Hi Akash

I'm very pleased to say that I managed to fix http://ascend4.org/b564. You can update your 'trunk' working copy to get the fixes. It was all due to a cast from long to int (which doesn't do anything on 32-bit systems, but can cause data loss on 64-bit systems). There is definitely some cleaning up that could be done in the code around root-finding (ascend/compiler/relation_util.c and surrounds).

Anyway, hopefully this means your 64-bit problems are gone, at least for the moment. Please let me know if you see any further issues.

Cheers
JP

On 27/03/15 00:39, Akash Deep Singhal wrote:
Thank you for the mini-driver and correcting my syntax. (Updated the repo <https://github.com/akigupta131/ascend-steam-reformation/commit/1a0de9f3e508caadeb91c7c12ba793d55f4ed0df>)

Unfortunately while I am trying to run that as you suggested, I am getting:-

PROGRAM ERROR: ascend/solver/solver.c:138:slv_lookup_client: Invalid engine name 'QRSlv' runqrslv: ascend/runqrslv.c:110: main: Assertion `qrslv_index != -1' failed.
Aborted (core dumped)

Also looking at the bug http://ascend4.org/b564, It's quite possible that maybe the case. I am using Ubuntu 14.04 LTS on 64 bit system. Good luck with the bug. I'm also trying to find some solution.

I'm looking for a 32 bit system. I'll soon start working on it. And get back to you in case of any progress.

Thanks for all the help :)

-Akash

ᐧ

On Thu, Mar 26, 2015 at 1:02 PM, John Pye <[email protected] <mailto:[email protected]>> wrote:

    Hi Akash

    An update on this bug. I found that it appears on my Ubuntu 12.04
    64-bit machine, but *not* on my Ubuntu 12.04 32-bit machine. So
    it's possible that this could be related to one of our most
    hard-to-solve (and very intermittent) bugs,
    http://ascend4.org/b564. This example model of yours is an even
    simpler test case than the last one, so I am optimistic that we my
    be able to track down this bug at last!

    As well as not crashing via the simplified 'runqrslv' driver, your
    model also now works fine under 32-bit in the PyGTK GUI. So it's
    clearly not a GUI-related problem.

    As far as development of your models goes, I can therefore only
    suggest for the moment that you try to do your development on a
    32-bit version of Linux (eg in a VM if necessary). Hope that's not
    too inconvenient to do that. I'll be doing my best to fix this
    tricky bug as soon as possible.

    Cheers
    JP


------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Ascend-sim-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ascend-sim-users

Reply via email to