Age | Commit message (Collapse) | Author | Files |
|
Poky)
|
|
I didn't touch the to be installed files, as we have GNU coreutils or
at least a compatible busybox version on our targets so far.
|
|
|
|
|
|
* there are both libgcc_s.so and libgcc_s.so.${someversion} in
${CROSS_DIR}/${TARGET_SYS}/lib
|
|
(from Poky)
|
|
|
|
use the locales from WORKDIR instead of /usr, closes #1821
|
|
GLIBC_GENERATE_LOCALES
to limit set of generated binary locales.
* If ENABLE_BINARY_LOCALE_GENERATION is set to "1", you can limit locales
generated to the list provided by GLIBC_GENERATE_LOCALES. This is huge
time-savior for developmental builds. Format: list of locale.encoding pairs
with spaces as separators, e.g.:
GLIBC_GENERATE_LOCALES = "en_GB.UTF-8 de_DE.UTF-8"
* Check your distro for minimal set of locales required. Old-time default is
"en_GB.UTF-8 de_DE.UTF-8 fr_FR.UTF-8".
* Angstrom requires en_GB only.
* This commit does not changes the default, which is still generation of all
available locales.
* Closes #1966.
|
|
only semantic change is what is being discussed in bug 1987
|
|
|
|
Oh use tabs for this method...
|
|
Use spaces in anonfunc, needed for BitBake trunk
to speedup parsing a bit
|
|
|
|
|
|
|
|
which allows to override kernel version report to guest apps (otherwise,
host kernel version is used, which causes "Kernel too old" errors in glibc
startup code in guess, if host kernel is older than 2.6.14-something).
Tested with qemu 0.8.2 (and yes, I couldn't find -r switch in qemu doc,
taken from a mail by Paul Brook, qemu co-maintainer).
|
|
|
|
|
|
|
|
|
|
which is clearer. Use .../* to match files and ignore directories.
|
|
it. Bump the PR on all the glibc files since they all include
glibc-package.bbclass. Closes #1220 "glibc ships spurious debug info".
|
|
|
|
|
|
stuff in it. Fixes i18n for non pregenerated locales
|
|
#729.
|
|
defaults to disabled. reinstate BINARY_LOCALE_ARCHES and make it a list of regular expressions.
- against contrary belief BINARY_LOCALE_ARCHES = "" did disable the new code.
- zapping that variable in glibc-package.bbclass as the changeset before
this did is not an option since we need to keep track of which arches
are supposed to have this enabled.
- also, rename USE_BINARY_LOCALE to GLIBC_INTERNAL_USE_BINARY_LOCALE to
prevent people who don't understand Python from shooting themselves in the
foot. DONT MESS WITH THAT VARIABLE. setting USE_BINARY_LOCALE = "0" will
_enable_ the new code since "0" (being a string) evaluates to True in
Python.
|
|
|
|
using qemu. based on a patch originally written by Phil Blundell. fixes hh.org Bug #1385 (localedef OOM issue).
- running localedef (in qemu...) for every locale takes a while. set
BINARY_LOCALE_ARCHES = "" to avoid the new behaviour.
|
|
|
|
into mebius.reciva.com:/home/pb/oe/oe
2005/04/23 08:57:40+01:00 reciva.com!pb
remove DEPENDS bogosity introduced in last patch
BKrev: 426a0021jAVi0ff_4yR6eV_d2ZCHqg
|
|
into mnz66.mn-solutions.de:/usr/src/mnci54/oe
2005/04/21 08:03:47+01:00 reciva.com!pb
move some more common glibc bits into glibc-package.bbclass
BKrev: 42675160QJABzuRNjNcLgszukeMUTQ
|
|
typo for catchsegv
BKrev: 42649657zxqJhJDJfG8f5EOHIyzHYA
|
|
into hyperion.kergoth.com:/home/kergoth/code/openembedded
2004/12/09 03:39:39-06:00 kergoth.com!kergoth
Break people's builds again.. this time moving the packages into a packages/ subdir to clean things up a bit.
BKrev: 41b81f3dvlp3rU7_8MUXLcI8LDdDoA
|