Re: [PATCH] Improve EXPAND_SUM handling in expand_expr_addr_expr* (PR middle-end/50074)

2011-11-28 Thread Richard Henderson
On 11/25/2011 12:58 PM, Jakub Jelinek wrote: PR middle-end/50074 * expr.c (expand_expr_addr_expr_1): Don't call force_operand for EXPAND_SUM modifier. Ok. r~

[PATCH] Improve EXPAND_SUM handling in expand_expr_addr_expr* (PR middle-end/50074)

2011-11-25 Thread Jakub Jelinek
Hi! While looking at this PR, I was first surprised that on i?86 we got pseudo = argp + 4 and mem_overlap* was called with that pseudo + 4 etc. I don't see why we should force the address into register for EXPAND_SUM modifier, with this mem_overlap* sees argp + 8 etc. directly (on i?86, of