Re: [PATCH v4, rs6000] Add a combine pattern for CA minus one [PR95737]

["Kewen.Lin" <linkw@linux.ibm.com>]

Hi Haochen,

on 2022/5/13 09:07, HAO CHEN GUI wrote:
> Hi,
>    This patch adds a combine pattern for "CA minus one". As CA only has two
> values (0 or 1), we could convert following pattern
>       (sign_extend:DI (plus:SI (reg:SI 98 ca)
>                 (const_int -1 [0xffffffffffffffff]))))
> to
>        (plus:DI (reg:DI 98 ca)
>             (const_int -1 [0xffffffffffffffff])))
>    With this patch, one unnecessary sign extend is eliminated.
> 
>    Bootstrapped and tested on powerpc64-linux BE and LE with no regressions.
> Is this okay for trunk? Any recommendations? Thanks a lot.
> 
> ChangeLog
> 2022-05-13 Haochen Gui <guihaoc@linux.ibm.com>
> 
> gcc/
> 	PR target/95737
> 	* config/rs6000/rs6000.md (extenddi_ca_minus_one): Define.
> 

Nit: (*extenddi_ca_minus_one): New define_insn_and_split.

> gcc/testsuite/
> 	PR target/95737
> 	* gcc.target/powerpc/pr95737.c: New.
> 
> patch.diff
> diff --git a/gcc/config/rs6000/rs6000.md b/gcc/config/rs6000/rs6000.md
> index 64049a6e521..483a93956f8 100644
> --- a/gcc/config/rs6000/rs6000.md
> +++ b/gcc/config/rs6000/rs6000.md
> @@ -2353,6 +2353,19 @@ (define_insn "subf<mode>3_carry_in_xx"
>    "subfe %0,%0,%0"
>    [(set_attr "type" "add")])
> 
> +(define_insn_and_split "*extenddi_ca_minus_one"
> +  [(set (match_operand:DI 0 "gpc_reg_operand" "=r")
> +	(sign_extend:DI (plus:SI (reg:SI CA_REGNO)
> +				 (const_int -1))))]
> +  ""
> +  "#"
> +  ""
> +  [(parallel [(set (match_dup 0)
> +		   (plus:DI (reg:DI CA_REGNO)
> +			    (const_int -1)))
> +	      (clobber (reg:DI CA_REGNO))])]
> +  ""
> +)
> 
>  (define_insn "@neg<mode>2"
>    [(set (match_operand:GPR 0 "gpc_reg_operand" "=r")
> diff --git a/gcc/testsuite/gcc.target/powerpc/pr95737.c b/gcc/testsuite/gcc.target/powerpc/pr95737.c
> new file mode 100644
> index 00000000000..d4d6a4198cf
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/powerpc/pr95737.c
> @@ -0,0 +1,10 @@
> +/* PR target/95737 */
> +/* { dg-do compile { target lp64 } } */> +/* { dg-options "-O2 -mno-isel" } */

Nit: It seems good to put one comment for the reason why we need this
special -mno-isel, like something: Power9 leverages isel for this case,
force -mno-isel to keep the test point valid on Power9 and later."

> +/* { dg-final { scan-assembler-not {\mextsw\M} } } */
> +
> +
> +unsigned long negativeLessThan (unsigned long a, unsigned long b)
> +{
> +   return -(a < b);
> +}

OK with those nits fixed, but please wait for a few days, just in case
Segher/David have more comments. 

BR,
Kewen