| From: | Vadim Ponomarev <vbponomarev(at)gmail(dot)com> |
|---|---|
| To: | Yura Sokolov <y(dot)sokolov(at)postgrespro(dot)ru> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Reduce SyncRepLock contention on the commit path |
| Date: | 2026-10-10 10:02:29 |
| Message-ID: | CANeUpr-5B7B+LnVeWeB2uJ=W4=fHKL_UahZFcakveMRrdOYhnQ@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
v6 attached, rebased onto f58ab18a904.
cfbot flagged v5 as needing a rebase: master took 058 and 059, and the
new test sat at the tail of src/test/recovery/meson.build, where every
new recovery test lands. 0003's two scenarios now live at the end of
007_sync_rep.pl, so 0003 no longer touches meson.build. On Windows 007
stops before them and runs its own tests as before, the way
019_replslot_limit.pl does.
Nothing else changed.
Regards,
Vadim Ponomarev
| Attachment | Content-Type | Size |
|---|---|---|
| v6-0001-Wake-the-released-sync-rep-waiters-after-the-queu.patch | text/x-patch | 6.5 KB |
| v6-0002-Compute-the-synced-positions-before-taking-the-sy.patch | text/x-patch | 3.1 KB |
| v6-0003-Release-the-sync-rep-waiters-once-per-drained-bat.patch | text/x-patch | 15.8 KB |
| v6-0004-Let-a-committer-whose-acknowledgement-already-arr.patch | text/x-patch | 7.9 KB |
| v6-0005-Add-stream-regress-test-for-synchronous-replicati.patch | text/x-patch | 12.1 KB |
| From | Date | Subject | |
|---|---|---|---|
| Previous Message | Zsolt Parragi | 2026-10-10 09:57:49 | Emptying oauth_validator_libraries doesn't stop OAuth logins |