BTW, kudos for bisecting this, here Michael and his very strong focus on benchmarks uncovered a real problem, that's why I'm sharing an article from his side.
Prikaži ovu nit
"[…] The apparent cause of at least some of the #Linux #kernel 5.5 slowdowns after bisecting turned out to be #AppArmor […] it explains why the likes of Red Hat's upstream developers and others weren't affected and thus not quickly jumping on the problem […]"https://twitter.com/phoronix/status/1211453420428759040 …
-
-
-
Out of curiosity: Does anyone know how (or where) the
#Apparmor performance regression(¹) was reported to the responsible developers? I can't find anything :-/ (¹) fixed in mainline (https://twitter.com/kernellogger/status/1213728241250385922 … ) and#Linux#kernel 5.5-rc5 (https://twitter.com/kernellogger/status/1214076414913368064 … )
Prikaži ovu nit
Kraj razgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.