lvs-devel
|
To: | Jiri Wiesner <jwiesner@xxxxxxx> |
---|---|
Subject: | Re: [RFC PATCHv6 3/7] ipvs: use u64_stats_t for the per-cpu counters |
Cc: | Simon Horman <horms@xxxxxxxxxxxx>, lvs-devel@xxxxxxxxxxxxxxx, yunhong-cgl jiang <xintian1976@xxxxxxxxx>, dust.li@xxxxxxxxxxxxxxxxx |
From: | Julian Anastasov <ja@xxxxxx> |
Date: | Mon, 14 Nov 2022 13:46:06 +0200 (EET) |
Hello, On Sat, 12 Nov 2022, Julian Anastasov wrote: > + for (;;) { > + start += u64_stats_fetch_begin(&c->syncp); Hm, this should be assignment and not addition... Not a problem for 64-bit tests. > + kconns += u64_stats_read(&c->cnt.conns); > + kinpkts += u64_stats_read(&c->cnt.inpkts); > + koutpkts += u64_stats_read(&c->cnt.outpkts); > + kinbytes += u64_stats_read(&c->cnt.inbytes); > + koutbytes += u64_stats_read(&c->cnt.outbytes); > + if (!u64_stats_fetch_retry(&c->syncp, start)) > + break; > +#if BITS_PER_LONG == 32 > + kconns = conns; > + kinpkts = inpkts; > + koutpkts = outpkts; > + kinbytes = inbytes; > + koutbytes = outbytes; > +#endif Regards -- Julian Anastasov <ja@xxxxxx> |
<Prev in Thread] | Current Thread | [Next in Thread> |
---|---|---|
|
Previous by Date: | Re: [RFC PATCHv6 3/7] ipvs: use u64_stats_t for the per-cpu counters, Julian Anastasov |
---|---|
Next by Date: | Re: [RFC PATCHv6 3/7] ipvs: use u64_stats_t for the per-cpu counters, Jiri Wiesner |
Previous by Thread: | Re: [RFC PATCHv6 3/7] ipvs: use u64_stats_t for the per-cpu counters, Julian Anastasov |
Next by Thread: | Re: [RFC PATCHv6 3/7] ipvs: use u64_stats_t for the per-cpu counters, Jiri Wiesner |
Indexes: | [Date] [Thread] [Top] [All Lists] |