Re: [Qemu-devel] [PATCH 1/3] sh: SE7750 board definition

[Shin-ichiro KAWASAKI]

Laurent Desnogues wrote:
> On Tue, Jan 13, 2009 at 3:32 AM, Nobuhiro Iwamatsu
> <address@hidden> wrote:
> > Because first of all, I am not so interested in the support of the board.
> > I am interested in emulation of CPU and the userland.
> > In the current situation, the number of supported real boards increases
> > when the support of CPU increases. The code of the board enters whenever
> > CPU is supported.
>
> I agree.  I was very surprised to find last Sunday that a simple instruction
> such as FMAC was not implemented.  I wonder how many holes are still
> there.

I've got anxious about implementation hole of instruction set.
Then I want to add a comment on it.

I reviewed target-sh4/translate.c and SH-4 (not SH-4A) instruction
set document again, and found that following four instructions are
not yet implemented.  (They will cause illegal instruction exceptions
on QEMU-SH.)

- STC SGR,Rn
- STC SRG,@-Rn
- FIPR
- FTRV  (Patch sent by Mans Rullgard before)

SGR is a register which holds stack pointer when exception happens.
But SH-Linux seems not to check SGR ever. These can be left safely
until other OS runs on QEMU-SH.

FIPR and FTRV provide floating point vector/matrix operations.
I have no idea which program or library runs these instructions...
Does anyone have it?


Regards,
Shin-ichiro KAWASAKI