>From the discussion on gcc-patches - I wonder how far should the port go
here to paper over basic brokenness in gcc to handle multi-hard-register
pointers.
On 25 October 2013 13:35, Joern Rennecke wrote:
> Well, it's not really a lie if you map hardware registers 22 and 23 to
> a single registe
On 24 October 2013 07:59, DJ Delorie wrote:
> In the past, if I use a register class that excludes the second half
> of register pairs, it can't do anything because it requires both parts
> of the register pair to be in the class (example: in_hard_reg_set_p
> checks this).
>
> Which way is the "r
The docs say to use HARD_REGNO_MODE_OK to enforce register pairs.
But reload (find_valid_class_1) rejects classes that include such
registers:
for (regno = 0; regno < FIRST_PSEUDO_REGISTER && !bad; regno++)
{
if (in_hard_reg_set_p (reg_class_contents[rclass], mode, regno)
