Re: [Chicken-hackers] CHICKEN in production

2014-10-10 Thread Felix Winkelmann
Ah, no, I meant store just the index in mmapped memory so it doesn't take up heap space. This does mean it gets treated as just a blob of data, and you need to write your own routines to access it. Some gc's allow moving objects out of the heap or otherwise marking them non-movable, we

Re: [Chicken-hackers] CHICKEN in production

2014-10-10 Thread Jörg F. Wittenberger
I don't think that I ever wrote that using memory debuggers like valgrind would work out of the box with chicken to any usable extend. To the contrary: when I did so, I posted fixes to runtime.c the mailing list to make this feasible. /Jerry Am 08.10.2014 um 23:17 schrieb Oleg Kolosov: We

Re: [Chicken-hackers] CHICKEN in production

2014-10-10 Thread Jörg F. Wittenberger
Am 08.10.2014 um 21:27 schrieb John Cowan: Peter Bex scripsit: Thanks for speaking up, I hadn't considered how big of a problem this would be for large, existing projects. We could possibly make it a runtime option, so the checking (or just the enforcing) could be enabled or disabled at will.

Re: [Chicken-hackers] CHICKEN in production

2014-10-10 Thread Christian Kellermann
Jörg F. Wittenberger joerg.wittenber...@softeyes.net writes: Since chicken 5 is already there (and I'm depending on the fixes to the current version to run our code at all, which forces more work to set up a compile your modified chicken build for the time being), I rather try porting those

Re: [Chicken-hackers] CHICKEN in production

2014-10-10 Thread Jörg F. Wittenberger
Am 10.10.2014 um 11:52 schrieb Christian Kellermann: Jörg F. Wittenberger joerg.wittenber...@softeyes.net writes: Since chicken 5 is already there (and I'm depending on the fixes to the current version to run our code at all, which forces more work to set up a compile your modified chicken

Re: [Chicken-hackers] CHICKEN in production

2014-10-10 Thread Christian Kellermann
Jörg F. Wittenberger joerg.wittenber...@softeyes.net writes: It's about 70KLOC of code (roughly doing what Ethereum will do once the latter works, just about 20 times a fast [if Ethereum switches to 12seconds per block as they are planning, otherwise it's even faster]) we are talking about.

Re: [Chicken-hackers] CHICKEN in production

2014-10-10 Thread Jörg F. Wittenberger
They are not *that* huge. Just the downside of what's usually a strength of chicken one C source file plays in badly. Let's take that to private mail for the time, not to spam the list. Am 10.10.2014 um 12:20 schrieb Christian Kellermann: Jörg F. Wittenberger joerg.wittenber...@softeyes.net

Re: [Chicken-hackers] CHICKEN in production

2014-10-10 Thread Alex Shinn
On Fri, Oct 10, 2014 at 5:34 PM, Felix Winkelmann felix.winkelm...@bevuta.com wrote: Ah, no, I meant store just the index in mmapped memory so it doesn't take up heap space. This does mean it gets treated as just a blob of data, and you need to write your own routines to access it.