From: will schmidt <will_schmidt@vnet.ibm.com>
To: Carl Love <cel@us.ibm.com>,
segher@gcc.gnu.org, dje.gcc@gmail.com, gcc-patches@gcc.gnu.org
Subject: Re: [PATCH 1/5] RS6000 Add 128-bit Binary Integer sign extend operations
Date: Wed, 07 Oct 2020 16:08:12 -0500 [thread overview]
Message-ID: <29f77863110b27d0f1bdc7123bc58811a43b2da4.camel@vnet.ibm.com> (raw)
In-Reply-To: <46d43691d5d946076439db905cd2d2c21b9cc8da.camel@us.ibm.com>
On Mon, 2020-10-05 at 11:51 -0700, Carl Love wrote:
> Will, Segher:
>
> Patch 1, adds the 128-bit sign extension instruction support and
> corresponding builtin support.
>
> I updated the change log per the comments from Will.
>
> Patch has been retested on Power 9 LE.
>
> Pet me know if it is ready to commit to mainline.
>
> Carl
>
> -----------------------------------------------------------
>
>
> gcc/ChangeLog
>
> 2020-10-05 Carl Love <cel@us.ibm.com>
> * config/rs6000/altivec.h (vec_signextll, vec_signexti): Add define
> for new builtins.
> * config/rs6000/rs6000-builtin.def (VSIGNEXTI, VSIGNEXTLL): Add
> overloaded builtin definitions.
> (VSIGNEXTSB2W, VSIGNEXTSH2W, VSIGNEXTSB2D, VSIGNEXTSH2D,VSIGNEXTSW2D):
> Add builtin expansions.
> * config/rs6000-call.c (P9V_BUILTIN_VEC_VSIGNEXTI,
> P9V_BUILTIN_VEC_VSIGNEXTLL): Add overloaded argument definitions.
> * config/rs6000/vsx.md: Make define_insn vsx_sign_extend_si_v2di
> visible.
> * doc/extend.texi: Add documentation for the vec_signexti and
> vec_signextll builtins.
>
> gcc/testsuite/ChangeLog
>
> 2020-10-05 Carl Love <cel@us.ibm.com>
> * gcc.target/powerpc/p9-sign_extend-runnable.c: New test case.
> ---
> gcc/config/rs6000/altivec.h | 3 +
> gcc/config/rs6000/rs6000-builtin.def | 9 ++
> gcc/config/rs6000/rs6000-call.c | 13 ++
> gcc/config/rs6000/vsx.md | 2 +-
> gcc/doc/extend.texi | 15 ++
> .../powerpc/p9-sign_extend-runnable.c | 128 ++++++++++++++++++
> 6 files changed, 169 insertions(+), 1 deletion(-)
> create mode 100644 gcc/testsuite/gcc.target/powerpc/p9-sign_extend-runnable.c
>
> diff --git a/gcc/config/rs6000/altivec.h b/gcc/config/rs6000/altivec.h
> index f7720d136c9..cfa5eda4cd5 100644
> --- a/gcc/config/rs6000/altivec.h
> +++ b/gcc/config/rs6000/altivec.h
> @@ -494,6 +494,9 @@
>
> #define vec_xlx __builtin_vec_vextulx
> #define vec_xrx __builtin_vec_vexturx
> +#define vec_signexti __builtin_vec_vsignexti
> +#define vec_signextll __builtin_vec_vsignextll
> +
> #endif
Can probably drop that blank line.
>
> /* Predicates.
> diff --git a/gcc/config/rs6000/rs6000-builtin.def b/gcc/config/rs6000/rs6000-builtin.def
> index e91a48ddf5f..4c2e9460949 100644
> --- a/gcc/config/rs6000/rs6000-builtin.def
> +++ b/gcc/config/rs6000/rs6000-builtin.def
> @@ -2715,6 +2715,8 @@ BU_P9V_OVERLOAD_1 (VPRTYBD, "vprtybd")
> BU_P9V_OVERLOAD_1 (VPRTYBQ, "vprtybq")
> BU_P9V_OVERLOAD_1 (VPRTYBW, "vprtybw")
> BU_P9V_OVERLOAD_1 (VPARITY_LSBB, "vparity_lsbb")
> +BU_P9V_OVERLOAD_1 (VSIGNEXTI, "vsignexti")
> +BU_P9V_OVERLOAD_1 (VSIGNEXTLL, "vsignextll")
>
> /* 2 argument functions added in ISA 3.0 (power9). */
> BU_P9_2 (CMPRB, "byte_in_range", CONST, cmprb)
> @@ -2726,6 +2728,13 @@ BU_P9_OVERLOAD_2 (CMPRB, "byte_in_range")
> BU_P9_OVERLOAD_2 (CMPRB2, "byte_in_either_range")
> BU_P9_OVERLOAD_2 (CMPEQB, "byte_in_set")
>
> +/* Sign extend builtins that work on ISA 3.0, but not defined until ISA 3.1. */
I have mixed feelings about straddling the ISA 3.0 and 3.1 ; but not
sure how to properly improve. (I defer).
The rest LGTM,
Thanks
-Will
> +BU_P9V_AV_1 (VSIGNEXTSB2W, "vsignextsb2w", CONST, vsx_sign_extend_qi_v4si)
> +BU_P9V_AV_1 (VSIGNEXTSH2W, "vsignextsh2w", CONST, vsx_sign_extend_hi_v4si)
> +BU_P9V_AV_1 (VSIGNEXTSB2D, "vsignextsb2d", CONST, vsx_sign_extend_qi_v2di)
> +BU_P9V_AV_1 (VSIGNEXTSH2D, "vsignextsh2d", CONST, vsx_sign_extend_hi_v2di)
> +BU_P9V_AV_1 (VSIGNEXTSW2D, "vsignextsw2d", CONST, vsx_sign_extend_si_v2di)
> +
> /* Builtins for scalar instructions added in ISA 3.1 (power10). */
> BU_P10_MISC_2 (CFUGED, "cfuged", CONST, cfuged)
> BU_P10_MISC_2 (CNTLZDM, "cntlzdm", CONST, cntlzdm)
> diff --git a/gcc/config/rs6000/rs6000-call.c b/gcc/config/rs6000/rs6000-call.c
> index a8b520834c7..9e514a01012 100644
> --- a/gcc/config/rs6000/rs6000-call.c
> +++ b/gcc/config/rs6000/rs6000-call.c
> @@ -5527,6 +5527,19 @@ const struct altivec_builtin_types altivec_overloaded_builtins[] = {
> RS6000_BTI_unsigned_V2DI, RS6000_BTI_unsigned_V2DI,
> RS6000_BTI_INTSI, RS6000_BTI_INTSI },
>
> + /* Sign extend builtins that work work on ISA 3.0, not added until ISA 3.1 */
> + { P9V_BUILTIN_VEC_VSIGNEXTI, P9V_BUILTIN_VSIGNEXTSB2W,
> + RS6000_BTI_V4SI, RS6000_BTI_V16QI, 0, 0 },
> + { P9V_BUILTIN_VEC_VSIGNEXTI, P9V_BUILTIN_VSIGNEXTSH2W,
> + RS6000_BTI_V4SI, RS6000_BTI_V8HI, 0, 0 },
> +
> + { P9V_BUILTIN_VEC_VSIGNEXTLL, P9V_BUILTIN_VSIGNEXTSB2D,
> + RS6000_BTI_V2DI, RS6000_BTI_V16QI, 0, 0 },
> + { P9V_BUILTIN_VEC_VSIGNEXTLL, P9V_BUILTIN_VSIGNEXTSH2D,
> + RS6000_BTI_V2DI, RS6000_BTI_V8HI, 0, 0 },
> + { P9V_BUILTIN_VEC_VSIGNEXTLL, P9V_BUILTIN_VSIGNEXTSW2D,
> + RS6000_BTI_V2DI, RS6000_BTI_V4SI, 0, 0 },
> +
> /* Overloaded built-in functions for ISA3.1 (power10). */
> { P10_BUILTIN_VEC_CLRL, P10V_BUILTIN_VCLRLB,
> RS6000_BTI_V16QI, RS6000_BTI_V16QI, RS6000_BTI_UINTSI, 0 },
> diff --git a/gcc/config/rs6000/vsx.md b/gcc/config/rs6000/vsx.md
> index 4ff52455fd3..31fcffe8f33 100644
> --- a/gcc/config/rs6000/vsx.md
> +++ b/gcc/config/rs6000/vsx.md
> @@ -4787,7 +4787,7 @@
> "vextsh2<wd> %0,%1"
> [(set_attr "type" "vecexts")])
>
> -(define_insn "*vsx_sign_extend_si_v2di"
> +(define_insn "vsx_sign_extend_si_v2di"
> [(set (match_operand:V2DI 0 "vsx_register_operand" "=v")
> (unspec:V2DI [(match_operand:V4SI 1 "vsx_register_operand" "v")]
> UNSPEC_VSX_SIGN_EXTEND))]
> diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
> index 5571c4f2ff2..c1c2c9f9bf7 100644
> --- a/gcc/doc/extend.texi
> +++ b/gcc/doc/extend.texi
> @@ -20800,6 +20800,21 @@ void vec_xst (vector unsigned char, int, vector unsigned char *);
> void vec_xst (vector unsigned char, int, unsigned char *);
> @end smallexample
>
> +uThe following sign extension builtins are provided.
> +
> +@smallexample
> +vector signed int vec_signexti (vector signed char a)
> +vector signed long long vec_signextll (vector signed char a)
> +vector signed int vec_signexti (vector signed short a)
> +vector signed long long vec_signextll (vector signed short a)
> +vector signed long long vec_signextll (vector signed int a)
> +@end smallexample
> +
> +Each element of the result is produced by sign-extending the element of the
> +input vector that would fall in the least significant portion of the result
> +element. For example, a sign-extension of a vector signed char to a vector
> +signed long long will sign extend the rightmost byte of each doubleword.
> +
> @node PowerPC AltiVec Built-in Functions Available on ISA 3.1
> @subsubsection PowerPC AltiVec Built-in Functions Available on ISA 3.1
>
> diff --git a/gcc/testsuite/gcc.target/powerpc/p9-sign_extend-runnable.c b/gcc/testsuite/gcc.target/powerpc/p9-sign_extend-runnable.c
> new file mode 100644
> index 00000000000..7bf979c6fd4
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/powerpc/p9-sign_extend-runnable.c
> @@ -0,0 +1,128 @@
> +/* { dg-do run { target { powerpc*-*-linux* && { lp64 && p9vector_hw } } } } */
> +/* { dg-require-effective-target powerpc_p9vector_ok } */
> +/* { dg-options "-O2 -mdejagnu-cpu=power9 -save-temps" } */
> +
> +/* These builtins were not defined until ISA 3.1 but only require ISA 3.0
> + support. */
> +
> +/* { dg-final { scan-assembler-times {\mvextsb2w\M} 1 } } */
> +/* { dg-final { scan-assembler-times {\mvextsb2d\M} 1 } } */
> +/* { dg-final { scan-assembler-times {\mvextsh2w\M} 1 } } */
> +/* { dg-final { scan-assembler-times {\mvextsh2d\M} 1 } } */
> +/* { dg-final { scan-assembler-times {\mvextsw2d\M} 1 } } */
> +
> +#include <altivec.h>
> +
> +#define DEBUG 0
> +
> +#if DEBUG
> +#include <stdio.h>
> +#include <stdlib.h>
> +#endif
> +
> +void abort (void);
> +
> +int main ()
> +{
> + int i;
> +
> + vector signed char vec_arg_qi, vec_result_qi;
> + vector signed short int vec_arg_hi, vec_result_hi, vec_expected_hi;
> + vector signed int vec_arg_wi, vec_result_wi, vec_expected_wi;
> + vector signed long long vec_result_di, vec_expected_di;
> +
> + /* test sign extend byte to word */
> + vec_arg_qi = (vector signed char) {1, 2, 3, 4, 5, 6, 7, 8,
> + -1, -2, -3, -4, -5, -6, -7, -8};
> + vec_expected_wi = (vector signed int) {1, 5, -1, -5};
> +
> + vec_result_wi = vec_signexti (vec_arg_qi);
> +
> + for (i = 0; i < 4; i++)
> + if (vec_result_wi[i] != vec_expected_wi[i]) {
> +#if DEBUG
> + printf("ERROR: vec_signexti(char, int): ");
> + printf("vec_result_wi[%d] != vec_expected_wi[%d]\n",
> + i, i);
> + printf("vec_result_wi[%d] = %d\n", i, vec_result_wi[i]);
> + printf("vec_expected_wi[%d] = %d\n", i, vec_expected_wi[i]);
> +#else
> + abort();
> +#endif
> + }
> +
> + /* test sign extend byte to double */
> + vec_arg_qi = (vector signed char){1, 2, 3, 4, 5, 6, 7, 8,
> + -1, -2, -3, -4, -5, -6, -7, -8};
> + vec_expected_di = (vector signed long long int){1, -1};
> +
> + vec_result_di = vec_signextll(vec_arg_qi);
> +
> + for (i = 0; i < 2; i++)
> + if (vec_result_di[i] != vec_expected_di[i]) {
> +#if DEBUG
> + printf("ERROR: vec_signextll(byte, long long int): ");
> + printf("vec_result_di[%d] != vec_expected_di[%d]\n", i, i);
> + printf("vec_result_di[%d] = %lld\n", i, vec_result_di[i]);
> + printf("vec_expected_di[%d] = %lld\n", i, vec_expected_di[i]);
> +#else
> + abort();
> +#endif
> + }
> +
> + /* test sign extend short to word */
> + vec_arg_hi = (vector signed short int){1, 2, 3, 4, -1, -2, -3, -4};
> + vec_expected_wi = (vector signed int){1, 3, -1, -3};
> +
> + vec_result_wi = vec_signexti(vec_arg_hi);
> +
> + for (i = 0; i < 4; i++)
> + if (vec_result_wi[i] != vec_expected_wi[i]) {
> +#if DEBUG
> + printf("ERROR: vec_signexti(short, int): ");
> + printf("vec_result_wi[%d] != vec_expected_wi[%d]\n", i, i);
> + printf("vec_result_wi[%d] = %d\n", i, vec_result_wi[i]);
> + printf("vec_expected_wi[%d] = %d\n", i, vec_expected_wi[i]);
> +#else
> + abort();
> +#endif
> + }
> +
> + /* test sign extend short to double word */
> + vec_arg_hi = (vector signed short int ){1, 3, 5, 7, -1, -3, -5, -7};
> + vec_expected_di = (vector signed long long int){1, -1};
> +
> + vec_result_di = vec_signextll(vec_arg_hi);
> +
> + for (i = 0; i < 2; i++)
> + if (vec_result_di[i] != vec_expected_di[i]) {
> +#if DEBUG
> + printf("ERROR: vec_signextll(short, double): ");
> + printf("vec_result_di[%d] != vec_expected_di[%d]\n", i, i);
> + printf("vec_result_di[%d] = %lld\n", i, vec_result_di[i]);
> + printf("vec_expected_di[%d] = %lld\n", i, vec_expected_di[i]);
> +#else
> + abort();
> +#endif
> + }
> +
> + /* test sign extend word to double word */
> + vec_arg_wi = (vector signed int ){1, 3, -1, -3};
> + vec_expected_di = (vector signed long long int){1, -1};
> +
> + vec_result_di = vec_signextll(vec_arg_wi);
> +
> + for (i = 0; i < 2; i++)
> + if (vec_result_di[i] != vec_expected_di[i]) {
> +#if DEBUG
> + printf("ERROR: vec_signextll(word, double): ");
> + printf("vec_result_di[%d] != vec_expected_di[%d]\n", i, i);
> + printf("vec_result_di[%d] = %lld\n", i, vec_result_di[i]);
> + printf("vec_expected_di[%d] = %lld\n", i, vec_expected_di[i]);
> +#else
> + abort();
> +#endif
> + }
> +
> + return 0;
> +}
next prev parent reply other threads:[~2020-10-07 21:08 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-21 21:17 [Patch 0/5] rs6000, 128-bit Binary Integer Operations Carl Love
2020-09-21 23:56 ` [PATCH 1/5] RS6000 Add 128-bit Binary Integer sign extend operations Carl Love
2020-09-24 18:20 ` will schmidt
2020-10-05 18:51 ` Carl Love
2020-10-07 21:08 ` will schmidt [this message]
2020-10-12 20:15 ` Carl Love
2020-10-12 20:51 ` Segher Boessenkool
2020-10-12 20:43 ` Segher Boessenkool
2020-10-12 21:06 ` Carl Love
2020-09-21 23:56 ` [PATCH 2/5] RS6000 add 128-bit Integer Operations Carl Love
2020-09-24 18:21 ` will schmidt
2020-10-05 18:52 ` [PATCH 2a/5] rs6000, vec_rlnm builtin fix arguments Carl Love
2020-10-07 21:08 ` will schmidt
2020-10-12 20:15 ` Carl Love
2020-10-12 23:37 ` Segher Boessenkool
2020-10-05 18:52 ` [PATCH 2b/5] RS6000 add 128-bit Integer Operations Carl Love
2020-10-07 21:53 ` will schmidt
2020-10-12 20:16 ` Carl Love
2020-10-13 0:23 ` Segher Boessenkool
2021-01-19 22:33 ` [PATCH 0/6 ver3] " Carl Love
2021-01-19 22:33 ` [PATCH 1/6 ver 3] rs6000, Fix arguments in altivec_vrlwmi and altivec_rlwdi builtins Carl Love
2021-01-19 22:33 ` [PATCH 2/6 ver 3] RS6000 Add 128-bit Binary Integer sign extend operations Carl Love
2021-01-19 22:33 ` [PATCH 3/6 ver 3] RS6000 add 128-bit Integer Operations part 1 Carl Love
2021-01-19 22:33 ` [PATCH 4/6 ver 3] Add TI to TD (128-bit DFP) and TD to TI support Carl Love
2021-01-19 22:33 ` [PATCH 5/6 ver 3] rs6000, Add test 128-bit shifts for just the int128 type Carl Love
2021-01-19 22:33 ` [PATCH 6/6 ver 3] Conversions between 128-bit integer and floating point values Carl Love
2020-09-21 23:56 ` [PATCH 3/5] Add TI to TD (128-bit DFP) and TD to TI support Carl Love
2020-09-24 18:21 ` will schmidt
2020-10-05 18:52 ` Carl Love
2020-10-08 14:22 ` will schmidt
2020-10-12 20:15 ` Carl Love
2020-09-21 23:56 ` [PATCH 4/5] Test 128-bit shifts for just the int128 type Carl Love
2020-09-24 18:21 ` will schmidt
2020-10-05 18:52 ` Carl Love
2020-10-08 14:50 ` will schmidt
2020-10-12 20:15 ` Carl Love
2020-09-21 23:57 ` [PATCH 5/5] Conversions between 128-bit integer and floating point values Carl Love
2020-09-24 18:22 ` will schmidt
2020-10-05 18:52 ` Carl Love
2020-10-08 15:35 ` will schmidt
2020-10-12 20:16 ` Carl Love
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=29f77863110b27d0f1bdc7123bc58811a43b2da4.camel@vnet.ibm.com \
--to=will_schmidt@vnet.ibm.com \
--cc=cel@us.ibm.com \
--cc=dje.gcc@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=segher@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).