[PATCH] MIPS EVA ASE Support

[Maciej W. Rozycki]

On Mon, 10 Jun 2013, Richard Sandiford wrote:

> >> Index: opcodes/micromips-opc.c
> >> ===================================================================
> >> --- opcodes/micromips-opc.c	2013-06-08 10:18:33.894842596 +0100
> >> +++ opcodes/micromips-opc.c	2013-06-08 10:18:44.227957964 +0100
> >> @@ -115,7 +115,7 @@ const struct mips_opcode micromips_opcod
> >>  /* These instructions appear first so that the disassembler will find
> >>     them first.  The assemblers uses a hash table based on the
> >>     instruction name anyhow.  */
> >> -/* name,    args,	match,      mask,	pinfo,			pinfo2,		membership,	[exclusions] */
> >> +/* name,    args,	match,      mask,	pinfo,			pinfo2,		membership,	[ase],	[exclusions] */
> >
> >  I think this should be:
> >
> > /* name,    args,	match,      mask,	pinfo,			pinfo2,		membership[[,	ase],	exclusions] */
> >
> > or suchlike as the use of "exclusions" requires "ase" to have been set 
> > too (possibly to 0).  Likewise in opcodes/mips-opc.c.
> 
> Don't you mean:
> 
>   [,	ase[,	exclusions]]

 D'oh, yes, of course, sorry -- what was I thinking?

> ?  But the original seems clearer to me.

 Not at all to me.  Note that a trailing comma in an aggregate type 
initialiser is accepted by GCC as a valid C syntax (I reckon it was also 
standardised in ISO C99).  So the previous description was syntactically 
correct even though no entries were updated to include the comma.  Now two 
consecutive commas with no intervening field initialiser are not accepted, 
so the new description is plain wrong.

 With the above in mind please feel free to express it as:

,     [ase,  [exclusions]]

though if you prefer.

  Maciej