Allow explicit data transfers to GPUs#156620
Draft
ZuseZ4 wants to merge 15 commits into
Draft
Conversation
ZuseZ4
added
A-LLVM
rustbot
added
the
S-waiting-on-author
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
from
e475c46 to
da102aa
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
Member
Author
|
Vendoring llvm/llvm-project#198033 for now. |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
from
abc274d to
1d8d1e7
Compare
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
from
1d8d1e7 to
a94ef31
Compare
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
2 times, most recently
from
4b77bad to
319ef7d
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
from
fba7eb2 to
358171b
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
from
2f1d614 to
bbe3882
Compare
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
from
bbe3882 to
d290591
Compare
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
from
d290591 to
e8ad696
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
ZuseZ4
force-pushed
the
offload-explicit-datatransfer
branch
from
4f5c325 to
6c8bec9
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
oli-obk
requested changes
Jun 1, 2026
Comment on lines
+39
to
+41
| // This exists so MIR creates Drop terminators for PreloadMut. | ||
| // rustc codegen intercepts those terminators and emits the | ||
| // offload return mapper. |
Contributor
There was a problem hiding this comment.
why is this not just an intrinsic call here?
Member
Author
There was a problem hiding this comment.
Partly just experimenting, partly because intrinsics recently changed a bit, they got updated for more explicit Place handling, about which I didn't want to think for my mvp. I'll update them to intrinsics after my deadline.
|
|
||
| #[lang = "preload"] | ||
| #[unstable(feature = "offload", issue = "124509")] | ||
| pub fn preload<'a, T: ?Sized>(x: &'a T) -> Preload<'a, T> { |
Contributor
There was a problem hiding this comment.
Yea I think these should just be intrinsics instead of catching lang item calls during codegen of call terminators.
Member
Author
There was a problem hiding this comment.
"Which lang items"? (my code who fails to catch an inlined terminator call :D)
Member
Author
There was a problem hiding this comment.
ok, with an intrinsic it actually seems to work in release.
Not sure if we want one intrinsic with 2 arguments (mut/const, init/drop) or 4 intrinsics. Right now I have two.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
Draft
This comment has been minimized.
This comment has been minimized.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
View all comments
So far we had our offload intrinsics handle data movement automatically to/from the gpu.
That's convenient (and reasonably fast once our LLVM opts land). However, Rust generally also allows being explicit. That might give perf benefits (where our LLVM opts fail), and it could also be nice for modelling, when passing data around but still preventing CPU users from accesing it.
Tracking Issue for GPU-offload #131513