On Tue, Jan 3, 2023 at 8:06 PM Khem Raj <[email protected]> wrote:
>
> The vendored copy of zlib undefines _FILE_OFFSET_BITS when
> _LARGEFILE64_SOURCE is defined and enabling 64bit time_t requires 64bit
> off_t ( _FILE_OFFSET_BITS=64 ), therefore remove this from the
> zlib module
>
> Signed-off-by: Khem Raj <[email protected]>
> ---
>  meta/recipes-devtools/rust/rust-source.inc    |  1 +
>  .../rust/rust/zlib-off64_t.patch              | 31 +++++++++++++++++++
>  2 files changed, 32 insertions(+)
>  create mode 100644 meta/recipes-devtools/rust/rust/zlib-off64_t.patch
>
> diff --git a/meta/recipes-devtools/rust/rust-source.inc 
> b/meta/recipes-devtools/rust/rust-source.inc
> index 643494b37e..bfb625fb36 100644
> --- a/meta/recipes-devtools/rust/rust-source.inc
> +++ b/meta/recipes-devtools/rust/rust-source.inc
> @@ -9,6 +9,7 @@ SRC_URI:append:class-target:pn-libstd-rs = "\
>  SRC_URI:append:class-target:pn-rust = " \
>      file://getrandom-open64.patch \
>      file://0001-Do-not-use-LFS64-on-linux-with-musl.patch \
> +    file://zlib-off64_t.patch \
>      file://hardcodepaths.patch \
>      file://crossbeam_atomic.patch"

I really think we've got this wrong... rustc-source is a vendored
tarball where anyone building out of it expects the same set of
sources, by carving it up, we're asking for trouble - in this case
libz-sys is consumed by libgit2-sys, which is consumed by cargo, which
wouldn't get this.

I'm going to play with collapsing all the patches across the consumers
of rust-source.inc into that include, as that feels like it should be
more correct.

--
Alex Kiernan
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#175371): 
https://lists.openembedded.org/g/openembedded-core/message/175371
Mute This Topic: https://lists.openembedded.org/mt/96036253/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to