Subj : Re: enigma on M68K linux
To   : vorlon
From : tenser
Date : Fri Jan 27 2023 03:40 am

On 26 Jan 2023 at 04:31p, vorlon pondered and said...
 
 vo>  > Ah. Too bad. I wonder how hard it would be to get it
 vo>  > to compile.
 vo> 
 vo> About this hard (This is the show stopper):
 vo> 
 vo> [snip]
 vo> jsutil.h:120:81: error: size -1 of array js_static_assert61 is negative  
 vo> 120 |         typedef int JS_STATIC_ASSERT_GLUE(js_static_assert,
 vo> __COUNTER__)[(cond) ? 1 : -1]    

Clever; they're using the creation of an array of
negative length as a means to force a error if some
constant condition is not true at compile-time.

 vo> 425 | JS_STATIC_ASSERT(sizeof(StackSegment) % sizeof(Value) == 0);

What this is saying is that the size of `StackSegment`
is not a multiple of the size of `Value`.  Let's see
what those things are.  `StackSegment` is a class that
appears to contain mostly a bunch of pointers; its
definition indicates it should be a multiple of 8-bytes
in length, and one sees that there's a preprocessor
conditional to add a padding word on 32-bit platforms
(like 68k); see line 249 of `jscntxt.h`.  I think the
first thing I'd check is whether `JS_BITS_PER_WORD`
is properly set to 32 in your build.

--- Mystic BBS v1.12 A48 (Linux/64)
 * Origin: Agency BBS | Dunedin, New Zealand | agency.bbs.nz (21:1/101)

.