I am blocking your proposal for a change to the pageadaemon.
Because:
1. You failed to label your proposal as a backout
2. Therefore people won't be aware of the statement that created the problem:
revision 1.131
date: 2024/11/25 13:06:25; author: mpi; state: Exp; lines: +12 -10;
commitid: 2TVEQnlW1pDAkPNp;
Account for in-flight pages being written to disk when computing page
shortage.
Due to its asynchronous design, on MP machines, the page daemon was
generally
over swapping memory resulting in a degenerative behavior when OOM.
To prevent swapping more pages than necessary, take the amount of in-flight
pages into account when calculating the page shortage.
Tested by sthen@ and miod@. ok claudio@, tb@
3. Since that previous CREATED a degenerative behaviour -- months of difficult
to diagnose deadlocks, after claiming it was fixing problems -- now is the
time to discuss the development process failure.
4. There are other backout diffs in the same area already under discussion
which must receive priority with good faith.
