This patch needs much more testing and I think is not sufficient either.
On Thu, Apr 9, 2020 at 1:29 AM Khem Raj <raj.k...@gmail.com> wrote:
>
> This is no longer needed as we can build multilibbed images with musl
> now
>
> Include bits/reg.h on musl instead of bits/wordsize.h where __WORDSIZE
> is defined
>
> [YOCTO #13122]
>
> Signed-off-by: Khem Raj <raj.k...@gmail.com>
> ---
> v2: include bits/reg.h on musl for __WORDSIZE
>
> meta/classes/multilib_header.bbclass | 7 -------
> scripts/multilib_header_wrapper.h | 4 ++++
> 2 files changed, 4 insertions(+), 7 deletions(-)
>
> diff --git a/meta/classes/multilib_header.bbclass
> b/meta/classes/multilib_header.bbclass
> index e03f5b13b2..48f231eed0 100644
> --- a/meta/classes/multilib_header.bbclass
> +++ b/meta/classes/multilib_header.bbclass
> @@ -6,13 +6,6 @@ inherit siteinfo
> # all of the ABI variants for that given architecture.
> #
> oe_multilib_header() {
> -
> - case ${HOST_OS} in
> - *-musl*)
> - return
> - ;;
> - *)
> - esac
> # For MIPS: "n32" is a special case, which needs to be
> # distinct from both 64-bit and 32-bit.
> case ${TARGET_ARCH} in
> diff --git a/scripts/multilib_header_wrapper.h
> b/scripts/multilib_header_wrapper.h
> index c81e7ee5e8..00ea5bf841 100644
> --- a/scripts/multilib_header_wrapper.h
> +++ b/scripts/multilib_header_wrapper.h
> @@ -16,7 +16,11 @@
> #elif defined (__aarch64__)
> #define __MHWORDSIZE 32
> #else
> +#if defined(__GLIBC__)
> #include <bits/wordsize.h>
> +#else
> +#include <bits/reg.h>
> +#endif
> #if defined (__WORDSIZE)
> #define __MHWORDSIZE __WORDSIZE
> #else
> --
> 2.26.0
>
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#137154):
https://lists.openembedded.org/g/openembedded-core/message/137154
Mute This Topic: https://lists.openembedded.org/mt/72892284/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
