Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[RFC\s+PATCHv2\s+0\/4\]\s+ipvs\:\s+Use\s+kthreads\s+for\s+stats\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[RFC\s+PATCHv2\s+0\/4\]\s+ipvs\:\s+Use\s+kthreads\s+for\s+stats\s*$/: 3 ]

Total 3 documents matching your query.

1. Re: [RFC PATCHv2 0/4] ipvs: Use kthreads for stats (score: 1): Author: Julian Anastasov <ja@xxxxxx>; Date: Sat, 10 Sep 2022 01:23:19 +0300 (EEST); Hello, v2 uses find_next_bit in for_each_set_bit which has cost. But we should not be surprised, if 268ms are for 50000 estimators on 104 CPUs (I guess this is also the number of possible CPUs we act; /html/lvs-devel/2022-09/msg00020.html (14,120 bytes)
2. Re: [RFC PATCHv2 0/4] ipvs: Use kthreads for stats (score: 1): Author: Jiri Wiesner <jwiesner@xxxxxxx>; Date: Fri, 9 Sep 2022 21:49:56 +0200; I couldn't wait so I have run tests on various machines and used the sched_switch tracepoint to measure the time needed to process one chain. The table contains a median time for processing one chain; /html/lvs-devel/2022-09/msg00019.html (11,082 bytes)
3. [RFC PATCHv2 0/4] ipvs: Use kthreads for stats (score: 1): Author: Julian Anastasov <ja@xxxxxx>; Date: Fri, 9 Sep 2022 01:21:05 +0300; Hello, This patchset implements stats estimation in kthread context. Simple tests do not show any problem. As it is late here, I'm sending v2 for review. It is interesting to know what value for IPVS; /html/lvs-devel/2022-09/msg00013.html (12,293 bytes)

Current List: 1 - 3
Page: [1]