--- Log opened Tue Jul 14 00:00:19 2020
01:52 < kcalvinalvin> https://github.com/Rjected/btcd/pull/1
01:53 < kcalvinalvin> rjected I made a PR on our btcd repo
02:00 -!- ghost43 [~daer@gateway/tor-sasl/daer] has quit [Remote host closed the connection]
02:01 -!- ghost43 [~daer@gateway/tor-sasl/daer] has joined #utreexo
02:05 < rjected> cool, taking a look
02:26 < rjected> I'm a bit confused as to why the scheduler would be choking from these calls, do you think you can explain that part to me?
02:28 < rjected> kcalvinalvin: is it just something you've noticed during profiling?
03:00 -!- digi_james [sid281632@gateway/web/irccloud.com/x-qxgppfsjpuhvoxcs] has quit [Ping timeout: 244 seconds]
03:01 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has quit [Ping timeout: 240 seconds]
03:02 -!- digi_james [sid281632@gateway/web/irccloud.com/x-llyljsqrlzyhgofo] has joined #utreexo
03:04 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has joined #utreexo
04:18 < kcalvinalvin> rjected I left a new comment in the PR
04:19 < rjected> 👍
04:20 < rjected> yeah that's super helpful
04:46 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has quit [Remote host closed the connection]
04:47 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has joined #utreexo
05:28 < adiabat> yeah 9:00, in 30 min
05:28 < adiabat> kcalvinalvin: the per-tx stuff looks cool; is it slower for fewer, larger txs?
05:33 < adiabat> like bb41a757f405890fb0f5856228e23b715702d714d59bf2b1feb70d8b2b4e3e08
05:34 < adiabat> there might be a threshold where it makes sense to make per-txin goroutines
05:34 < adiabat> like if you have 1000 txs with 5 inputs each, it makes sense to do per-tx, as 1000 routines is plenty to keep the cpu busy
05:35 < adiabat> but if you have 1 tx with 5000 inputs, maybe the scheduler hit of all the routines is worth it, because otherwise the CPU is idle?
05:36 < adiabat> then again the bb41 tx is quite rare, there are not many txs bigger than 100KB
06:45 < adiabat> 35.188.186.244
09:09 < dergoegge> adiabat: i can't reproduce #155, i tried on linux so it does not seem to be linux related. Or maybe it is what distro are you on?
09:13 < dergoegge> What cache size did you use? 20 like on master or 25?
09:55 < adiabat> I'm using the 20 in master
09:57 < adiabat> I trying it again now on the remote server
09:57 < adiabat> I had it happen twice on my local machine, but haven't tried on a different computer
11:15 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has quit [Remote host closed the connection]
11:17 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has joined #utreexo
11:28 < adiabat> huh it got through it on the larger server
11:28 < adiabat> I'll update the issue on github
11:31 < adiabat> I'll try again on the local machine and just let it run longer
11:31 < adiabat> I figured after a few minutes it was stuck, but maybe it just takes a really long time
12:33 < dergoegge> I used a cachesize of 25 so maybe thats why i couldn't reproduce.
12:33 < dergoegge> We are not using any of the ttl stuff currently right?
12:35 < dergoegge> I did a 20min run with -cpuprof and 50% of the is used for leveldb funtions
17:01 -!- adiabat [~adiabat@63.209.32.102] has quit [Ping timeout: 272 seconds]
17:57 -!- adiabat [~adiabat@63.209.32.102] has joined #utreexo
20:30 -!- aaronc2347 [uid451246@gateway/web/irccloud.com/x-dollwddpigmywvxk] has joined #utreexo
--- Log closed Wed Jul 15 00:00:21 2020