| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
Update meta-clang to match OE-Core changes to use recipe specific virtual
providers (without yet switching to the new switching mechanism).
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
This needs to get several recipes to work with clang-16, originally the
problem is in vala I think. Which should be fixed in due course
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Use clang again for qemux86
- Drop upstreamed patch [1]
- Ignore texrels with clang
QA Issue: busybox: ELF binary /usr/bin/busybox.nosuid has relocations in .text [textrel]
[1] https://git.busybox.net/busybox/commit/?id=117a8c9b7a50053964159c342af1f3810cbbd5b8
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
Remove the Os workaround, this is no longer needed with systemd 252+
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
The build issue is only seen on ppc32
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
| |
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Andreas Müller <schnitzeltony@gmail.com>
|
|
|
|
|
| |
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
clang12 can successfully compile and boot musl so remove the workaround
to use gcc
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
The hang in clang seems to gone with clang12
remove dependecy on gcc-cross if using clang to compile musl
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
This ensures that all buildessentials are also installed into images
which are needed when building on device
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
This commit fixes an issue when building a cmake based project where clang-tidy is configured as co-compiler. Without this option clang-tidy will fail as it cannot determinate the target architecture.
Signed-off-by: Tobias Bayer <tobiasbayer@outlook.de>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
cclang-tidy doesn't process compilation flags like "-mlittle-endian" and "sysroot". The current export tries to use those flags, which makes clang-tidy fail like in the following example:
> $ aarch64-poky-linux-clang-tidy fibonacci.cpp -mlittle-endian --sysroot=/opt/mgu22/sysroots/aarch64-poky-linux
> LLVM ERROR: CommonOptionsParser: failed to parse command-line arguments. [CommonOptionsParser]: aarch64-poky-linux-clang-tidy: Unknown command line argument '-mlittle-endian'. Try: 'aarch64-poky-linux-clan$
> aarch64-poky-linux-clang-tidy: Did you mean '--list-checks'?
> aarch64-poky-linux-clang-tidy: Unknown command line argument '--sysroot=/opt/mgu22/sysroots/aarch64-poky-linux'. Try: 'aarch64-poky-linux-clang-tidy --help'
> aarch64-poky-linux-clang-tidy: Did you mean '--color=/opt/mgu22/sysroots/aarch64-poky-linux'?
This change removes those compiler arguments:
> $ aarch64-poky-linux-clang-tidy fibonacci.cpp
> Error while trying to load a compilation database:
> Could not auto-detect compilation database for file "fibonacci.cpp"
> No compilation database found in /home/mqueiros/workspace/minimal-working-example/fibonacci or any parent directory
> fixed-compilation-database: Error while opening fixed database: No such file or directory
> json-compilation-database: Error while opening JSON database: No such file or directory
> Running without flags.
Signed-off-by: Mauro Queiros <maurofrqueiros@gmail.com>
|
|
|
|
|
|
| |
CLANGSDK is 1
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
This helps in ensuring that clang shows up in rootfs when tools-sdk is
selected via IMAGE_FEATURES and default TOOLCHAIN is set to be clang
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
| |
This helps defining CLANG* env vars in extensible SDK
Fixes Issue #370
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
This patch is alrady upstream for 1.32
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
| |
Only when CLANGSDK = 1
Fixes Issue #338
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
clang/master hangs (can you believe) while compiling musl on armv7ve
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
| |
As CLANGSDK defaults to '1', users who have not read the README may
think that setting it to '0' would disable adding Clang to the SDK.
Little do they know that you need to *unset* the value for this to work
(as bool('0') -> True).
Change the logic to use bb.utils.contains(), so that '0' is not true.
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
CLANGSDK ?= "1"
is default which means clang will be included into SDK
if one is using clang only for full build and dont intend to distribute
it as cross compiler SDK then set
CLANGSDK = ""
in local.conf or in another config metadata file
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
| |
This was noticed when /bin/ash crashed that there are more instances of
this optimization which causes same kind of failures in few other
applets
As a side benefit busybox compiled with clang works totally fine on
riscv64 now
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
Fails to build with builtins which are only in compiler-rt
but compiler-rt needs C library to build. Catch-22
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
system boot but not all the way to command prompt so we need to debug it
eventually
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
It crashes when compiled with clang
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
|
| |
clang needs built-ins from compiler-rt to build musl
and compiler-rt needs a full toolchain to build which means
we have a catch-22 on x32 with clang/musl, for now use gcc
to compile musl to break the jinx
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
i386 inline assembly runs out of registers on clang
so for now let gcc be used
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
helps with LTO enabled cross builds
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
define CLANG_TIDY_EXE which is used by some build systems e.g. cmake
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
see https://bugs.llvm.org/show_bug.cgi?id=39919
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
This helps build SDKs for risc-v when meta-clang is in layer mix
even though clang for risc-v may not fully work yet
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
There is a bug in busybox which is found by clang, we need to fix
busybox but until them just dont use thumb2 ISA for busybox
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
Not needed after meson migration
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
| |
As suggested in Issue #23
this will make sure we can override it with BAD_RECOMMENDATIONS
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
| |
when clang is not default compiler
Fixes issue #63
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
clang-5.0 supports the option now
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
| |
build llvm libunwind along with libcxx
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
|
| |
This is preferred way of building libc++ and
it also means we do not need musl specific tweaks
since top level build from llvm already know
musl and make right choices
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
| |
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
ncurses' configure script strips the arguments from $CC when checking
if -c and -o can be used together. Clang then doesn't have a default
sysroot argument to use, so it checks the host for the existence of
certain feature headers in libc. These checks fail, and it assumes
that clang can't handle -o and -c together.
Example config.log snippet:
configure:2524: checking whether arm-poky-linux-gnueabi-clang understands -c and -o together
configure:2540: arm-poky-linux-gnueabi-clang -c conftest.c -o conftest2.o >&5
In file included from conftest.c:1:
In file included from /usr/include/stdio.h:27:
In file included from /usr/include/features.h:399:
/usr/include/gnu/stubs.h:7:11: fatal error: 'gnu/stubs-32.h' file not found
# include <gnu/stubs-32.h>
^
1 error generated.
configure:2543: $? = 1
configure:2562: result: no
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com>
|