00:53:38 -!- fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has quit [Ping timeout: 264 seconds] 01:04:14 andreh [~andreh@186.213.204.71] has joined #sbcl 01:05:08 prxq_ [~mommer@mnhm-4d0103b5.pool.mediaWays.net] has joined #sbcl 01:08:39 -!- prxq [~mommer@mnhm-590c3849.pool.mediaWays.net] has quit [Ping timeout: 260 seconds] 01:09:02 -!- andreh [~andreh@186.213.204.71] has quit [Ping timeout: 240 seconds] 01:19:45 edgar-rft [~GOD@HSI-KBW-078-043-120-047.hsi4.kabel-badenwuerttemberg.de] has joined #sbcl 01:21:19 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 01:23:02 andreh [~andreh@186.213.204.71] has joined #sbcl 01:38:58 Vivitron [~Vivitron@c-50-172-44-193.hsd1.il.comcast.net] has joined #sbcl 01:47:30 -!- Bike [~Glossina@wl-nat99.it.wsu.edu] has quit [Ping timeout: 264 seconds] 01:56:47 Bike [~Glossina@wl-nat100.it.wsu.edu] has joined #sbcl 02:23:04 danlentz_ [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has joined #sbcl 02:23:17 -!- danlentz [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has quit [Read error: No route to host] 02:23:17 -!- danlentz_ is now known as danlentz 02:23:27 -!- danlentz [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has quit [Remote host closed the connection] 02:23:40 danlentz [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has joined #sbcl 02:25:58 Fare [fare@nat/google/x-uczibzapxkqqanzs] has joined #sbcl 02:38:29 -!- christoph_debian [~christoph@ppp-93-104-184-215.dynamic.mnet-online.de] has quit [Ping timeout: 248 seconds] 02:50:18 yacks [~py@103.6.159.99] has joined #sbcl 02:51:46 christoph_debian [~christoph@ppp-188-174-31-29.dynamic.mnet-online.de] has joined #sbcl 02:52:57 echo-area [~user@182.92.247.2] has joined #sbcl 03:07:20 -!- Fare [fare@nat/google/x-uczibzapxkqqanzs] has quit [Quit: Leaving] 03:19:31 pranavrc [~pranavrc@122.164.239.31] has joined #sbcl 03:19:31 -!- pranavrc [~pranavrc@122.164.239.31] has quit [Changing host] 03:19:31 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 03:34:17 Was my solution and tests acceptable? 03:34:18 https://bugs.launchpad.net/sbcl/+bug/1220084 03:39:02 momo-reina [~user@124.107.67.167] has joined #sbcl 04:10:05 -!- yacks [~py@103.6.159.99] has quit [Ping timeout: 268 seconds] 04:25:06 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Ping timeout: 264 seconds] 04:39:11 drmeister [~drmeister@pool-173-59-25-58.phlapa.fios.verizon.net] has joined #sbcl 04:40:22 -!- momo-reina [~user@124.107.67.167] has quit [Ping timeout: 246 seconds] 05:22:36 momo-reina [~user@112.207.194.43] has joined #sbcl 05:43:53 Bike_ [~Glossina@wl-nat100.it.wsu.edu] has joined #sbcl 05:44:18 -!- Bike [~Glossina@wl-nat100.it.wsu.edu] has quit [Ping timeout: 264 seconds] 05:50:31 -!- Bike_ is now known as Bike 05:59:49 -!- Bike [~Glossina@wl-nat100.it.wsu.edu] has quit [Read error: Connection reset by peer] 06:00:12 Bike [~Glossina@wl-nat100.it.wsu.edu] has joined #sbcl 06:17:59 Hydan [~hydan@ip-89-103-110-5.net.upcbroadband.cz] has joined #sbcl 06:27:25 -!- drmeister [~drmeister@pool-173-59-25-58.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 06:50:11 easye [~user@213.33.70.157] has joined #sbcl 07:00:30 -!- momo-reina [~user@112.207.194.43] has quit [Ping timeout: 264 seconds] 07:16:38 -!- Vivitron [~Vivitron@c-50-172-44-193.hsd1.il.comcast.net] has quit [Ping timeout: 240 seconds] 07:19:04 -!- edgar-rft [~GOD@HSI-KBW-078-043-120-047.hsi4.kabel-badenwuerttemberg.de] has quit [Read error: Operation timed out] 07:19:58 edgar-rft [~GOD@HSI-KBW-078-043-120-047.hsi4.kabel-badenwuerttemberg.de] has joined #sbcl 07:20:45 benkard [~benkard@2a01:198:6d5:0:4ea:293a:c9fb:b442] has joined #sbcl 07:46:55 -!- benkard [~benkard@2a01:198:6d5:0:4ea:293a:c9fb:b442] has quit [Ping timeout: 246 seconds] 07:48:01 benkard [~benkard@2a01:198:6d5:0:4ea:293a:c9fb:b442] has joined #sbcl 07:54:02 -!- milosn [~milosn@user-5af50c69.broadband.tesco.net] has quit [Read error: Operation timed out] 08:19:22 -!- benkard [~benkard@2a01:198:6d5:0:4ea:293a:c9fb:b442] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 08:22:32 -!- bege [~bege@S0106001d7e5132b0.ed.shawcable.net] has quit [Ping timeout: 260 seconds] 08:22:43 bege [~bege@S0106001d7e5132b0.ed.shawcable.net] has joined #sbcl 08:25:33 -!- ivan`` [~ivan@unaffiliated/ivan/x-000001] has quit [Ping timeout: 245 seconds] 08:25:46 -!- nicdev [user@2600:3c03::f03c:91ff:fedf:4986] has quit [Remote host closed the connection] 08:26:06 nicdev [user@2600:3c03::f03c:91ff:fedf:4986] has joined #sbcl 08:32:08 ivan`` [~ivan@unaffiliated/ivan/x-000001] has joined #sbcl 08:39:08 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 08:48:04 yacks [~py@103.6.159.99] has joined #sbcl 09:19:51 stassats [~stassats@wikipedia/stassats] has joined #sbcl 09:31:00 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Ping timeout: 240 seconds] 09:32:17 -!- kludge` [~comet@unaffiliated/espiral] has quit [Ping timeout: 256 seconds] 09:37:28 kludge` [~comet@unaffiliated/espiral] has joined #sbcl 10:21:52 benkard [~benkard@2a01:198:6d5:0:b8e9:7bca:a077:444c] has joined #sbcl 10:31:08 -!- benkard [~benkard@2a01:198:6d5:0:b8e9:7bca:a077:444c] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 10:43:51 momo-reina [~user@112.207.194.43] has joined #sbcl 10:46:23 -!- momo-reina [~user@112.207.194.43] has left #sbcl 10:49:34 -!- echo-area [~user@182.92.247.2] has quit [Remote host closed the connection] 10:58:18 -!- andreh [~andreh@186.213.204.71] has quit [Ping timeout: 264 seconds] 11:23:32 andreh [~andreh@186.213.204.71] has joined #sbcl 11:28:44 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Ping timeout: 240 seconds] 11:34:47 segv- [~mb@95-91-242-104-dynip.superkabel.de] has joined #sbcl 11:38:11 -!- Hydan [~hydan@ip-89-103-110-5.net.upcbroadband.cz] has quit [Remote host closed the connection] 11:44:46 ASau` [~user@p4FF9643B.dip0.t-ipconnect.de] has joined #sbcl 11:47:56 pranavrc [~pranavrc@122.164.233.66] has joined #sbcl 11:47:56 -!- pranavrc [~pranavrc@122.164.233.66] has quit [Changing host] 11:47:56 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 11:48:21 -!- ASau [~user@p4FF97D54.dip0.t-ipconnect.de] has quit [Ping timeout: 248 seconds] 11:52:58 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 12:16:47 -!- andreh [~andreh@186.213.204.71] has quit [Ping timeout: 246 seconds] 12:49:53 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 12:57:43 -!- psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has quit [Quit: Leaving.] 13:30:28 benkard [~benkard@2a01:198:6d5:0:1df1:4068:34d9:9bf0] has joined #sbcl 13:39:21 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 13:40:12 attila_lendvai [~attila_le@dsl51B6D29B.pool.t-online.hu] has joined #sbcl 13:40:12 -!- attila_lendvai [~attila_le@dsl51B6D29B.pool.t-online.hu] has quit [Changing host] 13:40:12 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 13:40:52 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Client Quit] 13:41:18 -!- benkard [~benkard@2a01:198:6d5:0:1df1:4068:34d9:9bf0] has quit [Write error: Broken pipe] 13:43:11 loke_erc [~user@bb115-66-249-26.singnet.com.sg] has joined #sbcl 13:43:33 stassats: thanks for applying the patch. That was fun. :-) 13:43:57 i'm now working on the same, but for EQUALP 13:44:05 I wonder if there are any more low-hanging fruit where some optimisations could be done... 13:46:05 benkard [~benkard@2a01:198:6d5:0:7902:dd53:83f2:a4e1] has joined #sbcl 13:48:34 drmeister [~drmeister@mobile-166-147-109-021.mycingular.net] has joined #sbcl 13:50:10 and turns out, all those functions called by EQUALP don't test for EQ first 13:50:42 psilord [~pkeller@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 13:50:52 and the cost of doing an additional EQ is negligible, but the speed-up is noticeable 13:51:49 (time (loop for i to 50000000 do (foo-calling-char-equal #\a #\a))) => without EQ 0.994, with EQ => 0.370 13:52:33 stassats: ah 13:52:39 stassats: so is it patched now? 13:52:44 will be 13:52:47 cool 13:53:21 When I'm bored (sometimes at work, as was the case with this one) I look at the generated assmebly just for kicks, to see if it's optimal :-) 13:53:51 that's what i do when i can't fall asleep 13:55:56 (char-equal #\a x) can be improved by precomputing equal-char-code 13:56:04 two-arg-char-equal, in particular 13:56:06 How much of a performance boost do you think you'd get if SBCL supported unboxed function arguments? 13:56:56 on non double floaty code, not that much 13:57:22 but for double floaty code? 13:57:42 if there's a lot of boxing, sure, it'll be noticeable 13:59:16 -!- drmeister [~drmeister@mobile-166-147-109-021.mycingular.net] has quit [Remote host closed the connection] 13:59:41 would an alternative be to pass doubles as two arguments? On amd64 it'd make sense, as arguments are passed in registers, yes? 14:00:21 doubles are usually passed in the floating point registers, XMM on x86-64 14:01:07 unboxed? 14:01:31 that's not on sbcl, usually as in x86-64 ABI 14:01:44 ah ok 14:02:27 anyway, then, wouldn't it be conceivable to have a system where the "normal" register just contains a tag and a reference to an xmm register, which is where the actual unboxed argument can be found? 14:03:29 that could be an option, but it would have to work with both kinds 14:03:49 just having unboxed function calls sounds easier to me. :p 14:03:49 and there's no space for any more lowtags on x86-64 14:04:26 stassats: there's plenty of space for wide tags though, and since you only need to have bits of data for the number of registers, not a big deal. 14:04:33 what if the double-float lowtag always meant "argument in XMM" when passed as argument? 14:04:49 there's no double-float lowtag 14:04:58 there's other-pointer-lowtag for double-float 14:05:17 unless you want to part with 63-bit fixnums, that is 14:05:18 oh, i see 14:05:45 You only need a set of distinct values though... 14:06:08 like NIL being a hardcoded value. 14:10:57 loke_erc: looks like your test just constant folds everything 14:11:31 you mean my test case? Strange, I tested it (confirmed that there are 360 calls to EQUAL) 14:12:14 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 14:13:04 it constant folds (equal "a" "a") 14:14:24 stassats: yes, isn't that happening as part of the same tranform? 14:14:43 no, it's happening during folding 14:15:01 That's why the list of test cases contains (copy-seq) and the same string. 14:15:17 so you get two difference instances of the same string (to deal with that case) 14:15:27 it's declared foldable, both arguments are constant, no transform is even invoked 14:15:31 Perhaps the compiler is too smart and realises they are the same anyway? 14:15:37 -!- ASau` is now known as ASau 14:15:53 it doesn't matter which are they, what matters is that they are constant 14:16:01 it just calls EQUAL at compile-time and inserts the result 14:16:35 oh 14:21:53 fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has joined #sbcl 14:23:55 fixed with COMPILE 14:24:12 OK, cool. :-) 14:27:52 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Remote host closed the connection] 14:33:07 i have found previously that the fastest way to do (string-equal "A Constant" string) is to precompute "a cONSTANT" and then just compare simply char= through both strings 14:33:21 not sure how that copes with unicode weirdness 14:33:53 Bike_ [~Glossina@69.166.47.103] has joined #sbcl 14:35:14 Bike__ [~Glossina@69.166.47.103] has joined #sbcl 14:35:25 the cl operators should at least be internally consistent 14:35:37 -!- Bike [~Glossina@wl-nat100.it.wsu.edu] has quit [Disconnected by services] 14:35:42 -!- Bike__ is now known as Bike 14:35:59 so if you precompute "a cONSTANT" by using char-upcase/downcase or whatever it's called, then the result you will get will be the same as if computed by string-equal 14:36:23 but of course that notion of STRING-EQUALity may not be terribly useful given the wonderful world of Unicode 14:37:05 reducing to (char-equal #\a char) => (or (char= #\a char)(char-= #\A char)) 14:37:53 might be worth doing that for > speed space 14:38:28 can be done with any space, just having (char-equal-constant #\a #\A char) 14:38:46 -!- Bike_ [~Glossina@69.166.47.103] has quit [Ping timeout: 245 seconds] 14:41:22 true 14:41:42 I wish we provided nice Unicode-aware string/character routines 14:41:50 I mean, we are nearly there with normalization forms 14:45:23 all these optimization and a simple CPU upgrade blows them all out of the water 14:46:08 teggi [~teggi@113.173.4.29] has joined #sbcl 14:46:18 *stassats* really misses boinkmarks 14:46:21 Hydan [~hydan@ip-89-103-110-5.net.upcbroadband.cz] has joined #sbcl 14:47:09 stassats: your shiny new Haswell ? 14:47:31 yeah 15:02:04 is this known: https://ci.cor-lab.org/job/sbcl-master/featureset=2,label=MAC_OS_lion_64bit/120/consoleFull? (search for string "fatal error") 15:03:01 sb-concurrency is known to fail 15:03:07 stassats: ok, thanks 15:03:35 on ppc linux it sometimes fails, on windows the test sometimes hang 15:04:01 stassats: I might have a bit of time this month to do not-too-intense sbcl work, like patch hoovering or something 15:04:09 is there something I can do that frees you up for more interesting work? 15:05:00 Krystof: maybe handling https://bugs.launchpad.net/sbcl/+bug/1206191 ? 15:05:53 lp 994528 15:05:53 https://bugs.launchpad.net/bugs/994528 15:05:57 would make me very happy :) 15:07:39 I shall make it my mission to bring very happiness to the world 15:08:47 Krystof: thank you 15:10:54 stassats: did you have time to fix the barriers in frlock? 15:11:35 pkhuong: i couldn't reproduce it reliably, and didn't want to resort to "think hard" 15:12:33 pretty sure the order swap I suggested is the correct one (used everywhere else) 15:13:24 i can't commit something without checking the logic, and i have no idea what that does 15:14:57 but yeah. The reason I haven't committed is that I can't make it fail. 15:15:58 and i now see why the equal transform needed to check for intersection with strings and bit-vectors, before doing returning NIL on (not (types-equal-or-intersect x-type y-type)) 15:16:15 because of different specialization, types will not intersect 15:16:25 even though EQUAL doesn't care about 15:16:45 *stassats* puts it back with an explanation comment 15:20:32 and it actually is broken for (equal (the (cons (or simple-bit-vector simple-base-string)) x) (the (cons (or (and bit-vector (not simple-array)) (simple-array character (*)))) y)) folds to NIL 15:20:44 even though (equal (list (string 'list)) (list "LIST")) => T 15:21:03 + ;; TODO question for reviewer: Is this correct? Sometimes, #!+sb-doc 15:21:04 + ;; (with "!") is used (applies to following docstring as well). 15:21:06 i.e. it was broken before i broke it 15:21:55 so, how do i check intersection disregarding specialization? or at least returning something more conservative? 15:22:13 scymtym: the answer to the question for the reviewer is "yes". It needs to be #!+sb-doc when the file in question is being cross-compiled (by an arbitrary host lisp) and the #! dispatches on the features of the sbcl-that-is-yet-to-be 15:22:28 it is #+sb-doc when the file is being compiled by the (not quite fully baked) sbcl-that-has-survived-cold-init 15:22:41 src/pcl/sequence.lisp is compiled after cold-init, so #+sb-doc is right 15:22:51 good thing people actually don't have types like (cons (or (and bit-vector (not simple-array)) (simple-array character (*)))) 15:23:00 (apparently) 15:24:57 stassats: the types-equal-or-intersect check is also in EQL. 15:25:50 and only makes sense for EQL-compared types, so if it must be anywhere, it's in the branch that converts EQUAL to EQL. 15:25:56 it'll work for eql 15:26:15 since for eql specialized arrays won't be eql, no problem 15:26:32 stassats: yes, and it only works for EQUAL when EQL would be correct. 15:28:39 Krystof: i learned about that in the meantime but forgot to update the patch 15:28:51 should i make a corrected version? 15:29:22 pkhuong: the EQL branch just checks for intersection with (or bit-vector string cons pathname), so this shouldn't affect it 15:29:31 drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has joined #sbcl 15:29:57 if either intersects with that union type, it refuses to convert to EQL, so specialized types won't matter 15:30:27 maybe the NIL branch should check for CONS too? 15:31:05 scymtym: no, I can do that 15:31:08 (types-equal-or-intersect (type-intersection x-type y-type) combination-type) would be more agressive. 15:31:46 Krystof: thanks 15:31:46 stassats: ... or the (not types-equal-or-intersect) check could just happen in the branch that reduces to EQL. 15:32:02 right 15:33:05 what's the difference between type-intersection and type-intersection2? 15:33:47 type-intersection2 can fail if there's no intersection type method. 15:34:03 type-intersection will return a (simplified) intersection type. 15:35:07 but (type-intersection (specifier-type '(cons (or simple-bit-vector simple-base-string))) (specifier-type '(cons (or (and bit-vector (not simple-array)) (simple-array character (*)))))) => NIL 15:35:44 argh 15:35:45 ok, (or (not (types-equal-or-intersect x-type combination-type)) (not (types-equal-or-intersect y-type combination-type))) should still here 15:36:00 s/here/work/ 15:36:59 I see. 15:37:24 it's # here 15:37:35 yes, i simplified 15:37:38 Krystof: I think/hope that's what stassats meant. 15:37:42 ok, the distinction is important 15:37:50 stassats: a straight NIL means something else. 15:37:57 NIL is how type-intersection2 says "I dunno" 15:38:50 (at which point the full TYPE-INTERSECTION says "OK, I'm going to return # then") 15:39:47 -!- drmeister [~drmeister@wirelessNAT188.wireless.temple.edu] has quit [Remote host closed the connection] 15:41:48 stassats: I'd go for if they're not both types-equal-or-intersect to any of cons, bit-vector, string or pathname, then reduce to EQL, and that branch can also check if the x and y type mismatch. 15:42:22 yeah, i'm going for (if (types-equal-or-intersect x-type y-type) '(eql x y) nil) inside the current '(eql x y) leg 15:42:51 *stassats* hopes it won't be like with APPEND 15:43:01 types are hard 15:43:09 also, I'm starting to wonder how much time we spend in specifier-type. It might make sense to cache types at the call site. 15:43:30 it would be such a pain to debug EQUAL always returning NIL 15:43:58 well then, better break the commit up to simplify bisection. 15:48:02 i wonder how many other places are perceptibly to such weird types 15:48:17 err, susceptible, wrong ispell button 15:48:54 the problem is that we deal with two arguments and only consider a coarser version of the type lattice. 15:49:53 and intersection doesn't commute safely with the coarsening step. Apart from EQUALP I don't think there are other *likely* suspects. 15:51:02 i actually found this out when applied to this transform to EQUALP, since it handles more types and allowed some tests to fail 15:53:52 -!- Bike [~Glossina@69.166.47.103] has quit [Ping timeout: 264 seconds] 15:58:17 Bike [~Glossina@69.166.47.105] has joined #sbcl 16:03:03 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 16:07:27 Vivitron [~Vivitron@c-50-172-44-193.hsd1.il.comcast.net] has joined #sbcl 16:54:06 sdemarre [~serge@179.81-64-87.adsl-dyn.isp.belgacom.be] has joined #sbcl 17:00:22 -!- teggi [~teggi@113.173.4.29] has quit [Remote host closed the connection] 17:12:38 -!- fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has quit [Quit: Leaving] 17:12:42 -!- Hydan [~hydan@ip-89-103-110-5.net.upcbroadband.cz] has quit [Ping timeout: 264 seconds] 17:20:59 fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has joined #sbcl 17:22:34 -!- fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has quit [Client Quit] 17:23:40 fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has joined #sbcl 17:25:36 deftransform char-equal ((a b) (base-char base-char)) is actually never triggered 17:47:02 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Read error: Connection reset by peer] 18:01:09 leuler [~user@p548F8289.dip0.t-ipconnect.de] has joined #sbcl 18:03:49 -!- segv- [~mb@95-91-242-104-dynip.superkabel.de] has quit [Remote host closed the connection] 18:04:22 segv- [~mb@95-91-242-104-dynip.superkabel.de] has joined #sbcl 18:13:54 -!- Bike [~Glossina@69.166.47.105] has quit [Ping timeout: 264 seconds] 18:15:57 Bike [~Glossina@69.166.47.103] has joined #sbcl 18:18:42 ordering issue, most probably. 18:19:02 it needs to be on two-arg-char-equal 18:20:02 transforming (char-equal #\a char) to (or (char= #\a char)(char-= #\A char)) via (char-equal-constant char #\a #\A) is up to 4 times faster 18:21:27 and (char-equal #\- y) is transformed to (char= #\- y) not sure who writes it like that, but anyway 18:21:58 there should be a transform to make sure the constant argument if any is the second one. 18:22:19 i check both 18:23:06 making a commutative-arg-swap transform can simplify the code 18:27:05 grep shows some code with (char-equal ch #\.), so, people do write it like that 18:32:43 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 18:44:15 opencoding to char= is even faster, but takes 25 bytes more 18:45:05 could be less if char= vops were smart enough to not do unboxing 18:52:18 and it's probably worth to do a base-char fastpath in two-arg-char-equal without opencoding 18:52:28 since it's much faster 18:55:10 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 19:06:49 -!- benkard [~benkard@2a01:198:6d5:0:7902:dd53:83f2:a4e1] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 19:18:23 optimizing (string-equalp "constant" ...), or even (equalp "constant" ...) would be the next step 19:34:16 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 19:37:56 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Client Quit] 19:42:27 -!- fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has quit [Ping timeout: 260 seconds] 19:53:52 -!- sdemarre [~serge@179.81-64-87.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 264 seconds] 19:58:00 benkard [~benkard@2a01:198:6d5:0:2446:aaff:60d6:d626] has joined #sbcl 19:59:09 Krystof: wow. unicode is interesting. 20:14:22 attila_lendvai [~attila_le@62-165-243-178.pool.digikabel.hu] has joined #sbcl 20:14:22 -!- attila_lendvai [~attila_le@62-165-243-178.pool.digikabel.hu] has quit [Changing host] 20:14:22 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 20:37:05 fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has joined #sbcl 20:39:15 -!- leuler [~user@p548F8289.dip0.t-ipconnect.de] has quit [Quit: ERC Version 5.1.2 $Revision: 1.796.2.6 $ (IRC client for Emacs)] 20:49:33 -!- prxq_ [~mommer@mnhm-4d0103b5.pool.mediaWays.net] has quit [Quit: Leaving] 20:50:30 -!- fisxoj [~fisxoj@192-0-131-151.cpe.teksavvy.com] has quit [Ping timeout: 264 seconds] 20:55:59 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 21:03:31 -!- benkard [~benkard@2a01:198:6d5:0:2446:aaff:60d6:d626] has quit [Ping timeout: 246 seconds] 21:08:14 drmeister [~drmeister@mobile-166-147-108-146.mycingular.net] has joined #sbcl 21:16:37 benkard [~benkard@2a01:198:6d5:0:58a5:7e6b:5b5c:b3d4] has joined #sbcl 21:18:12 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Remote host closed the connection] 21:28:49 -!- drmeister [~drmeister@mobile-166-147-108-146.mycingular.net] has quit [Remote host closed the connection] 21:49:59 -!- psilord [~pkeller@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: Leaving.] 21:54:30 klltkr [~klltkr@unaffiliated/klltkr] has joined #sbcl 22:00:53 drmeister [~drmeister@pool-173-59-25-58.phlapa.fios.verizon.net] has joined #sbcl 22:09:31 -!- joshe [~joshe@2001:470:e862::1:1] has quit [Ping timeout: 264 seconds] 22:09:55 joshe [~joshe@2001:470:e862::1:1] has joined #sbcl 22:12:49 -!- benkard [~benkard@2a01:198:6d5:0:58a5:7e6b:5b5c:b3d4] has quit [Ping timeout: 246 seconds] 22:33:29 -!- drmeister [~drmeister@pool-173-59-25-58.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 22:43:02 psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has joined #sbcl 22:51:30 -!- yacks [~py@103.6.159.99] has quit [Ping timeout: 264 seconds] 22:57:33 fisxoj [~fisxoj@192.0.131.151] has joined #sbcl 23:15:15 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 260 seconds] 23:21:42 -!- klltkr [~klltkr@unaffiliated/klltkr] has quit [Ping timeout: 264 seconds] 23:22:36 -!- Bike [~Glossina@69.166.47.103] has quit [Ping timeout: 276 seconds] 23:43:16 drmeister [~drmeister@pool-173-59-25-58.phlapa.fios.verizon.net] has joined #sbcl 23:50:32 Bike [~Glossina@wl-nat99.it.wsu.edu] has joined #sbcl