Re: Followup on PR/109279: large constants on RISCV

[Jeff Law <jeffreyalaw@gmail.com>]

On 6/12/23 13:32, Vineet Gupta wrote:

> 
> Gave this a try and it seems to fix Andrew's test, but then regresses 
> the actual large const case: 0x1010101_01010101 : the mem to const_int 
> transformation was being done in cse1 which no longer happens and the 
> const pool from initial expand remains all the way into asm generated. I 
> don't think we want to go back to that state
Was that with just -O1/-Og?  I have vague memories of seeing that kind 
of behavior.  If it was just with -O1 that might be a reasonable tradeoff.


> 
> Ok. From reading gccint it seems REG_EQUIV is a stronger form of 
> equivalence and seems to be prefered by post reload passes, while 
> REG_EQUAL is more of use in pre-reload.
That's a reasonable way to look at it.  In fact it's the allocators that 
promote REG_EQUAL to REG_EQUIV when its's safe to do so.

> 
> 
>>   I would also look at reload_cse_regs which should give us some 
>> chance at seeing the value reuse if/when IRA/LRA muck things up.
> 
> I'll be out of office for the rest of week, will look into this once I'm 
> back.
NP.

jeff