Handle all decommits/madvise(DONTNEED)s homogenously
#8587
Merged
Conversation
This file contains 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
All implementations are the same either way, and by treating them differently, it makes it difficult to prototype integration with batching syscalls that need to treat them homogenously. Co-Authored-By: Jamey Sharp <jsharp@fastly.com>
|
cc #4637 |
… interface This was already implemented by the existing `decommit_pages` function, we just need to determine what the behavior is for the left over mapping. Specifically, whether it is zeroed or restored to the original mapping (e.g. a CoW image). This allows us to handle `decommit_pages` and what previously were calls to `madvise_dontneed` homogenously, which enables us to prototype new system calls that batch decommits together and therefore must treat them all the same. Co-Authored-By: Jamey Sharp <jsharp@fastly.com>
fitzgen
added a commit
to fitzgen/wasmtime
that referenced
this pull request
May 14, 2024
…iance#8587) * Don't differentiate between decommitting table vs stack pages All implementations are the same either way, and by treating them differently, it makes it difficult to prototype integration with batching syscalls that need to treat them homogenously. Co-Authored-By: Jamey Sharp <jsharp@fastly.com> * Remove the `madvise_dontneed` function from the system virtual memory interface This was already implemented by the existing `decommit_pages` function, we just need to determine what the behavior is for the left over mapping. Specifically, whether it is zeroed or restored to the original mapping (e.g. a CoW image). This allows us to handle `decommit_pages` and what previously were calls to `madvise_dontneed` homogenously, which enables us to prototype new system calls that batch decommits together and therefore must treat them all the same. Co-Authored-By: Jamey Sharp <jsharp@fastly.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
We previously had different functions for tables vs memories vs stacks vs etc...
Now we use the same system virtual memory interface function for all of them. This enables follow up PRs where we will start batching these calls together (and potentially merging regions together in the further future, although that isn't a clear win as we've seen larger
madvises take longer than small ones in the past). Those future batching PRs will allow us to in turn start prototyping new syscalls that can take advantage of that batching, likemadvisev.