00:00:12 francisl [~flavoie@bas1-montreal48-1176173836.dsl.bell.ca] has joined #sbcl 00:01:20 attila_lendvai [~attila_le@apn-89-223-139-73.vodafone.hu] has joined #sbcl 00:01:20 -!- attila_lendvai [~attila_le@apn-89-223-139-73.vodafone.hu] has quit [Changing host] 00:01:20 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 00:13:45 -!- Kryztof [~user@81.174.155.115] has quit [Ping timeout: 260 seconds] 00:22:36 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 272 seconds] 00:24:18 attila_lendvai [~attila_le@apn-89-223-139-73.vodafone.hu] has joined #sbcl 00:24:19 -!- attila_lendvai [~attila_le@apn-89-223-139-73.vodafone.hu] has quit [Changing host] 00:24:19 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 00:30:09 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 252 seconds] 00:38:05 tsuru`` [~charlie@adsl-74-179-25-115.bna.bellsouth.net] has joined #sbcl 00:39:51 attila_lendvai [~attila_le@apn-89-223-139-73.vodafone.hu] has joined #sbcl 00:39:51 -!- attila_lendvai [~attila_le@apn-89-223-139-73.vodafone.hu] has quit [Changing host] 00:39:51 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 00:40:00 -!- tsuru` [~charlie@adsl-74-240-217-38.bna.bellsouth.net] has quit [Ping timeout: 260 seconds] 00:47:04 -!- kanru` [~user@199.195.142.182] has quit [Remote host closed the connection] 00:52:14 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 245 seconds] 00:52:40 -!- francisl [~flavoie@bas1-montreal48-1176173836.dsl.bell.ca] has quit [Quit: francisl] 00:54:13 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 00:55:10 francisl [~flavoie@bas1-montreal48-1176173836.dsl.bell.ca] has joined #sbcl 00:58:29 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 245 seconds] 01:37:49 tsuru``` [~charlie@adsl-98-87-51-144.bna.bellsouth.net] has joined #sbcl 01:39:30 -!- tsuru`` [~charlie@adsl-74-179-25-115.bna.bellsouth.net] has quit [Ping timeout: 260 seconds] 01:40:43 echo-area [~user@182.92.247.2] has joined #sbcl 02:03:33 tsuru```` [~charlie@adsl-74-179-198-223.bna.bellsouth.net] has joined #sbcl 02:04:10 -!- francisl [~flavoie@bas1-montreal48-1176173836.dsl.bell.ca] has quit [Quit: francisl] 02:05:10 -!- tsuru``` [~charlie@adsl-98-87-51-144.bna.bellsouth.net] has quit [Ping timeout: 260 seconds] 02:18:51 francisl [~flavoie@bas1-montreal48-1176173836.dsl.bell.ca] has joined #sbcl 04:03:30 TimKack`` [~user@c-2ec29d19-74736162.cust.telenor.se] has joined #sbcl 04:05:20 -!- TimKack` [~user@c-2ec2ae41-74736162.cust.telenor.se] has quit [Ping timeout: 260 seconds] 04:06:04 -!- francisl [~flavoie@bas1-montreal48-1176173836.dsl.bell.ca] has left #sbcl 05:35:10 -!- tsuru```` [~charlie@adsl-74-179-198-223.bna.bellsouth.net] has quit [Ping timeout: 260 seconds] 05:42:38 huangjs [~huangjs@190.8.100.83] has joined #sbcl 05:58:59 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 06:00:38 -!- echo-area [~user@182.92.247.2] has quit [Ping timeout: 240 seconds] 06:04:15 sdemarre [~serge@91.176.167.230] has joined #sbcl 06:04:51 Kryztof [~user@81.174.155.115] has joined #sbcl 06:04:52 -!- ChanServ has set mode +o Kryztof 06:14:30 echo-area [~user@182.92.247.2] has joined #sbcl 15:42:35 ccl-logbot [~ccl-logbo@setf.clozure.com] has joined #sbcl 15:42:35 15:42:35 -!- names: ccl-logbot blackwolf milanj antgreen TimKack` homie` slyrus_ scymtym Posterdati tsuru`` edgar-rft @Kryztof angavrilov stassats ivan`` les flip214 jsnell_ luis pkhuong slyrus pchrist redline6561_ |3b| |42|`` daimrod Neronus kanru lichtblau foom easye whoops Phoodus brown``` joshe kwmiebach__ froydnj DGASAU antifuchs antoszka borkman asedeno_work dlowe sshirokov Quadrescence fe[nl]ix cmm nyef specbot minion 15:42:50 -!- antifuchs [foobar@care.boinkor.net] has quit [Ping timeout: 272 seconds] 15:45:47 antifuchs [foobar@care.boinkor.net] has joined #sbcl 15:50:49 cmm- [~cmm@bzq-79-180-211-142.red.bezeqint.net] has joined #sbcl 15:53:18 -!- cmm [~cmm@bzq-79-183-234-200.red.bezeqint.net] has quit [Ping timeout: 260 seconds] 16:08:02 jdz [~jdz@host214-107-dynamic.5-87-r.retail.telecomitalia.it] has joined #sbcl 16:12:36 -!- antgreen [~user@bas3-toronto06-1177890281.dsl.bell.ca] has quit [Read error: Connection reset by peer] 16:19:27 -!- homie` [~levgue@xdsl-87-79-194-136.netcologne.de] has quit [Read error: Connection reset by peer] 16:20:09 homie` [~levgue@xdsl-87-79-194-136.netcologne.de] has joined #sbcl 17:18:43 christop` [~user@oteiza.siccegge.de] has joined #sbcl 17:27:56 -!- christop` [~user@oteiza.siccegge.de] has quit [Read error: Connection reset by peer] 17:31:47 there're two deftransforms for ceiling https://github.com/sbcl/sbcl/blob/master/src/code/numbers.lisp#L690 and https://github.com/sbcl/sbcl/blob/master/src/compiler/float-tran.lisp#L1605 17:32:10 the former expands to %ceiling which has the same body as the second transform, and the second transform doesn't appear to be ever triggered 17:34:08 and nikodemus is responsible for adding it 17:36:30 since he's not here, i'll file an lp-bug 17:42:50 sdemarre [~serge@91.176.167.230] has joined #sbcl 18:14:41 antgreen [~user@out-on-140.wireless.telus.com] has joined #sbcl 19:05:09 tcr [~tcr@host110-2.natpool.mwn.de] has joined #sbcl 19:07:52 -!- antgreen [~user@out-on-140.wireless.telus.com] has quit [Read error: Connection reset by peer] 19:22:29 (disassemble (lambda (x) (declare (fixnum x) (optimize speed)) (logand x #xFFFFFFFF))) produces a gazillion of NOPs for no apparent reason 19:23:28 -!- tcr [~tcr@host110-2.natpool.mwn.de] has quit [Read error: Connection reset by peer] 19:23:32 tcr [~tcr@host110-2.natpool.mwn.de] has joined #sbcl 19:23:44 looks like it stores #xFFFFFFFF after the nops 19:25:13 that's right 19:26:29 it would probably be good to teach the disassembler to stop (and also to print nice annotations for the constants) 19:26:31 but #xFFFFFFFF is 32 bits, and AND can be applied to 32-bit immediates 19:27:12 pkhuong or leuler might tell you that it's faster to apply it to memory. Dunno 19:28:03 #xFFFFFFF is inlined 19:28:09 (one F less) 19:31:13 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 252 seconds] 19:31:13 you could try changing some of the (unsigned-byte 31)s in src/compiler/x86/arith.lisp to (unsigned-byte 32) 19:31:15 see what happens 19:32:38 or: it could be that you need to mess with the instruction emitter, so that it understands that in an (unsigned-byte 32) quantity the top bit needs to be reinterpreted as the sign bit 19:32:58 and i apply it to an immediate 32-bit from an array, and its stored in a form suitable for applying to fixnums, so it first fixnumizes the 32-bit value 19:33:12 totally messed up 19:33:27 or: you can use (mask-signed-field 32 <>) 19:34:26 i got it why it wants 31 bits, so that it can shift it left, why does it need to shift it left, i don't know 19:34:37 no, that's not why it wants to shift it left 19:35:24 stassats: 32 bit immediates are sign-extended by x86-64 19:35:33 -!- edgar-rft [~user@HSI-KBW-078-043-123-191.hsi4.kabel-badenwuerttemberg.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 19:35:59 that'll be the one 19:36:27 and the nops are because you don't want instruction and data in the same cacheline -> 64 bytes worth of padding. 19:36:53 The padding is made of nops because some decoders are over eager and then get confused by non-instruction bytes. 19:38:05 -!- tcr [~tcr@host110-2.natpool.mwn.de] has quit [Ping timeout: 260 seconds] 19:39:51 why does it perform that dance when it's non-immediate then? 19:40:33 what dance? 19:40:54 shifting left my 32-bit integer 19:41:18 what shifting? 19:41:34 christop` [~user@oteiza.siccegge.de] has joined #sbcl 19:42:26 the mask is stored as #x1FFFFFFFE, so it shifts an immediate 32-bit integer from an array before ANDing it 19:42:37 stassats: that's a fixnum. 19:42:52 I don't know what array you're referring to. 19:43:17 a string, i do sap-ref-32 on it 19:43:55 that's a representation selection issue. 19:44:04 Probably because the result is stored in a fixnum. 19:45:08 i just do (zerop (logand (sap-ref-32 sap) #xFFFFFF00)) 19:52:03 (lambda () (let ((array (make-array 10 :element-type '(unsigned-byte 32)))) (zerop (logand (aref array 0) #xFFFFFF))))=> MOV EAX, [RCX+1]; SHL RAX, 1; AND RAX, #xFEFFFF01 19:52:16 yeah, well, the rest of logand is a fixnum 19:52:21 and the rest is history. 19:52:51 bummer 19:53:18 You could use logtest and fix OPTIMIZATIONS #29 19:54:01 still working on array stuff? 19:54:18 erh, string stuff 19:54:21 You probably want to use vector-raw-bits and handle 2 chars at a time. 19:54:55 yes, i'm checking if a string is ascii 19:58:16 ok, trying (zerop (logand (sap-ref-word sap) #xFFFFFF00FFFFFF00)) 19:58:49 at least write logtest like you're meant to. 20:03:51 plus, you can build the characters up that way as well.. (ldb (byte 16 0) (logior x (ash x -24))) 20:09:58 stassats` [~stassats@wikipedia/stassats] has joined #sbcl 20:10:24 -!- minion [~minion@pppoe.95-55-188-131.dynamic.avangarddsl.ru] has quit [Disconnected by services] 20:10:27 minion [~minion@pppoe.178-66-71-10.dynamic.avangarddsl.ru] has joined #sbcl 20:11:08 -!- specbot [~specbot@pppoe.95-55-188-131.dynamic.avangarddsl.ru] has quit [Disconnected by services] 20:11:11 specbot [~specbot@pppoe.178-66-71-10.dynamic.avangarddsl.ru] has joined #sbcl 20:13:38 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 252 seconds] 20:15:10 -!- homie` [~levgue@xdsl-87-79-194-136.netcologne.de] has quit [Read error: Connection reset by peer] 20:15:49 homie` [~levgue@xdsl-87-79-194-136.netcologne.de] has joined #sbcl 20:17:39 -!- TimKack` [~user@c-2ec20125-74736162.cust.telenor.se] has quit [Remote host closed the connection] 20:17:51 TimKack` [~user@c-2ec20125-74736162.cust.telenor.se] has joined #sbcl 20:18:07 alright, applying #xFFFFFF80FFFFFF80 is the fastest 20:18:16 and it doesn't shift anything 20:18:18 -!- fe[nl]ix [~quassel@pdpc/supporter/professional/fenlix] has quit [Remote host closed the connection] 20:18:48 would be faster if #xFFFFFF80FFFFFF80 was in a register? 20:20:25 fe[nl]ix [~quassel@pdpc/supporter/professional/fenlix] has joined #sbcl 20:22:33 -!- TimKack` is now known as TimKack 20:24:50 stassats`: yeah, but you'll have to contort your code to get that 20:25:03 contort? how? 20:26:31 I think the easiest way is to declare an ub64 variable, and setf it to the value you want 20:27:00 what happened to the peep-hole optimizer? 20:27:06 i see lots of redundant operations 20:29:48 didn't help 20:36:33 the next option is to use SSE 20:37:05 oh right... smarter constant propagation 20:37:21 peephole wouldn't help with hoisting this constant out of the loop 20:37:36 pkhuong: i was talking about different things 20:37:51 like MOV RCX, RAX MOV RCX, [RCX] 20:42:35 it would be nice if disassembly could label 537919567 and 537919511 as NIL and T 20:43:20 you get used to it ;) 20:59:08 -!- homie` [~levgue@xdsl-87-79-194-136.netcologne.de] has quit [Read error: Connection reset by peer] 20:59:55 homie` [~levgue@xdsl-87-79-194-136.netcologne.de] has joined #sbcl 21:00:09 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 21:00:17 -!- homie` [~levgue@xdsl-87-79-194-136.netcologne.de] has quit [Write error: Connection reset by peer] 21:03:19 homie [~levgue@xdsl-87-79-194-136.netcologne.de] has joined #sbcl 21:38:31 -!- jdz [~jdz@host214-107-dynamic.5-87-r.retail.telecomitalia.it] has quit [Quit: Byebye.] 21:45:05 huangjs [~huangjs@190.8.100.83] has joined #sbcl 21:49:41 -!- milanj [~milanj_@93-86-216-175.dynamic.isp.telekom.rs] has quit [Quit: Leaving] 21:52:02 stassats`: I have a branch for that kind of annotation. 21:54:38 stassats`: Have a look at http://repo.or.cz/w/sbcl/nyef.git/shortlog/refs/heads/x86oid-disassembler-fixes but remember that it's still a bit rough, not ready for commit upstream. 22:06:15 -!- sdemarre [~serge@91.176.167.230] has quit [Ping timeout: 260 seconds] 22:09:43 slyrus__ [~chatzilla@99-28-161-110.lightspeed.miamfl.sbcglobal.net] has joined #sbcl 22:12:55 -!- slyrus_ [~chatzilla@99-28-161-110.lightspeed.miamfl.sbcglobal.net] has quit [Read error: Connection reset by peer] 22:23:29 rstill [~rstill@12.104.144.2] has joined #sbcl 22:33:40 -!- homie [~levgue@xdsl-87-79-194-136.netcologne.de] has quit [Ping timeout: 260 seconds] 22:35:11 homie [~levgue@xdsl-78-35-146-156.netcologne.de] has joined #sbcl 23:29:36 -!- rstill [~rstill@12.104.144.2] has quit [Read error: Connection reset by peer] 23:54:09 attila_lendvai1 [~attila_le@188-143-58-245.pool.digikabel.hu] has joined #sbcl 23:54:09 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Disconnected by services] 23:54:29 echo-area [~user@182.92.247.2] has joined #sbcl 23:58:09 -!- huangjs [~huangjs@190.8.100.83] has quit [Ping timeout: 252 seconds]