rwmjones changed the topic of #ocaml to: Discussions about the OCaml programming language | http://caml.inria.fr/ | Grab Ocaml 3.10.1 from http://caml.inria.fr/ocaml/release.html (featuring new camlp4 and more!)
brooksbp has joined #ocaml
ahf has joined #ocaml
seafood_ has quit []
brooksbp has quit []
TheLittlePrince has joined #ocaml
TheLittlePrince has quit [Read error: 104 (Connection reset by peer)]
TheLittlePrince has joined #ocaml
Waleee has quit []
thermoplyae has quit [Remote closed the connection]
thermoplyae has joined #ocaml
Snrrrub has joined #ocaml
m3ga has joined #ocaml
brooksbp has joined #ocaml
AxleLonghorn has joined #ocaml
Morphous has quit ["shutdown"]
Amorphous has joined #ocaml
AxleLonghorn has left #ocaml []
TheLittlePrince has quit [Remote closed the connection]
brooksbp has quit []
mwc has joined #ocaml
mwc has left #ocaml []
middayc has joined #ocaml
delamarche has quit []
middayc has quit []
mrsolo has quit ["Leaving"]
m3ga has quit ["disappearing into the sunset"]
postalchris has joined #ocaml
Snrrrub has quit [Read error: 110 (Connection timed out)]
seafood_ has joined #ocaml
postalchris has quit [Read error: 110 (Connection timed out)]
goalieca has joined #ocaml
szell has quit [Client Quit]
m3ga has joined #ocaml
thermoplyae has quit ["daddy's in space"]
__suri has joined #ocaml
Snrrrub has joined #ocaml
bluestorm has joined #ocaml
<flux> neale, I'm in the process of writing bindings for libev, however I too have some gc issues at the moment
ecc has quit [Read error: 113 (No route to host)]
m3ga has quit ["disappearing into the sunset"]
jderque has joined #ocaml
Snrrrub has quit [Read error: 110 (Connection timed out)]
bluestorm has quit [Remote closed the connection]
ygrek has joined #ocaml
yminsky has joined #ocaml
hkBst has joined #ocaml
szell has joined #ocaml
|Catch22| has quit []
Linktim has joined #ocaml
seafood_ has quit []
l_a_m has joined #ocaml
marmottine has joined #ocaml
seafood_ has joined #ocaml
andrewdotnich has joined #ocaml
ttamttam has joined #ocaml
hkBst has quit ["Konversation terminated!"]
goalieca has quit [Remote closed the connection]
ygrek has left #ocaml []
ygrek has joined #ocaml
<ygrek> is there a way to copy a file not using Unix.system?
bongy has joined #ocaml
<pango> open source file for read, open target file for write, while you haven't reached source file eof read from it in a buffer and write buffer's content to target file. When done, close both files
dwmw2_gone is now known as dwmw2_GVA
<ygrek> thats an option, but no system call wrapper?
<pango> there's no system call to copy files
<ygrek> I do not mean syscall, but any system api - CopyFile on win32 etc
<flux> /bin/cp is the closest you can get
<flux> however, if you change your mind and want to use it, do it the safe way: Unix.putenv "SRC" sourcename; Unix.putenv "DST" dstname; Unix.system "/bin/cp -- \"$SRC\" \"$DST\""
<pango> that's not the safest way
<flux> hm
<flux> not?-(
<flux> what can go wrong?
<flux> oh, you mean refer to using Unix.exec?
<pango> the way you wrote it still involves a shell to expand environment variables
<flux> exactly, that was the point?
<flux> note the quotes, which also go to the shell
<ygrek> why is it safer than Unix.system (sprintf "/bin/cp %s %s" source dest) ?
<pango> ygrek: nope
<flux> ygrek, think source = "; rm -rf /"
<flux> pango, what kind of source name were you thinking that wouldn't be safe?
<ygrek> ah, I see
<pango> requirement is to pass arguments as array string... So I suppose exec* and create_process* should be fine
<flux> pango, Unix.putenv "SRC" "\"\nBLA'`\"\nbla"; Unix.system "/bin/echo \"$SRC\"" works exactly as expected
<flux> exec has the advantage of being slightly more efficient too
<pango> flux: but you must fork() first, if you want to keep control
<flux> well, system does that anyway; infact I think it'll likely do it twice, once for the shell and once for cp
<pango> yes
<pango> I was referring to your last msg about exec
<flux> it's not a drop-in replacement to system, but it's the essence of the alternative approach ;)
ttamttam has left #ocaml []
LordMetroid has joined #ocaml
lordmetroid_ has joined #ocaml
<ygrek> it copies through buffers
<ygrek> should work on win32 too
<pango> actually it'd be a good idea to add a "copy file" syscall... Filesystems doing COW (copy-on-write) could provide highly efficient implementation, etc.
bongy has quit [Read error: 110 (Connection timed out)]
LordMetroid has quit [Read error: 110 (Connection timed out)]
lordmetroid__ has joined #ocaml
lordmetroid_ has quit [Connection timed out]
Snrrrub has joined #ocaml
lordmetroid__ has quit ["Leaving"]
wy has joined #ocaml
dwmw2_GVA is now known as dwmw2_gone
Snrrrub has quit [Read error: 110 (Connection timed out)]
Linktim_ has joined #ocaml
andrewdotnich has quit []
Yoric[DT] has joined #ocaml
<Yoric[DT]> hi
Linktim- has joined #ocaml
andrewdotnich has joined #ocaml
Linktim has quit [Read error: 110 (Connection timed out)]
Linktim_ has quit [Read error: 110 (Connection timed out)]
andrewdotnich has quit []
Linktim- has quit [Read error: 110 (Connection timed out)]
marmottine has quit [Read error: 113 (No route to host)]
andrewdotnich has joined #ocaml
andrewdotnich has quit ["zzzz"]
marmottine has joined #ocaml
<wy> hello
<jlouis_> hep!
AxleLonghorn has joined #ocaml
bongy has joined #ocaml
seafood_ has quit []
bongy has quit ["Leaving"]
<jlouis_> yminsky: nice posts on phantom types and universal embeddings
postalchris has joined #ocaml
jderque has quit [Read error: 113 (No route to host)]
bzren has joined #ocaml
<yminsky> Thanks. We hope to post a bunch more in the same vein. I feel like we've built up a fair bit of material over the last few years.
tjoad94 has joined #ocaml
tjoad94 has quit [Client Quit]
AxleLonghorn has left #ocaml []
ygrek has quit [Remote closed the connection]
ygrek has joined #ocaml
pango has quit [Remote closed the connection]
pango has joined #ocaml
wy has quit ["Leaving"]
LordMetroid has joined #ocaml
Amorphous has quit [Read error: 110 (Connection timed out)]
Amorphous has joined #ocaml
postalchris has quit ["Leaving."]
<jlouis_> yminsky: Indeed! I happen to know Stephen from MLton and he tends to cook up amazing solutions to problems ;)
dwmw2_gone is now known as dwmw2_GVA
Linktim has joined #ocaml
dwmw2_GVA is now known as dwmw2_gone
uuuppz has quit [Read error: 113 (No route to host)]
Linktim_ has joined #ocaml
thermoplyae has joined #ocaml
Anarchos has joined #ocaml
Linktim- has joined #ocaml
bluestorm has joined #ocaml
Deknos has joined #ocaml
ecc has joined #ocaml
Linktim has quit [Read error: 110 (Connection timed out)]
Deknos is now known as menace
Linktim_ has quit [Read error: 110 (Connection timed out)]
mfp_ is now known as mfp
<Anarchos> is the ocaml cvs accessible right now ?
hsuh has joined #ocaml
<pango> http://caml.inria.fr at least looks unreachable
Linktim- has quit [Read error: 110 (Connection timed out)]
<Anarchos> ok that should be the ocaml servers which are down
<flux> should be distributed vcs.. ;-)
<flux> (I'm not sure if they'd prefer that, the developing process being a bit closed..)
tjoad94 has joined #ocaml
<thelema> flux, Anarchos: http://repo.or.cz/w/ocaml.git
tjoad94 has quit [Client Quit]
<thelema> the git branch suffers one problem from the conversion, and some configure files (magically copied from an upstream branch) didn't make it into the needed branches. Non-INRIA branches have the needed files checked in, the INRIA branches I'm afraid to do so with, for fear of breaking synchronization
LordMetroid has quit ["Leaving"]
middayc has joined #ocaml
middayc_ has joined #ocaml
middayc has quit [Read error: 110 (Connection timed out)]
middayc_ has quit []
middayc has joined #ocaml
Linktim- has joined #ocaml
hsuh has quit [Remote closed the connection]
ttamttam has joined #ocaml
ttamttam has left #ocaml []
thelema has quit ["Killed by frash (Requested by panasync)"]
marmottine has quit ["Quitte"]
hasone has joined #ocaml
letrec has quit [Connection timed out]
bluestorm has quit ["Konversation terminated!"]
mwc has joined #ocaml
evn_ has quit []
ygrek has quit [Remote closed the connection]
evn_ has joined #ocaml
<flux> hmh, I'm having hard time fixing ocaml-ev's gc issues.. maybe it's a plain simple bug somewhere, as afaik I've ensured the registration of c-referred ocaml values as gc roots already.
<Anarchos> i have plenty of segfault myself, due to multithread in my C code, even if i register global roots
szell` has joined #ocaml
<flux> my case is single threaded
<flux> atleast I've found a case that triggers the problem always
<flux> even if I keep references to the data I'm creating, so it shouldn't be a released-too-early-kind of problem but rather a stuff-gets-moved - maybe
<flux> actually, maybe not! I disabled heap compactions, and still the same.. oh well, maybe this points me to the source better.
<Anarchos> have you compiled ocaml with debug flags ?
<flux> ocaml itself or my code?
<flux> I've tried compiling my code with debug flags at times
<flux> it crashes at a function that converts stuff from a bitmask to a list
<flux> so mainly it's allocating memory..
<flux> OCAMLRUNPARAM=v255 makes the program say "Growing page table to 10335 entries" as its last thing.
<Anarchos> so this growing is eating some of your pointers
<Anarchos> have you printf'ed all your global roots you are registering ?
<flux> not
jlouis has joined #ocaml
<flux> at this point I actually register my roots to statically allocated buffers
<flux> with a fixed number of slots (and an assertion that I don't run over the number)
szell has quit [Read error: 110 (Connection timed out)]
<Anarchos> i don't know what is the effect to use statically allocated buffers with the GC
jlouis_ has quit [Read error: 110 (Connection timed out)]
Snrrrub has joined #ocaml
<flux> hm, I wonder if my test case is simply allocating too much memory
<flux> its memory space is one gigabyte..
<Anarchos> maybe your static memory area is growing upon the ocaml heap ? but i don't know how to detect it simply
<flux> maybe I'll put all the memory releasing stuff back in and test the behavior again
<flux> fun stuff, crashes under gdb at around 250M, doesn't crash without gdb
<flux> and .opt crashes :)
<flux> maybe I'll put together a minimal version of that to see the issue better.. but not tonight.
evn_ has quit []
l_a_m has quit [Remote closed the connection]
Linktim- has quit [Remote closed the connection]
Snrrrub has quit [Read error: 110 (Connection timed out)]
seafood_ has joined #ocaml
Yoric[DT] has quit ["Ex-Chat"]