Hi Haochen,
on 2024/7/9 15:18, HAO CHEN GUI wrote:
> Hi,
> This patch implemented optab_isinf for SFDF and IEEE128 by test
> data class instructions.
>
> Compared with previous version, the main changes are:
> 1 Define 3 mode attributes which are used for predicate, constraint
> and asm print selection. They help merge sp/dp/qp patterns to one.
> 2 Remove original sp/dp and qp patterns and combine them into one.
> 3 Rename corresponding icode name in rs6000-builtin.cc and
> rs6000-builtins.def.
> https://gcc.gnu.org/pipermail/gcc-patches/2024-June/655779.html
>
> The expand "isinf<mode>2" and following insn pattern for TF and
> KF mode should be guarded on "TARGET_FLOAT128_HW". It will be
> changed in sequential patch as some other "qp" insn patterns are
> also need to be changed.
>
> Bootstrapped and tested on powerpc64-linux BE and LE with no
> regressions. Is it OK for trunk?
>
> Thanks
> Gui Haochen
>
> ChangeLog
> rs6000: Implement optab_isinf for SFDF and IEEE128
>
> gcc/
> PR target/97786
> * config/rs6000/rs6000.md (constant VSX_TEST_DATA_CLASS_NAN,
> VSX_TEST_DATA_CLASS_POS_INF, VSX_TEST_DATA_CLASS_NEG_INF,
> VSX_TEST_DATA_CLASS_POS_ZERO, VSX_TEST_DATA_CLASS_NEG_ZERO,
> VSX_TEST_DATA_CLASS_POS_DENORMAL, VSX_TEST_DATA_CLASS_NEG_DENORMAL):
> Define.
> (mode_attr sdq, vsx_altivec, wa_v, x): Define.
> (mode_iterator IEEE_FP): Define.
> * config/rs6000/vsx.md (isinf<mode>2): New expand.
> (expand xststdcqp_<mode>, xststdc<sd>p): Combine into...
> (expand xststdc_<mode>): ...this.
> (insn *xststdcqp_<mode>, *xststdc<sd>p): Combine into...
> (insn *xststdc_<mode>): ...this.
> * config/rs6000/rs6000-builtin.cc (rs6000_expand_builtin): Rename
> CODE_FOR_xststdcqp_kf as CODE_FOR_xststdc_kf,
> CODE_FOR_xststdcqp_tf as CODE_FOR_xststdc_tf.
> * config/rs6000/rs6000-builtins.def: Rename xststdcdp as xststdc_df,
> xststdcsp as xststdc_sf, xststdcqp_kf as xststdc_kf.
>
> gcc/testsuite/
> PR target/97786
> * gcc.target/powerpc/pr97786-1.c: New test.
> * gcc.target/powerpc/pr97786-2.c: New test.
>
> patch.diff
> diff --git a/gcc/config/rs6000/rs6000-builtin.cc
> b/gcc/config/rs6000/rs6000-builtin.cc
> index bb9da68edc7..a62a5d4afa7 100644
> --- a/gcc/config/rs6000/rs6000-builtin.cc
> +++ b/gcc/config/rs6000/rs6000-builtin.cc
> @@ -3357,8 +3357,8 @@ rs6000_expand_builtin (tree exp, rtx target, rtx /*
> subtarget */,
> case CODE_FOR_xsiexpqpf_kf:
> icode = CODE_FOR_xsiexpqpf_tf;
> break;
> - case CODE_FOR_xststdcqp_kf:
> - icode = CODE_FOR_xststdcqp_tf;
> + case CODE_FOR_xststdc_kf:
> + icode = CODE_FOR_xststdc_tf;
> break;
> case CODE_FOR_xscmpexpqp_eq_kf:
> icode = CODE_FOR_xscmpexpqp_eq_tf;
> diff --git a/gcc/config/rs6000/rs6000-builtins.def
> b/gcc/config/rs6000/rs6000-builtins.def
> index 3bc7fed6956..8ac4cc200c9 100644
> --- a/gcc/config/rs6000/rs6000-builtins.def
> +++ b/gcc/config/rs6000/rs6000-builtins.def
> @@ -2752,11 +2752,11 @@
>
> const signed int \
> __builtin_vsx_scalar_test_data_class_dp (double, const int<7>);
> - VSTDCDP xststdcdp {}
> + VSTDCDP xststdc_df {}
>
> const signed int \
> __builtin_vsx_scalar_test_data_class_sp (float, const int<7>);
> - VSTDCSP xststdcsp {}
> + VSTDCSP xststdc_sf {}
>
> const signed int __builtin_vsx_scalar_test_neg_dp (double);
> VSTDCNDP xststdcnegdp {}
> @@ -2925,7 +2925,7 @@
>
> const signed int __builtin_vsx_scalar_test_data_class_qp (_Float128, \
> const int<7>);
> - VSTDCQP xststdcqp_kf {}
> + VSTDCQP xststdc_kf {}
>
> const signed int __builtin_vsx_scalar_test_neg_qp (_Float128);
> VSTDCNQP xststdcnegqp_kf {}
> diff --git a/gcc/config/rs6000/rs6000.md b/gcc/config/rs6000/rs6000.md
> index a5d20594789..2d7f227e362 100644
> --- a/gcc/config/rs6000/rs6000.md
> +++ b/gcc/config/rs6000/rs6000.md
> @@ -53,6 +53,20 @@ (define_constants
> (FRAME_POINTER_REGNUM 110)
> ])
>
> +;;
> +;; Test data class mask
Nit: s/mask/mask bits/
> +;;
> +
> +(define_constants
> + [(VSX_TEST_DATA_CLASS_NAN 0x40)
> + (VSX_TEST_DATA_CLASS_POS_INF 0x20)
> + (VSX_TEST_DATA_CLASS_NEG_INF 0x10)
Formatting nit: Move 0x20 and 0x10 align with the others?
> + (VSX_TEST_DATA_CLASS_POS_ZERO 0x8)
> + (VSX_TEST_DATA_CLASS_NEG_ZERO 0x4)
> + (VSX_TEST_DATA_CLASS_POS_DENORMAL 0x2)
> + (VSX_TEST_DATA_CLASS_NEG_DENORMAL 0x1)
> + ])
> +
> ;;
> ;; UNSPEC usage
> ;;
> @@ -605,6 +619,24 @@ (define_mode_iterator SFDF2 [SF DF])
> (define_mode_attr sd [(SF "s") (DF "d")
> (V4SF "s") (V2DF "d")])
>
> +; A generic s/d/q attribute, for sp/dp/qp for example.
> +(define_mode_attr sdq [(SF "s") (DF "d")
> + (TF "q") (KF "q")])
> +
> +; A predicate attribute, for IEEE floating point
> +(define_mode_attr vsx_altivec [(SF "vsx_register_operand")
> + (DF "vsx_register_operand")
> + (TF "altivec_register_operand")
> + (KF "altivec_register_operand")])
Nit: May be more meaningful with fp_register_op?
> +
> +; A constraint attribute, for IEEE floating point
Nit: Maybe "wa" or "v", for IEEE_FP constraints for example.
> +(define_mode_attr wa_v [(SF "wa") (DF "wa")
> + (TF "v") (KF "v")])
> +
> +; A register attribute, for IEEE floating point
Nit: Just follow the existing "s", say: "x" or nothing, for IEEE_FP for example.
> +(define_mode_attr x [(SF "x") (DF "x")
> + (TF "") (KF "")])
> +
Nit: it's would be good to move these three vsx_altivec, wa_v
and x to ...
> ; "s" or nothing, for fmuls/fmul for example.
> (define_mode_attr s [(SF "s") (DF "")])
>
> @@ -616,6 +648,9 @@ (define_mode_iterator IBM128 [(IF "FLOAT128_IBM_P
> (IFmode)")
> (define_mode_iterator IEEE128 [(KF "FLOAT128_IEEE_P (KFmode)")
> (TF "FLOAT128_IEEE_P (TFmode)")])
>
> +; Iterator for IEEE floating point
> +(define_mode_iterator IEEE_FP [SFDF IEEE128])
> +
... here, then they can closely follow the relevant IEEE_FP.
OK with these nits fixed, thanks!
BR,
Kewen
> ; Iterator for 128-bit floating point
> (define_mode_iterator FLOAT128 [(KF "TARGET_FLOAT128_TYPE")
> (IF "TARGET_FLOAT128_TYPE")
> diff --git a/gcc/config/rs6000/vsx.md b/gcc/config/rs6000/vsx.md
> index 48ba262f7e4..56d1d8c737e 100644
> --- a/gcc/config/rs6000/vsx.md
> +++ b/gcc/config/rs6000/vsx.md
> @@ -5300,36 +5300,15 @@ (define_insn "*xscmpexpqp"
> "xscmpexpqp %0,%1,%2"
> [(set_attr "type" "fpcompare")])
>
> -;; VSX Scalar Test Data Class Quad-Precision
> -;; (Expansion for scalar_test_data_class (__ieee128, int))
> -;; (Has side effect of setting the lt bit if operand 1 is negative,
> -;; setting the eq bit if any of the conditions tested by operand 2
> -;; are satisfied, and clearing the gt and undordered bits to zero.)
> -(define_expand "xststdcqp_<mode>"
> - [(set (match_dup 3)
> - (compare:CCFP
> - (unspec:IEEE128
> - [(match_operand:IEEE128 1 "altivec_register_operand" "v")
> - (match_operand:SI 2 "u7bit_cint_operand" "n")]
> - UNSPEC_VSX_STSTDC)
> - (const_int 0)))
> - (set (match_operand:SI 0 "register_operand" "=r")
> - (eq:SI (match_dup 3)
> - (const_int 0)))]
> - "TARGET_P9_VECTOR"
> -{
> - operands[3] = gen_reg_rtx (CCFPmode);
> -})
> -
> -;; VSX Scalar Test Data Class Double- and Single-Precision
> +;; VSX Scalar Test Data Class Quad-/Double-/Single-Precision
> ;; (The lt bit is set if operand 1 is negative. The eq bit is set
> ;; if any of the conditions tested by operand 2 are satisfied.
> ;; The gt and unordered bits are cleared to zero.)
> -(define_expand "xststdc<sd>p"
> +(define_expand "xststdc_<mode>"
> [(set (match_dup 3)
> (compare:CCFP
> - (unspec:SFDF
> - [(match_operand:SFDF 1 "vsx_register_operand" "wa")
> + (unspec:IEEE_FP
> + [(match_operand:IEEE_FP 1 "<vsx_altivec>" "<wa_v>")
> (match_operand:SI 2 "u7bit_cint_operand" "n")]
> UNSPEC_VSX_STSTDC)
> (match_dup 4)))
> @@ -5342,6 +5321,16 @@ (define_expand "xststdc<sd>p"
> operands[4] = CONST0_RTX (SImode);
> })
>
> +(define_expand "isinf<mode>2"
> + [(use (match_operand:SI 0 "gpc_reg_operand"))
> + (use (match_operand:IEEE_FP 1 "<vsx_altivec>"))]
> + "TARGET_HARD_FLOAT && TARGET_P9_VECTOR"
> +{
> + int mask = VSX_TEST_DATA_CLASS_POS_INF | VSX_TEST_DATA_CLASS_NEG_INF;
> + emit_insn (gen_xststdc_<mode> (operands[0], operands[1], GEN_INT (mask)));
> + DONE;
> +})
> +
> ;; The VSX Scalar Test Negative Quad-Precision
> (define_expand "xststdcnegqp_<mode>"
> [(set (match_dup 2)
> @@ -5377,27 +5366,16 @@ (define_expand "xststdcneg<sd>p"
> operands[3] = CONST0_RTX (SImode);
> })
>
> -(define_insn "*xststdcqp_<mode>"
> +(define_insn "*xststdc_<mode>"
> [(set (match_operand:CCFP 0 "" "=y")
> (compare:CCFP
> - (unspec:IEEE128
> - [(match_operand:IEEE128 1 "altivec_register_operand" "v")
> + (unspec:IEEE_FP
> + [(match_operand:IEEE_FP 1 "<vsx_altivec>" "<wa_v>")
> (match_operand:SI 2 "u7bit_cint_operand" "n")]
> UNSPEC_VSX_STSTDC)
> (const_int 0)))]
> "TARGET_P9_VECTOR"
> - "xststdcqp %0,%1,%2"
> - [(set_attr "type" "fpcompare")])
> -
> -(define_insn "*xststdc<sd>p"
> - [(set (match_operand:CCFP 0 "" "=y")
> - (compare:CCFP
> - (unspec:SFDF [(match_operand:SFDF 1 "vsx_register_operand" "wa")
> - (match_operand:SI 2 "u7bit_cint_operand" "n")]
> - UNSPEC_VSX_STSTDC)
> - (match_operand:SI 3 "zero_constant" "j")))]
> - "TARGET_P9_VECTOR"
> - "xststdc<sd>p %0,%x1,%2"
> + "xststdc<sdq>p %0,%<x>1,%2"
> [(set_attr "type" "fpcompare")])
>
> ;; VSX Vector Extract Exponent Double and Single Precision
> diff --git a/gcc/testsuite/gcc.target/powerpc/pr97786-1.c
> b/gcc/testsuite/gcc.target/powerpc/pr97786-1.c
> new file mode 100644
> index 00000000000..3951770f65b
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/powerpc/pr97786-1.c
> @@ -0,0 +1,22 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O2 -mdejagnu-cpu=power9" } */
> +/* { dg-require-effective-target powerpc_vsx } */
> +
> +int test1 (double x)
> +{
> + return __builtin_isinf (x);
> +}
> +
> +int test2 (float x)
> +{
> + return __builtin_isinf (x);
> +}
> +
> +int test3 (float x)
> +{
> + return __builtin_isinff (x);
> +}
> +
> +/* { dg-final { scan-assembler-not {\mfcmp} } } */
> +/* { dg-final { scan-assembler-times {\mxststdcsp\M} 2 } } */
> +/* { dg-final { scan-assembler-times {\mxststdcdp\M} 1 } } */
> diff --git a/gcc/testsuite/gcc.target/powerpc/pr97786-2.c
> b/gcc/testsuite/gcc.target/powerpc/pr97786-2.c
> new file mode 100644
> index 00000000000..cea2d0dd48e
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/powerpc/pr97786-2.c
> @@ -0,0 +1,17 @@
> +/* { dg-do compile } */
> +/* { dg-require-effective-target ppc_float128_hw } */
> +/* { dg-options "-O2 -mdejagnu-cpu=power9 -mabi=ieeelongdouble -Wno-psabi" }
> */
> +/* { dg-require-effective-target powerpc_vsx } */
> +
> +int test1 (long double x)
> +{
> + return __builtin_isinf (x);
> +}
> +
> +int test2 (long double x)
> +{
> + return __builtin_isinfl (x);
> +}
> +
> +/* { dg-final { scan-assembler-not {\mxscmpuqp\M} } } */
> +/* { dg-final { scan-assembler-times {\mxststdcqp\M} 2 } } */
>
>
