Re: [patch] MIPS16e support in simulator.

[David Ung <davidu@mips.com>]

On Mon, 2005-06-13 at 16:22 -0700, Eric Christopher wrote:
> On Mon, 2005-06-13 at 13:03 -0400, Daniel Jacobowitz wrote:
> > On Mon, Jun 13, 2005 at 10:02:45AM -0700, Eric Christopher wrote:
> > > 
> > > > So um... unfortunately, GDB does not have an active MIPS maintainer at
> > > > the moment.  I do not see anything (other than your followup about the
> > > > mipsisa32 configure stanza) obviously wrong with these changes, but I
> > > > can't provide any useful review of the mips-specific bits.  I don't
> > > > suppose someone on binutils@ could lend a hand?
> > > > 
> > > 
> > > I will, url to the original patch?
> > 
> > http://sourceware.org/ml/gdb-patches/2005-06/msg00078.html
> > http://sourceware.org/ml/gdb-patches/2005-06/msg00079.html
> 
> Looks like the jump instructions aren't executing the instructions in
> the delay slot? See the definition of jalr in m16.igen.

This is correct.  The MIPS16e jrc, jalrc etc does not have a delay slot.

> 
> I guess a good question would be asking how this patch was tested?
> 
> Otherwise it looks fine.

We've actually had MIPS16e simulator support for quite awhile at MIPS
for our internal use.  We've been doing nightly regressions test on
MIPS16e for gcc 3.4 and gdb for more that half year now.

David.