Stream: git-wasmtime

Topic: wasmtime / PR #11070 Prefix `Lift` and `Lower` trait meth...

Wasmtime GitHub notifications bot (Jun 18 2025 at 21:17):

fitzgen opened PR #11070 from fitzgen:add-memory-prefix-to-lift-lower to bytecodealliance:main:

This is to distinguish them from the GC versions that will be added in follow up commits.

No functional changes here, just renaming.

<!--
Please make sure you include the following information:

• If this work has been discussed elsewhere, please include a link to that
  conversation. If it was discussed in an issue, just mention "issue #...".

• Explain why this change is needed. If the details are in an issue already,
  this can be brief.

Our development process is documented in the Wasmtime book:
https://docs.wasmtime.dev/contributing-development-process.html

Please ensure all communication follows the code of conduct:
https://github.com/bytecodealliance/wasmtime/blob/main/CODE_OF_CONDUCT.md
-->

Wasmtime GitHub notifications bot (Jun 18 2025 at 21:17):

fitzgen requested alexcrichton for a review on PR #11070.

Wasmtime GitHub notifications bot (Jun 18 2025 at 21:17):

fitzgen requested wasmtime-core-reviewers for a review on PR #11070.

Wasmtime GitHub notifications bot (Jun 18 2025 at 21:21):

fitzgen updated PR #11070.

Wasmtime GitHub notifications bot (Jun 18 2025 at 22:28):

alexcrichton commented on PR #11070:

To motivate a bit of a bikeshed: these sorts of changes tend to have a very high sticking power where they stick around for a very long time so I'd ideally like to handle things in one fell swoop instead of going through multiple renames. To bikeshed: memory_lift and memory_load are IMO deceptively subtle in how they're different as "lift from memory" is something I would reasonably say should be synonymous with a load from memory. Given that I'd prefer to find different names that are more nuanced than just putting memory_* in front of what we currently have.

One observation as well is that I think there's probably not going to be a gc_load or a gc_store. Current design has lifting/lowering of GC values being mostly like the "flat" lifting/lowering of linear memory values. I realize the struct/array cases are a bit different but I think what we'll probably do is pass a parameter through which differentiates the local/struct/array source/destinations of values.

Naming-wise WDYT about {lift,lower}_{linear,gc} and otherwise leaving load and store as-is? (or maybe {load,store}_linear)

Wasmtime GitHub notifications bot (Jun 18 2025 at 23:36):

fitzgen commented on PR #11070:

For the GC methods I currently have:

• Lift::gc_lift_val for doing GC lifting from a value
• Lift::gc_lift_storage for doing GC lifting from a storage type (i.e. a struct field or array element)
• Lower::gc_lower_val for doing GC lowering into a value
• Lower::gc_lower_storage for doing GC lowering into a storage type

Wasmtime GitHub notifications bot (Jun 18 2025 at 23:42):

fitzgen commented on PR #11070:

How about

• Lift::lift_from_flat_linear
• Lift::lift_from_memory_linear
• Lift::lift_from_val_gc
• Lift::lift_from_storage_gc
• Lower::lower_to_flat_linear
• Lower::lower_to_memory_linear
• Lower::lower_to_val_gc
• Lower::lower_to_storage_gc

?

Wasmtime GitHub notifications bot (Jun 18 2025 at 23:48):

alexcrichton commented on PR #11070:

That sounds good to me :+1:

Wasmtime GitHub notifications bot (Jun 18 2025 at 23:49):

alexcrichton submitted PR review.

Wasmtime GitHub notifications bot (Jun 20 2025 at 17:03):

fitzgen updated PR #11070.

Wasmtime GitHub notifications bot (Jun 20 2025 at 17:04):

fitzgen has enabled auto merge for PR #11070.

Wasmtime GitHub notifications bot (Jun 20 2025 at 17:56):

pchickey commented on PR #11070:

https://github.com/bytecodealliance/wasmtime/pull/11089 fixes CI

Wasmtime GitHub notifications bot (Jun 20 2025 at 19:41):

alexcrichton merged PR #11070.

Last updated: Dec 06 2025 at 06:05 UTC