00:09:47 -!- slyrus [~chatzilla@172.1.59.137] has quit [Ping timeout: 255 seconds] 00:23:11 -!- wbooze [~wbooze@xdsl-78-35-177-175.netcologne.de] has quit [Ping timeout: 255 seconds] 00:38:07 -!- LiamH [~none@96.231.218.61] has quit [Quit: Leaving.] 00:38:37 -!- Qworkescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 01:04:57 There's another, more general question related to the one above. 01:05:20 Why are these tests executed at build time? 01:06:35 lichtblau: the build seems to have stuck after 01:06:51 ; disassembly for SB-SPROF::CONSALOT 01:06:51 ; 61C40D66: B80B001020 MOV EAX, 537919499 ; no-arg-parsing entry point 01:06:51 ... 01:15:54 LiamH [~none@96.231.218.61] has joined #sbcl 01:18:58 -!- tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has quit [Quit: Leaving.] 01:31:33 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 01:34:41 kanru [~kanru@118-163-10-190.HINET-IP.hinet.net] has joined #sbcl 02:03:06 that is an sb-sprof stress-test, which itself exercises signal handling 02:03:59 it's a bit depressing that just consing a lot and firing lots of signals crashes the system, though 03:46:54 -!- Fare [fare@nat/google/x-bkuadbibqlrzhyis] has quit [Read error: Operation timed out] 04:00:03 -!- Thra11 [~thrall@94.8.113.87.dyn.plus.net] has quit [Read error: Operation timed out] 04:03:54 Quadresce_ [~quad@unaffiliated/quadrescence] has joined #sbcl 04:04:28 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Disconnected by services] 04:04:38 -!- Quadresce_ is now known as Quadrescence 04:07:22 -!- Dalek_Baldwin [~Adium@108-225-26-178.lightspeed.irvnca.sbcglobal.net] has quit [Quit: Leaving.] 04:24:03 Dalek_Baldwin [~Adium@71-84-34-33.dhcp.mtpk.ca.charter.com] has joined #sbcl 04:33:14 -!- LiamH [~none@96.231.218.61] has quit [Quit: Leaving.] 04:44:44 -!- psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has quit [Quit: Leaving.] 05:52:15 Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has joined #sbcl 06:07:35 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 06:09:56 -!- Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has quit [Ping timeout: 248 seconds] 06:23:36 Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has joined #sbcl 06:25:19 nikodemus [~nikodemus@cs27127210.pp.htv.fi] has joined #sbcl 06:25:19 -!- ChanServ has set mode +o nikodemus 06:38:57 -!- Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has quit [Ping timeout: 248 seconds] 07:01:03 snowylike [~sn@91-67-171-156-dynip.superkabel.de] has joined #sbcl 07:03:14 -!- nikodemus [~nikodemus@cs27127210.pp.htv.fi] has quit [Quit: Leaving] 07:06:41 stassats [~stassats@wikipedia/stassats] has joined #sbcl 07:24:32 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 07:32:24 prxq [~mommer@mnhm-5f75e9dd.pool.mediaWays.net] has joined #sbcl 08:06:01 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 08:09:05 -!- ASau [~user@46.115.88.171] has quit [Ping timeout: 248 seconds] 09:26:35 tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has joined #sbcl 10:39:47 -!- tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has quit [Quit: Leaving.] 10:50:01 leoc [~leoc.git@p5DDB9F5F.dip.t-dialin.net] has joined #sbcl 11:54:56 cmm [~cmm@bzq-79-183-206-245.red.bezeqint.net] has joined #sbcl 11:56:40 -!- cmm- [~cmm@bzq-109-64-206-139.red.bezeqint.net] has quit [Ping timeout: 252 seconds] 11:58:30 -!- hydan [~udzinari@ip-89-102-13-27.net.upcbroadband.cz] has quit [Quit: Computer has gone to sleep.] 11:59:14 -!- cmm [~cmm@bzq-79-183-206-245.red.bezeqint.net] has quit [Ping timeout: 244 seconds] 11:59:58 cmm [~cmm@bzq-79-183-165-219.red.bezeqint.net] has joined #sbcl 12:18:11 hydan [~udzinari@ip-89-102-13-27.net.upcbroadband.cz] has joined #sbcl 12:20:17 probably not that important, but on os x 10.8.x the homebrew sbcl build fails because of sb-sprof as well (1.1.3) 12:23:26 it is important, but it's known already 12:27:22 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 12:32:37 Pretty weird situation that few people care about the vast majority of our test suite, but the few odd contrib tests are what everyone is concerned about. 12:33:35 How about if we re-evaluate the contrib concept and declare them to be a part of SBCL proper -- considering how users are actively pretending that they are part of SBCL anyway. 12:34:22 So far the SBCL view of things is that contribs can fail, and that's not a failure of the SBCL build. It's just a means of disabling a contrib, and a contrib is just something we're not truly responsible for. But that's not how users see them. 12:34:45 Maybe we should rm -rf the pointless, broken contribs (like sb-simple-streams). Anything that remains is to be treated as a part of SBCL proper. 12:35:19 Meaning that the remaining contribs must _build_ (hard failure if they don't). But their tests are hooked into run-tests.sh, not make.sh. 12:35:45 *lichtblau* not volunteering for the work; just thinking out loud 12:39:10 -!- hydan [~udzinari@ip-89-102-13-27.net.upcbroadband.cz] has quit [Quit: Computer has gone to sleep.] 12:41:58 i'm for putting contrib tests into the main test-suite 12:42:02 that means faster build! 12:48:42 hydan [~udzinari@ip-89-102-13-27.net.upcbroadband.cz] has joined #sbcl 13:20:17 LiamH [~none@129-2-129-145.wireless.umd.edu] has joined #sbcl 14:02:07 loke [~elias@bb115-66-85-121.singnet.com.sg] has joined #sbcl 14:09:07 -!- fe[nl]ix [~quassel@pdpc/supporter/professional/fenlix] has quit [Remote host closed the connection] 14:09:16 fe[nl]ix [~quassel@pdpc/supporter/professional/fenlix] has joined #sbcl 14:12:03 -!- LiamH [~none@129-2-129-145.wireless.umd.edu] has quit [Quit: Leaving.] 14:12:30 -!- fe[nl]ix [~quassel@pdpc/supporter/professional/fenlix] has quit [Remote host closed the connection] 14:12:39 fe[nl]ix [~quassel@pdpc/supporter/professional/fenlix] has joined #sbcl 14:15:22 -!- hydan [~udzinari@ip-89-102-13-27.net.upcbroadband.cz] has quit [Quit: Computer has gone to sleep.] 14:29:30 -!- kanru [~kanru@118-163-10-190.HINET-IP.hinet.net] has quit [Ping timeout: 264 seconds] 14:33:34 psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 14:51:34 wbooze [~wbooze@xdsl-78-35-165-20.netcologne.de] has joined #sbcl 15:06:01 -!- easye` [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 15:06:51 easye` [~user@213.33.70.157] has joined #sbcl 15:31:02 Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has joined #sbcl 15:34:57 -!- wbooze [~wbooze@xdsl-78-35-165-20.netcologne.de] has quit [Remote host closed the connection] 15:39:17 wbooze [~wbooze@xdsl-78-35-165-20.netcologne.de] has joined #sbcl 15:56:29 attila_lendvai [~attila_le@92.46.22.91] has joined #sbcl 15:56:29 -!- attila_lendvai [~attila_le@92.46.22.91] has quit [Changing host] 15:56:29 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 16:19:08 Thra11 [~thrall@87.112.186.70] has joined #sbcl 16:20:46 hydan [~udzinari@ip-89-102-13-27.net.upcbroadband.cz] has joined #sbcl 16:23:29 attila_lendvai1 [~attila_le@92.46.1.14] has joined #sbcl 16:23:29 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 16:33:21 leoc` [~leoc.git@p5DDBBA43.dip.t-dialin.net] has joined #sbcl 16:37:22 -!- leoc [~leoc.git@p5DDB9F5F.dip.t-dialin.net] has quit [Ping timeout: 276 seconds] 17:14:26 -!- Thra11 [~thrall@87.112.186.70] has quit [Quit: kthxbai] 17:23:08 Thra11 [~thrall@87.112.186.70] has joined #sbcl 17:28:40 -!- Fare [~fare@173-9-65-97-NewEngland.hfc.comcastbusiness.net] has quit [Ping timeout: 256 seconds] 17:39:19 ASau [~user@2.208.50.103] has joined #sbcl 17:40:53 Fare [fare@nat/google/x-yphnxmkqlchhapjf] has joined #sbcl 17:43:57 Krystof: the question is why those tests are executed at _build_ time at all. 17:44:18 Especially stress tests. 17:44:21 ASau: because dan_b designed the contribs that way 17:45:00 Are there objections if we rework it? 17:45:20 I made such a proposal earlier today (see the logs). 17:46:16 But I don't think it's something any random committer whould undertake without broad consensus among developers, because the contrib system as designed by dan_b back then has a long tradition. 17:46:18 Nice to see I'm not alone :) 17:47:42 What I'd be opposed to though, is to (entirely) disable only this particular test only in this particular contrib just because it's failing for someone. 17:49:10 FWIW, we have _much_ more annoying tests (long running time, vast amount of memory used) in our real test suite. 17:49:13 Fewer users seem to be running those tests though (or only on capable computers?). 17:49:25 one objection i can foresee: the core sbcl is actually tested during the build, i.e. when sbcl build itself 17:49:34 so tests do theirs too for parity 17:49:42 is there something like arbitrary floating point precision for lisp systems ? 17:49:51 wbooze: try #lisp 17:50:05 i saw pari uses a precision of 20080 on some server how can that be ? 17:50:18 wbooze: ECL uses GMP. 17:50:23 aah 17:50:41 ok thank you 17:50:47 stassats: all tests non-essential to build should be run as separate stage, IMO. 17:51:01 define essential 17:51:19 Not absolutely necessary. 17:52:10 lichtblau: i actually changed my mind, i'm for testing contribs during build 17:52:28 lichtblau: for the stated above reason 17:52:54 lichtblau: but main tests may include more comprehensive contrib tests 17:55:37 to reformulate, the core sbcl functionality is tested during the build, including the compiler, a substantial part of the library, the GC and other runtime features 17:56:30 What is wrong testing it _after_ the build? 17:56:37 but contribs are just built, not used, and many people usually just build (myself included), without running the whole test suite each time, since the build itself is a pretty good guarantee of a good state 17:57:47 so, not testing contribs will just result in too many confusing results after installation in case of trivial problems, which could be caught by testing 17:58:56 And what is the problem of running tests after the build? 17:59:09 please, read what i wrote 17:59:16 I have read it. 17:59:35 The problem is that there're many people who don't care of profiler 17:59:48 and some other rarely used library parts. 18:00:24 Insisting on correctness of those parts of SBCL isn't interesting to them. 18:00:47 If the need arises, they can run test suite separatly. 18:00:49 ASau: erm, no, that one's actually not a problem IMHO. If sb-sprof fails to build, users are not affected in any way whatsoever except that install.sh won't install this contrib. 18:01:15 ASau: then they should extend their not caring to the tests being run 18:01:17 argh, and there I've fallen into the trap of writing "sb-sprof fails to build" instead of "sb-sprof fails to test" myself. 18:01:45 It's important to understand that each contribs first builds, and then gets tested. Even if the tests fails, it has been built already. And the main part of SBCL has been built. 18:02:06 -!- wbooze [~wbooze@xdsl-78-35-165-20.netcologne.de] has quit [Ping timeout: 245 seconds] 18:02:19 In that particular case, the test doesn't even fail. 18:03:49 wbooze [~wbooze@xdsl-78-35-162-213.netcologne.de] has joined #sbcl 18:04:17 what does it do? 18:04:46 hang (if I'm reading the scrollback right) 18:04:54 admittedly that's not so nice 18:05:10 isn't the cause for improving the tests, not disabling them? 18:06:21 No. 18:06:24 hmm, when I had those problems on Windows, I just added #+affectedfeature (error "failing test explicitly, because it would hang") in front of the test to get clean test results. Not pretty, but better than hanging, and better than not reporting the failure. 18:06:31 It is the cause not to run tests at build time. 18:08:01 At least in the case of Darwin, this test failure seem to demonstrate a blatant signal safety problem. So why aren't people thankful I added the test and alerted them to the bug?! ;-) 18:09:00 ASau: that's an extreme position, and i see no compelling reason to do so 18:09:14 "i'm not happy" isn't quite convincing 18:12:11 lichtblau: I don't mind the test, but I'd rather have ability to deal with problems one by one rather than by patching the source to work around such cases. 18:12:26 -!- Thra11 [~thrall@87.112.186.70] has quit [Quit: kthxbai] 18:15:31 -!- attila_lendvai1 [~attila_le@92.46.1.14] has quit [Ping timeout: 276 seconds] 18:16:22 tests were being run for as long as i remember, they take a very small percentage of the whole build-time, a single problematic test doesn't call for removing all the tests 18:21:30 Thra11 [~thrall@87.112.186.70] has joined #sbcl 18:24:10 This is not the first SBCL test that doesn't even fail. 18:24:40 And don't argue as if I'm calling to remove tests. 18:24:56 I have never proposed that. 18:27:09 removing => removing them from the build process 18:27:43 I have never proposed to remove them from the build process. 18:27:53 I only proposed moving them to a separate stage. 18:28:27 besides, i weren't arguing with you 18:28:44 i were arguing with the idea that lichtblau proposed 18:29:00 was... 18:29:33 I haven't looked into logs, thus I can't comment on his proposal. 18:29:46 LiamH [~none@132.250.138.103] has joined #sbcl 18:30:03 well, then don't 18:30:38 I did not anyway. 18:34:24 then i don't see what you were arguing against, so you want the contribs to be tested during ./make.sh or not? 18:38:32 I want contribs not to be tested during make.sh, they should be tested at another stage. 18:39:30 so, you were proposing to remove them from the build process after all 18:40:20 but anyway, it doesn't really matter 18:40:56 No, I'm _not_ proposing to remove them from the build process. 18:41:29 ./make.sh is the build process 18:41:49 No, it is not the build process. 18:43:24 SBCL build process includes patching stage among the rest. 18:43:28 i'm not sure what are you talking about 18:44:21 I'm talking about building SBCL for end users. 18:45:16 Which includes various mathematicians and engineers who may not be interested in Common Lisp. 18:46:02 Currently, due to the way SBCL build process is organized, 18:46:27 I have to disable tests that don't affect anything for end users 18:46:39 in order to get it built at all. 18:47:34 I find it inconvenient and mildly annoying. 18:48:16 mathematicians and engineers should just grab a binary 18:48:38 Let people decide themselves. 18:49:49 you don't know whether they affect things for end users 18:49:55 SBCL is rare case where tests are intertwined with actual build process. 18:50:13 we don't know either 18:50:31 so we let the people who download our software decide, but we give them the information that they might need 18:51:02 (if you are talking about being a distributor, and knowing that no test failure in a contrib is ever sufficient reason not to install it, then patch the install script once instead of the contribs every time you discover a new test failure) 18:51:47 Again, the problem is that it is frequent enough that tests don't even fail. 18:51:52 and if you're talking about the hang, well, frankly a hang might well affect one of your users, and you do them a disservice in not helping to hunt the problem down 18:52:07 and I'm sorry that you think that that's "inconvenient" 18:52:16 ASau: i don't follow your argument, so you first give an example of people who supposedly don't care enough, and then you say that we should give them more choice, even though they might not be well-informed to make a proper choice? 18:52:39 If it were separate stage, then users can run testsuite. 18:52:54 stassats: what is not clear? 18:52:55 ASau: please, file a bug report onto launchpad or send an email to sbcl-bugs detailing your test problems 18:53:09 Users don't care enough until there is a problem. 18:53:32 one argument is that we therefore /make/ there be a problem so that there is motivation to fix it 18:53:59 so that the problem strikes when it is merely inconvenient (i.e. to you, the distributor) rather than massively disastrous (in the case of an end-user, with real data loss) 18:54:17 ASau: what do you propose, to hide the problem under the carpet, until it actually bits them during usage? 18:54:38 I propose to move tests to the stage where they belong. 18:54:49 I.e. test stage. 18:54:52 to move them to the stage when they are never run? 18:55:09 No. 18:55:23 on the basis of this discussion, I'd be more inclined to move the regular tests into the build stage and refuse to install if any test fails and isn't marked as expected to fail 18:55:29 I propose to move tests to the stage where they belong, not to the stage that is never run. 18:55:45 because otherwise it seems that distributors will never be motivated to help report and fix problems 18:55:52 Krystof: i'd support it if they didn't take so much time to run 18:56:11 Krystof: except that exactly the opposite happens. 18:56:29 (I also remember what this kind of discussion does to my motivation, so... 18:56:33 -!- Krystof [~user@81.174.155.115] has left #sbcl 18:57:06 Krystof: I'm not inclined to report problems when I have used all my time cutting through it just to get it build at least in some way. 18:58:01 well, you're not the only one who's using SBCL 18:58:23 Yes, I know it. 18:59:04 was the plan of sb-posix to provide _all_ posix-specified functions? some are missing, AFAICT 19:00:15 And there're some functions that should have been missing from POSIX rather. :) 19:00:50 flip214: no, that wasn't the plan 19:01:13 oh, ok. I plan to provide patches, if I can get something running ... 19:01:32 why not provide patches to osicat or iolib instead? 19:06:20 -!- milosn [~milosn@user-5AF500C6.broadband.tesco.net] has quit [Ping timeout: 260 seconds] 19:18:16 milosn [~milosn@user-5AF5045B.broadband.tesco.net] has joined #sbcl 19:21:28 -!- stassats [~stassats@wikipedia/stassats] has quit [Remote host closed the connection] 19:44:33 -!- snowylike [~sn@91-67-171-156-dynip.superkabel.de] has quit [Ping timeout: 248 seconds] 19:45:17 snowylike [~sn@91-67-171-156-dynip.superkabel.de] has joined #sbcl 19:46:46 Qworkescence [~quad@unaffiliated/quadrescence] has joined #sbcl 20:01:16 because it's not itching there. sorry, wouldn't know _what_ to patch. 20:09:04 -!- Dalek_Baldwin [~Adium@71-84-34-33.dhcp.mtpk.ca.charter.com] has left #sbcl 21:05:57 -!- snowylike [~sn@91-67-171-156-dynip.superkabel.de] has quit [Read error: Connection reset by peer] 21:21:43 milosn_ [~milosn@user-5AF50613.broadband.tesco.net] has joined #sbcl 21:23:26 -!- milosn [~milosn@user-5AF5045B.broadband.tesco.net] has quit [Ping timeout: 244 seconds] 21:30:02 slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has joined #sbcl 21:35:47 -!- slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has quit [Ping timeout: 255 seconds] 21:39:27 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 21:46:52 tcr1 [~tcr@84-72-21-32.dclient.hispeed.ch] has joined #sbcl 21:47:03 -!- wbooze [~wbooze@xdsl-78-35-162-213.netcologne.de] has quit [Ping timeout: 276 seconds] 21:56:16 slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has joined #sbcl 22:00:53 -!- slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has quit [Ping timeout: 248 seconds] 22:06:01 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: dead] 22:10:05 slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has joined #sbcl 22:20:29 -!- slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has quit [Ping timeout: 252 seconds] 22:22:43 -!- Qworkescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 22:23:23 slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has joined #sbcl 22:25:06 -!- prxq [~mommer@mnhm-5f75e9dd.pool.mediaWays.net] has quit [Quit: Leaving] 22:48:10 -!- slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has quit [Ping timeout: 244 seconds] 22:53:04 -!- psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: Leaving.] 23:25:03 -!- xymox [lechuck@unaffiliated/contempt] has quit [Ping timeout: 260 seconds] 23:28:21 xymox [~xymox@unaffiliated/contempt] has joined #sbcl 23:37:06 ASau` [~user@46.115.52.76] has joined #sbcl 23:40:54 -!- ASau [~user@2.208.50.103] has quit [Ping timeout: 264 seconds] 23:50:05 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 23:54:01 slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has joined #sbcl 23:54:55 Qworkescence [~quad@unaffiliated/quadrescence] has joined #sbcl 23:58:55 -!- slyrus [~chatzilla@70-90-161-58-ca.sfba.hfc.comcastbusiness.net] has quit [Ping timeout: 256 seconds]