Under some circumstances it is not desirable to create a combined locale
archive (/usr/lib/locale/locale-archive).
The new variable IMAGE_LOCALES_ARCHIVE defaults to '1', so the default
behaviour is not changed.
(From OE-Core rev: 8d78b819c2ec33fce3a34254fa90864ee5fa7617)
Signed-off-by: Michael Thalmeier <michael.thalmeier@hale.at>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
It incorrectly specified "cortexa55" instead of "cortexa65".
(From OE-Core rev: 25317c51d4d0ac922f4bd555aae9fbfe48a1aa40)
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Cortex-A32 is 32bit only, but the recent changes had it working as
64bit (causing build breaks). This reverts that change in this file.
(From OE-Core rev: 2957379c85c245f183e7e94e30c2b504fc25cedf)
Signed-off-by: Jon Mason <jon.mason@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Adds variables that can be used to allow a recipe to pass extra
arguments to `waf build` and `waf install`. In most cases, you want to
pass the same arguments to `build` and `install` (since install is a
superset of `build`), so by default setting EXTRA_OEWAF_BUILD also
affects `waf install`, but this can be overridded.
(From OE-Core rev: 493e17a2f5cbbbe3b1e435dadb281b007bca2cbf)
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There are autobuilder logs where the server commands are failing
but we have no debug info in the server log. Improve this to try and
understand what is failing.
(Bitbake rev: 04d3a79226c9ea448b22f4efbab33876a72c9bdb)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This fixes a warning about a trailing slash on ${S}.
(From OE-Core rev: 0685dac31a7ae614d9f75cd51b59c45dd050f52e)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Debug logs were only controlled by --debug flag while in --foreground mode.
In --daemon mode (the default for us) /var/log/message got stuffed with
details of entropy pool refilling, which is useless in production, and
hamful when log rotation then gets rid of the more useful logs.
This change makes the two modes consistently only produce debug logs when
--debug is specified.
(From OE-Core rev: 914526868656fd279b13ba8e4d721d27cb2a8792)
Signed-off-by: Yann Dirson <yann@blade-group.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The example recipe is setup to print a message using bb.plain() in the
"do_build" task but this task is "noexec" so the message never prints.
This might be confusing.
This moves the message printing into another "do_display_banner" task
and add it to the do_build "before" list.
(From OE-Core rev: 252385bef9b226f32691b8513869ea3e41813b40)
Signed-off-by: Yoann Congal <yoann.congal@smile.fr>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If --branch and --branch2 are the same, there is extra confusing data plotted
onto the graph. Only do that if the branches are actually different.
(From OE-Core rev: c263d810b7eb47ee90f2adeb5ab6decf7332bec2)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If buildstats is missing, avoid a backtrace.
(From OE-Core rev: 23c4b21f947c7aae1303ca6526fc5aaaa9fc7bb8)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* to make sure we won't hit such corner cases in future add a warning
to prevent trailing slashes and duplicated slashes even when they
in most cases don't cause harm
* only a few cases were found in layers included in my world builds:
oe-core: 1
meta-oe: 7
meta-python2: 1
meta-qt5: 1
meta-aws: 1
will send patches for these once this warning is approved for oe-core
(From OE-Core rev: 8a4c473c07cba159cf88ed775b9f073c6adf31d4)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* cannot use os.path.samefile, because S/B might not exist at this time yet
* there is issue with PSEUDO_IGNORE_PATHS when some recipe sets e.g.
S = "${WORKDIR}/"
whole WORKDIR gets added to PSEUDO_IGNORE_PATHS and then the build
can fail with various strange errors, in my case do_package was
failing when do_package calls:
fix_perms(.../1.0-r0/package/etc, 755, 0, 0, /etc)
and fails with "[Errno 1] Operation not permitted:"
(From OE-Core rev: 50b11a61ab29acb8ec990668353e0b7305114628)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The server may be displaying useful information for the user through log
messages so we should display anything that has been sent. Its either this
or expecting every UI to implement this code around every command call
which isn't good API.
[YOCTO #14054]
(Bitbake rev: 64ae9d7e2fad804dd9e12706c6d76b4b22f9586b)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Previous changes allowed BBHandledException to be detected but not exceptions
which inherit from it. Fix this. The code really needs totally reworking
to preserve the exceptions.
[YOCTO #14054]
(Bitbake rev: ef762d92df6c2554c6248e80212f984d9ec4c651)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This has been the same as OE-Core for a long time, drop our custom settings.
(From meta-yocto rev: 8c1084f3fd4ca987c3307f4c4c1475b40060ce3f)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Newer systems are using this currently unwrapped glibc function,
add it.
[YOCTO #14080]
(From OE-Core rev: cba4f51f0166e2c20e854ee0344be0c58769eaca)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
After the recent path ignore changes, there are invalid entries in exiting
pseudo databases which will trip up users with the new abort() (proving
how common the pseudo potential corrpution is!). Inform them a clean TMPDIR
is needed.
(From OE-Core rev: 016ee90e210c9b15b80e8370d83f41a14867a413)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add tunes for all the ARMv8.2a cores currently supported in GCC. This
is: Cortex-A65, Cortex-A65AE, Cortex-A75, Cortex-A76, Cortex-A76AE,
Cortex-A77, Neoverse-E1, Neoverse-N1, Cortex-A75-Cortex-A55, and
Cortex-A76-Cortex-A55.
(From OE-Core rev: f15e83df1d4ed4172b43d5b353cf42640862a28e)
Signed-off-by: Jon Mason <jon.mason@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add tunes for all the ARMv8a cores currently supported in GCC. This
is: Cortex-A34, Cortex-A73, and Cortex-A73-Cortex-A35.
(From OE-Core rev: fdf6921f145658b1f1c2e7413a2cde6857ddb768)
Signed-off-by: Jon Mason <jon.mason@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Uniformity clean-up, have the individual tunes reference the armv8a TUNE_FEATURES.
(From OE-Core rev: 3ac533127d444f39ad550ce923f494d0eb766ab5)
Signed-off-by: Jon Mason <jon.mason@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Move TUNECONFLICTS to the generic arm64 inc file, so that it can provide
coverage for more than just the big.LITTLE tunes.
(From OE-Core rev: 8022543d79b62b0daa4c988e17cd3e2c364d3cf1)
Signed-off-by: Jon Mason <jon.mason@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
In order to improve performance of qemumips autobuilder
test runs, let's use 34Kf-64tlb cpu type that was introduced
in OE version of qemu. 34Kf-64tlb cpu type is identical to
34Kf but it has 64 TLBs configure vs original 16 TLBs.
The change in number of TLBs in emulated CPU reduces
TLB trash and reduces number of times TLB refill kernel
code runs, and therefore siginificantly improves tests
execution time.
Note 34Kf-64tlb qemu cpu type does not exist in upstream,
so far it is added as patch in OE. See qemu 34Kf-64tlb
cpu addition commit for more details.
Fixes https://bugzilla.yoctoproject.org/show_bug.cgi?id=13992
(From OE-Core rev: 6a9568401cbf8bed89d1f63dae61a07a83a20b50)
Signed-off-by: Victor Kamensky <kamensky@cisco.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
In Yocto Project PR 13992 it was reported that qemumips
in autobuilder runs almost twice slower then qemumips64 and
some times hit time out.
Upon investigations of qemu-system with perf, gdb, and
SystemTap and comparing qemumips and qemumips64 machines
behavior it was noticed that qemu soft mmu code behaves
quite different and in case if qemumips tlbwr instruction
called 16 times more oftern. It happens that in qemumips64
case qemu runs with cpu type that contains 64 TLB, but in case
of qemumips qemu runs with cpu type that contains only
16 TLBs.
The idea of proposed qemu patch is to introduce fictitious
34Kf-64tlb cpu type that defined exactly as 34Kf but has
64 TLBs, instead of original 16 TLBs.
Testing of core-image-full-cmdline:do_testimage with
34Kf-64tlb shows 40% or so test execution real time
improvement.
Note for future porters of the patch: easiest way to update
the patch and be in sync with 34Kf definition is to copy
34Kf machine definition and apply the following changes to
it (just change 15 to 63 of CP0C1_MMU bits value)
[kamensky@coreos-lnx2 qemu]$ diff ~/34Kf.c ~/34Kf-64tlb.c
2c2
< .name = "34Kf",
> .name = "34Kf-64tlb",
6c6
< .CP0_Config1 = MIPS_CONFIG1 | (1 << CP0C1_FP) | (15 << CP0C1_MMU) |
> .CP0_Config1 = MIPS_CONFIG1 | (1 << CP0C1_FP) | (63 << CP0C1_MMU) |
Fixes https://bugzilla.yoctoproject.org/show_bug.cgi?id=13992
Upstream Status: Inappropriate
(From OE-Core rev: 4470a04943352224955f17e004962f0f9e1c9b0c)
Signed-off-by: Victor Kamensky <kamensky@cisco.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When the license identifier tags were added, an incorrect string was used
and the Share-Alike clause was lost. Fix this to match the license
description in the files and add back the lost piece (its clear from
the history it should be there)
(From yocto-docs rev: 8d30c3d792755a7bfdb74b331dad98f51d3516af)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add some new libraries to the list (fiber, headers, type_erasure).
Move context/coroutine to the list instead of using overrides as it
builds everywhere I can test it.
Remove the mips16e override for wave as Boost fails so dramatically with
mips16e enabled that this isn't even close to a fix. Someone who cares
can fix this properly.
(From OE-Core rev: 092228f2df6869e31b157ea08766b3e94bee6e29)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There's no need to specify an ancient GCC version here as Boost will
probe it.
(From OE-Core rev: 2339bf5b0aceb8e55f4b38e44b2383389e514393)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Boost is a huge unirepo made from multiple submodules. To bootstrap it we
need boost.build (previously bjam) which is also available as a solo
repository. This smaller repository can unpack/build/package faster than
the Boost unirepo can unpack.
Rename the recipe to the current name of Boost.Build that installs a b2
binary, use the solo repository, and update the Boost recipe to use
the b2 binary instead of bjam.
(From OE-Core rev: 76079f921e596125b0e281ca95e2394d7688aaf2)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Unfortunately, .pyc files can be generated in meta/lib/oe which corrupt the pseudo
database so we need to extend the ignore list to cover this as well.
(From OE-Core rev: e0cb6dd689a362d8433caa14cc5a9fdd5eb44923)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Brings following changes, supports linux 5.8 syscalls [1]
* a5aff197 avoid set*id/setrlimit misbehavior and hang in vforked/cloned child
* 55fb9a17 use new SYS_faccessat2 syscall to implement faccessat with flags
* f035c7b1 netinet/if_ether.h: add ETH_P_MRP from linux v5.8
* 6b1741aa elf.h: add .note.gnu.property related definitions
* 9b7ed970 bits/syscall.h: add __NR_faccessat2 from linux v5.8
* 616a8bf6 netinet/tcp.h: update to linux v5.7
* a6c302be sys/mman.h: add MREMAP_DONTUNMAP from linux v5.7
* 8adf42f7 sys/fanotify.h: update to linux v5.7
* 94ab68c2 aarch64: add new HWCAP2_ macros from linux v5.6
* 0296baff aarch64: add HWCAP2_ macros from linux v5.3
* 43b640c2 sched.h: add CLONE_NEWTIME from linux v5.6
* 3da18e69 sys/random.h: add GRND_INSECURE from linux v5.6
* 8f4aa78a sys/prctl.h: add PR_{SET,GET}_IO_FLUSHER from linux v5.6
* 1ab341e8 netinet/udp.h: add TCP_ENCAP_ESPINTCP from linux v5.6
* c6321616 netinet/tcp.h: update for linux v5.6
* 8fca0ea4 netinet/in.h: add IPPROTO_ macros from linux v5.6
* d9900903 add pidfd_getfd and openat2 syscall numbers from linux v5.6
* d4f29814 netinet/tcp.h: update tcp_info for linux v5.5
[1] https://git.musl-libc.org/cgit/musl/log/?qt=range&q=ffac0c229986725c0d0f3c806bafa7e3ca409f3b..a5aff1972c9e3981566414b09a28e331ccd2be5d
(From OE-Core rev: e104d235cb14f0d58bfefe48e3888430503ba619)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If a recipe applies patches which are in machine-specific override
directories, devtool will fail to fetch the patches that don't match the
default configuration. For example where there are patches at
qemux86/x86.patch and qemuarm/arm.patch:
SRC_URI = "file://source"
SRC_URI_append_qemuarm = " file://arm.patch"
SRC_URI_append_qemux86 = " file://x86.patch"
The patch apply phase sets OVERRIDES but does not set FILESOVERRIDES, so
it cannot find the patch files as the search path isn't correct. Fix
this by setting FILESOVERRIDES too.
Also when iterating through the overrides we need to be sure that other
overrides that are used are not enabled, so extend no_overrides instead of
simply appending the current override.
Fixes most but not all of [ YOCTO #14060 ].
(From OE-Core rev: a372cdf8e175423c47faeecc98ad076ee26bbec8)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
devtool doesn't quite behave right when a recipe has patches applied in overrides, so
add a test case to exercise that behaviour.
(From OE-Core rev: ad89c3254cc6c06026f1bebe23ec3717c1546633)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The tests are actually skipped if meta-oe isn't present which isn't
quite the same thing, but hopefully close enough.
(From OE-Core rev: d22ed015d8f38241acb783e1a468fb15d4317670)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There is magic in the posts where specific variable names have specific
magic. For that magic to work, "path" needs to be used not "pathname" as
is currently there. Fix this, which fixes path issues on systems using
statx (Ubuntu 20.04 in particular).
(From OE-Core rev: f9ae80aee98a7bb59c83ac9ebf9d317302507cec)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Adjust wic to correctly handle the new PSEUDO_IGNORE_PATH variable and avoid
inode corruption issues.
(From OE-Core rev: 13500f5234361385c365c7c35e83f99435500481)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is a pretty big change to the way pseudo operates when used in OpenEmbedded.
Normally, pseudo monitors and logs (adds to its database) any file created or
modified whilst in a fakeroot environment. There are large numbers of files
we simply don't care about the permissions of whilst in that fakeroot context,
for example ${S}, ${B}, ${T}, ${SSTATE_DIR}, the central sstate control
directories,
This change uses new functionality in pseudo to ignore these directory trees,
resulting in a cleaner database with less chance of "stray" mismatches if files
are modified outside pseudo context. It also should reduce some overhead from
pseudo as the interprocess round trip to the server is avoided.
There is a possible complication where some existing recipe may break, for
example, we found a recipe which was writing to "${B}/install" for
"make install" in do_install and since we listed ${B} as not to be tracked,
there were errors trying to chown root for files in this location.
This patch fixes a few corner cases in OE-Core when used with this new
ignore list:
* The archiver directory matched a "${WORKDIR}/deploy*" pattern so was renamed
to something else since that directory does need its root permissions
* The ${S} and ${B} ignoring is conditional on them being different to ${WORKDIR}
* package_write_* task output (the debs/rpms/ipks) are now owned by the build
user so we don't want the file ownership information in the hashequiv outhash
calculation even if they are built under pseudo.
* The fontcache postinstall intercept is run under qemu outside of pseudo context
so delete files it may delete up front where pseudo can see this.
* SSTATE_DIR is in PSEUDO_PATHS_IGNORE, which is in FAKEROOTENV which is cached
by bitbake. We therefore need to trigger reparsing if this changes, which means
SSTATE_DIR can be in BB_HASHBASE_WHITELIST but not BB_HASHCONFIG_WHITELIST.
Rework the variables to handle this. This otherwise breaks some of our sstate
tests in oe-selftest.
* Ignore the temp directory wic uses for rebuilding rootfs.
(From OE-Core rev: ad8f5532ffaead9a5ad13e1034fe9e5e1b7979f4)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add a path to fix up handling of dirfd being passed as a full file
and with path="".
(From OE-Core rev: 3006bbe7768e4efe33d3d2d3f5786a561ecbe96f)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark files which are unlinked (nlink == 0) but open with fd's as
"may-unlink" to avoid problematic database entries.
(From OE-Core rev: e1e481f3608c05ab14c61bf45cd0837d7287b6a5)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fix a NULL pointer dereference exposed by the path ignore code in
xattr handling.
(From OE-Core rev: 929a27bf6cbca94d1141d2094ae0c915d93bd3f4)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Where files are link()'d and one is unlink()'d, pseudo's fd mappings
can become confused. Add a patch to try and improve this for the common
usecases we see.
(From OE-Core rev: 9ce621fa2099608ca0ccbb8420b31d71cdd7b00e)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Rather than doing what turns out to be a rather dangerous "fixup" if
we see a file with a different path but the same inode as another file
we've previously seen, throw and abort. Direct the user to a wiki page
where we can maintain information about what this error means.
(From OE-Core rev: 2db491d97da08d44ebd257f98489550a82a7935c)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
