On Sat, 2025-07-05 at 14:10 -0500, Segher Boessenkool wrote:
> Hi!
> 
> On Sat, Jul 05, 2025 at 11:10:05PM +0800, Xi Ruoyao wrote:
> > Possibly this is https://gcc.gnu.org/PR101882.  Specifically comment 5
> > from Segher:
> > 
> > "The LRA change is correct AFAICS.  But combine makes a change that
> > violates the earlyclobber...  I need to do something about that, too."
> 
> See my comment in that PR.  The loongarch machine description explicitly
> allows operand 1 to be the same as operand 0.  combine has notyhing to
> do with this, the pattern it ended up with is recognised by recog(),
> exactly as it should be (according to the pattern here, is that
> "<optab>_alsl_reversesi_extended"?)

We don't allow operand **4** to be same as operand 0 here, but combine
makes them same :(.


-- 
Xi Ruoyao <xry...@xry111.site>

Reply via email to