On Thu, Oct 28, 2010 at 8:54 PM, Dr. David Kirkby
<david.kir...@onetel.net>wrote:

> I'm having a hard time on Solaris and OpenSolaris building the libecl.so
> library
> without text relocations against non-writable segments. This basically
> makes the
> ECL library unusable on 64-bit Solaris/OpenSolaris.
>

I know. I am slowly progressing through the list of bug reports.


> Hence I'm trying to build the latest git snapshot (downloaded around 1900
> GMT on
> 28th October 2010) using the Sun compiler. But the Sun compiler will not
> compile
> this - it is failing with:
>
> Unknown symbol: mp::interrupt-process
>

You have hit some ugly point between commits. Bad luck :-)


> line 97: #error: "We cannot use the mmap code without siginfo"


mmap is used in ECL to help in delaying interrupts. The trick is simple: ECL
gets an interrupt while executing a C function that can not be interrupted,
it marks (using mmap) the Lisp environment as read-only, ECL tries to write
to the environment after the C function finished, this causes a signal and
now it can be processed together with the pending interrupts. But this only
works when mmap implements interrupts with information.

I am really surprised that the Sun platform does not support SA_SIGINFO :-/
Would one need additional compiler flags to allow the headers provide this
facility?

Juanjo

-- 
Instituto de FĂ­sica Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest
Create new apps & games for the Nokia N8 for consumers in  U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store 
http://p.sf.net/sfu/nokia-dev2dev
_______________________________________________
Ecls-list mailing list
Ecls-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ecls-list

Reply via email to