lucybun has quit [Remote host closed the connection]
lucybun has joined #ocaml
shinnya has quit [Ping timeout: 240 seconds]
pnathan has left #ocaml ["ERC (IRC client for Emacs 24.5.1)"]
slash^ has joined #ocaml
Simn has joined #ocaml
babygetoboy has joined #ocaml
babygetoboy has quit [Client Quit]
Fistine has quit [Ping timeout: 260 seconds]
Fistine has joined #ocaml
Anarchos has joined #ocaml
M-Illandan has quit [Ping timeout: 260 seconds]
Riviera_ has quit [Ping timeout: 260 seconds]
M-jimt has quit [Ping timeout: 240 seconds]
M-jimt has joined #ocaml
M-Illandan has joined #ocaml
govg has quit [Ping timeout: 240 seconds]
shinnya has joined #ocaml
orbifx has joined #ocaml
nightmared has quit [Quit: WeeChat 1.7]
nightmared has joined #ocaml
dmi3y has joined #ocaml
dmi3y has quit [Client Quit]
dmi3y has joined #ocaml
cdidd has quit [Read error: Connection reset by peer]
cdidd has joined #ocaml
segmond has quit [Quit: l8r]
Muzer has quit [Ping timeout: 258 seconds]
dmi3y has quit [Quit: dmi3y]
_whitelogger has joined #ocaml
Fistine has quit [Ping timeout: 258 seconds]
sepp2k has joined #ocaml
Fistine has joined #ocaml
Muzer has joined #ocaml
maxirater has joined #ocaml
Anarchos has quit [Quit: Vision[0.9.7-H-20140108]: i've been blurred!]
mfp has joined #ocaml
cbot_ has quit [Quit: Leaving]
maker has quit [Quit: = ""]
maker has joined #ocaml
sepp2k has quit [Quit: Leaving.]
sepp2k has joined #ocaml
sepp2k has quit [Client Quit]
sepp2k has joined #ocaml
<Simn>
That's funny, I removed an unused argument from a variant (from 2 args to 1) and now my benchmark runs slightly slower than before. Turns out this change causes one additional major GC collection. I guess that's just "bad luck"?
<companion_cube>
that's weird, removing an argument should trigger fewer GCs, not more
fraggle_ has joined #ocaml
<Simn>
That's what I thought as well.
argent_smith has joined #ocaml
<companion_cube>
are you sure it's the only change?
<jerith>
Maybe the thing you removed is more ephemeral and prevented the thing that's left from being promoted?
argent_smith has joined #ocaml
<Simn>
I didn't think that could happen, hmm
<jerith>
I know very little about GC and general and OCaml's in particular.
<Simn>
I have less promoted_words in the new version too though. However I do have more fragments
<Simn>
With a strategically placed Gc.compact() things even out. I still don't really understand it though.
<jerith>
Maybe it's just interactions between the number of things you're allocating and various thresholds in the GC?
<Simn>
Must be yeah. That's what I meant by "bad luck". :)
<jerith>
With less stuff being allocated, some of it hangs around longer and thus gets promoted.
<jerith>
Even though less stuff is being promoted overall, it might be promoted just before it becomes garbage and thus triggers extra collections.
<Simn>
Yes I'll go with that explanation. Thanks!
andreas__ has joined #ocaml
<def`>
Another problem that you can see is fragments: because the promotted things are too small, they cannot be reallocated easily and cause increased fragmentation.
<def`>
"type t = V of a * b" has the same size as a list cell
<def`>
but "type t = V of a" is too small for holding a list cell
<jerith>
Do you require at least one of them, or will the program work without either?
<andreas__>
The library works with none, either or both, but the tests require both, since I test all of the functionality of the library.
<jerith>
Ah.
<andreas__>
Trying EXTRA_DEPS now.
<jerith>
So you want to make sure it compiles without them, but you only want to run tests with tem.
<jerith>
*them
<andreas__>
That would be optimal :)
<jerith>
Why not add them as test dependencies the same as alcotest?
<jerith>
EXTRA_DEPS is for things you need during the build, but don't want to put in your opam file.
<andreas__>
Thanks, I'll give that a shot also -- feels like I've tried so many different things by now, but maybe not that one..
<jerith>
You probably want those test dependencies on the opam file anyway, otherwise the tests will be broken for anyone who hasn't installed the optional deps.
<hannes>
andreas__: you might want to add lwt {test} (etc.) in your opam depends section...
<jerith>
hannes: That's what I suggested, but your phrasing is better. :-)
moei has joined #ocaml
<hannes>
jerith: excuse my ignorance ;)
<jerith>
No, I was probably unclear.
mengu has joined #ocaml
<andreas__>
Thanks both, new Travis build is running now with the test deps :)
<jerith>
That's a lot of builds.
<andreas__>
It's testing all combinations of depopts on 4.03 and 4.04 now..
<jerith>
It would be nice if there were a way to say "here are the combinations of depopts I want to use" for a single build.
grayswandyr has joined #ocaml
<grayswandyr>
hi I'm trying to use ppx_deriving.show and get a customized output for a (toy) polymorphic, recursive type but I can't succeed. Here is what I wrote:
<grayswandyr>
type 'a prop = A of 'a | And of 'a prop * 'a prop [@polyprinter fun ppa out a b -> fprintf out "%a & %a" (pp_prop ppa) a (pp_prop ppa) b] [@@deriving show];;
<grayswandyr>
pp_prop Format.pp_print_string Format.std_formatter @@ And (And (A "a", A "b"), A "c");;
<grayswandyr>
Which gives: (And ((And ((A a), (A b))), (A c)))
<sh0t>
Hi guys. I am trying to use why3 in my code but i keep getting: Unbounded module: Why3. But I did as the manual says for the installation of the why3 API.
maxirater has joined #ocaml
maxirater has quit [Quit: Leaving]
richi235 has joined #ocaml
<orbitz>
Does exhaustivness checking and GADTs not play well together? The compiler is telling me I'm not matching on a case that should be impossible base don the types, unelss I'm doing something wrong.
maarhart has quit [Remote host closed the connection]
diphuser has joined #ocaml
<orbitz>
def`: Am I correct in understanding that the current checking will give false errors here. What is the suggested way to deal with this in code? Add the unused cases to silence the warning?
Xadnem has joined #ocaml
<lyxia>
that's a sane solution
<def`>
| Counter_example -> .
orbifx has joined #ocaml
slash^ has quit [Read error: Connection reset by peer]
jbrown has joined #ocaml
kamog has joined #ocaml
jbrown has quit [Client Quit]
copy` has joined #ocaml
richi235 has quit [Ping timeout: 240 seconds]
segmond has joined #ocaml
richi235 has joined #ocaml
MercurialAlchemi has quit [Ping timeout: 252 seconds]
average has quit [Remote host closed the connection]
larhat has joined #ocaml
diphuser has quit [Ping timeout: 240 seconds]
two_wheels has joined #ocaml
larhat has quit [Quit: Leaving.]
TheLemonMan has quit [Quit: "It's now safe to turn off your computer."]
Simn has quit [Quit: Leaving]
mars0i has joined #ocaml
mars0i has quit [Client Quit]
sh0t has quit [Quit: Leaving]
argent_smith1 has quit [Quit: Leaving.]
mars0i has joined #ocaml
mars0i has left #ocaml [#ocaml]
ryanartecona has joined #ocaml
two_wheels has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
kakadu has quit [Remote host closed the connection]
kakadu has joined #ocaml
Xadnem has quit [Quit: leaving]
two_wheels has joined #ocaml
Fistine has quit [Ping timeout: 258 seconds]
Fistine has joined #ocaml
average has joined #ocaml
StatelessCat has quit [Quit: WeeChat 1.7]
sepp2k has quit [Quit: Leaving.]
average has quit [Quit: Lost terminal]
ryanartecona has quit [Quit: ryanartecona]
cbot has joined #ocaml
spew has joined #ocaml
mengu has quit [Remote host closed the connection]