00:43:48 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 00:45:10 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 00:46:33 does anyone know what the complaints about SBCL not closing streams are about? 01:32:05 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 01:45:22 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 02:14:18 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Ping timeout: 256 seconds] 02:39:28 -!- christoph_debian [~christoph@ppp-188-174-59-77.dynamic.mnet-online.de] has quit [Ping timeout: 264 seconds] 02:52:33 christoph_debian [~christoph@ppp-188-174-111-253.dynamic.mnet-online.de] has joined #sbcl 02:54:39 echo-area [~user@123.120.239.34] has joined #sbcl 03:03:08 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 03:31:02 attila_lendvai [~attila_le@95.56.68.98] has joined #sbcl 03:31:02 -!- attila_lendvai [~attila_le@95.56.68.98] has quit [Changing host] 03:31:02 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 03:43:08 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 04:01:03 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 04:03:01 -!- Bike [~Glossina@67-5-251-178.ptld.qwest.net] has quit [Ping timeout: 276 seconds] 04:04:28 Bike [~Glossina@67-5-251-178.ptld.qwest.net] has joined #sbcl 04:49:35 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has left #sbcl 05:36:13 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Read error: Connection reset by peer] 05:36:25 attila_lendvai [~attila_le@95.56.68.98] has joined #sbcl 05:36:25 -!- attila_lendvai [~attila_le@95.56.68.98] has quit [Changing host] 05:36:25 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 05:45:53 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Read error: Connection reset by peer] 05:46:06 attila_lendvai [~attila_le@95.56.68.98] has joined #sbcl 05:46:06 -!- attila_lendvai [~attila_le@95.56.68.98] has quit [Changing host] 05:46:06 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 05:49:16 slyrus [~chatzilla@107.200.11.156] has joined #sbcl 05:50:16 -!- Bike [~Glossina@67-5-251-178.ptld.qwest.net] has quit [Ping timeout: 276 seconds] 05:50:53 Bike [~Glossina@67-5-251-178.ptld.qwest.net] has joined #sbcl 05:52:29 Bike_ [~Glossina@67-5-251-178.ptld.qwest.net] has joined #sbcl 05:55:28 -!- Bike [~Glossina@67-5-251-178.ptld.qwest.net] has quit [Ping timeout: 276 seconds] 06:17:52 -!- Bike_ is now known as Bike 07:01:12 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 07:03:11 stassats [~stassats@wikipedia/stassats] has joined #sbcl 07:11:37 prxq [~mommer@mnhm-590c2f28.pool.mediaWays.net] has joined #sbcl 07:52:53 -!- oleo [~oleo@xdsl-78-35-164-53.netcologne.de] has quit [Ping timeout: 240 seconds] 07:54:13 oleo [~oleo@xdsl-78-35-130-183.netcologne.de] has joined #sbcl 08:13:45 Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 08:34:32 -!- Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 08:38:53 -!- slyrus [~chatzilla@107.200.11.156] has quit [Ping timeout: 240 seconds] 08:42:17 -!- oleo [~oleo@xdsl-78-35-130-183.netcologne.de] has quit [Quit: Verlassend] 08:43:27 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 08:48:23 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep] 08:57:25 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 09:28:24 oleo [~oleo@xdsl-78-35-130-183.netcologne.de] has joined #sbcl 09:32:24 -!- oleo [~oleo@xdsl-78-35-130-183.netcologne.de] has quit [Client Quit] 09:33:46 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 09:44:47 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 09:52:01 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 10:14:47 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 10:16:02 scymtym_ [~user@ip-5-147-122-209.unitymediagroup.de] has joined #sbcl 10:26:25 oleo [~oleo@xdsl-78-35-130-183.netcologne.de] has joined #sbcl 11:10:43 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 11:15:07 pkhuong: this thing could also be used to fix https://bugs.launchpad.net/sbcl/+bug/668543 11:40:00 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 12:10:46 stassats: it's different. That one needs a hook in constraints. 12:12:57 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 12:18:06 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Ping timeout: 264 seconds] 12:20:05 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 13:02:41 pkhuong: but structure predicates seem to be implemented as (typep x 'structure) 13:05:44 I don't know if they're always inlined 13:06:38 seems like it is 13:06:49 except for typed ones 13:42:18 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 14:06:40 -!- yacks [~py@103.6.159.2] has quit [Read error: Connection reset by peer] 14:22:59 yacks [~py@103.6.158.102] has joined #sbcl 14:40:58 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 14:45:14 scymtym__ [~user@ip-5-147-122-209.unitymediagroup.de] has joined #sbcl 14:46:09 -!- scymtym_ [~user@ip-5-147-122-209.unitymediagroup.de] has quit [Read error: Operation timed out] 15:04:30 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 15:09:04 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Client Quit] 15:16:24 Quadrescence [~quad@unaffiliated/quadrescence] has joined #sbcl 15:20:59 Does anyone (pkhuong probably) have any thoughts on "unboxing" user-defined types, or even built in ones such as cons? Could DX be used for stack allocation of structs? Are they used for that? 15:22:55 The motivation for asking is: I have a type called STRING-SLICE, which represents a slice of a string via an offset + length, but allocating/passing around the data structures sort of defeats the point of the efficiency hack. But at the same time, it's not really desirable to change the API of everything to take those extra parameters explicitly. 15:23:23 structs do get DX allocated. 15:23:50 nikodemus's sb-cga has an interesting macroified approach to a similar problem. 15:23:56 you can get scalarisation. 15:24:25 Cool, I'll go look. 15:24:37 unboxing won't happen, except perhaps for local functions. if someday we gain specialised calling conventions for full calls, things might change. 15:26:08 and, on that note, my brain shall be offline for 4-5 days. 15:27:18 ah, I think you'll need to declare the constructor inline for DX to work. 15:28:01 pkhuong, Curious. Why? 15:28:42 otherwise DX doesn't make sense: you're declaring th eresult of a full call to be DX 15:28:49 but the result will be received as a pointer 15:29:27 it could work with a particularly clever default calling convention. 15:30:00 I see. 15:57:19 -!- yacks [~py@103.6.158.102] has quit [Quit: Leaving] 16:01:24 pkhuong: i didn't quite catch which part causes problem with the GC, having an untagged pointer? 16:06:15 stassats: yes. 16:07:17 but that already happens 16:07:24 there is some code on precise platforms that tries to get one such case (a raw pointer that's linked to a tagged pointer) right (there's no workaround for jump/call targets)... I'm not sure it's bug-free yet. 16:07:31 where? 16:07:57 %test-lowtag already does lea eax, [x-15] 16:07:59 even if that's true, not adding additional instances is a worthy goal. 16:08:24 stassats: and? 16:08:34 it's not using that value as a register. 16:09:21 uses it as what then? 16:09:35 I don't have a problem with performing arithmetic on pointers 16:09:43 I have a problem with dereferencing such pointers. 16:10:39 *not using that value (the eax register) as an address. 16:10:50 how does the GC know that? 16:11:02 stassats: it doesn't have to. 16:11:32 all that test-lowtag does is test the lowtag. 16:11:36 GC won't change the lowtag. 16:13:11 but how does the GC know that the code uses it as an effective address? 16:13:22 what? That question makes no sense. 16:14:00 well, i ask before what you're telling me makes no sense to me, so, it may in turn not make any sense either 16:14:21 how using [ECX] is worse than using [ECX-7] for the GC 16:14:22 you added code that dereferences an untagged pointer. 16:14:38 a tagged ECX ia recognized as a pointer by the GC. 16:14:42 an untagged ECX isn't. 16:15:16 is dereferncing recorded during compilation? 16:15:21 [ECX-7] means ECX remains tagged. [ECX] means ECX is a raw pointer that will not be considered as such by the GC. 16:15:39 GC only keeps track of potential tagged pointers. 16:16:01 yes, i understand that part, the technical side of how the gc decides how to treat which registers is unclear to me 16:16:20 same as the stack. 16:16:29 If it looks like a tagged pointer to the Lisp heap, it's a conservative root. 16:16:33 Otherwise, it's not. 16:16:40 A raw pointer is not a tagged pointer. 16:17:25 ok then why is doing LEA EAX, [ECX-7] ok? 16:17:40 because we're not using eax as a pointer 16:17:43 it's just arithmetic. 16:18:03 indeed we aren't, but how does the gc know that we don't? 16:18:05 we don't care if eax doesn't get fixed up. 16:18:17 WHAT? 16:19:14 GC assumes that everything that doesn't look like a tagged pointer isn't. 16:19:40 then what's the problem with using EAX as a pointer if ECX still contains a tagged pointer? 16:20:55 "This only works because we're conservative on x86." 16:21:46 then non-conservativeness will take into account that EAX is in fact dereferenced? 16:21:58 how? it's a raw pointer. 16:22:37 the compiler may encode which registers are dereferenced 16:23:31 i'm just trying to imagine a scenario how that would break with a non-conservative GC 16:24:19 i just need to understand what's going on 16:24:19 <|3b|> is the problem that a precise GC might move the data and adjust (only) teh pointers it knows about? 16:24:20 because we have an untagged pointer to the lisp heap that we must somehow infer is a pointer and not a fixnum? 16:24:46 |3b|: yes. 16:25:04 so, if it stops in-between, that starts to make sense 16:27:07 i will back up this and an upcoming optimtization which also does similar pointer arithmetic 16:38:58 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 16:42:29 some type-check vops don't seem to be ever invoked, the type-test ones are used instead 16:42:37 why do we even need type-check then? 16:45:24 pretty sure "simple" casts turn into type checks. 16:45:31 Now I'm offline for real. 16:56:39 sdemarre [~serge@109.134.136.172] has joined #sbcl 16:58:31 -!- oleo [~oleo@xdsl-78-35-130-183.netcologne.de] has quit [Read error: Connection reset by peer] 17:11:08 oleo [~oleo@xdsl-78-35-130-183.netcologne.de] has joined #sbcl 17:11:10 milosn_ [~milosn@user-5af50b88.broadband.tesco.net] has joined #sbcl 17:13:09 -!- milosn [~milosn@user-5af50bca.broadband.tesco.net] has quit [Ping timeout: 248 seconds] 17:19:18 -!- sdemarre [~serge@109.134.136.172] has quit [Ping timeout: 264 seconds] 17:53:02 sdemarre [~serge@109.134.136.172] has joined #sbcl 17:56:37 -!- echo-area [~user@123.120.239.34] has quit [Remote host closed the connection] 17:56:57 echo-area [~user@123.120.239.34] has joined #sbcl 18:01:24 -!- jsnell_ [~jsnell@ash.snellman.net] has quit [Read error: Operation timed out] 18:09:38 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 18:17:19 jsnell [~jsnell@ash.snellman.net] has joined #sbcl 18:37:36 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 18:38:00 ok, no more bit-fiddling, i should do something more substantial this release round 18:38:29 like finally finishing fast octets-to-string 18:43:48 thanks for the patch 18:46:14 fe[nl]ix: using two swap-bytes-32 for -64 on 32 bits is quite an improvement 18:46:22 maybe a portable version should that too 18:49:27 I'll keep that in mind 18:58:50 yacks [~py@103.6.158.102] has joined #sbcl 19:08:55 drmeister [~drmeister@mobile-166-147-109-113.mycingular.net] has joined #sbcl 19:15:55 davazp [~user@204.Red-79-153-96.dynamicIP.rima-tde.net] has joined #sbcl 19:17:42 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 256 seconds] 19:25:12 -!- davazp [~user@204.Red-79-153-96.dynamicIP.rima-tde.net] has quit [Remote host closed the connection] 19:27:28 -!- flip214 [~marek@unaffiliated/flip214] has quit [Quit: Lost terminal] 19:36:36 -!- drmeister [~drmeister@mobile-166-147-109-113.mycingular.net] has quit [Remote host closed the connection] 19:49:58 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 19:54:21 -!- pipping [~pipping@exherbo/developer/pipping] has left #sbcl 20:06:10 ehaliewi` [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 20:06:20 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Read error: No route to host] 20:06:58 -!- |3b| [bbb@2600:3c00::f03c:91ff:fedf:5b65] has quit [Remote host closed the connection] 20:07:44 |3b| [bbb@2600:3c00::f03c:91ff:fedf:5b65] has joined #sbcl 20:15:54 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 20:17:30 -!- Bike [~Glossina@67-5-251-178.ptld.qwest.net] has quit [Ping timeout: 264 seconds] 20:21:22 -!- ehaliewi` [~user@50-0-51-11.dsl.static.sonic.net] has quit [Remote host closed the connection] 20:22:41 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 20:23:46 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 20:25:37 Bike [~Glossina@67-5-251-178.ptld.qwest.net] has joined #sbcl 20:39:15 -!- oleo [~oleo@xdsl-78-35-130-183.netcologne.de] has quit [Read error: Connection reset by peer] 20:40:13 oleo [~oleo@xdsl-84-44-154-236.netcologne.de] has joined #sbcl 20:46:47 -!- sdemarre [~serge@109.134.136.172] has quit [Ping timeout: 268 seconds] 21:00:06 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Read error: Connection reset by peer] 21:14:02 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 21:16:43 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Client Quit] 21:19:11 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 21:21:58 gabnet [~gabnet@138.195.197.77.rev.sfr.net] has joined #sbcl 21:34:57 -!- gabnet [~gabnet@138.195.197.77.rev.sfr.net] has quit [Quit: Quitte] 21:54:15 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Read error: Connection reset by peer] 21:58:45 benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has joined #sbcl 22:03:49 -!- benkard [~benkard@mnch-4d04dcfd.pool.mediaWays.net] has quit [Quit: Textual IRC Client: www.textualapp.com] 22:05:32 -!- vi1 [~vi1@93.92.216.186] has quit [Ping timeout: 246 seconds] 22:12:22 fe[nl]ix: you can try pshufb with sb-simd 22:15:33 is it really worth for just 4/8 byes ? 22:16:48 jarod_ch_ [~jarod_che@115.205.239.204] has joined #sbcl 22:18:51 for 8, perhaps 22:19:40 -!- prxq [~mommer@mnhm-590c2f28.pool.mediaWays.net] has quit [Remote host closed the connection] 22:40:33 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 22:40:50 -!- Quadrescence [~quad@unaffiliated/quadrescence] has quit [Quit: This computer has gone to sleep]