[Bug rtl-optimization/104914] [MIPS] wrong comparison with scrabbled int value
syq at gcc dot gnu.org
gcc-bugzilla@gcc.gnu.org
Fri Jul 7 04:21:52 GMT 2023
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104914
--- Comment #8 from YunQiang Su <syq at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #7)
> The initial RTL has a signed extend in there:
>
>
> (insn 20 19 23 2 (set (reg/v:DI 200 [ val+-4 ])
> (sign_extend:DI (subreg:SI (reg/v:DI 200 [ val+-4 ]) 4)))
> "/app/example.cpp":7:29 -1
> (nil))
> (jump_insn 23 20 24 2 (set (pc)
> (if_then_else (le (subreg/s/u:SI (reg/v:DI 200 [ val+-4 ]) 4)
> (const_int 0 [0]))
> (label_ref 32)
> (pc))) "/app/example.cpp":8:5 -1
> (int_list:REG_BR_PROB 440234148 (nil))
> -> 32)
>
>
> Before combine also looks fine:
> (insn 20 19 23 2 (set (reg/v:DI 200 [ val+-4 ])
> (sign_extend:DI (subreg:SI (reg/v:DI 200 [ val+-4 ]) 4)))
> "/app/example.cpp":7:29 235 {extendsidi2}
> (nil))
Yes. I noticed it. while in mips.md, extendsidi2 is expanded to no
instructions at all.
```
;; Extension insns.
;; Those for integer source operand are ordered widest source type first.
;; When TARGET_64BIT, all SImode integer and accumulator registers
;; should already be in sign-extended form (see TARGET_TRULY_NOOP_TRUNCATION
;; and truncdisi2). We can therefore get rid of register->register
;; instructions if we constrain the source to be in the same register as
;; the destination.
;;
;; Only the pre-reload scheduler sees the type of the register alternatives;
;; we split them into nothing before the post-reload scheduler runs.
;; These alternatives therefore have type "move" in order to reflect
;; what happens if the two pre-reload operands cannot be tied, and are
;; instead allocated two separate GPRs. We don't distinguish between
;; the GPR and LO cases because we don't usually know during pre-reload
;; scheduling whether an operand will be LO or not.
(define_insn_and_split "extendsidi2"
[(set (match_operand:DI 0 "register_operand" "=d,l,d")
(sign_extend:DI (match_operand:SI 1 "nonimmediate_operand" "0,0,m")))]
"TARGET_64BIT"
"@
#
#
lw\t%0,%1"
"&& reload_completed && register_operand (operands[1], VOIDmode)"
[(const_int 0)]
{
emit_note (NOTE_INSN_DELETED);
DONE;
}
[(set_attr "move_type" "move,move,load")
(set_attr "mode" "DI")])
```
More information about the Gcc-bugs
mailing list