00:26:14 tsuru`` [~charlie@adsl-98-87-49-34.bna.bellsouth.net] has joined #sbcl 00:28:05 -!- tsuru` [~charlie@adsl-74-179-30-15.bna.bellsouth.net] has quit [Ping timeout: 258 seconds] 02:57:36 -!- tsuru`` [~charlie@adsl-98-87-49-34.bna.bellsouth.net] has quit [Ping timeout: 240 seconds] 03:14:59 that's odd: hunchentoot is ~25% faster on sbcl/wine/amd64 than on sbcl/linux/amd64 built from the same sources... 03:52:13 -!- loke [~elias@bb115-66-85-121.singnet.com.sg] has quit [Ping timeout: 240 seconds] 03:57:09 loke [~elias@bb115-66-85-121.singnet.com.sg] has joined #sbcl 06:03:12 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 276 seconds] 06:20:35 Shucks, I didn't realize threading wasn't supported on Darwin PPC 06:38:04 sdemarre [~serge@91.176.50.149] has joined #sbcl 07:23:19 akovalenko: what's "faster" ... more throughput? less latency? less CPU? 07:32:20 flip214: more throughput (req/s and byte/s). I didn't look too deep (it can be threading, network, gc, or countless other things), just wanted to know if I've done something foolish w.r.t. performance. Seems unlikely :) 07:37:07 do you have a pcap output? perhaps we could see whether it's increased parallization ... 07:39:08 no (i've looked at "ab" output only, with slightly varied -c and absent/present -k). 07:39:45 -!- loke [~elias@bb115-66-85-121.singnet.com.sg] has quit [Ping timeout: 260 seconds] 07:39:50 well, I've known that wine is sometimes faster than windows (because of better disk caching in linux), but wine better than linux? strange ... 07:40:27 are this perhaps very small requests, so that via wine more caching on the sockets is done, or perhaps O_NODELAY defaults in wine/linux? 07:41:52 As of defaults, it may be so. Requests are 10-20 kbytes, so not too large indeed. 07:44:43 hmmm, but much larger than the MTU ... are you going via loopback? 07:48:20 It was with loopback (my home wifi latency drowns any possible differences caused by /hunchentoot/ otherwise..) 07:51:36 loke [~elias@bb115-66-85-121.singnet.com.sg] has joined #sbcl 07:53:31 hmmm ... with a pcap file we might see whether the difference is network or implementation related 07:55:00 With loopback, we instantly see that it's totally implementation-related, don't we? (when I said that network performance might be involed here, I meant SBCL's or Wine's I/O code) 07:56:24 real investigation of the issue requires more work than I'm ready to do at this point (and it all doesn't seem too useful..). 08:05:32 even with loopback the packets have to get into the kernel and out again ... so if wine is _much_ faster (more than, well, 2-10%) I'd think there's some explaining to do ... 08:06:06 especially as IO is what sbcl should be (get) very good at ... I'd like to use it for high-performance network things 08:32:36 Blkt [~user@89-96-199-46.ip13.fastwebnet.it] has joined #sbcl 08:40:05 nikodemus [~Nikodemus@cs181063174.pp.htv.fi] has joined #sbcl 08:40:05 -!- ChanServ has set mode +o nikodemus 08:56:00 -!- nikodemus [~Nikodemus@cs181063174.pp.htv.fi] has quit [Remote host closed the connection] 09:07:31 hlavaty [~user@91-65-217-112-dynip.superkabel.de] has joined #sbcl 09:09:55 nikodemus [~nikodemus@cs181056239.pp.htv.fi] has joined #sbcl 09:09:55 -!- ChanServ has set mode +o nikodemus 10:20:31 -!- drdo` [~drdo@85.207.54.77.rev.vodafone.pt] has quit [Remote host closed the connection] 10:20:46 drdo`` [~drdo@85.207.54.77.rev.vodafone.pt] has joined #sbcl 10:23:21 akovalen` [~anton@95.72.168.23] has joined #sbcl 10:24:39 -!- pchrist [~spirit@gentoo/developer/pchrist] has quit [Ping timeout: 252 seconds] 10:25:09 -!- akovalenko [~anton@95.73.104.74] has quit [Ping timeout: 255 seconds] 10:25:51 pchrist_ [~spirit@gentoo/developer/pchrist] has joined #sbcl 10:26:30 oiiii [~oiiii@82.71.241.25] has joined #sbcl 10:28:22 -!- spacebat [~spacebat@ubermonkey.net] has quit [*.net *.split] 18:09:52 ccl-logbot [~ccl-logbo@setf.clozure.com] has joined #sbcl 18:09:52 18:09:52 -!- names: ccl-logbot antgreen Qworkescence joshe homie tsuru` nyef fe[nl]ix lichtblau slyrus Guest30207 jsnell deepfire luis flip214 antoszka danlarkin dsp_ loke @nikodemus akovalenko pchrist_ drdo`` sdemarre jiacobucci cmm reb` christoph_debian sbryant whoops Phoodus blumbri antifuchs ljos daimrod kanru @Xof foom cow-orker redline6561 |3b| spacebat sshirokov MikeSeth pkhuong 18:14:31 -!- ljos [~mozzyb@login1.uib.no] has quit [Remote host closed the connection] 18:47:12 ljos [~mozzyb@login1.uib.no] has joined #sbcl 18:49:04 homie` [~levgue@xdsl-78-35-177-246.netcologne.de] has joined #sbcl 18:51:40 -!- homie [~levgue@xdsl-78-35-143-123.netcologne.de] has quit [Ping timeout: 255 seconds] 18:53:28 https://github.com/nikodemus/sb-texinfo # not a contrib yet, but this has docstrings.lisp with all the bells and whistles i could find 19:01:09 -!- nikodemus [~nikodemus@cs181056239.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 20:02:09 tsuru`` [~charlie@adsl-98-87-46-38.bna.bellsouth.net] has joined #sbcl 20:03:51 -!- tsuru` [~charlie@adsl-98-87-47-198.bna.bellsouth.net] has quit [Read error: Connection reset by peer] 20:14:42 peddie [~peddie@repl.esden.net] has joined #sbcl 20:33:49 -!- homie` [~levgue@xdsl-78-35-177-246.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 20:45:23 -!- nyef [~nyef@c-174-63-105-188.hsd1.ma.comcast.net] has quit [Quit: G'night all.] 21:30:05 prxq [~mommer@mnhm-4d01223f.pool.mediaWays.net] has joined #sbcl 21:51:17 -!- sdemarre [~serge@91.176.50.149] has quit [Ping timeout: 240 seconds] 22:12:41 -!- prxq [~mommer@mnhm-4d01223f.pool.mediaWays.net] has quit [Quit: Leaving] 22:50:34 udzinari [~user@ip-89-102-12-6.net.upcbroadband.cz] has joined #sbcl 23:00:23 tsuru``` [~charlie@adsl-74-179-31-75.bna.bellsouth.net] has joined #sbcl 23:01:41 -!- tsuru`` [~charlie@adsl-98-87-46-38.bna.bellsouth.net] has quit [Ping timeout: 240 seconds]