summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* nspr: fix package splitingDexuan Cui2012-04-141-3/+4
| | | | | | | | | | Here /usr/lib/lib*.so files are binaries rather than symbol links. We should package them into ${PN} rather than ${PN}-dev, or else, when a package, that rdepends on nspr, is packaged, we get a "non-dev package rdepends on nspr-dev" ERROR. Signed-off-by: Dexuan Cui <dexuan.cui@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* systemtap: disable document generation by defaultTom Zanussi2012-04-141-0/+4
| | | | | | | | Building the systemtap documentation adds significantly to the build time, so disable it by default. Signed-off-by: Tom Zanussi <tom.zanussi@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* systemtap: fix docproc build errorTom Zanussi2012-04-143-2/+27
| | | | | | | | | | | When building docs in systemtap, docproc is used to generate the tapset documentation, but it gets built for the target, while it needs to be build for the host instead. This change causes that to happen. Fixes [YOCTO #2193]. Signed-off-by: Tom Zanussi <tom.zanussi@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* linux-yocto: allow .cfg, .scc, .patch and defconfigs to be processed in orderBruce Ashfield2012-04-132-68/+8
| | | | | | | | | | | | | | | | | | | | | | | | During testing/extension of the linux-yocto-tiny kernel it was found that defconfigs were not always properly applied. This was due to two issues: - not being able to fully control the order of objects applied to the git tree on the SRC_URI - defconfigs triggering --allnoconfig before being applied To fix this, the recipe space code that previously detected and generated automatic features moves back to the kernel tools (where it was before) and is updated to also process .cfg and defconfigs. Moving this back to the tools allow other recipes to automatically benefit from the additional support. The second issue is addressed by allowing configme to take --alldefconfig when a recipe wishes to pass a defconfig and override the default behaviour. Fixes [YOCTO: 2250] Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* linux-yocto/3.2: add igb support to romleyBruce Ashfield2012-04-133-3/+3
| | | | | | | | | | | | | | | | Updating the 3.2 recipe SRCREVs to pickup the following meta change: [ meta: Add igb.scc to Romley Romley machine has 82580 Giga bit Ethernet Controller. Add the relavent Nic driver to it. Signed-off-by: Kishore Bodke <kishore.k.bodke@intel.com> ] Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* Revert "eglibc: fix perl path in target scripts on fedora 17 alpha"Richard Purdie2012-04-133-4/+2
| | | | | | | | | | | This reverts commit 7c95d554c431451f975e1e5d0336d1fb7d0ce7ae. eglibc uses the PERL to execute things as well as encoding the path in target scripts. We will therefore have to resort to sed on the scripts at do_install time to fix the problems originally reported. This patch causes various weird build failures and needs to be removed. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* rpm-native: Compile python rpm module (with-python)Andrei Gherzan2012-04-131-3/+1
| | | | | | | | | | This module is needed by python scripts from createrepo native package. [YOCTO #2134] RP: Add missing DEPENDS on python-native Signed-off-by: Andrei Gherzan <andrei@gherzan.ro> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* package-index: Force NATIVE python to use modules from STAGING_DIR_NATIVEAndrei Gherzan2012-04-131-0/+3
| | | | | | | | | | Export PYTHONHOME to ${STAGING_DIR_NATIVE}/usr. In this way python will search for modules in native sysroot. [YOCTO #2134] Signed-off-by: Andrei Gherzan <andrei@gherzan.ro> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* createrepo: Python scripts should use the python interpreter from envAndrei Gherzan2012-04-132-1/+49
| | | | | | | | | | | | Added path to modify the hardcoded path of python interpreter from python scripts. The python interpreter should be the one from native sysroot and not the one from host machine. [YOCTO #2134] Signed-off-by: Andrei Gherzan <andrei@gherzan.ro> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* eglibc: fix perl path in target scripts on fedora 17 alphaNitin A Kamble2012-04-133-2/+4
| | | | | | | | | | | | | fedora 17 has /usr/bin/perl as well as /bin/perl, which results in the perl path set in thetarget perl scripts as /bin/perl While we install perl on target at ${bindir}/perl i.e. /usr/bin/perl Hence the target perl scripts are broken when build is done on fedora 17. Work around the issue by providing cached value of perl path to eglibc This fixes bug [YOCTO #2286] Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* eglibc: fix re-execution of taskVenkata ramana gollamudi2012-04-132-8/+8
| | | | | | | | | | | | | Task do_patch_append calling do_fix_ia_headers is removing files using "rm" not "rm -f". So first time execution of patch task is success, while re-execution of patch task fails as it tries to remove the files already removed. So changed "rm" to "rm -f". [Yocto #2194] Signed-off-by: Venkata Ramana Gollamudi <ramana.gollamudi@huawei.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* boost: fix re-execution of taskVenkata ramana gollamudi2012-04-131-1/+5
| | | | | | | | | | | | | | | After building boost package, re-execution of boostconfig task followed by re-execution of compile task is giving following error "error: duplicate initialization of gcc with the following parameters" during compilation It is because multiple entries of gcc are being added during boostconfig re-execution there by failing the compilation. The patch fixes adding multiple "Using gcc" entries into /tools/build/v2/user-config.jam [Yocto #2194] Signed-off-by: Venkata Ramana Gollamudi <ramana.gollamudi@huawei.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* PR bump packages with gdbm in DEPENDSAndrei Gherzan2012-04-134-4/+4
| | | | | | | | This is done because of this change in gdbm: "gdbm: Package compat libs in gdbm-compat" Signed-off-by: Andrei Gherzan <andrei@gherzan.ro> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* gdbm: Package compat libs in gdbm-compatAndrei Gherzan2012-04-131-1/+6
| | | | | | | | In order to avoid breaking packages which depend on old package name libgdbm4 (>= 1.10), compat libs are packaged into a separate package named gdbm-compat. Signed-off-by: Andrei Gherzan <andrei@gherzan.ro> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* Revert "default-distrovars: remove NO32LIBS setting"Richard Purdie2012-04-131-0/+2
| | | | | | | We still need this value to keep the sanity testing working. The other part of the change is still valid. This reverts commit cb940f46efacddc7200581edba034e4201abb94a.
* opkg-utils: Update to include a fix for version comparision issuesRichard Purdie2012-04-131-2/+2
| | | | | | | | | This brings the version comparision within opkg-utils into sync with that in opkg itself. [YOCTO #2233] Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* gcc-4.6: Add fix for relocation problem and ccacheRichard Purdie2012-04-132-1/+42
| | | | | | | | | | | | | | | | | If the toolchain is reused from sstate and ccache is installed, build failures were occuring due to gcc trying to access the original sysroot rather than the new one, particularly if the old sysroot existed but was not readable by the current user. This turns out of the an issue inside gcc to do with preservation of the sysroot option. See the gcc patch for more details. It only triggers when preprocessed sources are used which happens when ccache is used. The same issue occurs with c++ and c++-cpp-output so the same fix is applied there. [YOCTO #2074] Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* default-distrovars: remove NO32LIBS settingPaul Eggleton2012-04-131-2/+0
| | | | | | | | The ??= assignment in pseudo.inc effectively nullifies this ??= assignment here, so remove it. Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* pseudo: default NO32LIBS to 1Paul Eggleton2012-04-131-1/+1
| | | | | | | | | | If this value is not set to 1, then systems with some 32-bit libraries but no 32-bit version of libgcc installed will have pseudo-native fail at do_compile. It should only really be set to 0 by those who know what they are doing. Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* base.bbclass: Fix PACKAGECONFIG issues with native and nativesdk ↵Richard Purdie2012-04-131-5/+31
| | | | | | | | | | | | | | | | | | | | | | | BBCLASSEXTEND recipes (and multilib) This patch fixes up the issues that were being seen where BBCLASSEXTEND and PACKAGECONFIG were interacting badly. It also ensures PACKAGECONFIG interacts properly with multilib builds. Ideally some of this code will be abstracted into lib/oe/classextend.py but at this point in release more invasive changes like this are inappropriate. This patch also removed empty strings from expressions rather than passing them around as this was complicating the additional code unnecessarily. The patch was verified against the OE-Core metadata where the return values of expandFilter() were sanity checked by hand for native/nativesdk and multilib combinations. [YOCTO #2225] Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* mklibs-native 0.1.33: include unistd.h to fix build for gcc 4.7Robert Yang2012-04-132-1/+45
| | | | | | | | | | | | | | | | | | Many of the standard C++ library include files have been edited to no longer include <unistd.h> to remove namespace pollution in gcc 4.7, so we need include <unistd.h> manually sometimes. More information: http://gcc.gnu.org/gcc-4.7/porting_to.html [YOCTO #2278] Signed-off-by: Robert Yang <liezhi.yang@windriver.com> [Updated patch header - sgw] Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* rpm: Ensure that we check both providename and filepathsMark Hatle2012-04-132-1/+38
| | | | | | | | | | | In lib/rpmts.c we need to make sure to look for missing dependencies, such as /bin/sh, in both the providename and filepaths DB. Previously the system would key off an initial '/' and only look in the filepaths DB. This causes problems when a package such as "bash", has a Provides: /bin/sh. Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* pseudo: Tell pseudo to avoid specifying an RPATHMark Hatle2012-04-133-6/+6
| | | | | | | | | | [Yocto #2251] Add --without-rpath to avoid embedding rpaths into the pseudo components. Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* package_rpm.bbclass: Set tmppath for rpm to somewhere which won't conflict ↵Richard Purdie2012-04-121-0/+1
| | | | | | | | | | | | | | | with the rootfs If the rootfs has a specific layout in mind for /var/tmp, it should be free to create this. Preinst scripts trigger rpm to create its tmppath directory which could potentially conflict with this. We're seeing issues if the provider of the /var/tmp wants to use a symlink and a preinst script executes before it has been installed. This change places tmppath as used by rpm somewhere safe which is already automatically cleaned up too. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* libunistring: Fix parallel make issueRichard Purdie2012-04-122-2/+29
| | | | | | See patch header for the full description of the parallel make issue this resolves. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* linux-dtb: Ensure dtb files are covered by sstateRichard Purdie2012-04-121-3/+3
| | | | | | | | | | | The dtb files were not being installed into a location compatible with sstate and the do_deploy task. This means in builds just using sstate, the dtb files disappeared. This patch fixes the code to use the correct location for deploy files. [YOCTO #2190] Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* package_rpm.bbclass: Use the correct macros file to avoid empty solvedb path ↵Richard Purdie2012-04-121-2/+1
| | | | | | issues Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* package_rpm: Fix useradd preinst ordering issuesRichard Purdie2012-04-121-16/+47
| | | | | | | | | | | | | | | | | | | | | | | | | | We were already having occasional ordering issues with package_rpm. Fixing the ldconfig postinstall issue pushed rpm over the cliff and totally broke rpm builds with the packages getting installed in effectively a random order and the useradd preinstalls getting executed out of order and breaking. The only explanation I can find for this is that rpm is special. It will happily run a preinst for a package without any of that package's dependencies being present regardless of whether there are any circular dependency issues or not. I attempted various ways of solving this such as ordering the total_solution.manifest in creative ways but the bottom line is RPM ignores this. It takes little account of any request to ignore /bin/sh dependencies for the purposes of constructing the final image. The end result is we're having to install the base-passwd, base-files and shadow packages first (if there is a request to install them), then install any other packages. It this wasn't in the middle of a release I'd be rewriting this bbclass file, its horrible. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* distro-tracking: Fix syslinux NO_UPDATE_REASONSaul Wold2012-04-111-1/+1
| | | | | Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* insane.bbclass: Add missing variable bpnNoor Ahsan2012-04-111-0/+1
| | | | | | | * bpn variable was used but it was not defined. Add bpn to resolve that issue. Signed-off-by: Noor Ahsan <noor_ahsan@mentor.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* "The suite of statements in a function definition executes with a local ↵Andrei Gherzan2012-04-111-1/+4
| | | | | | | | | | | | | | | | | | namespace that is different from the global namespace. This means that all variables created within a function are local to that function. When the suite finishes, these working variables are discarded." In this way the needs_ldconfig variable in linux_so never gets True in the statements below this function. As global statement is generally discouraged, a return value would be a clean and fast way to solve this issue. [YOCTO #2205] RP: Added logic to ensure the value doesn't get overwritten once set Signed-off-by: Andrei Gherzan <andrei@gherzan.ro>--- Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* python: multilib header support.Lianhao Lu2012-04-114-4/+6
| | | | | | | | Add intercept multilib header for pyconfig.h in python. This is part of the bug fixing [YOCTO #2216]. Signed-off-by: Lianhao Lu <lianhao.lu@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* initrdscripts: fix init-live.sh and use unionfsYang Shi2012-04-112-7/+30
| | | | | | | | | | | | | | | | [YOCTO #1487] When booting up with liveCD image, init scripts can't work well on read-only filesystem. Unionfs, which is supported in Yocto kernel, allows a filesystem to appear as writeable, but without actually allowing writes to change the filesystem. Use unionfs to mount rootfs and make root file system can be writen when using liveCD to boot up. Set UNION_FS variable depending on kernel config, so that it can work with kernel which doesn't have unionfs feature. [RP: Mark recipe as machine specific due to kernel dependency] Signed-off-by: Yang Shi <yang.shi@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* tclibc-eglibc.inc: make locale packages dependency conditionalNitin A Kamble2012-04-112-8/+17
| | | | | | | | | | Only add locale package dependencies if the eglibc is configured with locale support. This avoids dependencies issues for distros such as poky-tiny [RP: Add PR bump] Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* python: Fix host contamination issueRichard Purdie2012-04-111-1/+4
| | | | | | | | | | | | | | | | | | python-nativesdk could fail with: | /srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-ppc/build/build/tmp/sysroots/x86_64-linux/usr/bin/python: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by /srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-ppc/build/build/tmp/work/x86_64-nativesdk-pokysdk-linux/python-nativesdk-2.7.2-r1.9/Python-2.7.2/libpython2.7.so.1.0) | make: *** [sharedmods] Error 1 which is caused by the fact LD_LIBRARY_PATH is being set to include WORKDIR when calling HOSTPYTHON. HOSTPYTHON is from python-native and needs no such help so the easiest fix here is not to set crazy LD_LIBRARY_PATH values since we should never be running anything from WORKDIR given the way we build things. The patch clears out the RUNSHARED variable which would contain the LD_LIBRARY_PATH value prior to this patch. Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* archiver.bbclass: Amend the problem for moving tarballs in ${DL_DIR} to ↵Xiaofeng Yan2012-04-111-0/+1
| | | | | | | | | | | | | | | | | | ${DEPLOY_DIR}/source when enable archiver When running "bitbake core-imamge-minmal", the error information like the following will appear: ERROR: Error executing a python function in /buildarea2/yzhao-test/poky-test/meta/recipes-core/zlib/zlib_1.2.6.bb: IOError: [Errno 2] No such file or directory: '/buildarea2/yzhao-test/poky-test/build-archive/downloads/zlib-1.2.6.tar.bz2' An absolute path from variable "file" pointed to tarballs in ${DL_DIR} \ cause this problem. So return base-name of "file" for fixing this bug here. [YOCTO #2272] Signed-off-by: Xiaofeng Yan <xiaofeng.yan@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* busybox: fix options -b, -a and -PAndreas Oberritter2012-04-112-7/+14
| | | | | | | | | * busybox-udhcpc-no_deconfig.patch broke the options, preventing udhcpc from forking into the background if no lease can be obtained. Signed-off-by: Andreas Oberritter <obi@opendreambox.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* classes/package_deb: create .gz index instead of .bz2Paul Eggleton2012-04-111-1/+1
| | | | | | | | | | | | | | | apt is looking for Packages.gz files instead of the .bz2 files we are currently creating and failing when they cannot be found. It is not immediately obvious how to make the current version use the .bz2 indexes; thus create .gz indexes for now which allows us to successfully create images. Tested on both a Fedora 14 and an Ubuntu 11.10 host machine. Fixes [YOCTO #1858]. Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* task-self-hosted: Add vte and eglibc-gconv-ibm850Saul Wold2012-04-111-1/+3
| | | | | | | | Adding vte for a more full featured terminal adding eglibc-gconv-ibm850 for mcopy Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* boot-directdisk: remove the default setting of SYSLINUXCFG and SYSLINUXMENUSaul Wold2012-04-111-3/+0
| | | | | | | | | These two should get setup in {S} and installed into {HDDDIR}, by having them defined with {HDDDIR} they try to get created when the directory does not exist yet. Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* connman: add ofono dependency.Lianhao Lu2012-04-111-0/+1
| | | | | | | | | | | | Added dependency to ofono because we always enable ofono plug-in. We did this because the 1.2 release cycle is coming to an end. We should use PACKAGECONFIG in 1.3. This is part of the 1.2 bug fixing [YOCTO #2216]. Signed-off-by: Lianhao Lu <lianhao.lu@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* hello-mod: Move hello-mod from meta to meta-skeletonDarren Hart2012-04-104-0/+0
| | | | | | | | | | Fixes [YOCTO #1501] hello-mod is an example kernel module, and does not provide any real functionality. As such, it would be better placed under meta-skeleton than meta. Signed-off-by: Darren Hart <dvhart@linux.intel.com>
* syslinux: Update distro-tracking fieldsDarren Hart2012-04-101-2/+3
| | | | | | Add reason for not updating syslinux this release. Signed-off-by: Darren Hart <dvhart@linux.intel.com>
* at: Fix origin location of install pam.conf from from ${P} -> ${BP}Saul Wold2012-04-101-2/+2
| | | | | | | | | This patch changes P to BP to address where a file is installed from when building with PAM enabled and using multi-lib. [YOCTO #2224] Signed-off-by: Saul Wold <sgw@linux.intel.com>
* gdb: add --with-libexpat-prefixMartin Jansa2012-04-101-2/+2
| | | | | | | * otherwise it sometimes finds host's expat Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
* opkg: Add the condition for the content of arch.conf when enable multilibXiaofeng Yan2012-04-101-1/+6
| | | | | | | | | | | | | | | | After successfully installed some lib32 multilib packages into the x86-64 image, we just found that the file content of /var/lib/opkg/status in rootfs changed after the very 1st boot, many lib32 related packages information are missing in that file. The missing arch "x86" in arch.conf cause the above problem. Adding the condition for the content of arch.conf when enable multilib. If build multilib image, "ALL_MULTILIB_PACKAGE_ARCHS" will be used instead of "PACKAGE_ARCHS". [YOCTO #1522] Signed-off-by: Xiaofeng Yan <xiaofeng.yan@windriver.com>
* adt-installer: Fixed ppc kernel naming.Lianhao Lu2012-04-103-13/+13
| | | | | | | | | | | 1. Fixed the ppc kernel naming. 2. Disabled opkg shared library to avoid runtime opkg-cl launching error. 3. Adjusted the variable sequence in adt-installer.conf Fixed bug [YOCTO #2233] Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
* meta-toolchain: runqemu falied on FC16/Opensuse12.1 x86_64Robert Yang2012-04-101-2/+2
| | | | | | | | | | | | | | | runqemu can't launch a target image on Fedora 16 64bit or Opensuse 12.1 64bit, this is because runqemu needs the host's libGL.so, which requires GLIBC_2.14 which is defined in libc.so.6, but our default libc.so.6 is version 2.13, here is the message from Richard: The easiest solution would be to change the nativesdk libc to 2.15. I don't think we plan to do this for the target libc for 1.2 but we could change nativesdk's version if its well tested [YOCTO #1968] Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
* git 1.7.7: remove perl.mak before compileRobert Yang2012-04-101-0/+6
| | | | | | | | | | | | | | | | | The git may fail to rebuild when perl's Config.pm or config.h changes, this is because Makefile detects that perl/perl.mak is out of date. Remove perl.mak to let Makefile regenerate it would fix the error. Both git and git-native have this problem. To reproduce the error: (On x86_64 host) $ bitbake git-native $ touch tmp/sysroots/x86_64-linux/usr/lib/perl-native/perl/5.14.2/Config.pm $ bitbake git-native -ccompile -f [YOCTO #2156] Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
* grub-efi: Include GPT partition EXT FS supportDarren Hart2012-04-101-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes [YOCTO 2257] GPT partitions are common for EFI systems. Add support for them by including the part_gpt partition module in the grub-efi image. In order to allow for loading a Linux kernel from an EXT* filesystem, include the ext2 module as well. With this fix applied, I was able to boot from a USB key using a GPT partition table with the following layout: $ sudo gdisk -l /dev/sdc GPT fdisk (gdisk) version 0.8.2 Partition table scan: MBR: protective BSD: not present APM: not present GPT: present Found valid GPT with protective MBR; using GPT. Disk /dev/sdc: 7669824 sectors, 3.7 GiB Logical sector size: 512 bytes Disk identifier (GUID): 68FA7CD4-E0C3-4A8E-82B5-1331C9B17A3C Partition table holds up to 128 entries First usable sector is 34, last usable sector is 7669790 Partitions will be aligned on 2-sector boundaries Total free space is 7428816 sectors (3.5 GiB) Number Start (sector) End (sector) Size Code Name 1 34 32801 16.0 MiB 0700 # FAT16 2 32802 240974 101.6 MiB 0700 # EXT3 From within GRUB, booted as bootia32.efi from the BOOT partition, I booted the OS with the following commands: grub> linux (hd0,gpt2)/vmlinuz rootwait root=/dev/sda2 console=ttyS0,115200 grub> boot This change will enable BSP developers to use the grub-efi image in their own images as well as enable upcoming changes to the installer to support EFI. Signed-off-by: Darren Hart <dvhart@linux.intel.com> CC: Kishore Bodke <kishore.k.bodke@intel.com> CC: Rahul Saxena <rahul.saxena@intel.com>