00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:07  * ircretaryjoined
00:00:32  * indutny&
00:00:33  <LOUDBOT>PLZ ELABORATE OH THE LEGOZ K THX
00:00:34  <indutny>ttyl
00:30:02  * kazuponjoined
00:34:45  * kazuponquit (Ping timeout: 248 seconds)
00:42:10  * TooTallNatequit (Quit: Computer has gone to sleep.)
00:42:30  * pooyaquit (Quit: pooya)
01:05:18  * timoxleyquit (Ping timeout: 252 seconds)
01:05:30  * TooTallNatejoined
01:05:32  * TooTallNatequit (Client Quit)
01:07:39  * timoxleyjoined
01:23:40  * dominictarrquit (Quit: dominictarr)
01:30:37  * kazuponjoined
01:35:08  * kazuponquit (Ping timeout: 245 seconds)
01:58:19  * st_lukequit (Remote host closed the connection)
02:31:12  * kazuponjoined
02:36:06  * kazuponquit (Ping timeout: 264 seconds)
02:39:39  * c4milojoined
02:59:39  * brsonquit (Quit: leaving)
03:05:46  * st_lukejoined
03:31:51  * kazuponjoined
03:36:21  * kazuponquit (Ping timeout: 252 seconds)
03:58:15  * st_lukequit (Remote host closed the connection)
03:59:06  * st_lukejoined
04:12:39  * loladiroquit (Quit: loladiro)
04:15:15  * brsonjoined
04:32:30  * kazuponjoined
04:34:54  * brsonquit (Ping timeout: 264 seconds)
04:35:35  * c4miloquit (Remote host closed the connection)
04:36:01  * c4milojoined
04:36:43  * brsonjoined
04:37:24  * kazuponquit (Ping timeout: 264 seconds)
04:38:56  * timoxleyquit (Quit: Computer has gone to sleep.)
04:41:02  * c4miloquit (Ping timeout: 272 seconds)
04:44:09  * kazuponjoined
04:48:37  * kazuponquit (Ping timeout: 248 seconds)
05:06:03  * st_lukequit (Remote host closed the connection)
05:38:15  * timoxleyjoined
06:03:17  * mikealquit (Quit: Leaving.)
06:06:50  * timoxleyquit (Quit: Computer has gone to sleep.)
06:12:17  * stagasjoined
06:24:09  <MI6>joyent/node: Fedor Indutny master * f2d5cea : benchmark: add chunked-encoding benchmark (+4 more commits) - http://git.io/uFcBvA
06:24:45  <tjfontaine>dun dun dun
06:27:52  <isaacs>:D
06:29:23  <tjfontaine>windows build slave is still misbehaving
06:42:22  <MI6>joyent/node: isaacs master * 9cfc929 : doc: stream writev cleanups - http://git.io/bKtqRg
06:45:21  <isaacs>indutny: landed, thanks
07:04:51  * rendarjoined
07:10:38  <MI6>nodejs-master: #180 FAILURE smartos-x64 (2/584) osx-ia32 (1/584) windows-ia32 (151/584) smartos-ia32 (1/584) windows-x64 (145/584) http://jenkins.nodejs.org/job/nodejs-master/180/
07:50:34  <indutny>kewl
07:50:35  <indutny>thank you
07:56:49  * brsonquit (Quit: leaving)
08:04:15  * stolsmajoined
08:06:13  * pooyajoined
08:21:41  * pooyaquit (Quit: pooya)
08:24:29  * bnoordhuisjoined
08:32:57  * stagasquit (Read error: Connection reset by peer)
09:05:52  * benoitcquit (Ping timeout: 245 seconds)
09:07:44  * stagasjoined
09:08:49  * benoitcjoined
09:18:44  * hzjoined
09:20:27  * bnoordhuisquit (Ping timeout: 245 seconds)
09:38:45  * dominictarrjoined
10:23:36  * stagasquit (Ping timeout: 264 seconds)
12:16:12  * benoitcquit (Excess Flood)
12:19:28  * benoitcjoined
13:20:53  * hzquit
13:46:52  * AvianFlujoined
14:09:47  * abraxasjoined
14:10:50  * defunctzombie_zzchanged nick to defunctzombie
14:14:10  * c4milojoined
14:14:14  * abraxasquit (Ping timeout: 252 seconds)
14:45:53  * defunctzombiechanged nick to defunctzombie_zz
14:45:56  * kazuponjoined
14:52:19  * jmar777joined
15:18:58  * bnoordhuisjoined
15:21:40  * jmar777quit (Read error: Connection reset by peer)
15:22:18  * jmar777joined
15:23:58  * jmar777quit (Remote host closed the connection)
15:30:43  * defunctzombie_zzchanged nick to defunctzombie
15:32:58  * defunctzombiechanged nick to defunctzombie_zz
15:43:09  * kenperkinsquit (Quit: Computer has gone to sleep.)
15:52:22  * kenperkinsjoined
16:10:04  * bnoordhuisquit (Ping timeout: 260 seconds)
16:14:54  * timoxleyjoined
16:16:32  * loladirojoined
16:21:23  * kazuponquit (Remote host closed the connection)
16:43:29  * loladiroquit (Quit: loladiro)
16:50:28  <indutny>hoya
16:52:49  * bnoordhuisjoined
17:00:15  * timoxleyquit (Quit: Computer has gone to sleep.)
17:01:49  * timoxleyjoined
17:11:48  * kenperkinsquit (Quit: Computer has gone to sleep.)
17:31:52  * kenperkinsjoined
17:40:08  * loladirojoined
17:50:58  * loladiroquit (Quit: loladiro)
18:26:03  * TooTallNatejoined
18:30:36  * mikealjoined
18:32:52  * loladirojoined
18:54:41  * timoxleyquit (Quit: Computer has gone to sleep.)
19:02:14  * timoxleyjoined
19:21:19  * timoxleyquit (Quit: Computer has gone to sleep.)
19:36:03  * mikealquit (Quit: Leaving.)
19:37:00  <MI6>nodejs-master: #182 UNSTABLE smartos-x64 (3/584) windows-ia32 (145/584) smartos-ia32 (1/584) windows-x64 (144/584) linux-ia32 (1/584) http://jenkins.nodejs.org/job/nodejs-master/182/
19:39:03  <MI6>libuv-master: #84 UNSTABLE windows (4/189) smartos (3/188) linux (1/188) osx (1/188) http://jenkins.nodejs.org/job/libuv-master/84/
19:48:21  * stolsmaquit (Ping timeout: 248 seconds)
19:54:00  * jmar777joined
19:57:33  <tjfontaine>it takes the same amount of time for {smartos,linux,osx}{ia32,x64} to complete a build/test cycle run serially on each platform, as it does for one windows build/test cycle
19:58:58  <indutny>well
19:59:02  <indutny>are you surpsied?
19:59:11  <tjfontaine>no, not really
19:59:17  <indutny>neither I
19:59:18  <tjfontaine>just frustrated :)
19:59:32  * indutnyis working on property cache
19:59:40  <indutny>almost figured it out
19:59:45  <indutny>just need to hook it up into runtime
19:59:54  <indutny>without copying all code
20:00:12  <tjfontaine>cute, have you done any benches with it?
20:00:24  * mikealjoined
20:00:43  <indutny>its not ready for it :)
20:00:45  <indutny>it just caches
20:00:54  <indutny>now I need to use that cache
20:01:08  <indutny>also, there seems to be one shared cache for all property lookup
20:01:12  <indutny>but its too small
20:01:15  <indutny>just 64 entries
20:01:52  <tjfontaine>ah
20:02:05  <tjfontaine>are you having fun with it?
20:07:37  * jmar777quit (Remote host closed the connection)
20:10:40  * abraxasjoined
20:12:09  * c4miloquit (Remote host closed the connection)
20:12:35  * c4milojoined
20:12:53  * brsonjoined
20:15:04  * abraxasquit (Ping timeout: 260 seconds)
20:17:44  * c4miloquit (Ping timeout: 272 seconds)
20:23:51  * c4milojoined
20:25:46  <indutny>ok, it seems to be faster
20:26:00  <indutny>by 20%
20:26:03  <indutny>but that's debug build
20:26:17  <indutny>lets try release build
20:26:27  <indutny>and I was benchmarking ->Get()
20:26:32  <MI6>libuv-node-integration: #35 UNSTABLE osx-x64 (1/584) osx-ia32 (1/584) smartos-x64 (3/584) windows-x64 (145/584) smartos-ia32 (1/584) windows-ia32 (145/584) http://jenkins.nodejs.org/job/libuv-node-integration/35/
20:32:33  <indutny>23% faster
20:32:34  <indutny>yay!
20:32:35  <indutny>great
20:34:18  * Raynosquit (Ping timeout: 256 seconds)
20:34:58  <tjfontaine>yay :)
20:36:15  * c4miloquit (Remote host closed the connection)
20:36:42  * c4milojoined
20:38:38  * stagasjoined
20:41:30  * c4miloquit (Ping timeout: 258 seconds)
20:44:33  <indutny>its probably buggy :)
20:44:36  <indutny>but works so far
20:44:44  <indutny>finishing caching for ->Set()
20:45:03  <indutny>I wonder how much it'll affect node.js performance
20:47:16  * loladiroquit (Quit: loladiro)
20:48:54  <tjfontaine>indeed
20:53:11  * dominictarrquit (Quit: dominictarr)
20:56:49  * rendarquit
21:07:40  <indutny>tjfontaine: https://codereview.chromium.org/14425011
21:07:51  <indutny>its not finished yet
21:11:22  * dominictarrjoined
21:14:03  <tjfontaine>seems like a clean patch thus far
21:29:36  <indutny>heh
21:29:41  <indutny>it took a lot of time
21:29:47  <indutny>running v8 tests right now
21:29:58  <indutny>http_parser seems to be running a bit faster with it
21:30:07  <indutny>but I'm not really sure
21:30:08  <indutny>yet
21:30:18  <indutny>also, I haven't changed it much
21:30:32  <indutny>just cached lookups to various callbacks
21:30:38  <indutny>like headers_complete, etc
21:39:55  <indutny>bnoordhuis: oh here you are
21:40:51  * Raynosjoined
21:41:05  * loladirojoined
21:41:40  * loladiroquit (Client Quit)
21:42:21  * saghulquit (Quit: ["Textual IRC Client: www.textualapp.com"])
21:44:22  <bnoordhuis>indutny: indeed, there i am
21:44:27  <indutny>https://codereview.chromium.org/14425011/
21:44:31  <indutny>take a look at it
21:44:39  <bnoordhuis>what's the magic word?
21:45:57  * mikealquit (Quit: Leaving.)
21:48:05  <indutny>please :)
21:48:08  <indutny>but
21:48:12  <indutny>its not like I want you to review it
21:48:15  <indutny>I just share it with you
21:48:19  <indutny>and its like
21:48:26  <indutny>hey man, take a look at this stuff, its awesome
21:48:28  <indutny>so
21:48:33  <indutny>please doesn't really fit it
21:50:47  <bnoordhuis>ah, you mean something like 'check this out'
21:50:55  <indutny>yes
21:51:36  <bnoordhuis>though i cannot read that and not hear run dmc in my head
21:52:06  <bnoordhuis>looks neat, fedor
21:52:42  * defunctzombie_zzchanged nick to defunctzombie
21:52:55  <indutny>heh
21:53:04  <indutny>I need to try it with node
21:53:18  <indutny>to figure out if it gives us anything
21:53:23  <indutny>do you want to help me with it?
21:55:17  <trevnorris>indutny: looking at the patch. have a one sentence description?
21:55:44  * mikealjoined
21:55:54  <indutny>trevnorris: just one word https://gist.github.com/indutny/44586daf809530f8b6f0
21:58:55  <bnoordhuis>indutny: Real Men would use something like `int* count_field = set_property ? &hidden_count_ : &count_;` and then do `*count_field = count;`
21:59:07  <indutny>bnoordhuis: I did it
21:59:08  <indutny>but then
21:59:22  <indutny>I decided that I don't want to pass pointer into loop condition
21:59:54  <bnoordhuis>yeah, sensible
22:01:50  <indutny>mraleph told me that its too complex
22:02:01  <indutny>so it might not get into upstream
22:02:08  <indutny>but we should figure out if it gives anything to us
22:02:12  <indutny>and probably float-patch it
22:02:19  <trevnorris>heh. looking at this, have no idea how I'd implement the SetIndexedPropertiesToExternalArrayData been wanting to work on.
22:02:29  <trevnorris>so many parts to consider.
22:04:17  <indutny>wooow it crashes
22:05:07  <indutny>ah, GC old fag
22:06:14  <trevnorris>indutny: looks cool. so would this be faster than GetAlignedPointerFromInternalField?
22:06:28  <indutny>I don't know
22:06:32  <indutny>but you could benchmark it ;)
22:07:09  <trevnorris>sounds good to me. :)
22:07:37  <trevnorris>just wrap the class in an Extern and pass that in. hopefully it is. happens enough.
22:14:23  * wolfeidauquit (Remote host closed the connection)
22:16:30  <trevnorris>nice. applied cleanly to master. now time to give it a test drive.
22:16:41  * mikealquit (Quit: Leaving.)
22:19:45  <indutny>wait a minute
22:19:47  <indutny>reapply it
22:19:52  <indutny>:)
22:19:54  <indutny>I've some fixes
22:20:03  <indutny>otherwise you'll get crashes
22:20:10  <indutny>well, you probably will get them anyway
22:20:15  <indutny>I'm still polishing some stuff
22:20:28  <indutny>yeah, its still crashing
22:20:30  <indutny>shit
22:21:39  <trevnorris>indutny: when? i'm running it right now.
22:21:48  <indutny>well, I did some patches to node
22:21:53  <indutny>its hard to tell what exactly crashes so far
22:21:57  <indutny>but it should mostly work for you
22:22:06  <trevnorris>hm. same here. I patched stream_wrap
22:22:07  * indutnybuilds dbg version of node
22:22:16  <indutny>crashed?
22:22:27  <trevnorris>nope. just ran http simple test
22:22:31  <trevnorris>*benchmark
22:23:03  <trevnorris>(though, just realized I'm running this on top of my buffer-buffet branch. not just master)
22:23:48  <trevnorris>but that shouldn't be affecting this.
22:24:29  <indutny>ok
22:25:50  <indutny>ah, I believe I'm just doing it a bit incorrectly
22:26:02  <indutny>I think I'm not taking in account some transition changes
22:26:23  <trevnorris>hm. ok. I am getting a few crashes in the test. let me log an make sure the http bench is hitting the code I changed.
22:29:27  <indutny>or I'm misusing LookupTarget
22:29:31  <indutny>err
22:29:35  <indutny>LookupResult
22:29:42  <indutny>anyway, I hope v8 team will help me to figure out
22:29:45  * defunctzombiechanged nick to defunctzombie_zz
22:31:10  <indutny>trevnorris: can you try building node with bleeding edge v8?
22:31:13  <indutny>with this patch
22:31:14  <indutny>?
22:31:21  <trevnorris>sure
22:31:30  <indutny>thank you
22:31:36  * wolfeidaujoined
22:34:20  * mikealjoined
22:34:43  * bnoordhuisquit (Ping timeout: 252 seconds)
22:36:23  <trevnorris>indutny: anything specific for me to look at, or just do what I was going to do?
22:36:34  * bnoordhuisjoined
22:39:44  <indutny>I think just do the same
22:41:12  * stagasquit (Quit: ChatZilla 0.9.90-rdmsoft [XULRunner 1.9.0.17/2009122204])
22:41:15  <trevnorris>coolio
22:42:51  * saghuljoined
22:43:31  <trevnorris>indutny: hm. might have to get off in a minute (kid just woke up) but I think bleeding edge is making the http tests ~5% slower than master.
22:43:39  <trevnorris>will double check, but just fyi if I have to jam.
22:44:38  <trevnorris>ok, yeah. fluke.
22:48:42  <trevnorris>indutny: fyi, first change was to replace it in "StreamWrap::UpdateWriteQueueSize". seems to work fine. tests are running now.
22:53:13  <indutny>kewl
22:53:20  <indutny>basically, there's some problem with transitions
22:53:24  <indutny>it caches transition id
22:53:38  <indutny>but then, suddenly, the same map has different transitions count
22:53:43  <indutny>and it fails
22:55:59  * bnoordhuisquit (Ping timeout: 252 seconds)
22:56:38  <indutny>ah, map could be evacuated
22:56:40  <indutny>oh god
22:59:53  <trevnorris>well, fwiw replaced every instance of ->Set in stream_wrap to use 'em and all tests are passing.
23:01:49  <indutny>interesting
23:01:51  <indutny>what about speed?
23:02:01  <indutny>is there any improvement?
23:02:20  <trevnorris>running those now.
23:06:47  <trevnorris>ugh. ok. need to re-run the tests. one against bleeding edge, then against your patch.
23:07:01  <trevnorris>was running against master but realized that'll be a useless test.
23:08:29  <indutny>yeah
23:08:33  <indutny>ok, I'm going to sleep
23:08:35  <indutny>ttyl
23:08:37  <trevnorris>night.
23:11:04  <indutny>thanks
23:17:13  <trevnorris>indutny: from changing all Set in stream_wrap. no perf changes I can immediately see.
23:20:10  <kellabyte>anyone familiar with http_parser_parse_url() from http_parser? I'm using it in on_url but it's not parsing the url correctly, it only gets the path like /one/two but not the port, host, etc
23:20:22  * timoxleyjoined
23:22:30  * mikealquit (Quit: Leaving.)
23:23:00  * mikealjoined
23:26:30  * loladirojoined
23:42:42  * timoxleyquit (Ping timeout: 264 seconds)
23:48:05  * skebcioquit (Read error: Connection reset by peer)
23:48:12  * skebciojoined
23:54:47  * timoxleyjoined
23:54:47  * timoxleyquit (Client Quit)
23:56:19  * brsonquit (Remote host closed the connection)