summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* bitbake/fetcher: Deal with a ton of different bugsRichard Purdie2010-10-192-36/+38
| | | | | | | | | | | | | | | | | | | | | | The more we try and patch up the fetcher code, the more things break. The code blocks in question are practically unreadable and are full of corner cases where fetching could fail. In summary the issues noticed included: a) Always fetching strange broken urls from the premirror for "noclone" git repositories b) Not creating or rewriting .md5 stamp files inconsistently c) Always fetching git source mirror tarballs from the premirror even if they already exist but the checkout directory does now d) Passing "None" values to os.access() and os.path.extsts() checks under certain circumstances e) Not using fetched git mirror tarballs if the preexist and always try and fetch them. This patch rewrites the sections of code in question to be simpler and more readable, fixing the above problems and most likely other odd corner cases. Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* bitbake/fetch/git.py: Fix git fetcher to correctly use mirror tarballsRichard Purdie2010-10-191-1/+5
| | | | Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* rpm: Change pcre to use internal library [BUGID #404]Saul Wold2010-10-181-2/+2
| | | | Signed-off-by: Saul Wold <Saul.Wold@intel.com>
* sed: use ALTERNATIVE_LINKKevin Tian2010-10-182-2/+4
| | | | | | | | | | | | | update-alternatives.bbclass assumes /usr/bin/sed if ALTERNATIVE_LINK is not specified, however sed has changed its default installation directory to /bin/sed. This causes below errors at the first boot of lsb image: update-alternatives: Error: cannot register alternative sed to /usr/bin/sed since it is already registered to /bin/sed This fixes [BUGID #478] Signed-off-by: Kevin Tian <kevin.tian@intel.com>
* gstreamer: add support for theora pluginsDongxiao Xu2010-10-184-4/+32
| | | | | | | Add theora codec and its corresponding lib recipe (import from OE). This fixes [BUGID #476]. Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
* wrs_meta: update SRCREV for netfilter optionsBruce Ashfield2010-10-181-1/+1
| | | | | | | | | | | Fixes [BUGID #199] Updating the SRCREV for wrs_meta to import a set of default netfilter options, which can then be applied to any BSP. This keeps the default BSP configs small, while allowing the functionality to be available. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* kernel tools: update SRCREVBruce Ashfield2010-10-181-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | Fixes [BUGID #199] To allow optional features to be enabled for a given BSP, we need to re-enable this functionality in the kernel tools. The following commit in that repo fixes a problem when passing feature names: [ commit 140693a6b0d81b7ba7175b6cfce11c6c22f81e24 Author: Bruce Ashfield <bruce.ashfield@windriver.com> Date: Sun Oct 17 01:22:00 2010 -0400 updateme: do not expand directories If a directory is passed to updateme, do NOT expand it to a full path, the compilation phase will search and do this while applying some special logic to find properly named features. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> ] Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* linux-wrs: pass netfilter options as an addon featureBruce Ashfield2010-10-181-0/+1
| | | | | | | | | | | | Fixes [BUGID #199] Although an optional feature to a base BSP, most of the machines currently supported want/need the netfilter options. To enforce this configuration, it is temporarily forced in the kernel recipe itself, but can move to machine configurations in the future. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* linux-wrs: allow optional features via KERNEL_FEATURES variableBruce Ashfield2010-10-181-2/+5
| | | | | | | | | | | | | | | | | | | | Fixes [BUGID #199] Reinstate the ability to specify optional/additional kernel features when updating the tree. This is done via the variable KERNEL_FEATURES which specifies a list of features to be appended to the current branch and config. These features are part of the wrs_meta branch in the kernel repository and hence are self contained within the kernel tree waiting to be activated. This saves multiple branches simply to allow a machine to have many profiles. The kernel patching/configuration phases will locate these features and add them to the meta_series, which in turn modifies the tree. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* distrodata: Modify Distro tracking spreadsheetSaul Wold2010-10-153-19/+60
| | | | Signed-off-by: Saul Wold <Saul.Wold@intel.com>
* distro_tracking: Update Distro AliasesSaul Wold2010-10-151-32/+60
| | | | Signed-off-by: Saul Wold <Saul.Wold@intel.com>
* task-poky-sdk: Add lttng-ustRichard Purdie2010-10-153-2/+11
| | | | Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* libtasn1: Comment out the RREPLACES as it causes problems with rpm backendRichard Purdie2010-10-151-2/+2
| | | | | | | | | Since we've never distributed binary feeds this isn't a problem for now for Poky and as it stands this line breaks images. [BUGID #442] Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* poky.conf: Add qemu-config dependency for qemux86-64Richard Purdie2010-10-151-0/+1
| | | | | | | | | After booting qemux86-64 sato image, there is no icon for shutdown on the desktop which is caused by missing the qemu-config package on this machine. [BUGID #458] Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* shared-mime-info: Don't run the util for the native version of the packageRichard Purdie2010-10-151-5/+11
| | | | | | | | | | | The whole point of building a native version of this package is to provide the share-mime-info binary so using it in do_install of the native package is incorrect. Add a modified do_install for the native version of the package to avoid this. [BUGID #466] Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* documentation/poky-ref-manual: Various tweaks to the textRichard Purdie2010-10-159-55/+39
| | | | Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* documentation/tools: Add comments to scriptRichard Purdie2010-10-151-1/+3
| | | | Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* libffi: fix the uriJoshua Lock2010-10-151-1/+1
| | | | | | It had the previous version hard-coded Signed-off-by: Joshua Lock <josh@linux.intel.com>
* documentation: Complete poky-handbook -> poky-ref-manual renameRichard Purdie2010-10-157-27/+14
| | | | Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* documentation/handbook: Tweak Makefiles and paths after path changesRichard Purdie2010-10-155-8/+12
| | | | Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* handbook/poky-ref-manual: Makefile tweaksRichard Purdie2010-10-151-2/+2
| | | | Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* handbook: Move into documentation directoryRichard Purdie2010-10-1546-43/+5
| | | | Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* Use gdk_threads_enter/gdk_threads_leave to ensure gtk multithread safeDongxiao Xu2010-10-159-1/+40
| | | | | | | | | | | If the video player tries to play an unknown type file, it will enter the error_cb() function, which is called in another thread. Use gdk_threads_enter/gdk_threads_leave to ensure the safe of gtk multi-thread operation. This fixes [BUGID #474] Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
* libffi: bumpt to version 3.0.9Joshua Lock2010-10-151-0/+0
| | | | | | 3.0.8 is incompatible with the autotools versions we use in Poky Signed-off-by: Joshua Lock <josh@linux.intel.com>
* make: enable make-nativeJoshua Lock2010-10-151-0/+2
| | | | | | | | | Fedora 14 ships with a version of make which doesn't work with the Makefiles shipped in our busybox and eglibc versions. Enable make native as a workaround. This is a low risk fix for [BUGID #467] Signed-off-by: Joshua Lock <josh@linux.intel.com>
* handbook: Fix the Ubuntu version requirementScott Rifenbark2010-10-151-1/+1
| | | | Signed-off-by: Scott Rifenbark <scott.m.rifenbark@intel.com>
* handbook: Fix section numberingScott Rifenbark2010-10-151-0/+1
| | | | Signed-off-by: Scott Rifenbark <scott.m.rifenbark@intel.com>
* atom-pc: remove kernel video mode parametersDarren Hart2010-10-151-1/+1
| | | | | | | | | | | | | | | | | The kernel video mode parameters are not needed with the atom-pc linux-wrs configuration. Without them, we boot up with a standard 80x25 text display until the Intel i915 driver is initialized, at which point we switch to a higher resolution text mode. Afterwards, a sato image will start the splash screen and successfully boot to X. Tested on the Toshiba NB305 netbook and the Black Sand development board. [BUGID #450] Signed-off-by: Darren Hart <dvhart@linux.intel.com> CC: Tom Zanussi <tom.zanussi@intel.com> CC: Bruce Ashfield <bruce.ashfield@windriver.com> CC: Richard Purdie <rpurdie@linux.intel.com>
* linux-wrs: update SRCREV for mpc8315e-rdbBruce Ashfield2010-10-151-1/+1
| | | | | | | | | | | Fixes [BUGID: 423] The mpc8315e-rdb BSP failed to build due to -Os being used in the boot code. The fix that doesn't involve changing the compiler is to remove -Os in the powerpc code and keep it contained on the BSP branch. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* gupnp: fix broken SRC_URIsScott Garman2010-10-155-5/+5
| | | | | | This fixes [BUGID #469] Signed-off-by: Scott Garman <scott.a.garman@intel.com>
* emenlow: add 8250 serial optionsBruce Ashfield2010-10-142-2/+2
| | | | | | | | The emenlow didn't have the default 8250 serial ports. Including the common serial config fragment fixes this problem. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* poky-qemu-internal: make the -no-reboot qemu option mandatoryScott Garman2010-10-141-2/+3
| | | | | | This fixes [BUGID #100] Signed-off-by: Scott Garman <scott.a.garman@intel.com>
* poky-qemu: miscellaneous bugfixesScott Garman2010-10-142-14/+19
| | | | | | | | | | | In scenarios where the POKY_NATIVE_SYSROOT env variable hasn't been set up, bug #427 can still be triggered. This fixes it by running setup_tmpdir(). This fixes [BUGID #427]. Also, the qemu tap dev lock directory needs to be chmod 777 so that multiple users can create and delete lock files. Signed-off-by: Scott Garman <scott.a.garman@intel.com>
* poky-qemu: add nographic option for disabling video consoleScott Garman2010-10-141-4/+6
| | | | | | This fixes [BUGID #408] Signed-off-by: Scott Garman <scott.a.garman@intel.com>
* atom-pc: get rid of undefined video mode prompt on bootupTom Zanussi2010-10-141-1/+1
| | | | | | | | | | Temporary fix for [BUGID #450] This fixes the video mode number to the expected value for this video setting, but the setting itself needs to be fixed along with matching config options - this patch is just a temporary fix for RC2. Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
* sysvinit: add check for no X11 to disbale psplashSaul Wold2010-10-141-6/+6
| | | | | | | | | | Add a check to the rc script to check if the /etc/init.d/xserver-nodm script exists and is executable and not disable psplash if it is, otherwise disable pspalsh since we do not have X installed. Fixed [BUG #457] Signed-off-by: Saul Wold <Saul.Wold@intel.com>
* task-poky-qt: Re-Enable MIPS Build and disable fotowallSaul Wold2010-10-141-3/+2
| | | | | | | This address a number if issues with fotowall not being ready and [BUGID #462] Qt missing for Mips build Signed-off-by: Saul Wold <Saul.Wold@intel.com>
* bitbake-runtask: Ensure logging settings take effect in the worker process ↵Richard Purdie2010-10-142-1/+5
| | | | | | (and pass verbose setting) Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* pseudo: Update to revision 489ed5e89f040febcb605f22187968f597211205Richard Purdie2010-10-141-1/+1
| | | | | | | | | This fixes various issues with modern coreutils which use *xattr() syscalls which pseudo wasn't catching causing permissions problems. [BUGID #463] Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* eglibc-nativesdk: Add patch to change the search path of the dynamic linkerRichard Purdie2010-10-142-1/+56
| | | | | | | | | Change the search path of the dynamic linker to search ld.so.cache as a last resort when trying to find libraries at runtime. This means libs in /opt/poky are used in preference but any host libs such as libGL can also be found. See the patch for a more detailed description. Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* Revert "meta-toolchain: Update the way ld.so.cache is handled"Richard Purdie2010-10-142-13/+2
| | | | | | | This change didn't work as expected as ld.so.cache doesn't preseve any search path ordering, therefore revert the change. This reverts commit b2c3ce6d17520d43b10777dd306449876cb40819.
* linux-wrs: update SRCREVsBruce Ashfield2010-10-132-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes [BUGID #432, #438, #437, #422] [atom-pc: switch to the atom-pc branch The atom-pc has a dedicated branch and merged kernel commits. Swtich the kernel recipe to use the fully integrated BSP Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>] [atom-pc: add kernel config options needed for -live images Fixes [BUGID #432] Live images require VFAT filesystem and loop device support - this adds the required kernel config options for them. Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>] [atom-pc: Add netbook wireless networking support for NB305 and eee901 Add the atom-pc-wifi.cfg to enable wireless networking support for the netbooks mentioned in the atom-pc machine config file. Note: this adds a staging tree driver (rt2860sta) for the eee901 wifi. [BUGID #438] Signed-off-by: Darren Hart <dvhart@linux.intel.com> CC: Tom Zanussi <tom.zanussi@intel.com> CC: Richard Purdie <rpurdie@linux.intel.com> CC: Bruce Ashfield <bruce.ashfield@windriver.com>] [atom-pc: Add netbook wired networking support for NB305, Aspire One, and eee901 Add the atom-pc-eth.cfg to enable wired networking support for the three netbooks mentioned in the atom-pc machine config file. [BUGID #437] Signed-off-by: Darren Hart <dvhart@linux.intel.com> CC: Tom Zanussi <tom.zanussi@intel.com> CC: Richard Purdie <rpurdie@linux.intel.com> CC: Bruce Ashfield <bruce.ashfield@windriver.com>] [routerstationpro: add missing kernel patches [BUGID: #422] Add the kernel patches required for routerstaion pro boot Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>] [netbook: allow hardisk booting [BUGID: #445] Signed-off-by: Darren Hart <dvhart@linux.intel.com>] Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* atom-pc: switch to the atom-pc branchBruce Ashfield2010-10-133-31/+2
| | | | | | | The atom-pc has a dedicated branch and merged kernel commits. Swtich the kernel recipe to use the fully integrated BSP Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* emenlow/linux-wrs: merge emenlow configurationBruce Ashfield2010-10-134-119/+1
| | | | | | | | | | Complete the merge of the emenlow BSP, by removing the external configurations from the emelnow layer and merging them into the wrs_meta branch of the kernel tree. This also allows the emenlow to drop the tools config fragement and use the common definitions shared by BSPs maintained in the kernel. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* linux-wrs: remove external x86 config fragmentBruce Ashfield2010-10-133-5/+1
| | | | | | | | | [BUGID #363] Integrate the x86 gfx options into the existing x86 graphics options for the common_pc and common_pc_64 boards. Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
* emenlow: update formfactor keyboard settingTom Zanussi2010-10-131-0/+2
| | | | | | | | | Fixes [BUGID #447] We expect the emenlow to normally have a keyboard so tell the formfactor so and avoid the keyboard pop-up. Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
* libowl-av: Fix music player crash if adding unknown type filesDongxiao Xu2010-10-132-2/+43
| | | | | | | | | | | | | Using bus message to call function flush_head/feed_head in main thread to protect private data operation. A better fix may be to listen to bus message "CODEC_NOT_FOUND" and "MISSING_PLUGIN" directly to avoid hooking into "unknown-type" signal. We will revisit it in next period. This fixes [BUGID #296] Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
* tzdata: install zone.tab and iso1366.tabZhai Edwin2010-10-131-1/+7
| | | | | | | | | | | | These files are removed in libc-package.bbclass. Missing of zone.tab cause get_zone_directory in libical return NULL for strlen, which leads seg fault of dates & tasks. Openembedded has similar patch. [BUGID #420] got fixed by this. Signed-off-by: Zhai Edwin <edwin.zhai@intel.com>
* meta-toolchain: Update the way ld.so.cache is handledRichard Purdie2010-10-132-2/+13
| | | | | | | | Fix the library search path so that libs in the toolchain in /opt/ have priority over those in base system but ones from the base system can still be found. Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
* sanity.bbclass: ensure 32it libc is correctly detectedJoshua Lock2010-10-131-1/+5
| | | | | | | | Fix to work for both Ubuntu and Fedora multilib paths. Fixes [BUGID #454] Signed-off-by: Joshua Lock <josh@linux.intel.com>