aboutsummaryrefslogtreecommitdiffstats
path: root/meta/conf
AgeCommit message (Collapse)Author
2016-02-18no-static-libs.inc: Add libcap-nativeRichard Purdie
We need to disable static flags for libcap-native as well. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-18bitbake.conf: use target path as compile dir in debugging infoHongxu Jia
In debugging information, it uses target paths rather than build ones as compile dir. ... -fdebug-prefix-map=old=new When compiling files in directory old, record debugging information describing them as in new instead. ... Compile without this fix: objdump -g git/test.o ... The Directory Table (offset 0x1b): | 1 /buildarea/raid0/hjia/build-20160119-yocto-buildpath/tmp/sysroots/x86_64-linux/usr/lib/ i686-pokymllib32-linux.lib32-gcc-cross-initial-i686/gcc/i686-pokymllib32-linux/5.3.0/include | 2 /buildarea/raid0/hjia/build-20160119-yocto-buildpath/tmp/sysroots/lib32-qemux86-64/usr/include/bits | 3 /buildarea/raid0/hjia/build-20160119-yocto-buildpath/tmp/sysroots/lib32-qemux86-64/usr/include ... Compile with this fix: objdump -g git/test.o ... The Directory Table (offset 0x1b): | 1 /usr/lib/i686-pokymllib32-linux.lib32-gcc-cross-initial-i686/gcc/i686-pokymllib32-linux/ 5.3.0/include | 2 /usr/include/bits | 3 /usr/include ... [YOCTO #7058] Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-16layer.conf: Add gstreamer1.0-meta-base to SIGGEN_EXCLUDERECIPES_ABISAFERichard Purdie
This recipe and its packages are by design like a packagegroup and can be safely depended not to change names upon despite it being machine specific. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-16conf/no-static-libs: add explicit rule for libicalRoss Burton
cmake doesn't have a standard for disabling static libraries so libical needs an explicit statement. Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-15security_flags: Add SECURITY_CFLAGS to TARGET_CC_ARCH for binutilsKhem Raj
libtool decides to filter out -fstack-protector-strong on its own and its documented here https://www.gnu.org/software/libtool/manual/html_node/Stripped-link-flags.html this causes linking errors when linking libbfd.so since objects (.o) are compiled using -fstack-protector-strong so they are expecting to link with libssp but the option goes missing in linker flags. With this patch the security flags are hoisted upto CC itself and libtool thankfully does not touch CC. Adding to CC also means that we can now remove it from LDFLAGS since when gcc driver is used to do linking then we have LD = CC and this option makes to linker cmdline Since CC is used without CFLAGS in configure tests, some tests fail complaining that -Olevel is not used while using _FORTIFY_SOURCE therefore added SELECTED_OPTIMIZATION to TARGET_CC_ARCH as well Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-15bitbake.conf: Remove unhelpful default value for EXTRA_OEMAKEMike Crowe
Setting EXTRA_OEMAKE to "-e MAKEFLAGS=" by default is a historical accident and many classes (e.g. autotools.bbclass, module.bbclass) and recipes have to override this default in order to work with sensible build systems. Now that openssl and pciutils have been fixed to set EXTRA_OEMAKE explicitly it is possible to set EXTRA_OEMAKE = "". Signed-off-by: Mike Crowe <mac@mcrowe.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-15conf/distro/include: Add no-static-libs.incRichard Purdie
Static libraries are old technology. We've left them around since in previous tests, they only added around 5% to the overall build time. With new and interesting uses of OE, they're becoming more problematic. For example, sstate becomes much larger with static libs enabled which increases the size of eSDK and increases the time taken for sstate operations. Since the static libs contain all the debugging symbols, they're also pretty huge. This patch adds a common include file which allows the user or distro to disable the static libraries in the majority of cases. There are some libs where we do need the static lib, a good example being pseudo-native which uses sqlite3-native static libs. These are left enabled by the include file, as are recipes where --disable-static doesn't work, or isn't supported. This list can likely be reduced over time as individual corner cases are addressed. Maintaining this list in a common location is more desireable than everyone doing it themselves. Poky will switch to using this, OE-Core will need to discuss that as its default. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-15qemu.inc: Add rng-tools to qemu imagesSaul Wold
This patch adds rng-tools to MACHINE_EXTRA_RRECOMMENDS so that can be used to provide the additional entropy to prevent hangs in getrandom() for qemu images [YOCTO #8681] [YOCTO #8816] Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-12sanity: Bump minimum version to 1.29.0Richard Purdie
This requires the python expansion changes and allow SRCPV to work correctly without errors. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-11bitbake.conf: remove unused ALLOWED_FLAGSRoss Burton
This variable hasn't been used for a *long* time, remove it from bitbake.conf. Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-11Remove obsolete references to exmapRandy MacLeod
Exmap was a useful tool but it appears to be unsupported. Remove it from the sample local.conf and remove the commented lines from the profile tools packagegroup. Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-11meta/conf/layer.conf: adapt to more flexible initramfs-framework RDEPENDSPatrick Ohly
initramfs-framework now RDEPENDS on ${VIRTUAL-RUNTIME_base-utils}, which can be busybox or some alternative like toybox. Making the SIGGEN_EXCLUDE_SAFE_RECIPE_DEPS exception flexible, too, ensures that distros using toybox still pass the selftests. Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-11tune-corei7.inc: tell qemu to emulate a matching processorRoss Burton
If tune-corei7 is in use then the target binaries may contain instructions that qemu-x86-64 can't execute by default, resulting in errors on rootfs construction: NOTE: Running intercept scripts: NOTE: > Executing update_font_cache intercept ... qemu: uncaught target signal 4 (Illegal instruction) - core dumped In this case the instruction is popcnt, part of SSE4.2, so tell Qemu to emulate the CPU that the tune targets (in this case, Nehalem). Also pass check=false as the Nehalem machine supports VME but user-space qemu doesn't, which produces a warning unless CPUID checking is disabled. [ YOCTO #8888 ] Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-10bitbake.conf/base: Improve handling of SRCPVRichard Purdie
If SRCPV is set, it can be expanded when SRC_URI doesn't support it leading to errors. Avoid doing this by setting it only when it makes sense. This patch depends on the bitbake python expansion patch series. [YCOTO #7772] Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-07security_flags: wipe security flags for gcc/glibc and related librariesKhem Raj
It causes a catch-22 situation where we build libssp in gcc-runtime but also pass -fstack-protector flags which require libssp Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-07security_flags: use -fstack-protector-strongKhem Raj
This is a better version of -fstack-protector-all with reduced stack usage and better performance yet giving same amount of coverage. It's available in gcc 4.9 onwards. https://outflux.net/blog/archives/2014/01/27/fstack-protector-strong/ has more details. Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-07security_flags: ensure security flags only apply to target buildsKhem Raj
As otherwise the security flags can leak into target builds. This can result in flags that the host compiler doesn't support, causing build failures. Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-07security_flags.inc: don't do -pie for syslinuxRoss Burton
sysroots/x86_64-linux/usr/libexec/x86_64-poky-linux/gcc/x86_64-poky-linux/5.3.0/ld: syslinux.o: relocation R_X86_64_32 against `.rodata.str1.1' can not be used when making a shared object; recompile with -fPIC Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-06gtk-theme-torturer: remove from oe-coreRoss Burton
This recipe is very old, unmaintained, not used at all in OE-Core, and not useful in a world that has moved to GTK+ 3 (even if Sato is slow at catching up). Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-06bitbake.conf/native/nativesdk: Set PKG_CONFIG_SYSTEM_ at top levelRichard Purdie
Setting PKG_CONFIG_SYSTEM_{HEADERS/INCLUDE}_PATH for nativesdk isn't enough, we also need to deal with multlib cases where libdir from pkg-config-native isn't correct. Native builds are about the only case where this variable shouldn't be set. Therefore move the code from nativesdk to bitbake.conf and unexport it in the native case. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-05local.conf.sample.extended: Document HOW-TO enable systemd or busbox for ↵Khem Raj
init system OE core provides systemd,busybox as options for system init system along with sysvinit being the default. Document the needed settings to enable systemd and busybox as options if user wishes. Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-05local.conf.sample: Remove trailing whitespacesKhem Raj
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-04meta: add ASSUME_PROVIDED dependency on wget-native for http fetchesRoss Burton
For clarity and consistency, add a dependency on wget-native for any URIs that will be fetched using the wget fetcher, and add wget-native to ASSUME_PROVIDED. Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-02-02documentation.conf: align the documentation for DEBUG_OPTIMIZATION and ↵Pascal Bach
FULL_OPTIMIZATION with bitbake.conf Signed-off-by: Pascal Bach <pascal.bach@siemens.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-02-02machine/include: drop tune-cortexm*.inc and tune-cortexr4.incAndre McCurdy
The Cortex M1, M3 and R4 CPU tuning files are poorly tested (if at all). They have no obvious users either inside or outside oe-core. Until OE officially gains support for CPUs without an MMU, these tuning files are probably better maintained outside of oe-core (e.g. in a separate meta-nommu layer). Signed-off-by: Andre McCurdy <armccurdy@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-31binutils: Upgrade to 2.26Khem Raj
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-30gcc, qemuppc: Explicitly disable forcing SPE flagsKhem Raj
G4 does not have SPE, so we make that explicit in the tune files and since we emulate G4 when building Qemu, we ensure it for qemuppc as well. GCC config for powerpc-linux is made to include SPE by default which is equivalent if the tripet was powerpc-linux*spe, this forces gcc to configure assembler to enable -mspe by default, when we do that then the kernel fails to compile with binutils 2.26, since newer assembler is smart to detect the tlbia instructions are not compatible with SPE and hence the kernel build breaks rightly. We configure the kernel for G4 as well where it enables tlbia instrucitons rightly so because it thinks its being configured for power4. So we keep the options but do not force -mspe down to assembler as default. Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-29bitbake.conf: stop exporting PATCH_GET = "0"Andre McCurdy
Exporting PATCH_GET = "0" has been redundant since patch 2.6.0 was released in 2009: http://git.savannah.gnu.org/cgit/patch.git/commit/?id=b008dece18e6b94b8a13ea44a253855bf407ed01 Host distros which shipped with patch 2.5.x (e.g. Centos 5) are no longer supported, so this export can be retired from bitbake.conf. Signed-off-by: Andre McCurdy <armccurdy@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-29libc-package.bbclass: add LOCALE_UTF8_IS_DEFAULTRichard Tollerton
python hard-codes the encoding of many locales; for instance, en_US is always assumed to be ISO-8859-1, regardless of the actual encoding of the en_US locale on the system. cf https://hg.python.org/cpython/file/7841e9b614eb/Lib/locale.py#l1049, getdefaultlocale(), etc. This code appears to date back to python 2.0. The source of this hard-coding is Xorg's locale.alias but is ultimately justified by glibc's SUPPORTED. This causes problems on OE, because any locale lacking an explicit encoding suffix (e.g. en_US) is UTF-8. It has been this way from the beginning (svn r1). That is not a bug, per se -- no specification prohibits this AFAIK. But it seems to be at odds with virtually every other glibc-based distribution in existence. To avoid needlessly aggravating hidden bugs that nobody else might hit, it makes sense to disable this behavior such that locales are named precisely as specified by SUPPORTED. I suppose that reasonable minds may disagree on whether or not the current behavior is prudent; at the very least, this is likely to break IMAGE_LINGUAS settings. So let's create a new distro variable LOCALE_UTF8_IS_DEFAULT to allow either behavior. Set it to 0 and all your locales get named exactly like they are in SUPPORTED. Leave it at 1 to preserve current OE locale naming conventions. Signed-off-by: Richard Tollerton <rich.tollerton@ni.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-29local.conf.sample.extended: Update the info about BBMASKPeter Kjellerstedt
Update the info about BBMASK to reflect that it can now contain multiple regular expressions. Also correct a misspelled PREFERRED_VERSION in a comment. Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-29documentation.conf: Update the help for BBMASKPeter Kjellerstedt
Since it is now possible to concatenate multiple regular expressions into BBMASK, there is no longer any real reason to limit it to be specified only in local.conf. Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-26image.bbclass: check INITRAMFS_MAXSIZERobert Yang
Usually, the initramfs' maxsize can be 1/2 of ram size since modern kernel uses tmpfs as initramfs by dafault, and tmpfs allocates 1/2 of ram by default at boot time, which will be used to locate the initramfs. Set INITRAMFS_MAXSIZE to 131072K (128M) by default (ram 256M), the initramfs is small usually, for example, core-image-minimal-initramfs is about 21M (uncompressed, 17M * 1.3) by default, but if the user add a lot pkgs to initramfs, we can error and stop to let the user know ealier rather than fail to boot (e.g., OOM-killer) at boot time. Please see the bug for more info: https://bugzilla.yoctoproject.org/show_bug.cgi?id=5963 [YOCTO #5963] Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-22bitbake: set default libexecdir to $prefix/libexecRoss Burton
The use of $libdir/$BPN as libexecdir is contrary to all other mainstream distributions (which either use $prefix/libexec or $libdir), and the GNU Coding Standards[1] which suggests $prefix/libexec and notes that any package-specific nesting should be done by the package itself. Finally, having libexecdir change between recipes makes it very difficult for different recipes to invoke binaries that have been installed into libexecdir. The File System Hierarchy[2] now recognises the use of $prefix/libexec/, giving distributions the choice of $prefix/lib or $prefix/libexec without breaking FHS. Change bitbake.conf to use $prefix/libexec for libexecdir, so that the binaries are separated from the libraries. This should avoid complications with multilib configurations. [1] https://www.gnu.org/prep/standards/html_node/Directory-Variables.html [2] http://refspecs.linuxfoundation.org/FHS_3.0/fhs/ch04s07.html [ YOCTO #6398 ] Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-22linux-libc-headers: update to 4.4Bruce Ashfield
Updating the lib-headers to match the 4.4 LTSI kernel version. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-22bitbake.conf: Remove horrible variable expansion hacksRichard Purdie
We used to need these hacks to make things work. Rework the variables to remove the horrible hacks and make things slightly less ugly. This does mean PE and PRAUTO are given default empty values but this is preferred to the other ugliness. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-19python: Upgrade 2.7.9 > 2.7.11Alejandro Hernandez
- no license change, just dates Rebased: - check-if-target-is-64b-not-host.patch - add-CROSSPYTHONPATH-for-PYTHON_FOR_BUILD.patch Signed-off-by: Alejandro Hernandez <alejandro.hernandez@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-19security_flags.inc: remove obsolete workarounds for curlAndre McCurdy
The curl configure script contains sanity checks for unexpected options being passed via CFLAGS, LDFLAGS, etc. environment variables. These sanity checks catch -Dxxx options in CFLAGS, which clashes with OE's approach of using CFLAGS to pass -D_FORTIFY_SOURCE (curl's configure script suggests, quite correctly, that -Dxxx options should be passed via CPPFLAGS instead). These sanity checks previously generated fatal errors, but have been downgraded to warnings since curl v7.32. Therefore the workaround of avoiding -D_FORTIFY_SOURCE for curl is obsolete and can be removed. https://github.com/bagder/curl/commit/5d3cbde72ece7d83c280492957a26e26ab4e5cca Signed-off-by: Andre McCurdy <armccurdy@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-17bitbake.conf: rename python-native-runtimeEd Bartosh
The code in native.bbclass adds -native suffix to the package names that don't have it. Renamed python-native-runtime -> hostpython-runtime-native to avoid mangling it and to conform with the naming convetion for native packages. Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-13tcmode-default.inc: Fix preferred provider nativesdk-sdk_prefix-libc-initialMark Hatle
Similar to the libc-for-gcc preferred provider, we also need a libc-initial version. Layers such as meta-mingw need the ability to override these values in order to generate an SDK that works on non Linux environments. Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-13gdb: upgrade to 7.10.1Richard Purdie
The PPC inferior patch was dropped since an equivalent fix was merged upstream. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-11local.conf.sample: add qemumips64Yi Zhao
Add the missing example machine configuration for qemumips64 Signed-off-by: Yi Zhao <yi.zhao@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-11meta: rename perl-native-runtimeEd Bartosh
The code in native.bbclass adds -native suffix to the package names that don't have it. perl-native-runtime becomes perl-native-runtime-native because of this. Renamed perl-native-runtime -> hostperl-runtime-native to avoid mangling it and to conform with the naming convetion for native packages. Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-11bitbake.conf: remove 'stamp-base'Chen Qi
Remove 'stamp-base' from this file as this flag is no longer used. [YOCTO #8468] Signed-off-by: Chen Qi <Qi.Chen@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-11bitbake.conf: add virtual/libiconv-native to ASSUME_PROVIDEDRoss Burton
It's possible for a native recipe to have virtual/libiconv-native as a build dependency, but as we expect that the host provides that add it to ASSUME_PROVIDED. Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2016-01-07x11vnc: remove all references to moved packageIoan-Adrian Ratiu
Together with the move to meta-oe, all references to x11vnc should be removed from oe-core. There are three of these: a distro alias, a packagegroup rdepends and a runtime test. Signed-off-by: Ioan-Adrian Ratiu <adrian.ratiu@ni.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-07tune-*: use mcpu instead of mtune for ARM tunesMartin Jansa
* since: commit cffda9a821a3b83a8529d643c567859e091c6846 Author: Martin Jansa <Martin.Jansa@gmail.com> Date: Tue Sep 11 17:05:45 2012 +0000 arch-arm: define different ARMPKGARCH when different CCARGS are used we don't need to worry about e.g. cortexa7 device upgrading binary package from armv7a feed which would be built with -mcpu=cortexa15, so we can use -mcpu instead of -mtune, because we won't share the binary feed with MACHINEs built with different tune. Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-07arch-armv7ve: add tune include for armv7ve and use it from cortexa7 and ↵Martin Jansa
cortexa15 * be aware that this -march value is available only in gcc-4.9 and newer: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57907 * -mcpu=cortex15 and -mcpu=cortexa7 conflict with -march=armv7a We either have to stop putting -march in default CCARGS or at least set it compatible one like this patch does. Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-07cortexa{7,15,17}: add VFPv4 tunesMartin Jansa
* it was added only to hf cortexa7 in: commit e97d152ca13556b41a236c1a4cfb11e77ff857d7 Author: Kristof Robot <krirobo@gmail.com> Date: Sun Jan 26 10:03:56 2014 +0100 Add Cortex A7 support for NEONv2 & FPv4 * add it to softfp cortexa7 and both versions for cortexa15 and cortexa17 tunes Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-07feature-arm-vfp.inc: Further simplify with TUNE_CCARGS_MFLOATMartin Jansa
* add TUNE_CCARGS_MFLOAT variable which is used to set -mfloat-abi parameter as well as ARMPKGSFX_EABI suffix in TUNE_PKGARCH and TARGET_FPU * TARGET_FPU was using ARMPKGSFX_FPU, but in most cases we use it only to distinguish between hard and soft abi, not various -mfpu variants which can appear in ARMPKGSFX_FPU Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
2016-01-07feature-arm-{neon,vfp}.inc: refactor and fix issuesMartin Jansa
* respect all 4 vfp options ('vfp', 'vfpv3d16', 'vfpv3', 'vfpv4') when setting -mfloat-abi and ARMPKGSFX_EABI, without this change it wasn't possible to use call-convention hard together with vfpv4 * move 'vfpv3d16', 'vfpv3', 'vfpv4' support from feature-arm-vfp.inc to feature-arm-neon.inc the main difference is that feature-arm-vfp.inc is included in arch-armv5.inc while feature-arm-neon.inc only in armv7*.inc, so these options should be added to TUNEVALID also only for armv7* MACHINEs. * support vfpv4 with or without neon when both vfpv4 and neon are in TUNE_FEATURES we want to set only one -mfpu parameter and to neon-vfpv4 * prevent multiple appends to ARMPKGSFX_FPU, we don't want to include e.g. -vfp as well as -vfpv4 when both "vfp" and "vfpv4" are in TUNE_FEATURES * add -mfpu=vfp for tunes with "vfp" in TUNE_FEATURES - before that we were only adding -vfp to ARMPKGSFX_FPU * add TUNE_CCARGS_MFPU variable which is used to set -mfpu parameter as well as ARMPKGSFX_FPU suffix in TUNE_PKGARCH, all enabled values are appended to it based on TUNE_FEATURES and then the last one is used in the actual param and suffix * this prevents multiple -mfpu options in TUNE_CCARGS * !!! This means we need to change TUNE_PKGARCH and PACKAGE_EXTRA_ARCHS for vfpv4, vfpv3d16, vfpv3 tunes, because the -vfp* isn't prependend multiple times. If you're using one of these new DEFAULTTUNES (which were at least partially broken anyway) and depend on working binary package feed upgrade-path, then don't forget to migrate PR service database to new TUNE_PKGARCH. Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
generated by cgit 1.2.3-korg (git 2.39.0) at 2024-04-16 23:21:24 +0000