[Mono-dev] Debugging Native Client (was Re: more support for Google Native Client)

[Elijah Taylor]

Hi,

I've not seen anything like that myself, but I'll admit we haven't run NaCl
Mono through many production level environments yet, so there may be
something there that's been previously undetected.

However, we've run embedded in Unity3D (their entire scripting engine uses
Mono) and only had to modify some scripts to get rid of null exceptions and
floating point errors (div by zero).  I've also personally run gmcs.exe and
the output from that compiler all embedded in a NaCl module (and have done
the same thing with Microsoft's F# compiler).  And we pass all the tests in
mono/mini/'*.cs that we are designed to pass (minus hardware exceptions,
really).

Because we've run successfully in these various environments, I would still
not discount possible bugs in your code that could be overwriting memory.

Can you provide more info?  Are you using my github fork, or are you using
Mono 2.10 with my scripts? NaCl changes merged to 2.10 or HEAD might be less
stable than elijahtaylor/mono on github currently.  I'm assuming you're
embedding mono in your executable, is that the case?  If so, can you
possibly send me your driver code that sets up the environment and runs
things?  What architecture are you running (32- or 64-bit), and what OS?
Are you running your nexe in sel_ldr or as a PPAPI plugin inside of Chrome?
What do your linker options look like?  Sorry for the deluge of questions :)

-Elijah



On Sat, Apr 2, 2011 at 9:24 AM, C.Rivlaldo <vladimir at neoaxisgroup.com>wrote:

> Hello!
>
> Firstly I must say big thanks to Elijah again. I had successfully run a
> part
> of our C# code with Mono on NaCl.
>
> But occasionally I'm finding some bugs with global variables in Mono, which
> value is equal 0xffffffff. For example:
> * assemly_search_hook in assembly.c,
> * global_codeman in mini.c.
>
> I think NaCl corrupts the values of variables during the initialization or
> during running application.
>
> May be somebody had this problem before?
>
> --
> View this message in context:
> http://mono.1490590.n4.nabble.com/PATCH-more-support-for-Google-Native-Client-tp3159583p3422375.html
> Sent from the Mono - Dev mailing list archive at Nabble.com.
> _______________________________________________
> Mono-devel-list mailing list
> Mono-devel-list at lists.ximian.com
> http://lists.ximian.com/mailman/listinfo/mono-devel-list
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ximian.com/pipermail/mono-devel-list/attachments/20110404/ed8e9b1a/attachment.html