Re-embed multilist_t storage. #12158
Merged
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Outdated suggestions cannot be applied.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 "7968 multi-threaded spa_sync()" commit multilists were stored
in caller's memory same as regular lists. It is not clear why it was
changed, considering that the structure is still public, but I found
it reduced performance by additional uncached pointer dereference.
This commit partially reverts the mentioned one and adds some cache
line alignments to separate read-only multilists and heavily modified
refcount's to different cache lines.
How Has This Been Tested?
On 40-thread FreeBSD system heavy uncached 4KB reads from 8 ZVOLs show ~2% IOPS increase, while profiler shows reduction of time spent in dbuf_evict_one() and descendants from 8.2% to 6.8%.
Types of changes
Checklist:
Signed-off-by
.