Skip to content

[ML] Upgrade to Boost 1.86.0 on ARM macOS#2779

Merged
edsavage merged 28 commits intoelastic:mainfrom
edsavage:macos_boost_186
Dec 3, 2024
Merged

[ML] Upgrade to Boost 1.86.0 on ARM macOS#2779
edsavage merged 28 commits intoelastic:mainfrom
edsavage:macos_boost_186

Conversation

@edsavage
Copy link
Copy Markdown
Contributor

Update build scripts, docs and license information to refer to Boost 1.86.0

Depends on #2774

From version 9.0.0 we drop support for macos x86_64 builds.
Therefore all support for cross compiling for macos x86_64 has also been
dropped.
Builds have been migrated to be executed solely on macos 13 (Ventura)
aarch64. In CI, these builds are performed on updated Orka VMs.

In addition, 3rd part dependencies (boost, pytorch) are now baked into
the Orka VM, so downloading these dependencies at run time is no longer
required,
Update build scripts, docs and license information to refer to Boost
1.86.0
Also a couple of compilation/test fixes
@edsavage
Copy link
Copy Markdown
Contributor Author

buildkite build this

@edsavage edsavage requested a review from valeriy42 November 27, 2024 23:24
@edsavage edsavage marked this pull request as ready for review November 27, 2024 23:31
Copy link
Copy Markdown
Contributor

@valeriy42 valeriy42 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

Comment on lines +1290 to +1295
// Interestingly we used to assert extraMem > 0 here as it used to be the case
// that once a boost::small_vector had switched
// off of internal storage it would NEVER go back to internal storage.
// Arguably that was a bug, and this assertion started failing after
// upgrading Boost to 1.86.0, meaning that boost::small_vector has been improved.
BOOST_TEST_REQUIRE(extraMem >= 0); // Change to `==` once upgraded to Boost 1.86 on all platforms
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nice! :-)

@valeriy42
Copy link
Copy Markdown
Contributor

Cmake cannot find the new boost in the build. The new orka image seems to contain the old version of Boost.

@elastic-sonarqube
Copy link
Copy Markdown

@edsavage edsavage merged commit d5c60e2 into elastic:main Dec 3, 2024
edsavage added a commit to edsavage/ml-cpp that referenced this pull request Dec 6, 2024
Update build scripts, docs and license information to refer to Boost 1.86.0

Follow up to elastic#2774
edsavage added a commit that referenced this pull request Dec 8, 2024
…#2794)

Upgrade build scripts, docs etc. to refer to PyTorch 2.5 and Boost 1.86.0 for macOS ARM builds.

Backports #2778 and #2779
@edsavage edsavage deleted the macos_boost_186 branch March 20, 2025 03:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants