[Patch] [Aarch64] PR rtl-optimization/87763 - this patch fixes gcc.target/aarch64/lsl_asr_sbfiz.c

Segher Boessenkool segher@kernel.crashing.org
Fri Apr 26 23:59:00 GMT 2019


On Fri, Apr 26, 2019 at 03:54:49PM -0600, Jeff Law wrote:
> > It makes most sense if the mode for extv<mode> is the same both in and out
> > (it has only one mode in the pattern name, to start with), and for
> > sign_extract to be similar.  The docs aren't quite clear, but defining it
> > to have multiple modes doesn't really solve anything afaics, subregs work
> > just fine here?
> You'd have to scatter them in the MD file.  That's generally frowned upon.

Why would you need that?  You just say  match_operand:M "register_operand"
and that will match a  subreg:M (reg:N)  just fine?

And on the outside...  Do you have insns that act on DImode regs but only
output to a SImode one?!

> My argument is that we have very clear semantics here.  We're taking a
> field from an object in a mode.  We zero or sign extract it to the mode
> of the destination.   Two modes actually make reasonable sense here.
> 
> Think of it like zero_extend or sign_extend, but for a bitfield.

The pattern with the standard name extv<mode> can only have one mode
obviously...  Do you want to define a new standard one (and handle it
everywhere you need to handle it)?  Or just have the aarch64 port define
its own non-standard pattern?  Or make extv<mode> have semantics that
are different per target, like extv?


Segher



More information about the Gcc-patches mailing list