[Bug 66524] resource cache eviction is MRU not LRU

2023-03-14 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66524 Christopher Schultz changed: What|Removed |Added Resolution|--- |FIXED Status|NEW

[Bug 66524] resource cache eviction is MRU not LRU

2023-03-13 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66524 --- Comment #4 from Mark Thomas --- +1 -- You are receiving this mail because: You are the assignee for the bug. - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org F

[Bug 66524] resource cache eviction is MRU not LRU

2023-03-13 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66524 --- Comment #3 from Christopher Schultz --- (In reply to Steven Kearns from comment #2) > The field CachedResource.nextCheck is what is being sorted, it is also a > proxy for lastUsedTime. Yup. I got the reversal reversed in my head when readi

[Bug 66524] resource cache eviction is MRU not LRU

2023-03-11 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66524 --- Comment #2 from Steven Kearns --- The field CachedResource.nextCheck is what is being sorted, it is also a proxy for lastUsedTime. since items will be evicted in iteration order, then sorting by "reverse nextCheck" makes the iteration orde

[Bug 66524] resource cache eviction is MRU not LRU

2023-03-11 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=66524 --- Comment #1 from Christopher Schultz --- Why reverse the reverse? What is being compared here? What do you think ought to be compared? -- You are receiving this mail because: You are the assignee for the bug. --