Gambit-list Avril 2013

gambit-list@iro.umontreal.ca

21 participants
51 discussions

Termite
by Ralph Möritz 28 Avr '13

28 Avr '13

Hi, I just skimmed through the Termite paper - it looks awesome. Looking at the commit log of the repo, however, reveals the code hasn't been touched since 2009. Yikes! What a pity, it seems like such an interesting project. Sorry if this seems like a pointless e-mail. Cheers, Ralph

5 4

Two questions about GambitREPL & iOS
by Jason Felice 26 Avr '13

26 Avr '13

1. Is there a reason (something to be wary of) why gambit is set up in the view controller rather than in main()? 2. Is the ___enable_interrupts()/___disable_interrupts() currently only used for communicating events to gambit? I guess what I'm wondering, in other words, is whether there would be some kind of problem with running gambit on the main thread, invoking (c-define ...) functions, if gambit will never wait on a port? -Jason

1 0

decreasing GC
by Zhen Shen 24 Avr '13

24 Avr '13

Hi I have this simple numerical algorithm that computes an integral. I compile it with -prelude "(declare (standard-bindings) (block) (not safe))" -cc-options "-O4". All functions are tail-call optimized. The very last line of generated by ##gc-report-set! is *** GC: 0 ms, 18.7G alloc, 1.52M heap, 45.3K live (3% 22896+23496) The live area is consistently 45K throughout the entire execution. Timing report indicates that 18% is spent on GC: 7904 ms real time 7784 ms cpu time (3684 user, 4100 system) 13048 collections accounting for 1404 ms real time (584 user, 724 system) 20078526600 bytes allocated 480 minor faults no major faults Besides tail-calls, what else can i do to reduce GC?

6 11

iOS Embedding -- Don't see errors / exceptions
by Axis Sivitz 24 Avr '13

24 Avr '13

Errors thrown from scheme code embedded in an iOS app don't get displayed anywhere. Instead, the the program just hangs. I see the same behavior in the example GambitREPL app. If I add in an error (for example, calling (cons 1), wrong is the number of arguments) to the program.scm file, it will hang on launch, and the error message won't be displayed in stdout (which is ideally where I'd like it to go). I'm using the following debug_settings: debug_settings = (debug_settings & ~___DEBUG_SETTINGS_UNCAUGHT_MASK) | (___DEBUG_SETTINGS_UNCAUGHT_ALL << ___DEBUG_SETTINGS_UNCAUGHT_SHIFT); debug_settings = (debug_settings & ~___DEBUG_SETTINGS_ERROR_MASK) | (___DEBUG_SETTINGS_ERROR_REPL << ___DEBUG_SETTINGS_ERROR_SHIFT); but I've also tried debug_settings = (debug_settings & ~___DEBUG_SETTINGS_REPL_MASK) | (___DEBUG_SETTINGS_REPL_STDIO << ___DEBUG_SETTINGS_REPL_SHIFT); and I've also tried wrapping my scheme calls in ___BEGIN_TRY ___END_TRY I apologize for all the questions regarding setting up an iOS/Scheme development environment, but I'll be sure to document everything once I get it all working to help other folks in the future. Thanks, Axis

2 2

Re: [gambit-list] Gambit-list Digest, Vol 103, Issue 27
by Estevo 23 Avr '13

23 Avr '13

> All of these issues can be abstracted in a macro for defining foreign > structures. This is explained in the following message: > > https://mercure.iro.umontreal.ca/pipermail/gambit-list/2009-May/003475.html > > Marc Geez, I so wish I'd found that one about one year ago, when I was trying to do the same thing! :) My best effort attempt (don't use it! it never got out of playground/experimental stage) was here: https://github.com/euccastro/gambit-SDL2/blob/master/ffi-macro.scm That one also tries to overcome the following issue, which Marc's macro (perhaps wisely) leaves alone: ;; IMPORTANT NOTE: the foreign pointer returned by s2-coordinates-ref ;; is a pointer to a C struct within another C struct (the ;; "container"). The foreign pointer is only valid as long as the ;; container is allocated. Be wary that in the case of a C structure ;; allocated from Scheme, such as with a call to alloc-s2, the garbage ;; collector could reclaim the structure as soon as the foreign ;; pointer to it is dropped. I tried to solve that problem by keeping a table of references with weak keys corresponding to contained/child structs and strong values corresponding to container/parent values. The big problem was how to make the references table unique since it was defined in macro-defining code, i.e., which was meant to be included, not loaded. I suppose the cleaner solution would have been to make a separate ffi library to `load`. I don't remember why I didn't just do this. The hacky approach in that file (checking for the current type of the symbol) seemed to work, but I guess that relies on undocumented behavior. In retrospect, that would have been made so much easier if you could assign an arbitrary Scheme reference to a foreign pointer. I think that would address a genuine need: in general, ffi pointers may point to objects that 'depend' on other objects managed in the scheme world (probably other ffi pointers), in such a way that it makes sense to keep the dependencies alive while the pointer is in the scheme world. Something like ___set_data_rc, but you can't use this function since that messes with memory in the vicinity of the object pointed to; most often, this memory belongs to the container/parent struct. So it would need to be something like a Scheme call (set-foreign-reference! pointer dependency) that would set a ___SCMOBJ field in the foreign object itself. A getter is not needed for this use case, but maybe it would enable other uses (?). Alternatively, there could be a different type of foreign object that has this functionality, so applications that don't use it won't incur the space cost of the additional pointer.

1 0

Generic functions in 45 lines
by Jason Felice 22 Avr '13

22 Avr '13

Clearly, there could be more features and it could be more efficient, but in terms of producing the simplest thing that works for me, I'm really happy with this: https://gist.github.com/eraserhd/5435505 Usage: (define-generic (length object)) (add-method (length (vector? v)) (vector-length s)) (add-method (length (string? s)) (string-length s)) (add-method (length (list? l)) (##length l)) (length "hello") => 5 (length '(1 2 3)) => 3 (length '#(a b)) => 2

2 1

Re: [gambit-list] FFI C struct
by sales.creditscore＠creditscore.co.nz 22 Avr '13

22 Avr '13

Thanks Marc, In my poking about I did notice that post, but it uses a macro. H'mmm... at this early stage I'm afraid I'm still living in imperative land :-| I think there's a long way to go to reach macro heaven. Right now just making it out of imperative land into functional land is pretty tough. One day, perhaps, I'll make it through the pearly gates into macro heaven! Thanks again for helping me out. Cheers, Rob. ---- Marc Feeley <feeley(a)iro.umontreal.ca> wrote: > > As far as I can tell it is memory safe. However, it might be simpler to put your free_point function as a release function on the Sstruct* foreign type. That way the free_point function will be called by the Gambit garbage collector when the Scheme program no longer has a reference to a given structure. If it is possible for your C code to keep pointers to your structure, then it will be convenient to use ___alloc_rc instead of malloc, that way you can increment the reference count in that structure to prevent the garbage collector from freeing the structure. > > All of these issues can be abstracted in a macro for defining foreign structures. This is explained in the following message: > > https://mercure.iro.umontreal.ca/pipermail/gambit-list/2009-May/003475.html > > Marc >

1 0

FFI C struct
by sales.creditscore＠creditscore.co.nz 22 Avr '13

22 Avr '13

Hi, I've spent many many frustrating hours trying to talk to a C struct using the FFI. Looking around at other examples, there seem to be many ways of doing this..., but none of them worked for me. In the end I wrote a short example shown below that appears to work. It relies on all the heavy lifting being done in C and just simple wrappers being written in Scheme. As written, I believe it makes Scheme and C share the same struct, no copies being taken (this is crucial for me, as the C code will alter the C struct and likewise my Scheme code will alter it too). Before I go too far with this approach, I just want to check that it is memory safe. Here it is: ; ;Example where Gambit Scheme and C are sharing the SAME struct in memory using a pointer. ; (c-declare #<<end-of-c-declare typedef struct { int x; int y; } Cstruct; Cstruct* make_point( int x, int y ) { Cstruct* p = (Cstruct*) malloc(sizeof(Cstruct)); p->x = x; p->y = y; return p; } int point_x( Cstruct* p ) { return p->x; } int point_y( Cstruct* p ) { return p->y; } //if you create new memory using malloc you must free it again when you've finished to avoid a dangling pointer memory leak void free_point( Cstruct* p ){free(p);} end-of-c-declare ) (c-define-type Sstruct* (pointer "Cstruct")) (define make-point (c-lambda (int int) Sstruct* "make_point")) (define free-point (c-lambda (Sstruct*) void "free_point")) (define point-x (c-lambda (Sstruct*) int "point_x")) (define point-y (c-lambda (Sstruct*) int "point_y")) (define SchemeStruct (make-point 12345 6789)) (println "p->x in C struct holds the value " (point-x SchemeStruct)) (println "p->y in C struct holds the value " (point-y SchemeStruct)) (free-point SchemeStruct); included for completeness, but not needed because program is ending anyway in this example

2 1

gsc parameter "-check"
by Zhen Shen 19 Avr '13

19 Avr '13

I've noticed some sample projects invoke gsc with -check paramter, which is AFAIK not documented. What does it do?

1 0

embedding GSI
by Zhen Shen 19 Avr '13

19 Avr '13

are there any examples on how to embed GSI within a C/C++ program? in my case the C core must do some initialization before starting the GSI/REPL.

2 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Gambit-list Avril 2013