No, I made a mistake there -- I removed a for by accident, when the real
problem was that the original for loop wasn't using 'in-range'. But Dmitry
figured that out with the latest version of his code.
On Thu, Feb 21, 2013 at 7:02 PM, Robby Findler
wrote:
>
>
>
> On Thu, Feb 21, 2013 at 4:47 PM
On Thu, Feb 21, 2013 at 4:47 PM, James Swaine <
james.swa...@eecs.northwestern.edu> wrote:
> I think the problem (in addition to the ones I mentioned in my previous
> message) is the use of let*, etc. inside your future thunk.
>
This seems suspicious to me. Probably if that is really the case, w
Never mind, it was from running it in DrRacket. From the command line,
it doesn't block, so it parallelizes a bit, but it syncs a lot on
allocations. The code:
http://pastebin.com/dEE1JfXF
It uses two futures because that's how my randomized search for
something that worked turned out. I'm
After converting everything to Typed Racket and using no `for' macros,
I'm also getting a block on "prim_indirect" in the dummy renderer. It
would be nice to know which primitive it's referring to, especially
since I can't find anything but unsafe operations in the expanded code.
Neil ⊥
On 02
Sam, hello.
On 2013 Feb 21, at 22:56, Sam Tobin-Hochstadt wrote:
> Thinking about it further, it's possible to enforce this invariant
> statically. Unfortunately TR isn't smart enough to let you express
> this in the nicest way possible, but if you look at the above gist
> now, it shows an exa
On Thu, Feb 21, 2013 at 5:47 PM, Sam Tobin-Hochstadt wrote:
> On Thu, Feb 21, 2013 at 5:28 PM, Norman Gray wrote:
>>
>>
>> But I'm now quite securely stuck, I think.
>>
>>
>>
>> In case anyone is curious (and in case the clear and obvious solution is
>> indeed clear and obvious), what I'm t
Just an update.
I've improved[1] a code a bit and it seems that values primitive has went
away: http://ompldr.org/vaGpweQ
but now it's [prim-indirect] stuff that blocks everything.
[1] code: http://pastebin.com/csst2NCy
--
With best regards,
Dmitry
Racket Users list:
ht
I think the problem (in addition to the ones I mentioned in my previous
message) is the use of let*, etc. inside your future thunk. These were
expanding to define-values, which seemed to be the source of the `values'
barricading. If you rewrite `render-scene-dummy' like so:
(define (render-scene
On Thu, Feb 21, 2013 at 5:28 PM, Norman Gray wrote:
>
>
> But I'm now quite securely stuck, I think.
>
>
>
> In case anyone is curious (and in case the clear and obvious solution is
> indeed clear and obvious), what I'm trying to do is produce a typed version
> of the procedure below:
I th
Hi James,
> sph-list is a list, and you are using for/fold to iterate over the
> elements in the list (inside a future). This is problematic for a number of
> reasons;
>
could you pls elaborate on that? why there would be any performance
implications on iterating over immutable list inside the f
Greetings, again.
On 2013 Feb 21, at 19:34, Norman Gray wrote:
> It might also be worth looking at the docs for 'lambda:', since I can't see
> any way of adding keywords to an anonymous function.
...and following up with tangent to my own question, it appears that
make-keyword-procedure and
I'm looking into this now, but there are a few modifications you can make
to eliminate some blocking calls, namely:
sph-list is a list, and you are using for/fold to iterate over the elements
in the list (inside a future). This is problematic for a number of reasons;
instead, you could convert it
At Fri, 22 Feb 2013 00:50:27 +0300, Dmitry Cherkassov wrote:
> I don't use ``pair?'' though.
>
> Is there a simple method to find out what function calls (or macroexpands
> to) it?
You could try the macro stepper or `raco expand' to see expanded code.
You might even resort to `raco decompile' to
Hi.
> Well, just a quick look at the screenshot: I think pair? will block only
when it used in a higher-order way, ie:
Make sense, thanks.
I don't use ``pair?'' though.
Is there a simple method to find out what function calls (or macroexpands
to) it?
--
With best regards,
Dmitry
__
Ignore this, it was due to an errant redefinition of FigureTarget in my
style.tex
\renewcommand{\FigureTarget}[1]{}
On 02/21/2013 01:28 PM, Jon Rafkind wrote:
> Scribble is producing a figure that contains _ in the name without preceding
> \ characters. I will debug the scribble source unless s
Well, just a quick look at the screenshot: I think pair? will block only
when it used in a higher-order way, ie:
(define (f g) (g 17))
(f pair?)
instead of:
(pair? 17)
or, for that matter:
(define (f g) (g 17))
(g (lambda (x) (pair? x)))
Robby
On Thu, Feb 21, 2013 at 3:03 PM, Dmit
Hi list.
I've been doing a simple ray tracer[1] and decided to parallelize it using
futures.
I've tried to use flonums everywhere and added (in-range) to loops [3]
(over x and y coordinates).
The problem is that execution of future is blocked seriously.
(apparently by the value primitive) [2]
Ar
Scribble is producing a figure that contains _ in the name without preceding \
characters. I will debug the scribble source unless someone has a better idea.
\Centertext{\Legend{\FigureTarget{\label{t:x28counter_x28x22figurex22_x22enforestx2dmodelx22x29x29}Figure~1:
}{t:x28counter_x28x22figu
Vincent, hello.
On 2013 Feb 21, at 15:20, Vincent St-Amour wrote:
> I agree, the docs could be a lot clearer.
>
> I'll add examples that use keyword arguments.
It might also be worth looking at the docs for 'lambda:', since I can't see any
way of adding keywords to an anonymous function. I
Specifically, Matthew Flatt's RacketCon presentation:
http://news.ycombinator.com/item?id=5256999
Racket Users list:
http://lists.racket-lang.org/users
Vincent, hello.
On 2013 Feb 21, at 15:20, Vincent St-Amour wrote:
> I agree, the docs could be a lot clearer.
>
> I'll add examples that use keyword arguments.
Great -- thanks.
Norman
--
Norman Gray : http://nxg.me.uk
SUPA School of Physics and Astronomy, University of Glasgow, UK
I agree, the docs could be a lot clearer.
I'll add examples that use keyword arguments.
Thanks for the report!
Vincent
At Thu, 21 Feb 2013 09:18:39 +,
Norman Gray wrote:
>
>
> Neil, hello.
>
> On 2013 Feb 21, at 03:49, Neil Toronto wrote:
>
> > Well, you can't do it like that, but yo
At Wed, 20 Feb 2013 20:49:02 -0700,
Neil Toronto wrote:
>
> Well, you can't do it like that, but you can like this:
>
> #lang typed/racket
>
> (: test0 (Integer [#:zero Integer] -> Boolean))
> (define (test0 x #:zero [v 0])
>(= x v))
>
> I think the `Keyword' type is for symbols that happen
Neil, hello.
On 2013 Feb 21, at 03:49, Neil Toronto wrote:
> Well, you can't do it like that, but you can like this:
>
> #lang typed/racket
>
> (: test0 (Integer [#:zero Integer] -> Boolean))
> (define (test0 x #:zero [v 0])
> (= x v))
Aha -- that's certainly a neater way of doing it. Than
24 matches
Mail list logo
