYSTEM
-+include_directories(
- ${ATSPI_INCLUDE_DIRS}
- ${GLIB_INCLUDE_DIRS}
- ${GSTREAMER_INCLUDE_DIRS}
-diff --git a/Tools/WebKitTestRunner/CMakeLists.txt b/Tools/WebKitTestRunner/CMakeLists.txt
-index 7db90f2..a4f917f 100644
---- a/Tools/WebKitTestRunner/CMakeLists.txt
-+++ b/Tools/WebKitTestRunner/CMakeLists.txt
-@@ -116,7 +116,7 @@ GENERATE_BINDINGS(WebKitTestRunner_SOURCES
- WEBKIT_INCLUDE_CONFIG_FILES_IF_EXISTS()
-
- include_directories(${WebKitTestRunner_INCLUDE_DIRECTORIES})
--include_directories(SYSTEM ${WebKitTestRunner_SYSTEM_INCLUDE_DIRECTORIES})
-+include_directories(${WebKitTestRunner_SYSTEM_INCLUDE_DIRECTORIES})
-
- add_library(TestRunnerInjectedBundle SHARED ${WebKitTestRunnerInjectedBundle_SOURCES})
- target_link_libraries(TestRunnerInjectedBundle ${WebKitTestRunner_LIBRARIES})
---
-2.9.3
-
diff --git a/meta/recipes-sato/webkit/webkitgtk_2.14.1.bb b/meta/recipes-sato/webkit/webkitgtk_2.14.1.bb
index a2586de..69c9f11 100644
--- a/meta/recipes-sato/webkit/webkitgtk_2.14.1.bb
+++ b/meta/recipes-sato/webkit/webkitgtk_2.14.1.bb
@@ -18,7 +18,6 @@ SRC_URI = "http://www.webkitgtk.org/releases/${BPN}-${PV}.tar.xz \
file://ppc-musl-fix.patch \
file://0001-Fix-racy-parallel-build-of-WebKit2-4.0.gir.patch \
file://0001-Tweak-gtkdoc-settings-so-that-gtkdoc-generation-work.patch \
- file://0001-WebKitMacros-Append-to-I-and-not-to-isystem.patch \
"
SRC_URI[md5sum] = "8d6c60dc41604d3bbd43165a674c07e5"
--
2.1.4
^ permalink raw reply related
* Re: [Patch v2 00/14] MIPS: Use MACHINEOVERRIDES and reduce duplication
From: Khem Raj @ 2016-11-08 17:44 UTC (permalink / raw)
To: Zubair Lutfullah Kakakhel, openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
[-- Attachment #1.1: Type: text/plain, Size: 2516 bytes --]
On 11/8/16 9:12 AM, Zubair Lutfullah Kakakhel wrote:
> Hi,
>
> This patch series adds a few MACHINEOVERRIDES options in arch-mips
> to reduce duplicate configurations in various recipes.
>
> Further info in
> https://bugzilla.yoctoproject.org/show_bug.cgi?id=10404
>
This series looks good to me.
Thanks for doing this.
> Regards,
> ZubairLK
>
> V1 -> V2
> Added more options for endian independent and big/little endian
> Added another patch for o32 in TUNE_FEATURES for MIPS32R6.
>
> Zubair Lutfullah Kakakhel (14):
> arch-mips: Add o32 in TUNE_FEATURES for MIPS32R6
> arch-mips: Add MACHINEOVERRIDES variables to reduce duplication
> bitbake.conf: Reduce duplication in MIPS variants
> fts: Reduce duplication in MIPS variants.
> glibc: Reduce duplication in MIPS variants
> packagegroup: Reduce duplication in MIPS variants.
> gcc-runtime: Reduce duplication in MIPS variants.
> gdb: Reduce duplication in MIPS variants.
> mmc-utils: Reduce duplication in MIPS variants.
> valgrind: Reduce duplication in MIPS variants.
> ghostscript: Reduce duplication in MIPS variants.
> ltp: Reduce duplication in MIPS variants.
> mdadm: Reduce duplication in MIPS variants.
> webkit: Reduce duplication in MIPS variants.
>
> meta/conf/bitbake.conf | 11 +----------
> meta/conf/machine/include/mips/README | 3 +++
> meta/conf/machine/include/mips/arch-mips.inc | 12 ++++++++++++
> meta/conf/machine/include/tune-mips32r6.inc | 8 ++++----
> meta/recipes-core/fts/fts.bb | 5 +----
> meta/recipes-core/glibc/glibc-ld.inc | 17 ++---------------
> .../recipes-core/packagegroups/packagegroup-core-sdk.bb | 5 +----
> .../packagegroups/packagegroup-core-tools-profile.bb | 10 ++--------
> meta/recipes-devtools/gcc/gcc-runtime.inc | 9 +--------
> meta/recipes-devtools/gdb/gdb-common.inc | 7 +------
> meta/recipes-devtools/mmc/mmc-utils_git.bb | 4 ++--
> meta/recipes-devtools/valgrind/valgrind_3.12.0.bb | 3 +--
> meta/recipes-extended/ghostscript/ghostscript_9.19.bb | 3 +--
> meta/recipes-extended/ltp/ltp_20160126.bb | 2 +-
> meta/recipes-extended/mdadm/mdadm_3.4.bb | 4 ++--
> meta/recipes-sato/webkit/webkitgtk_2.14.1.bb | 3 +--
> 16 files changed, 36 insertions(+), 70 deletions(-)
>
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 211 bytes --]
^ permalink raw reply
* Re: [PATCH v2 2/2] runqemu: Allow the user to specity no kernel or rootFS
From: Alistair Francis @ 2016-11-08 17:23 UTC (permalink / raw)
To: Nathan Rossi; +Cc: OE-core
In-Reply-To: <CA+aJhH2_HGwTCuOy8ahGpygxXWBJLLOb6O6Cao9scvUAWYrFbw@mail.gmail.com>
On Tue, Nov 8, 2016 at 3:15 AM, Nathan Rossi <nathan@nathanrossi.com> wrote:
> On 3 November 2016 at 10:17, Alistair Francis
> <alistair.francis@xilinx.com> wrote:
>> In some cirsumstances the user doesn't want to supply a kernel, rootFS
>> or DTB to QEMU. This will occur more now that QEMU supports loading
>> images using a '-device loader' method.
>>
>> Allow users to specify 'none' for QB_DEFAULT_FSTYPE or QB_DEFAULT_KERNEL
>> to avoid supplying these options to QEMU.
>
> Just wondering if it would be worth adding a "QB_KERNEL_OPT" similar
> to how QB_ROOTFS_OPT works. This way the "none" would not be needed
> and it would allow for using the "-device loader" arg for machines
> that boot that way.
>
> e.g.
>
> QB_KERNEL_OPT = "-kernel @KERNEL@"
> or
> QB_KERNEL_OPT = "-device loader,addr=0xdeadbeef,file=@KERNEL@"
> or (if not able to load kernel, aka 'none')
> QB_KERNEL_OPT = ""
That way is fine with me, but that would then break compatibility with
other layers using this (they would all need to set the KERNEL_OPT so
I tried to avoid that.
I guess you could set the default to "-kernel @KERNEL@" which probably
wouldn't break anything though.
I'm happy to update the patch if that is the preference. Do people
think that's a better option?
Thanks,
Alistair
>
> Regards,
> Nathan
>
>>
>> Signed-off-by: Alistair Francis <alistair.francis@xilinx.com>
>> ---
>> scripts/runqemu | 9 ++++++++-
>> 1 file changed, 8 insertions(+), 1 deletion(-)
>>
>> diff --git a/scripts/runqemu b/scripts/runqemu
>> index 922ebaf..5f60312 100755
>> --- a/scripts/runqemu
>> +++ b/scripts/runqemu
>> @@ -453,7 +453,7 @@ class BaseConfig(object):
>> def check_rootfs(self):
>> """Check and set rootfs"""
>>
>> - if self.fstype == 'nfs':
>> + if self.fstype == 'nfs' or self.fstype == "none":
>> return
>>
>> if self.rootfs and not os.path.exists(self.rootfs):
>> @@ -481,6 +481,10 @@ class BaseConfig(object):
>> # QB_DEFAULT_KERNEL is always a full file path
>> kernel_name = os.path.basename(self.get('QB_DEFAULT_KERNEL'))
>>
>> + # The user didn't want a kernel to be loaded
>> + if kernel_name == "none":
>> + return
>> +
>> deploy_dir_image = self.get('DEPLOY_DIR_IMAGE')
>> if not self.kernel:
>> kernel_match_name = "%s/%s" % (deploy_dir_image, kernel_name)
>> @@ -857,6 +861,9 @@ class BaseConfig(object):
>> k_root = '/dev/nfs nfsroot=%s:%s,%s' % (self.nfs_server, self.nfs_dir, self.unfs_opts)
>> self.kernel_cmdline = 'root=%s rw highres=off' % k_root
>>
>> + if self.fstype == 'none':
>> + self.rootfs_options = ''
>> +
>> self.set('ROOTFS_OPTIONS', self.rootfs_options)
>>
>> def guess_qb_system(self):
>> --
>> 2.7.4
>>
^ permalink raw reply
* [Patch v2 14/14] webkit: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-sato/webkit/webkitgtk_2.14.1.bb | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/meta/recipes-sato/webkit/webkitgtk_2.14.1.bb b/meta/recipes-sato/webkit/webkitgtk_2.14.1.bb
index a2586de..a67ff0a 100644
--- a/meta/recipes-sato/webkit/webkitgtk_2.14.1.bb
+++ b/meta/recipes-sato/webkit/webkitgtk_2.14.1.bb
@@ -82,8 +82,7 @@ EXTRA_OECMAKE_append_mips64 = " -DUSE_LD_GOLD=OFF "
EXTRA_OECMAKE_append_toolchain-clang = " -DUSE_LD_GOLD=OFF "
# JIT not supported on MIPS either
-EXTRA_OECMAKE_append_mips = " -DENABLE_JIT=OFF "
-EXTRA_OECMAKE_append_mips64 = " -DENABLE_JIT=OFF "
+EXTRA_OECMAKE_append_mipsarch = " -DENABLE_JIT=OFF "
SECURITY_CFLAGS_remove_aarch64 = "-fpie"
SECURITY_CFLAGS_append_aarch64 = " -fPIE"
--
2.10.2
^ permalink raw reply related
* [Patch v2 13/14] mdadm: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-extended/mdadm/mdadm_3.4.bb | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/meta/recipes-extended/mdadm/mdadm_3.4.bb b/meta/recipes-extended/mdadm/mdadm_3.4.bb
index 261054e..d34b885 100644
--- a/meta/recipes-extended/mdadm/mdadm_3.4.bb
+++ b/meta/recipes-extended/mdadm/mdadm_3.4.bb
@@ -34,8 +34,8 @@ EXTRA_OEMAKE = 'CHECK_RUN_DIR=0 CXFLAGS="${CFLAGS}"'
# to u64 == long in userspace. Define __SANE_USERSPACE_TYPES__ to get
# int-ll64.h included
CFLAGS_append_powerpc64 = ' -D__SANE_USERSPACE_TYPES__'
-CFLAGS_append_mips64 = ' -D__SANE_USERSPACE_TYPES__'
-CFLAGS_append_mips64n32 = ' -D__SANE_USERSPACE_TYPES__'
+CFLAGS_append_mipsarchn64 = ' -D__SANE_USERSPACE_TYPES__'
+CFLAGS_append_mipsarchn32 = ' -D__SANE_USERSPACE_TYPES__'
do_compile() {
oe_runmake SYSROOT="${STAGING_DIR_TARGET}"
--
2.10.2
^ permalink raw reply related
* [Patch v2 12/14] ltp: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-extended/ltp/ltp_20160126.bb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/recipes-extended/ltp/ltp_20160126.bb b/meta/recipes-extended/ltp/ltp_20160126.bb
index 7631e0e..7ee84e3 100644
--- a/meta/recipes-extended/ltp/ltp_20160126.bb
+++ b/meta/recipes-extended/ltp/ltp_20160126.bb
@@ -23,7 +23,7 @@ DEPENDS = "attr libaio libcap acl openssl zip-native"
DEPENDS_append_libc-musl = " fts "
EXTRA_OEMAKE_append_libc-musl = " LIBC=musl "
CFLAGS_append_powerpc64 = " -D__SANE_USERSPACE_TYPES__"
-CFLAGS_append_mips64 = " -D__SANE_USERSPACE_TYPES__"
+CFLAGS_append_mipsarchn64 = " -D__SANE_USERSPACE_TYPES__"
SRCREV = "fce797676b14f50406718e7ef640b50da66c9b36"
SRC_URI = "git://github.com/linux-test-project/ltp.git \
--
2.10.2
^ permalink raw reply related
* [Patch v2 11/14] ghostscript: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-extended/ghostscript/ghostscript_9.19.bb | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/meta/recipes-extended/ghostscript/ghostscript_9.19.bb b/meta/recipes-extended/ghostscript/ghostscript_9.19.bb
index fe2016b..8524591 100644
--- a/meta/recipes-extended/ghostscript/ghostscript_9.19.bb
+++ b/meta/recipes-extended/ghostscript/ghostscript_9.19.bb
@@ -46,8 +46,7 @@ EXTRA_OECONF = "--without-x --with-system-libtiff --without-jbig2dec \
--with-cups-datadir=${datadir}/cups \
"
-EXTRA_OECONF_append_mips = " --with-large_color_index=0"
-EXTRA_OECONF_append_mipsel = " --with-large_color_index=0"
+EXTRA_OECONF_append_mipsarcho32 = " --with-large_color_index=0"
# Explicity disable libtiff, fontconfig,
# freetype, cups for ghostscript-native
--
2.10.2
^ permalink raw reply related
* [Patch v2 10/14] valgrind: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-devtools/valgrind/valgrind_3.12.0.bb | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/meta/recipes-devtools/valgrind/valgrind_3.12.0.bb b/meta/recipes-devtools/valgrind/valgrind_3.12.0.bb
index a113880..2474bbc 100644
--- a/meta/recipes-devtools/valgrind/valgrind_3.12.0.bb
+++ b/meta/recipes-devtools/valgrind/valgrind_3.12.0.bb
@@ -37,8 +37,7 @@ COMPATIBLE_HOST_armv5 = 'null'
COMPATIBLE_HOST_armv6 = 'null'
# valgrind doesn't like mips soft float
-COMPATIBLE_HOST_mips = "${@bb.utils.contains("TARGET_FPU", "soft", "null", ".*-linux", d)}"
-COMPATIBLE_HOST_mipsel = "${@bb.utils.contains("TARGET_FPU", "soft", "null", ".*-linux", d)}"
+COMPATIBLE_HOST_mipsarch = "${@bb.utils.contains("TARGET_FPU", "soft", "null", ".*-linux", d)}"
inherit autotools ptest
--
2.10.2
^ permalink raw reply related
* [Patch v2 09/14] mmc-utils: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-devtools/mmc/mmc-utils_git.bb | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/meta/recipes-devtools/mmc/mmc-utils_git.bb b/meta/recipes-devtools/mmc/mmc-utils_git.bb
index 2cba860..7858819 100644
--- a/meta/recipes-devtools/mmc/mmc-utils_git.bb
+++ b/meta/recipes-devtools/mmc/mmc-utils_git.bb
@@ -13,8 +13,8 @@ SRC_URI = "git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc-utils.git;branc
S = "${WORKDIR}/git"
CFLAGS_append_powerpc64 = " -D__SANE_USERSPACE_TYPES__"
-CFLAGS_append_mips64 = " -D__SANE_USERSPACE_TYPES__"
-CFLAGS_append_mips64n32 = " -D__SANE_USERSPACE_TYPES__"
+CFLAGS_append_mipsarchn64 = " -D__SANE_USERSPACE_TYPES__"
+CFLAGS_append_mipsarchn32 = " -D__SANE_USERSPACE_TYPES__"
do_install() {
install -d ${D}${bindir}
--
2.10.2
^ permalink raw reply related
* [Patch v2 08/14] gdb: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-devtools/gdb/gdb-common.inc | 7 +------
1 file changed, 1 insertion(+), 6 deletions(-)
diff --git a/meta/recipes-devtools/gdb/gdb-common.inc b/meta/recipes-devtools/gdb/gdb-common.inc
index 33a5ce9..5b8087c 100644
--- a/meta/recipes-devtools/gdb/gdb-common.inc
+++ b/meta/recipes-devtools/gdb/gdb-common.inc
@@ -6,12 +6,7 @@ DEPENDS = "expat zlib ncurses virtual/libiconv ${LTTNGUST}"
LTTNGUST = "lttng-ust"
LTTNGUST_aarch64 = ""
LTTNGUST_libc-uclibc = ""
-LTTNGUST_mips = ""
-LTTNGUST_mipsel = ""
-LTTNGUST_mips64 = ""
-LTTNGUST_mips64el = ""
-LTTNGUST_mips64n32 = ""
-LTTNGUST_mips64eln32 = ""
+LTTNGUST_mipsarch = ""
LTTNGUST_sh4 = ""
LTTNGUST_libc-musl = ""
--
2.10.2
^ permalink raw reply related
* [Patch v2 07/14] gcc-runtime: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-devtools/gcc/gcc-runtime.inc | 9 +--------
1 file changed, 1 insertion(+), 8 deletions(-)
diff --git a/meta/recipes-devtools/gcc/gcc-runtime.inc b/meta/recipes-devtools/gcc/gcc-runtime.inc
index 15252f1..d56f82a 100644
--- a/meta/recipes-devtools/gcc/gcc-runtime.inc
+++ b/meta/recipes-devtools/gcc/gcc-runtime.inc
@@ -17,14 +17,7 @@ EXTRA_OECONF_PATHS = "\
EXTRA_OECONF_append_linuxstdbase = " --enable-clocale=gnu"
RUNTIMELIBITM = "libitm"
-RUNTIMELIBITM_mips = ""
-RUNTIMELIBITM_mipsel = ""
-RUNTIMELIBITM_mips64 = ""
-RUNTIMELIBITM_mips64el = ""
-RUNTIMELIBITM_mipsisa32r6 = ""
-RUNTIMELIBITM_mipsisa32r6el = ""
-RUNTIMELIBITM_mipsisa64r6 = ""
-RUNTIMELIBITM_mipsisa64r6el = ""
+RUNTIMELIBITM_mipsarch = ""
RUNTIMELIBITM_nios2 = ""
RUNTIMELIBITM_microblaze = ""
--
2.10.2
^ permalink raw reply related
* [Patch v2 06/14] packagegroup: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-core/packagegroups/packagegroup-core-sdk.bb | 5 +----
.../packagegroups/packagegroup-core-tools-profile.bb | 10 ++--------
2 files changed, 3 insertions(+), 12 deletions(-)
diff --git a/meta/recipes-core/packagegroups/packagegroup-core-sdk.bb b/meta/recipes-core/packagegroups/packagegroup-core-sdk.bb
index aceba78..327cd8e 100644
--- a/meta/recipes-core/packagegroups/packagegroup-core-sdk.bb
+++ b/meta/recipes-core/packagegroups/packagegroup-core-sdk.bb
@@ -28,10 +28,7 @@ RDEPENDS_packagegroup-core-sdk = "\
SANITIZERS = "libasan-dev libubsan-dev"
SANITIZERS_aarch64 = ""
-SANITIZERS_mips = ""
-SANITIZERS_mipsel = ""
-SANITIZERS_mips64 = ""
-SANITIZERS_mips64n32 = ""
+SANITIZERS_mipsarch = ""
SANITIZERS_nios2 = ""
SANITIZERS_powerpc64 = ""
SANITIZERS_sparc = ""
diff --git a/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb b/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
index dd98445..31ad82b 100644
--- a/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
+++ b/meta/recipes-core/packagegroups/packagegroup-core-tools-profile.bb
@@ -36,10 +36,7 @@ PERF_libc-musl = ""
SYSTEMTAP = "systemtap"
SYSTEMTAP_libc-uclibc = ""
SYSTEMTAP_libc-musl = ""
-SYSTEMTAP_mips = ""
-SYSTEMTAP_mipsel = ""
-SYSTEMTAP_mips64 = ""
-SYSTEMTAP_mips64n32 = ""
+SYSTEMTAP_mipsarch = ""
SYSTEMTAP_nios2 = ""
SYSTEMTAP_aarch64 = ""
@@ -65,10 +62,7 @@ BABELTRACE = "babeltrace"
VALGRIND = "valgrind"
VALGRIND_libc-uclibc = ""
VALGRIND_libc-musl = ""
-VALGRIND_mips = ""
-VALGRIND_mipsel = ""
-VALGRIND_mips64 = ""
-VALGRIND_mips64n32 = ""
+VALGRIND_mipsarch = ""
VALGRIND_nios2 = ""
VALGRIND_armv4 = ""
VALGRIND_armv5 = ""
--
2.10.2
^ permalink raw reply related
* [Patch v2 05/14] glibc: Reduce duplication in MIPS variants
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-core/glibc/glibc-ld.inc | 17 ++---------------
1 file changed, 2 insertions(+), 15 deletions(-)
diff --git a/meta/recipes-core/glibc/glibc-ld.inc b/meta/recipes-core/glibc/glibc-ld.inc
index b982368..e2e2474 100644
--- a/meta/recipes-core/glibc/glibc-ld.inc
+++ b/meta/recipes-core/glibc/glibc-ld.inc
@@ -7,21 +7,8 @@ def ld_append_if_tune_exists(d, infos, dict):
def glibc_dl_info(d):
ld_info_all = {
- "mips": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mips64-n32": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mips64": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mipsel": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mips64el-n32": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mips64el": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mips-nf": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mipsisa32r6": ["ld-linux-mipsn8.so.1", "FLAG_ELF_LIBC6"],
- "mipsisa32r6el": ["ld-linux-mipsn8.so.1", "FLAG_ELF_LIBC6"],
- "mips64-nf-n32": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mips64-nf": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mips64el-nf-n32": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mips64el-nf": ["ld.so.1", "FLAG_ELF_LIBC6"],
- "mipsisa64r6": ["ld-linux-mipsn8.so.1", "FLAG_ELF_LIBC6"],
- "mipsisa64r6el": ["ld-linux-mipsn8.so.1", "FLAG_ELF_LIBC6"],
+ "mipsarch": ["ld.so.1", "FLAG_ELF_LIBC6"],
+ "mipsarchr6": ["ld-linux-mipsn8.so.1", "FLAG_ELF_LIBC6"],
"powerpc": ["ld.so.1", "FLAG_ELF_LIBC6"],
"powerpc-nf": ["ld.so.1", "FLAG_ELF_LIBC6"],
"powerpc64": ["ld64.so.1", "FLAG_ELF_LIBC6"],
--
2.10.2
^ permalink raw reply related
* [Patch v2 04/14] fts: Reduce duplication in MIPS variants.
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
Reduce duplication in MIPS variants now that the MACHINEOVERRIDES
variable is defined
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/recipes-core/fts/fts.bb | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/meta/recipes-core/fts/fts.bb b/meta/recipes-core/fts/fts.bb
index 9d8230f..de9297e 100644
--- a/meta/recipes-core/fts/fts.bb
+++ b/meta/recipes-core/fts/fts.bb
@@ -22,10 +22,7 @@ S = "${WORKDIR}/${BPN}"
do_configure[noexec] = "1"
-HASHSTYLE_mips = "sysv"
-HASHSTYLE_mipsel = "sysv"
-HASHSTYLE_mips64 = "sysv"
-HASHSTYLE_mips64el = "sysv"
+HASHSTYLE_mipsarch = "sysv"
HASHSTYLE = "gnu"
VER = "0"
--
2.10.2
^ permalink raw reply related
* [Patch v2 01/14] arch-mips: Add o32 in TUNE_FEATURES for MIPS32R6
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
mips32r6 tunings should have o32 ABI flag in TUNE_FEATURES
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
Applies to Morty as well
---
meta/conf/machine/include/tune-mips32r6.inc | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/meta/conf/machine/include/tune-mips32r6.inc b/meta/conf/machine/include/tune-mips32r6.inc
index 4721319..dea33ea 100644
--- a/meta/conf/machine/include/tune-mips32r6.inc
+++ b/meta/conf/machine/include/tune-mips32r6.inc
@@ -9,21 +9,21 @@ TUNE_CCARGS .= "${@bb.utils.contains('TUNE_FEATURES', 'mipsisa32r6', ' -march=mi
# Base Tunes
AVAILTUNES += "mipsisa32r6 mipsisa32r6el"
-TUNE_FEATURES_tune-mipsisa32r6 = "bigendian mipsisa32r6 fpu-hard r6"
+TUNE_FEATURES_tune-mipsisa32r6 = "o32 bigendian mipsisa32r6 fpu-hard r6"
MIPSPKGSFX_VARIANT_tune-mipsisa32r6 = "${TUNE_ARCH}"
PACKAGE_EXTRA_ARCHS_tune-mipsisa32r6 = "mipsisa32r6"
-TUNE_FEATURES_tune-mipsisa32r6el = "fpu-hard mipsisa32r6 r6"
+TUNE_FEATURES_tune-mipsisa32r6el = "o32 fpu-hard mipsisa32r6 r6"
MIPSPKGSFX_VARIANT_tune-mipsisa32r6el = "${TUNE_ARCH}"
PACKAGE_EXTRA_ARCHS_tune-mipsisa32r6el = "mipsisa32r6el"
# Soft Float
AVAILTUNES += "mipsisa32r6-nf mipsisa32r6el-nf"
-TUNE_FEATURES_tune-mipsisa32r6-nf = "bigendian mipsisa32r6 r6"
+TUNE_FEATURES_tune-mipsisa32r6-nf = "o32 bigendian mipsisa32r6 r6"
MIPSPKGSFX_VARIANT_tune-mipsisa32r6-nf = "${TUNE_ARCH}"
PACKAGE_EXTRA_ARCHS_tune-mipsisa32r6-nf = "mipsisa32r6-nf"
-TUNE_FEATURES_tune-mipsisa32r6el-nf = "mipsisa32r6 r6"
+TUNE_FEATURES_tune-mipsisa32r6el-nf = "o32 mipsisa32r6 r6"
MIPSPKGSFX_VARIANT_tune-mipsisa32r6el-nf = "${TUNE_ARCH}"
PACKAGE_EXTRA_ARCHS_tune-mipsisa32r6el-nf = "mipsisa32r6el-nf"
--
2.10.2
^ permalink raw reply related
* [Patch v2 02/14] arch-mips: Add MACHINEOVERRIDES variables to reduce duplication
From: Zubair Lutfullah Kakakhel @ 2016-11-08 17:12 UTC (permalink / raw)
To: openembedded-core
In-Reply-To: <20161108171235.41837-1-Zubair.Kakakhel@imgtec.com>
In some cases, each MIPS variant in a recipe requires a duplicate
line. Even if the passed flag is the same.
Add global MACHINEOVERRIDES variables for the following
* mipsarch : All MIPS
* mipsarch{eb,el} : All MIPS Big/Little Endian
* mipsarchr6 : All MIPS R6
* mipsarcho32 : All MIPS o32 Endian Independent
* mipsarchn32 : All MIPS n32 Endian Independent
* mipsarchn64 : All MIPS n64 Endian Independent
* mipsarcho32{eb,el} : All MIPS o32 Big/Little Endian
* mipsarchn32{eb,el} : All MIPS n32 Big/Little Endian
* mipsarchn64{eb,el} : All MIPS n64 Big/Little Endian
This is intended to reduce duplications in recipes
[YOCTO #10404]
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com>
---
meta/conf/machine/include/mips/README | 3 +++
meta/conf/machine/include/mips/arch-mips.inc | 12 ++++++++++++
2 files changed, 15 insertions(+)
diff --git a/meta/conf/machine/include/mips/README b/meta/conf/machine/include/mips/README
index 62fa156..f36e87b 100644
--- a/meta/conf/machine/include/mips/README
+++ b/meta/conf/machine/include/mips/README
@@ -32,6 +32,9 @@ MIPSPKGSFX_BYTE - This is defined as either blank and "64" for MIPS64 CPUs.
MIPSPKGSFX_ENDIAN - For bigendian hardware this is blank, otherwise it's
defined as "el".
+MIPSPKGSFX_ENDIAN2 - For bigendian hardware this is "eb", otherwise it's
+defined as "el".
+
MIPSPKGSFX_VARIANT_tune-<tune> - In the default tunings it is set to the
same value as TUNE_ARCH. In custom, optimized tunings, the value should
be modified to more precisely describe the tuning.
diff --git a/meta/conf/machine/include/mips/arch-mips.inc b/meta/conf/machine/include/mips/arch-mips.inc
index 6069ca1..d8b2bcb 100644
--- a/meta/conf/machine/include/mips/arch-mips.inc
+++ b/meta/conf/machine/include/mips/arch-mips.inc
@@ -39,6 +39,7 @@ TUNEVALID[mipsisa32r6] = "Use 32r6"
# Package naming
MIPSPKGSFX_ENDIAN = "${@bb.utils.contains('TUNE_FEATURES', 'bigendian', '', 'el', d)}"
+MIPSPKGSFX_ENDIAN2 = "${@bb.utils.contains('TUNE_FEATURES', 'bigendian', 'eb', 'el', d)}"
MIPSPKGSFX_BYTE = "${@bb.utils.contains('TUNE_FEATURES', 'n64' , '64', '', d)}"
MIPSPKGSFX_BYTE .= "${@bb.utils.contains('TUNE_FEATURES', 'n32' , '64', '', d)}"
MIPSPKGSFX_FPU = "${@bb.utils.contains('TUNE_FEATURES', 'fpu-hard' , '', '-nf', d)}"
@@ -50,6 +51,17 @@ MIPSPKGSFX_32R6 = "${@bb.utils.contains('TUNE_FEATURES', 'mipsisa32r6', 'isa32',
TUNE_ARCH = "mips${MIPSPKGSFX_32R6}${MIPSPKGSFX_64R6}${MIPSPKGSFX_BYTE}${MIPSPKGSFX_R6}${MIPSPKGSFX_ENDIAN}"
TUNE_PKGARCH = "${MIPSPKGSFX_VARIANT_tune-${DEFAULTTUNE}}${MIPSPKGSFX_FPU}${MIPSPKGSFX_ABI}"
+# Various Global Machine Overrides
+MACHINEOVERRIDES =. "mipsarch:"
+MACHINEOVERRIDES =. "mipsarch${MIPSPKGSFX_ENDIAN2}:"
+MACHINEOVERRIDES =. "${@bb.utils.contains('TUNE_FEATURES', 'r6', 'mipsarchr6:', '' ,d)}"
+MACHINEOVERRIDES =. "${@bb.utils.contains('TUNE_FEATURES', 'n32', 'mipsarchn32:', '' ,d)}"
+MACHINEOVERRIDES =. "${@bb.utils.contains('TUNE_FEATURES', 'o32', 'mipsarcho32:', '' ,d)}"
+MACHINEOVERRIDES =. "${@bb.utils.contains('TUNE_FEATURES', 'n64', 'mipsarchn64:', '' ,d)}"
+MACHINEOVERRIDES =. "${@bb.utils.contains('TUNE_FEATURES', 'n32', 'mipsarchn32${MIPSPKGSFX_ENDIAN2}:', '' ,d)}"
+MACHINEOVERRIDES =. "${@bb.utils.contains('TUNE_FEATURES', 'o32', 'mipsarcho32${MIPSPKGSFX_ENDIAN2}:', '' ,d)}"
+MACHINEOVERRIDES =. "${@bb.utils.contains('TUNE_FEATURES', 'n64', 'mipsarchn64${MIPSPKGSFX_ENDIAN2}:', '' ,d)}"
+
# Base tunes
AVAILTUNES += "mips mips64-n32 mips64 mipsel mips64el-n32 mips64el mips-nf mips64-nf-n32 mips64-nf mipsel-nf mips64el-nf-n32 mips64el-nf"
TUNE_FEATURES_tune-mips = "o32 bigendian fpu-hard"
--
2.10.2
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox