remotecache: Only visit each item once when walking results. #1577
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
Before this change, walkAllResults did not skip an item if it had already been visited, so walking the graph of results actually followed every single possible path in the graph instead of just visiting each item once. On small graphs this wasn't noticeable, but on sufficiently large remote cache imports this rapidly scaled out of control. For example, I first encountered this when importing a max-cache from a build of a full linux rootfs from source (which takes about 30 minutes to build w/out cache on an 18-core machine) and after 30 minutes the cache import was still running with all CPUs pegged at 100% To fix this, walkAllResults now keeps a map of already visited items (keyed by their pointer) and skips visiting an item if it's already been visited, making it usable on large remote cache imports. Signed-off-by: Erik Sipsma <[email protected]>
tonistiigi
approved these changes
Jul 17, 2020
This was referenced Jul 25, 2020
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Outdated suggestions cannot be applied.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.
Before this change, walkAllResults did not skip an item if it had already been
visited, so walking the graph of results actually followed every single possible
path in the graph instead of just visiting each item once.
On small graphs this wasn't noticeable, but on sufficiently large remote cache
imports this rapidly scaled out of control. For example, I first encountered
this when importing a max-cache from a build of a full linux rootfs from source
(which takes about 30 minutes to build w/out cache on an 18-core machine) and
after 30 minutes the cache import was still running with all CPUs pegged at 100%
To fix this, walkAllResults now keeps a map of already visited items (keyed by
their pointer) and skips visiting an item if it's already been visited, making
it usable on large remote cache imports.
Signed-off-by: Erik Sipsma [email protected]