Commit ab5194e

committed

Improve LWLock scalability.

The old LWLock implementation had the problem that concurrent lock acquisitions required exclusively acquiring a spinlock. Often that could lead to acquirers waiting behind the spinlock, even if the actual LWLock was free. The new implementation doesn't acquire the spinlock when acquiring the lock itself. Instead the new atomic operations are used to atomically manipulate the state. Only the waitqueue, used solely in the slow path, is still protected by the spinlock. Check lwlock.c's header for an explanation about the used algorithm. For some common workloads on larger machines this can yield significant performance improvements. Particularly in read mostly workloads. Reviewed-By: Amit Kapila and Robert Haas Author: Andres Freund Discussion: 20130926225545.GB26663@awork2.anarazel.de

1 parent 7882c3b commit ab5194eCopy full SHA for ab5194e

2 files changed

+653

-302

lines changed

src
- backend/storage/lmgr
  - lwlock.c
- include/storage
  - lwlock.h

2 files changed

+653

-302

lines changed

Comments

(0)

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commit ab5194e

2 files changed

2 files changed

pFad - (p)hone/(F)rame/(a)nonymizer/(d)eclutterfier! Saves Data!

File tree

2 files changed

2 files changed

0 commit comments

pFad - (p)hone/(F)rame/(a)nonymizer/(d)eclutterfier! Saves Data!