00:05:09 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 00:13:13 i don't suppose there's a "your first VOP" somewhere? 00:13:28 nope 00:13:35 'course not 00:17:10 Bike: http://paste.lisp.org/display/138679 00:19:29 it is not allowed to hold on to a fd-stream like the "resolved" value of *debug-io* across image save/restart and then use the stream in the restarted image, right? 00:20:21 scymtym: right. 00:22:05 good. because log4cl seems to do that which seems to cause a memory fault that took me the last day to debug 00:22:28 "%primitive doesn't give the right return sequence, but i can't be bothered to write a defknown" 00:23:36 MOV ECX, 2 CMP RCX, 2 is particularly interesting 00:24:36 (defun bsf (x) (declare ((signed-byte 64) x) (optimize speed (safety 0)) (values (integer 0 64) &optional)) (%primitive bsf x)) is better 00:25:37 a defknown would really help ;) 00:26:07 this is just as good as a defknown! 00:26:15 i mean, generated code-wise 00:30:26 pkhuong: did you manage to look at the recursive types branch? 00:32:58 scymtym: no. But I'm really not convinced that going for circular types is the best way to do this. 00:34:24 pkhuong: can you point me in alternate directions? 00:34:27 The type people seem to have settled on a fixpoint combinator, so you say something like (mu x (or null (cons fixnum x))). 00:35:03 I think that choice makes it easier to compare types, and avoids having to deal with circularity. 00:35:41 so you'd represent a recursive type as a variable name along with a noncircular type using the name? 00:36:02 Bike: right. 00:37:20 in SBCL, would this entail two new type classes, one corresponding to mu and one corresponding variables like x? 00:37:40 scymtym: probably just to the mu. 00:39:03 in your example, what would be the cons-type-cdr-type? 00:39:04 scymtym: also, parameterising structure types looks like it might require a completely different approach. Maybe that second approach can be simpler, and may well be good enough. 00:41:19 scymtym: not sure. A distinguished object with a de bruijn index attached to it? 00:46:41 pkhuong: i will try to research your suggestions and update the code 00:46:52 many thanks 00:47:10 no more lisp test failures on Solaris x86-64/no-threads 00:47:37 (some .sh test have a problem with passing arguments, apparently) 00:49:19 scymtym: the SSE and early GSoC involvement was the last I could do before focusing on my thesis... and it's getting really close to my deadlines ;) 00:50:00 stassats: can you try running them in bash? 00:50:45 i maybe just running it wrong, ./run-test.sh foreign.test.sh is not the right invocation 00:51:50 no, it's the correct one 00:53:54 pkhuong: i appreciate the feedback all the more 00:54:42 bash got a problem too: /home/stas/sbcl/tests/foreign.test.sh: line 32: $1: unbound variable 00:59:37 the problem is actually with sed 00:59:56 "$(basename "${TEST_BASENAME}" | sed 's/\.sh$// ; s/\./-/g')" sed: command garbled: s/\.sh$// ; s/\./-/g 01:02:57 I don't think the semicolon is necessary 01:03:16 echo "$(echo foo.sh | sed 's/\.sh$//' | sed 's/\./-/g')" works 01:03:51 "sed -e 's///' -e 's///'" ought to work too, if you want to save a process 01:04:37 right 01:04:56 "our test suite is the most optimized test suite ever!" 01:08:12 echo-area [~user@182.92.247.2] has joined #sbcl 01:08:36 now, no failures of any kind 01:09:49 enter sb-thread 01:12:07 wait what. Why aren'y we using basename to you know, find the base name without its extension? Not that it matters, but basename "${TEST_BASENAME}" .sh | sed -e 's/./-/g' would work just as well, I think. 01:13:36 is sb-thread even ported to solaris? 01:13:43 I think so. 01:14:20 #ifdef LISP_FEATURE_SB_THREAD #error "Threading is not supported for Solaris running on x86-64." #endif 01:15:08 so, testing threads would involve first porting them, and that's a task for some other night 01:15:30 only x86? 01:15:36 can at least test -m32 01:15:52 yep, on sparc too #error "Define threading support functions" 01:22:06 *stassats* tries to download solaris studio, and c whether it can be compile sbcl 01:23:42 i also haven't tried intel c compiler at it 01:27:13 i should probably test BSDs too sometime 01:27:39 because apparently, nobody bothers testing non-linux-x86oids 01:28:17 oh, mixed gnu/sun toolchains. I'll light a candle for you. 01:28:40 I test and build fbsd/x86oids semi regularly. 01:29:58 *stassats* thinks how to better automate testing with virtualbox 01:30:40 (i can even pirate some windows without feeling bad about it) 01:30:47 stassats: scymtym might have ideas ;) 01:36:42 -!- davazp [~user@92.251.232.228.threembb.ie] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 01:40:21 parallel_studio_xe_2013_update4.tgz weighs at 2G, SBCL has a long way to go 01:43:46 -!- Bike [~Glossina@wl-nat100.it.wsu.edu] has quit [Ping timeout: 240 seconds] 01:54:10 no threading failures on 32, but i get some float exceptions failures again 01:59:55 this time more weird: (funcall (fdefinition 'float-radix) "notfloat") set bad traps 02:00:16 i mean, what the hell 02:03:50 (list (ignore-errors (funcall (fdefinition '+) 'b)) (scale-float 1.0 most-positive-fixnum)) => (NIL #.SB-EXT:SINGLE-FLOAT-POSITIVE-INFINITY) 02:06:02 edgar-rft [~GOD@HSI-KBW-078-043-120-047.hsi4.kabel-badenwuerttemberg.de] has joined #sbcl 02:17:21 any error does that 02:17:24 fascinating 02:33:07 or rather, errors generate through INT 02:33:10 d 02:38:10 -!- christoph_debian [~christoph@ppp-188-174-93-89.dynamic.mnet-online.de] has quit [Read error: Operation timed out] 02:40:38 Bike [~Glossina@wl-nat100.it.wsu.edu] has joined #sbcl 02:42:48 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 02:52:19 christoph_debian [~christoph@ppp-93-104-22-106.dynamic.mnet-online.de] has joined #sbcl 02:54:32 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 03:16:22 sun c choked on assembly 03:24:40 The first time I tried to build on SBCL/SPARC, I had to first build gcc. For some reason, gcc spawned the wrong as, and nothing worked. Good luck ;) 03:25:45 i'm not really interested in getting it working 03:25:53 so far sbcl can be compiled with gcc, clang and icc 03:27:13 not depending on mingw on windows would be interesting, if not for .S 03:28:12 yeah. I don't think we want to maintain a masm fork. 03:29:38 maybe .S could be rolled into assembly routines 03:29:55 pranavrc [~pranavrc@122.164.235.15] has joined #sbcl 03:29:55 -!- pranavrc [~pranavrc@122.164.235.15] has quit [Changing host] 03:29:55 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 03:32:19 though, some float tests fail on icc 03:32:36 it probably optimizes away something important 04:02:46 tcr [~tcr@46-126-110-164.dynamic.hispeed.ch] has joined #sbcl 04:10:00 -!- echo-area [~user@182.92.247.2] has quit [Remote host closed the connection] 04:19:10 -!- tcr [~tcr@46-126-110-164.dynamic.hispeed.ch] has quit [Quit: Leaving.] 04:32:01 -!- stassats [~stassats@wikipedia/stassats] has quit [Read error: Operation timed out] 04:33:15 -!- yacks [~py@103.6.159.99] has quit [Quit: Leaving] 04:40:52 yacks [~py@103.6.159.99] has joined #sbcl 05:47:38 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Ping timeout: 264 seconds] 06:10:04 prxq [~mommer@mnhm-4d012655.pool.mediaWays.net] has joined #sbcl 06:16:07 -!- yacks [~py@103.6.159.99] has quit [Quit: Leaving] 06:17:34 -!- edgar-rft [~GOD@HSI-KBW-078-043-120-047.hsi4.kabel-badenwuerttemberg.de] has quit [Quit: maintainance finished by nonexistent authorization] 06:57:24 Krystof [~user@81.174.155.115] has joined #sbcl 06:57:24 -!- ChanServ has set mode +o Krystof 06:59:13 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 07:03:51 edgar-rft [~GOD@HSI-KBW-078-043-120-047.hsi4.kabel-badenwuerttemberg.de] has joined #sbcl 07:30:21 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 08:07:57 daimrod [daimrod@sbrk.org] has joined #sbcl 08:13:00 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 08:56:05 yacks [~py@103.6.159.99] has joined #sbcl 09:02:52 pranavrc [~pranavrc@122.164.249.193] has joined #sbcl 09:02:53 -!- pranavrc [~pranavrc@122.164.249.193] has quit [Changing host] 09:02:53 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 09:13:16 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Ping timeout: 246 seconds] 09:15:34 segv- [~mb@95-91-242-152-dynip.superkabel.de] has joined #sbcl 09:19:45 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 09:31:49 -!- kludge` [~comet@unaffiliated/espiral] has quit [Ping timeout: 248 seconds] 09:36:42 kludge` [~comet@unaffiliated/espiral] has joined #sbcl 10:25:21 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 10:29:55 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Ping timeout: 246 seconds] 10:54:41 pranavrc [~pranavrc@122.164.37.76] has joined #sbcl 10:54:41 -!- pranavrc [~pranavrc@122.164.37.76] has quit [Changing host] 10:54:41 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 11:23:27 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 11:40:24 -!- yacks [~py@103.6.159.99] has quit [Quit: Leaving] 11:44:19 ASau` [~user@p5797E212.dip0.t-ipconnect.de] has joined #sbcl 11:46:32 -!- danlentz [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has quit [Remote host closed the connection] 11:48:06 -!- ASau [~user@p4FF97B8F.dip0.t-ipconnect.de] has quit [Ping timeout: 264 seconds] 11:52:54 pranavrc [~pranavrc@122.164.37.76] has joined #sbcl 11:52:54 -!- pranavrc [~pranavrc@122.164.37.76] has quit [Changing host] 11:52:54 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 11:54:40 yacks [~py@103.6.159.99] has joined #sbcl 12:33:41 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 13:21:44 scymtym_ [~user@ip-5-147-120-181.unitymediagroup.de] has joined #sbcl 13:33:01 drmeister [~drmeister@c-68-82-220-150.hsd1.pa.comcast.net] has joined #sbcl 13:34:51 -!- drmeister [~drmeister@c-68-82-220-150.hsd1.pa.comcast.net] has quit [Remote host closed the connection] 14:01:51 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Ping timeout: 276 seconds] 14:14:37 pranavrc [~pranavrc@122.164.74.175] has joined #sbcl 14:14:37 -!- pranavrc [~pranavrc@122.164.74.175] has quit [Changing host] 14:14:37 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 14:19:10 -!- Krystof [~user@81.174.155.115] has quit [Ping timeout: 245 seconds] 14:20:02 -!- yacks [~py@103.6.159.99] has quit [Ping timeout: 264 seconds] 14:20:33 -!- jdz [~jdz@85.254.212.34] has quit [Ping timeout: 245 seconds] 14:26:54 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 14:41:17 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Remote host closed the connection] 14:43:14 teggi [~teggi@123.20.116.221] has joined #sbcl 14:52:19 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 15:13:17 yacks [~py@103.6.159.99] has joined #sbcl 15:20:26 pnpuff [~l@unaffiliated/pnpuff] has joined #sbcl 15:20:36 -!- teggi [~teggi@123.20.116.221] has quit [Remote host closed the connection] 15:20:52 -!- pnpuff [~l@unaffiliated/pnpuff] has left #sbcl 15:24:53 eeezkil [~eeezkil@unaffiliated/eeezkil] has joined #sbcl 15:39:26 stassats [~stassats@wikipedia/stassats] has joined #sbcl 15:48:46 -!- billitch [~billitch@men75-12-88-183-197-206.fbx.proxad.net] has quit [Ping timeout: 246 seconds] 16:10:14 billitch [~billitch@men75-12-88-183-197-206.fbx.proxad.net] has joined #sbcl 16:17:14 -!- ASau` is now known as ASau 16:17:41 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 16:18:56 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 16:40:35 Vivitron [~Vivitron@c-50-172-44-193.hsd1.il.comcast.net] has joined #sbcl 16:44:30 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 16:47:12 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 16:50:14 -!- billitch [~billitch@men75-12-88-183-197-206.fbx.proxad.net] has quit [Ping timeout: 240 seconds] 16:50:44 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Client Quit] 16:59:18 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 17:00:19 -!- Bike [~Glossina@wl-nat100.it.wsu.edu] has quit [Ping timeout: 260 seconds] 17:10:53 Bike [~Glossina@wl-nat115.it.wsu.edu] has joined #sbcl 17:12:34 fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has joined #sbcl 17:22:48 LiamH [~none@aes048148.nrl.navy.mil] has joined #sbcl 17:42:20 Krystof [~user@81.174.155.115] has joined #sbcl 17:42:20 -!- ChanServ has set mode +o Krystof 18:03:28 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 18:04:20 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 18:05:50 -!- Bike [~Glossina@wl-nat115.it.wsu.edu] has quit [Ping timeout: 245 seconds] 18:06:27 Bike [~Glossina@wl-nat115.it.wsu.edu] has joined #sbcl 18:10:16 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Ping timeout: 240 seconds] 18:31:30 billitch [~billitch@men75-12-88-183-197-206.fbx.proxad.net] has joined #sbcl 18:37:52 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Read error: Connection reset by peer] 18:38:17 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 18:40:48 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 18:42:56 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Remote host closed the connection] 18:47:31 -!- Vivitron [~Vivitron@c-50-172-44-193.hsd1.il.comcast.net] has quit [Read error: Connection reset by peer] 18:49:11 Vivitron [~Vivitron@c-50-172-44-193.hsd1.il.comcast.net] has joined #sbcl 19:04:08 -!- Bike [~Glossina@wl-nat115.it.wsu.edu] has quit [Ping timeout: 268 seconds] 19:05:35 Bike [~Glossina@wl-nat100.it.wsu.edu] has joined #sbcl 19:28:54 -!- slyrus [~chatzilla@107.201.4.22] has quit [Ping timeout: 264 seconds] 19:29:49 slyrus [~chatzilla@107.201.4.22] has joined #sbcl 19:52:51 prxq_ [~mommer@mnhm-5f75eddf.pool.mediaWays.net] has joined #sbcl 19:55:51 -!- prxq [~mommer@mnhm-4d012655.pool.mediaWays.net] has quit [Ping timeout: 245 seconds] 20:00:51 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 20:02:07 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Read error: Connection reset by peer] 20:02:26 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 20:08:20 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Read error: Connection reset by peer] 20:08:50 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 20:28:48 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Read error: Connection reset by peer] 20:29:18 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 20:50:05 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Remote host closed the connection] 20:53:28 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 20:59:32 Well... 20:59:49 Changing CC default value from "cc" to "gcc" is wrong in modern time. 21:00:37 stassats: I do test non-linux from time to time. 21:01:01 Status: 21:01:01 Unexpected success: float.pure.lisp / (ADDITION-OVERFLOW BUG-372) 21:01:01 Unexpected success: float.pure.lisp / (ADDITION-OVERFLOW BUG-372 TAKE-2) 21:01:02 Failure: interface.pure.lisp / (SLEEP NON-CONSING) 21:01:05 (25 tests skipped for this combination of platform and features) 21:01:38 Latest report for NetBSD/i486. 21:01:41 ASau: we have some gccism, I believe. 21:02:02 pkhuong: Clang supports some of them. 21:02:57 yeah, we also build on clang. 21:03:05 or purport to at least 21:03:26 clang built and passed all test on x86-64 21:03:28 some time ago 21:04:41 and "modern time" is funny 21:29:19 -!- eeezkil [~eeezkil@unaffiliated/eeezkil] has quit [Quit: ^D] 21:32:15 -!- prxq_ [~mommer@mnhm-5f75eddf.pool.mediaWays.net] has quit [Remote host closed the connection] 21:32:17 except that clang stopped understanding the .end directive 21:34:44 in .S, that is 21:36:44 *stassats* tries clang's analyzer 21:36:56 scan-build: 10 bugs found 21:41:40 -!- segv- [~mb@95-91-242-152-dynip.superkabel.de] has quit [Remote host closed the connection] 21:45:50 -!- LiamH [~none@aes048148.nrl.navy.mil] has quit [Quit: Leaving.] 21:58:34 found a memory leak, nice 21:59:01 probably 21:59:40 nope, it's freed on the lisp side and i looked at the wrong htop row 22:06:40 -!- fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has quit [Ping timeout: 256 seconds] 22:09:27 but another one looks real 22:15:26 do we still mmap in os x signal handlers? 22:23:17 hm, nope, schedule_thread_post_mortem doesn't seem to have a leak, even though creating lots of threads increases VIRT 22:24:44 and i don't get what it's trying to achieve by delaying the postmortem 22:33:33 do static and volatile make a variable thread specific? 22:33:38 as in static struct thread_post_mortem * volatile pending_thread_post_mortem = 0; 22:33:57 which is then used in: swap_lispobjs((lispobj *)(void *)&pending_thread_post_mortem, (lispobj)post_mortem) 22:34:05 danlentz [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has joined #sbcl 22:34:09 i can see how post_mortem can be lost when lots of threads concurrently run schedule_thread_post_mortem 22:37:03 though, swap_lispobjs is a lock xchg 22:39:34 /proc/maps suggests that some stacks are not cleaned up after (time (loop repeat 100000 do (sb-thread:make-thread (lambda ())))) 22:43:46 -!- Bike [~Glossina@wl-nat100.it.wsu.edu] has quit [Ping timeout: 245 seconds] 22:46:27 no, all the threads are cleaned up 22:51:37 except for the last one, because of the delay, so, there's always one thread stuck in limbo 23:01:25 Bike [~Glossina@wl-nat100.it.wsu.edu] has joined #sbcl 23:09:30 -!- Vivitron [~Vivitron@c-50-172-44-193.hsd1.il.comcast.net] has quit [Ping timeout: 264 seconds] 23:11:28 strange, just doing swap_lispobjs((lispobj *)(void *)&pending_thread_post_mortem, (lispobj)post_mortem) seems to create a 64MB memory area, according to /proc/maps 23:14:43 (or it just stops plan_thread_post_mortem(corpse) call from being optimized away) 23:16:56 i'm just on a wild goose chase 23:26:51 stassats: no, static volatile isn't thread local. 23:27:07 yeah, i figured that 23:27:30 mem-mem lock xchg is useful because it enqueues in a single instruction. 23:27:34 and that 64MB area looks like a space for malloc 23:27:45 sounds plausible. 23:27:56 so, everything is right 23:28:37 the memory leaks analysis of clang is not that great then 23:31:42 got a couple of unused variables right 23:43:57 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl