Re: Issues with compiler failures

[maya%netbsd.org@localhost]

I accidentally came across the same failure while working on GCC.
It's saying "we are seeing a pseudo-register in post-reload".

It's possible that the PSW needs to be considered a hard register, but I
am not sure yet!

adding a printf("%d\n",regno); will probably result in printing 16 just
before the failure, by this guess.

Please try -save-temps too because it should be possible to reproduce
this bug with the cross-compiler too (and then have faster testing for
changes).

That should produce a blake2b.i file and running gcc -O blake2b.i will
fail too. But it's standalone.