diff --git a/server/commit_diff.txt b/server/commit_diff.txt index 590883743616a..96be0bf522a09 100644 --- a/server/commit_diff.txt +++ b/server/commit_diff.txt @@ -162,41 +162,6 @@ Date: Tue Jun 30 23:53:10 2020 -0400 Also related: #3388 #4077 -commit 93104e616f53ed46503577032c9eacf9be31575a -Merge: 6c689992 1dbe5cd1 -Author: Brandon Simmons -Date: Fri Jun 26 18:24:53 2020 -0400 - - Merge pull request #5214 from jberryman/5190-improve-websockets-idle-memory-v2-only-kc8 - - Lower stack chunk size in RTS to reduce thread STACK memory (closes #5190) - -commit 1dbe5cd1ab4a68dd2518f9e57a656517f708e900 -Merge: f9b5b838 6c689992 -Author: Brandon Simmons -Date: Fri Jun 26 15:02:49 2020 -0400 - - Merge branch 'master' into 5190-improve-websockets-idle-memory-v2-only-kc8 - -commit f9b5b8382cbd41a589d1096c720805d3184153d4 -Author: Brandon Simmons -Date: Tue Jun 23 20:35:36 2020 -0400 - - Lower stack chunk size in RTS to reduce thread STACK memory (closes #5190) - - This reduces memory consumption for new idle subscriptions significantly - (see linked ticket). - - The hypothesis is: we fork a lot of threads per websocket, and some of - these use slightly more than the initial 1K stack size, so the first - overflow balloons to 32K, when significantly less is required. - - However: running with `+RTS -K1K -xc` did not seem to show evidence of - any overflows! So it's a mystery why this improves things. - - GHC should probably also be doubling the stack buffer at each overflow - or doing something even smarter; the knobs we have aren't so helpful. - (Done, but we should re-visit this, if we do query plan caching) commit 20cbe9cfd3e90b91d3f4faf370b081fc3859cbde Author: Auke Booij