00:33:06  * bnoordhuisquit (Ping timeout: 240 seconds)
01:22:58  * C-Manquit (Quit: Connection reset by beer)
01:29:03  * CIA-70joined
02:27:55  * jroessleinquit (Quit: Leaving...)
03:36:19  * joshthecoderjoined
03:49:58  * xaqjoined
04:07:47  * xaqquit (Remote host closed the connection)
04:10:38  * xaqjoined
04:15:35  * xaqquit (Remote host closed the connection)
07:31:15  * rendarjoined
07:50:16  * luitejoined
07:52:41  <luite>i'm trying to optimize some loop overan an array that stores both numbers and objects (functions). it's really fast when i look up only functions, but when i look up a number (integer) it's a lot slower. how are these mixed arrays implemented in v8? is this expected?
09:18:04  * joshthecoderquit (Quit: Linkinus - http://linkinus.com)
09:23:31  * Raynosquit (Excess Flood)
09:24:51  * Raynosjoined
09:42:48  <CIA-70>yangguo@chromium.org * r12535 /branches/bleeding_edge/src/ (x64/code-stubs-x64.cc ia32/code-stubs-ia32.cc): (log message trimmed)
09:42:48  <CIA-70>Improve the assembly code for power function with integer exponential on x64
09:42:48  <CIA-70>The change removes one unused multiply and reschedules
09:42:48  <CIA-70>the shift, multiply and jump instructions to reduce
09:42:48  <CIA-70>stall. Experiment shows it improve about 20% performance
09:42:48  <CIA-70>on x64 for exponetials from about 100 to 2000.
09:42:49  <CIA-70>Review URL: https://chromiumcodereview.appspot.com/10939013
10:33:31  * C-Manjoined
10:54:25  * elliottcablejoined
10:54:45  * Net147joined
11:00:43  * bnoordhuisjoined
11:03:13  * [[zzz]]joined
11:07:15  * [[zz]]quit (Ping timeout: 272 seconds)
11:33:00  * xan_joined
13:02:56  * [[zzz]]changed nick to [[zz]]
13:06:46  * Net147quit (Quit: HydraIRC -> http://www.hydrairc.com <- \o/)
13:17:22  <CIA-70>ulan@chromium.org * r12536 /data/benchmarks/v7/run.html:
13:17:22  <CIA-70>Show hint to Octane.
13:17:22  <CIA-70>R=stefanoc@chromium.org
13:17:22  <CIA-70>Review URL: https://codereview.appspot.com/6497145
13:17:23  <CIA-70>mstarzinger@chromium.org * r12537 /branches/bleeding_edge/src/hydrogen-instructions.cc:
13:17:23  <CIA-70>Fix printing of HCallKnownGlobal.
13:17:23  <CIA-70>R=jkummerow@chromium.org
13:17:23  <CIA-70>Review URL: https://codereview.chromium.org/10941009
14:48:52  * TheJHjoined
15:23:16  <CIA-70>verwaest@chromium.org * r12538 /branches/bleeding_edge/ (8 files in 2 dirs):
15:23:16  <CIA-70>Preallocate space in descriptor arrays.
15:23:16  <CIA-70>Review URL: https://chromiumcodereview.appspot.com/10916336
15:23:17  <CIA-70>verwaest@chromium.org * r12539 /branches/bleeding_edge/src/mips/ (3 files):
15:23:17  <CIA-70>MIPS: Do not go to slow mode and back to fast in initializer blocks.
15:23:17  <CIA-70>Port r12534 (4acfb92e)
15:23:17  <CIA-70>BUG=
15:23:17  <CIA-70>TEST=
15:23:18  <CIA-70>Review URL: https://chromiumcodereview.appspot.com/10905313
15:23:19  <CIA-70>Patch from Akos Palfi <palfia@homejinni.com>.
15:23:19  <CIA-70>verwaest@chromium.org * r12540 /branches/bleeding_edge/src/mips/ (3 files):
15:23:20  <CIA-70>Revert due to invalid description / CL combination.
15:23:20  <CIA-70>Revert "MIPS: Do not go to slow mode and back to fast in initializer blocks."
15:23:21  <CIA-70>This reverts commit 9b05931ec130e831f7496aed6f7ae32e8f2da934.
15:57:20  * RT|Chatzillaquit (Quit: ChatZilla 0.9.86.1 [Firefox 2.0.0.22pre/2009081014])
16:09:57  * TheJH_joined
16:09:59  * TheJH_quit (Read error: Connection reset by peer)
16:40:33  * xaqjoined
16:49:34  * eohjoined
16:50:44  * xaqquit (Remote host closed the connection)
17:00:13  * xaqjoined
17:02:52  * xaq_joined
17:03:07  * xaqquit (Read error: Connection reset by peer)
17:24:13  * joshthecoderjoined
17:46:14  * eoh|joined
17:46:49  * eoh|quit (Read error: Connection reset by peer)
17:47:09  * eoh|joined
17:47:21  * eohquit (Read error: Connection reset by peer)
17:47:35  * AndreasMadsenjoined
17:59:08  * xaq_quit (Remote host closed the connection)
17:59:45  * xaqjoined
18:00:06  * xaqquit (Remote host closed the connection)
19:21:45  * AndreasMadsenquit (Remote host closed the connection)
19:42:07  * joshthecoderquit (Quit: Leaving...)
19:55:17  * mralephjoined
19:56:18  <mraleph>luite: this is not expected load speed should be the same. but it is impossible to say what happens without knowning more about the code.
20:02:21  <luite>mraleph: oh i found the problem, some data in the array was wrong and it tried to compare numbers to functions sometimes
20:03:19  <mraleph>luite: yeah that would not be fast :-)
20:03:42  <luite>still the function only has 25% type information (from node --trace-opt), is there a way to find out what it knows and what it doesn't?
20:06:20  <luite>oh one other thing, i use a lot of functions with a few extra properties set (on the function object), is that just as fast as properties on regular objects?
20:14:22  * joshthecoderjoined
20:19:59  <mraleph>luite: that is no simple way to find that out. in general it does not no anything about the parts of the function that were not executed.
20:20:24  <mraleph>s/does not no/does not know/
20:21:01  <mraleph>lute: it should be as fast as on normal objects. internally functions are just a variation of an object.
20:24:24  <luite>great, thanks.
20:27:10  * xaqjoined
21:22:18  * TheJHquit (Quit: goodbye)
21:31:14  * rendarquit
21:53:14  * xan_quit (Ping timeout: 240 seconds)
21:54:38  * xan_joined
22:13:54  * xaqquit (Remote host closed the connection)
22:17:46  * RT|Chatzillajoined
23:16:22  * xaqjoined
23:18:11  * xaqquit (Remote host closed the connection)