-
Notifications
You must be signed in to change notification settings - Fork 44
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Added mimalloc dev and dev-slice support #75
Conversation
b812d30
to
62318a5
Compare
I fixed the |
@octavonce, is this something on which you are open to merge? Because if you are, I will update the code and try to make it all build/pass. |
- Added support to use the v1 dev branch or v2 dev-slice branch - Added option to update the git submodules to be able to use the latest commits from upstream mimalloc.
This commit should solve the CI errors. - Using `ctest2` instead of `ctest` which has not been updated for a while - Fixed extended feature because `mi_option_reserve_huge_os_pages_at` was added
62318a5
to
26fc715
Compare
@octavonce I think I fixed the CI workflows. |
@octavonce any feedback on this pr?? |
I'm also interested in seeing this merged as dev slice (ie: version 2) has memory features that would benefit heavy usage. |
@octavonce shall i create a separate PR for the workflow fixes? So, i was thinking about having a feature like Just add some good comments that people are on there own using that, and that it's not supported/stable. |
The You can verify this by checking If you think about this, it makes sense, since if you publish a crate in some workspace, it's not like the whole workspace should be published. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The submodule_update parts of this don't work, as mentioned. I think they should be removed, since there's no way to make them work either.
I also think that having dev/dev-slice branches has much less value given this, since it's still tied to the versions that exist when when published.
Also, we now use 2.x anyway, so I don't really see the point.
@thomcc You are totally right. I didn't checked out what would be sent to crates.io and how that would work. I will change this PR (or close it) and just fix the CI part. |
Closing this in favor of #81 which only fixes the CI workflow, including fixes the extended.rs. |
dev
branch or v2dev-slice
branchResolves #62
Closes #66
It would be really nice if this PR could be merged so that we are able to use the updated mimalloc code from the
dev
branch, or even the v2 versiondev-slice
. Themaster
branch is the stable v1 branch and thus does not contain the latest patches.Both the
dev
anddev-slice
branches have updated code so that it will work correctly on MUSL ARM platforms (see microsoft/mimalloc#495).With the added feature of executing a
git submodule update
users of this crate can also be sure that they are able to use the latest commits from upstream mimalloc and not depend on updates from this crate/repo it self if they really want to.I know that the mimalloc lib isn't fully ABI stable yet, but the current commits for both
dev
anddev-slice
work just fine using this crate, and i have added a warning within the comments.Thanks in advance!