subreg:HI of PSI HW register issue

[Bernd Schmidt]

On 03/09/2012 11:20 AM, Aurelien Buhrig wrote:
> Hi,
> 
> It seems there is an issue around subreg:HI of PSI hardware register,
> which occurs either during expand or reload (GCC 4.6.1).
> 
> For my big endian target,
> (subreg:HI (reg:PSI A0_REGNO) 0) is not representable but
> (subreg:HI (reg:PSI A0_REGNO) 2) is (reg:HI A0_REGNO).

> - Is it correct that gcc emits such a subreg pattern in Pmode=PSI during
> expand ? Or should it be in ptr_mode=SImode (in this case, both
> (subreg:HI (reg:SI) 0/2) are representable)?

This, I think. I have a port with a rather similar situation, and I'm
betting you also have a failure in gcc.c-torture/execute/20040625-1.c. I
don't want to promise anything but I may have something for you next
week-ish.


Bernd