[PATCH 0/2] extend opindex and fx_pcrel_adjust to 16 bits

[Alan Modra]

On Wed, May 25, 2022 at 12:09:15AM +0300, Dmitry Selyutin wrote:
> Hello all, just in case the previous message was missed, I'd like to ping
> about the patches again.

Sorry, I haven't been very active over the last few weeks.  I'm about
to apply the ppc_opindex_t patches, squashed into one, and your isel
patch.  I've made some comment changes, plus gone back to masking
fx_pcrel_adjust, the reason being that if someone needed to reduce
fx_pcrel_adjust to, say, 12 bits to accomodate some new struct fix
field, they could do so with only a change to PPC_OPINDEX_MAX in the
ppc target code.  A cast to ppc_opindex_t wouldn't work in that
situation I believe, you'd get the 12-bit signed value being extended
to an int, then masked to 16 bits by the cast.

-- 
Alan Modra
Australia Development Lab, IBM