00:03:07 scymtym_: seems to be a libm issue. (compile nil '(lambda () (asin -1.0f0))) works here. 00:04:00 pkhuong: did you try with debug 3 and safety 3? 00:04:56 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 00:05:32 yup 00:05:48 ok, anything else i can try? 00:09:36 btw: (compile nil '(lambda () (asin -1.0d0))) works fine 00:09:45 but -1.0f0 fails 00:11:25 (and *read-default-float-format* is double-float, no idea whether that matters) 00:12:08 but calling asin works? 00:12:42 actually, no. i just didn't try yet :) 00:12:43 CL-USER> (asin -1.0f0) 00:12:43 ; Evaluation aborted on # (DOUBLE-FLOAT -1.5707963267948966 1.5707963267948966) 00:12:46 datum: -1.5707963267948966>. 00:13:48 ...wtf? 00:14:28 and without slime as well? 00:15:09 yes 00:16:18 i can try on a different x86 system tomorrow 00:16:28 maybe this machine has some problem 00:17:10 long shot, but can you disassemble asin, and find the addresses for FLDD, then sap-ref-double there? 00:18:18 i will try 00:18:40 should be around L5, two adjacent addresses 00:19:23 this one? 00:19:24 ; 6ED: L5: DDD8 FSTPD FR0 00:19:24 ; 6EF: DD05982DEE09 FLDD [#x9EE2D98] 00:19:37 right, and the other is probably D90 00:22:53 how do i turn an integer into a sap? 00:24:12 got it 00:24:14 (values (sb-int::sap-ref-double (sb-int::int-sap #x9EE2D90) 0) 00:24:15 (sb-int::sap-ref-double (sb-int::int-sap #x9EE2D98) 0)) 00:24:15 -1.5707963267948966 00:24:18 1.5707963267948966 00:27:44 would it be possible that/make any difference if the *read-default-float-format* = double-float somehow leaked into the host compiler? 00:32:14 I'm pretty sure we set that variable to a default value before bootstrapping 00:34:14 if i recompile asin in the running image, the problem disappears 00:35:45 -!- wbooze [~wbooze@xdsl-78-35-189-86.netcologne.de] has quit [Ping timeout: 248 seconds] 00:35:45 there goes my hypothesis of libm setting the precision field to single float ;) 00:43:01 recompiling only helps when debug is 3 00:46:32 no, there are other combinations which yield a working asin 00:46:49 but the compiler policy definitely has an impact 00:54:37 scymtym_: just a bit before L5, there should be a CALL EBX. After that, a sequence of FLDZ and FXCH. On working builds, is the value written to the stack soon after? 00:58:33 pkhuong: i can't see a difference in that area 00:58:44 trying to diff the disassemblies 00:58:51 or, can you paste the disassemblies from working and broken asin? 01:00:20 i would have suggested that :) 01:02:00 http://paste.lisp.org/display/137344 01:02:24 around the original l5, there is a difference 01:03:16 the broken version operates on fr1 and the working version operates on a stack slot 01:04:37 -!- davazp [~user@178.167.175.194.threembb.ie] has quit [Ping timeout: 246 seconds] 01:05:33 DD55D8 FSTD [EBP-40] 01:06:19 yup. So on working builds, the value in FR0/1 is stored on the stack, rounding it to 64 bit 01:07:29 just to clarify, i don't have a working /build/ at the moment, i can "repair" asin by recompiling with suitable policy in the running image 01:08:37 yeah yeah. 01:10:33 :) 01:10:49 bisecting 1.1.7..HEAD 01:12:35 right... really doesn't look like a straight regression 01:13:27 We can force rounding by evicting to memory and back, but I'm really not sure I agree about the double rounding libm is making us do. 01:14:27 I'd be surprised if there's any difference with 1.1.7. 01:15:28 (sorry for the sharpness, I was getting really angry at the wifi for a while here) 01:15:34 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Ping timeout: 256 seconds] 01:16:01 (no problem) 01:17:32 if bisecting does not help, i could reboot this machine and repeat the cl-test-grid runs for 1.1.7 and master and check if that still indicates a regression for cl-geo 02:04:55 -!- nyef [~nyef@pool-70-109-132-87.cncdnh.east.myfairpoint.net] has quit [Quit: G'night all.] 02:23:23 robgssp [~user@cpe-24-93-28-218.rochester.res.rr.com] has joined #sbcl 02:23:39 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 02:42:49 -!- christoph4 [~christoph@ppp-188-174-116-180.dynamic.mnet-online.de] has quit [Ping timeout: 252 seconds] 02:50:38 pkhuong: i think, i bisected the asin problem to "34f08ec * bad Allow inlining more calls to INVOKE-WITH-SAVED-FP-AND-PC during XC." 02:53:23 christoph4 [~christoph@ppp-188-174-143-5.dynamic.mnet-online.de] has joined #sbcl 03:05:10 echo-area [~user@182.92.247.2] has joined #sbcl 03:18:37 attila_lendvai [~attila_le@92.47.191.212] has joined #sbcl 03:18:37 -!- attila_lendvai [~attila_le@92.47.191.212] has quit [Changing host] 03:18:37 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 03:28:49 minion: memo for pkhuong: i think, i bisected the asin problem to "34f08ec * bad Allow inlining more calls to INVOKE-WITH-SAVED-FP-AND-PC during XC." 03:28:50 Remembered. I'll tell pkhuong when he/she/it next speaks. 03:54:32 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Read error: Connection reset by peer] 03:55:11 attila_lendvai [~attila_le@92.47.191.212] has joined #sbcl 03:55:11 -!- attila_lendvai [~attila_le@92.47.191.212] has quit [Changing host] 03:55:11 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 04:31:40 teggi [~teggi@123.20.101.86] has joined #sbcl 04:56:19 minion: forget my memos 04:56:20 OK, I threw it out. 04:57:34 scymtym_: right that makes sense... So, it might be experienced as a regression. We'll see if anyone (nyef?) has an opinion on evicting FP results to memory. 05:03:49 -!- Bike [~Glossina@75-175-78-112.ptld.qwest.net] has quit [Ping timeout: 245 seconds] 05:05:13 Bike [~Glossina@75-175-78-112.ptld.qwest.net] has joined #sbcl 05:06:57 slyrus [~chatzilla@adsl-99-183-240-185.dsl.pltn13.sbcglobal.net] has joined #sbcl 05:21:19 -!- slyrus [~chatzilla@adsl-99-183-240-185.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 252 seconds] 05:32:23 gor[e] [~svr@79.165.187.105] has joined #sbcl 06:29:57 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Ping timeout: 248 seconds] 06:36:42 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 06:39:24 prxq [~mommer@mnhm-590c0d6f.pool.mediaWays.net] has joined #sbcl 06:40:53 pkhuong: fwiw, I think it is imprecise: (values character &optional) means "definitely one-value, which is a character" -- it's different from (values character) which means "if there is a first value, then it's a character" 06:41:34 so (or (values character &optional) (values &optional)) means "either exactly no values or exactly one value which is a character" 06:42:19 but. Let's see what happens :) 06:44:12 also, yes, I think an RC later in the week 06:46:23 (duh, your patch of course handles the distinction correctly. Huh, ok, it really is "how did this work for a decade?") 07:03:33 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 07:10:12 attila_lendvai [~attila_le@87.247.13.73] has joined #sbcl 07:10:12 -!- attila_lendvai [~attila_le@87.247.13.73] has quit [Changing host] 07:10:12 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 07:39:22 -!- psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has quit [Quit: Leaving.] 08:34:22 rudi [~rudi@1x-193-157-251-78.uio.no] has joined #sbcl 09:38:28 -!- yacks [~py@180.151.36.168] has quit [Read error: Operation timed out] 09:55:46 -!- scymtym_ [~user@ip-5-147-122-209.unitymediagroup.de] has quit [Ping timeout: 252 seconds] 10:06:10 yacks [~py@180.151.36.168] has joined #sbcl 10:08:10 stassats [~stassats@wikipedia/stassats] has joined #sbcl 11:13:52 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 11:21:59 davazp [~user@31.200.190.32] has joined #sbcl 11:45:40 -!- scymtym [~user@2001:638:504:2093:226:b9ff:fe7d:3e1f] has quit [Remote host closed the connection] 12:24:28 -!- yacks [~py@180.151.36.168] has quit [Ping timeout: 276 seconds] 12:26:37 -!- echo-area [~user@182.92.247.2] has quit [Read error: Connection reset by peer] 12:29:34 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 12:53:23 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 12:55:00 yacks [~py@180.151.36.168] has joined #sbcl 13:04:13 LiamH [~none@pdp8.nrl.navy.mil] has joined #sbcl 13:04:34 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 13:10:49 stassats [~stassats@wikipedia/stassats] has joined #sbcl 13:12:43 segv- [~mb@95-91-241-75-dynip.superkabel.de] has joined #sbcl 13:21:18 foreignFunction [~niksaak@ip-4761.sunline.net.ua] has joined #sbcl 13:26:00 scymtym [~user@2001:638:504:2093:226:b9ff:fe7d:3e1f] has joined #sbcl 13:26:21 sagane [~sagane@sao-paulo.lrde.epita.fr] has joined #sbcl 13:29:46 -!- rudi [~rudi@1x-193-157-251-78.uio.no] has quit [Quit: Client exciting.] 14:15:45 maxm- [~user@unaffiliated/maxm] has joined #sbcl 14:16:33 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 14:17:28 -!- maxm [~user@openchat.com] has quit [Write error: Connection reset by peer] 14:18:00 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 249 seconds] 14:18:00 -!- yacks [~py@180.151.36.168] has quit [Ping timeout: 249 seconds] 14:18:02 -!- christoph4 [~christoph@ppp-188-174-143-5.dynamic.mnet-online.de] has quit [Ping timeout: 249 seconds] 14:18:03 -!- pkhuong [~pkhuong@modemcable086.239-37-24.mc.videotron.ca] has quit [Ping timeout: 249 seconds] 14:18:24 christoph4 [~christoph@ppp-188-174-143-5.dynamic.mnet-online.de] has joined #sbcl 14:18:25 yacks [~py@180.151.36.168] has joined #sbcl 14:18:49 pkhuong_ [~pkhuong@24.37.239.86] has joined #sbcl 14:18:51 slyrus [~chatzilla@adsl-99-183-240-185.dsl.pltn13.sbcglobal.net] has joined #sbcl 14:19:05 -!- pkhuong_ is now known as pkhuong 14:20:12 -!- nicdev [user@2600:3c03::f03c:91ff:fedf:4986] has quit [Remote host closed the connection] 14:20:31 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 14:24:01 drmeiste_ [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 14:27:12 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Ping timeout: 256 seconds] 14:27:23 attila_lendvai [~attila_le@92.47.191.212] has joined #sbcl 14:27:23 -!- attila_lendvai [~attila_le@92.47.191.212] has quit [Changing host] 14:27:23 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 14:31:14 wbooze [~wbooze@xdsl-84-44-179-234.netcologne.de] has joined #sbcl 14:37:30 -!- sagane [~sagane@sao-paulo.lrde.epita.fr] has quit [Quit: Leaving] 15:05:30 psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has joined #sbcl 15:40:52 so, for the record: we have two students accepted onto GSoC 15:41:10 one working on register allocation; one on efficient interpretation 15:41:45 jsnell / pkhuong: I think it would be good to encourage them to be here as a matter of course 15:42:04 (which means of course we'll need a side-channel later on) 15:45:30 woo, students 15:49:08 what is register allocation good for if we can't even do proper instruction selection? 15:49:25 what is "proper instruction selection" 15:49:57 foom: in the presence of multiple VOPs for a single function, the selection is not optimal 15:50:14 and may result in unnecessary boxing/unboxing 15:50:31 -!- teggi [~teggi@123.20.101.86] has quit [Remote host closed the connection] 15:50:38 ah, indeed. 15:51:20 thought, it's not the largest problem, and affects chiefly integer operations 15:58:41 -!- Bike [~Glossina@75-175-78-112.ptld.qwest.net] has quit [Ping timeout: 256 seconds] 16:00:03 For those who are interested in the MPS GC stuff, there is now a #memorypoolsystem on this IRC network. 16:02:53 -!- foreignFunction [~niksaak@ip-4761.sunline.net.ua] has quit [Remote host closed the connection] 16:03:13 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 256 seconds] 16:20:55 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 16:22:15 foreignFunction [~niksaak@ip-4761.sunline.net.ua] has joined #sbcl 16:34:28 brucem: oh, good-oh 16:35:00 Krystof: did you end up with a GSOC project working on that? 16:35:16 I've been drugged into unconsciousness for some weeks, so I haven't followed. 16:35:44 sadly not 16:35:52 one was high but not quite high enough on our list 16:36:12 Too bad  the student should do it out of love and desire to learn anyway. 16:36:53 heh 16:37:04 I would love some students with desire to learn :-) 16:37:14 me too. they're rare though. 16:46:38 Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has joined #sbcl 16:59:16 -!- slyrus [~chatzilla@adsl-99-183-240-185.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 252 seconds] 17:25:02 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 17:57:31 -!- tcr1 [~tcr@77-56-40-229.dclient.hispeed.ch] has quit [Quit: Leaving.] 17:57:52 <|3b|> Krystof: is there a public API for the unicode normalization stuff? 18:01:13 tcr1 [~tcr@77-56-40-229.dclient.hispeed.ch] has joined #sbcl 18:03:32 no 18:03:47 try (sb-impl::normalize-string ...) and see if it works for you 18:03:55 see if it works at all, too 18:06:05 -!- tcr1 [~tcr@77-56-40-229.dclient.hispeed.ch] has quit [Client Quit] 18:07:09 <|3b|> yeah, that's what i had found so far, seems to do what i want 18:07:46 yay one satisfied customer 18:07:58 only 999,999 more and I can retire a millionnaire 18:12:05 *|3b|* might also find sb-impl::decompose-char useful 18:13:03 worth knowing 18:43:07 slyrus [~chatzilla@173-228-44-92.dsl.static.sonic.net] has joined #sbcl 19:05:19 Bike [~Glossina@67-5-251-162.ptld.qwest.net] has joined #sbcl 19:10:48 -!- yacks [~py@180.151.36.168] has quit [Quit: Leaving] 19:11:14 yacks [~py@180.151.36.168] has joined #sbcl 19:13:03 Sagane [~Sagane@177.100-226-89.dsl.completel.net] has joined #sbcl 19:34:40 Heh. A google compliance person pointed out that there is a GPLv2'd file, with a missing license, in SBCL: git/post-receive-email. 19:35:04 I've seen the following error twice recently when running all the SBCL tests. It appears to be intermittent. Platform is 64-bit x86 / Linux. 19:35:10 -!- Guest7689 is now known as reb 19:35:27 Invalid exit status: timer.impure.lisp 19:36:15 I have not yet seen it when I've run "run-tests.sh timer.impure.lisp". 19:36:17 -!- Bike [~Glossina@67-5-251-162.ptld.qwest.net] has quit [Ping timeout: 252 seconds] 19:37:59 Bike [~Glossina@75-164-174-173.ptld.qwest.net] has joined #sbcl 19:42:03 <|3b|> reb: if you can repeat it (or if you saved the output from failing runs) can you tell how far it got? 19:43:48 I just ran timer.impure successfully about 10 times in a row ... running all tests now, which is how I've seen the failure before. 19:45:48 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 19:51:42 -!- davazp [~user@31.200.190.32] has quit [Remote host closed the connection] 19:53:02 sdemarre [~serge@133.83-64-87.adsl-dyn.isp.belgacom.be] has joined #sbcl 20:29:27 foom: haha 20:29:53 is that the only problematic (from the point of view of your compliance process) file? 20:38:29 ASau` [~user@p4FF96C29.dip0.t-ipconnect.de] has joined #sbcl 20:41:44 -!- ASau [~user@p5797EA99.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds] 20:48:13 -!- sdemarre [~serge@133.83-64-87.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 276 seconds] 21:01:02 Krystof: It's the only one that's been pointed out. 21:01:50 The problem is easily solved on my end by deleting the file, but since it's technically against the license to not distribute it with the license, I thought I'd mention it 21:03:39 of course we have made several non-trivial modifications to it 21:07:17 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 21:33:50 -!- Sagane [~Sagane@177.100-226-89.dsl.completel.net] has quit [Read error: Connection reset by peer] 21:40:21 -!- ASau` is now known as ASau 21:43:24 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 21:48:55 momo-reina [~user@212.156.132.218] has joined #sbcl 21:53:56 Bike_ [~Glossina@67-5-198-145.ptld.qwest.net] has joined #sbcl 21:54:54 -!- Bike [~Glossina@75-164-174-173.ptld.qwest.net] has quit [Ping timeout: 252 seconds] 21:55:23 -!- Bike_ is now known as Bike 21:56:29 -!- drmeiste_ [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 21:59:02 I can manage to consintently lock up SBCL via world lock thing I think 21:59:24 I thought it was my logic with my locking 21:59:40 -!- prxq [~mommer@mnhm-590c0d6f.pool.mediaWays.net] has quit [Quit: Leaving] 21:59:45 but really its SBCL it seems 22:02:58 -!- momo-reina [~user@212.156.132.218] has quit [Remote host closed the connection] 22:03:09 momo-reina [~user@212.156.132.218] has joined #sbcl 22:10:34 this is getting better and better 22:10:57 I can consintenly manage to screw up SBCL, so that anything that calls (get-universal-time) blocks forever 22:11:04 wtf is that? 22:11:44 thought it was my logging stuff, finally tried simple layout -> works.. Stated adding stuff one by time, its the printing of time it seems 22:14:17 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.] 22:15:13 aha! (decode-universal-time (get-universal-time)) <- this hangs from REPL 22:15:37 and no stack trace, says "bogus frame" and thats it 22:15:52 but this (decode-universal-time (get-universal-time) 0) <- returns 22:18:47 Krystof: Well, minimal fix is to add a copy of the GPLV2 to the directory, and mentioning in the file that it's under GPLv2. 22:30:16 -!- segv- [~mb@95-91-241-75-dynip.superkabel.de] has quit [Ping timeout: 276 seconds] 22:32:13 -!- robgssp [~user@cpe-24-93-28-218.rochester.res.rr.com] has quit [Read error: Operation timed out] 22:33:38 farther traced it to (sb-unix::get-timezone 1369780334) hanging 22:45:22 -!- momo-reina [~user@212.156.132.218] has quit [Remote host closed the connection] 23:33:47 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 260 seconds] 23:43:24 cmm- [~cmm@109.67.119.85] has joined #sbcl 23:44:33 Blkt [~user@dynamic-adsl-62-10-11-72.clienti.tiscali.it] has joined #sbcl 23:45:10 -!- cmm [~cmm@bzq-79-179-155-88.red.bezeqint.net] has quit [Ping timeout: 252 seconds] 23:56:31 -!- Blkt [~user@dynamic-adsl-62-10-11-72.clienti.tiscali.it] has quit [Quit: good night!]