From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 1984) id 932F43858D28; Fri, 6 Jan 2023 12:59:09 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 932F43858D28 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1673009949; bh=6NGx6TISguUmKhWDA9PwhKNsOikIz7PNq828CUt2BKE=; h=From:To:Subject:Date:From; b=ud6BGbpmnyNU1tAYGXn9NVSISSRVVoGwpRsAAZ6H/1ZvJSTEhENve8c0VavKfKg2k 054ab/dpAFnnMHHHD3WNnlKzDLH8JN1COdR1JcfAQdnv1oW8ubn1Ktp9Q3K020nGMi 2e+jKHIN4AnpPr3gRI4VvDLknS89eBLEWEidYKks= MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="utf-8" From: Tamar Christina To: gcc-cvs@gcc.gnu.org Subject: [gcc r13-5046] Revert "aarch64: Make existing V2HF be usable." X-Act-Checkin: gcc X-Git-Author: Tamar Christina X-Git-Refname: refs/heads/master X-Git-Oldrev: 17176326de45cc7de4f8cedf4414b40df0df39bb X-Git-Newrev: a40c22c377c7cc657b0feaf0119d84d0d142a318 Message-Id: <20230106125909.932F43858D28@sourceware.org> Date: Fri, 6 Jan 2023 12:59:09 +0000 (GMT) List-Id: https://gcc.gnu.org/g:a40c22c377c7cc657b0feaf0119d84d0d142a318 commit r13-5046-ga40c22c377c7cc657b0feaf0119d84d0d142a318 Author: Tamar Christina Date: Fri Jan 6 12:57:50 2023 +0000 Revert "aarch64: Make existing V2HF be usable." This reverts commit 2cba118e538ba0b7582af7f9fb5ba2dfbb772f8e. Diff: --- gcc/config/aarch64/aarch64-simd.md | 88 +++++++++------------------- gcc/config/aarch64/aarch64.cc | 1 - gcc/config/aarch64/iterators.md | 30 +++------- gcc/config/arm/types.md | 6 +- gcc/testsuite/gcc.target/aarch64/sve/slp_1.c | 10 ++-- 5 files changed, 45 insertions(+), 90 deletions(-) diff --git a/gcc/config/aarch64/aarch64-simd.md b/gcc/config/aarch64/aarch64-simd.md index c0e6164b3bd..104088f67d2 100644 --- a/gcc/config/aarch64/aarch64-simd.md +++ b/gcc/config/aarch64/aarch64-simd.md @@ -19,10 +19,10 @@ ;; . (define_expand "mov" - [(set (match_operand:VMOVE 0 "nonimmediate_operand") - (match_operand:VMOVE 1 "general_operand"))] + [(set (match_operand:VALL_F16 0 "nonimmediate_operand") + (match_operand:VALL_F16 1 "general_operand"))] "TARGET_FLOAT" -{ + " /* Force the operand into a register if it is not an immediate whose use can be replaced with xzr. If the mode is 16 bytes wide, then we will be doing @@ -46,11 +46,12 @@ aarch64_expand_vector_init (operands[0], operands[1]); DONE; } -}) + " +) (define_expand "movmisalign" - [(set (match_operand:VMOVE 0 "nonimmediate_operand") - (match_operand:VMOVE 1 "general_operand"))] + [(set (match_operand:VALL_F16 0 "nonimmediate_operand") + (match_operand:VALL_F16 1 "general_operand"))] "TARGET_FLOAT && !STRICT_ALIGNMENT" { /* This pattern is not permitted to fail during expansion: if both arguments @@ -72,16 +73,6 @@ [(set_attr "type" "neon_dup, neon_from_gp")] ) -(define_insn "aarch64_simd_dupv2hf" - [(set (match_operand:V2HF 0 "register_operand" "=w") - (vec_duplicate:V2HF - (match_operand:HF 1 "register_operand" "0")))] - "TARGET_SIMD" - "@ - sli\\t%d0, %d0, 16" - [(set_attr "type" "neon_shift_imm")] -) - (define_insn "aarch64_simd_dup" [(set (match_operand:VDQF_F16 0 "register_operand" "=w,w") (vec_duplicate:VDQF_F16 @@ -94,10 +85,10 @@ ) (define_insn "aarch64_dup_lane" - [(set (match_operand:VMOVE 0 "register_operand" "=w") - (vec_duplicate:VMOVE + [(set (match_operand:VALL_F16 0 "register_operand" "=w") + (vec_duplicate:VALL_F16 (vec_select: - (match_operand:VMOVE 1 "register_operand" "w") + (match_operand:VALL_F16 1 "register_operand" "w") (parallel [(match_operand:SI 2 "immediate_operand" "i")]) )))] "TARGET_SIMD" @@ -159,29 +150,6 @@ (set_attr "arch" "*,*,*,*,*,*,*,simd,*")] ) -(define_insn "*aarch64_simd_movv2hf" - [(set (match_operand:V2HF 0 "nonimmediate_operand" - "=w, m, m, w, ?r, ?w, ?r, w, w") - (match_operand:V2HF 1 "general_operand" - "m, Dz, w, w, w, r, r, Dz, Dn"))] - "TARGET_SIMD_F16INST - && (register_operand (operands[0], V2HFmode) - || aarch64_simd_reg_or_zero (operands[1], V2HFmode))" - "@ - ldr\\t%s0, %1 - str\\twzr, %0 - str\\t%s1, %0 - mov\\t%0.2s[0], %1.2s[0] - umov\\t%w0, %1.s[0] - fmov\\t%s0, %w1 - mov\\t%w0, %w1 - movi\\t%d0, 0 - * return aarch64_output_simd_mov_immediate (operands[1], 32);" - [(set_attr "type" "neon_load1_1reg, store_8, neon_store1_1reg,\ - neon_logic, neon_to_gp, f_mcr,\ - mov_reg, neon_move, neon_move")] -) - (define_insn "*aarch64_simd_mov" [(set (match_operand:VQMOV 0 "nonimmediate_operand" "=w, Umn, m, w, ?r, ?w, ?r, w, w") @@ -225,7 +193,7 @@ (define_insn "aarch64_store_lane0" [(set (match_operand: 0 "memory_operand" "=m") - (vec_select: (match_operand:VMOVE 1 "register_operand" "w") + (vec_select: (match_operand:VALL_F16 1 "register_operand" "w") (parallel [(match_operand 2 "const_int_operand" "n")])))] "TARGET_SIMD && ENDIAN_LANE_N (, INTVAL (operands[2])) == 0" @@ -1090,11 +1058,11 @@ ) (define_insn "aarch64_simd_vec_set" - [(set (match_operand:VMOVE 0 "register_operand" "=w,w,w") - (vec_merge:VMOVE - (vec_duplicate:VMOVE + [(set (match_operand:VALL_F16 0 "register_operand" "=w,w,w") + (vec_merge:VALL_F16 + (vec_duplicate:VALL_F16 (match_operand: 1 "aarch64_simd_nonimmediate_operand" "w,?r,Utv")) - (match_operand:VMOVE 3 "register_operand" "0,0,0") + (match_operand:VALL_F16 3 "register_operand" "0,0,0") (match_operand:SI 2 "immediate_operand" "i,i,i")))] "TARGET_SIMD" { @@ -1116,14 +1084,14 @@ ) (define_insn "@aarch64_simd_vec_copy_lane" - [(set (match_operand:VMOVE 0 "register_operand" "=w") - (vec_merge:VMOVE - (vec_duplicate:VMOVE + [(set (match_operand:VALL_F16 0 "register_operand" "=w") + (vec_merge:VALL_F16 + (vec_duplicate:VALL_F16 (vec_select: - (match_operand:VMOVE 3 "register_operand" "w") + (match_operand:VALL_F16 3 "register_operand" "w") (parallel [(match_operand:SI 4 "immediate_operand" "i")]))) - (match_operand:VMOVE 1 "register_operand" "0") + (match_operand:VALL_F16 1 "register_operand" "0") (match_operand:SI 2 "immediate_operand" "i")))] "TARGET_SIMD" { @@ -1431,7 +1399,7 @@ ) (define_expand "vec_set" - [(match_operand:VMOVE 0 "register_operand") + [(match_operand:VALL_F16 0 "register_operand") (match_operand: 1 "aarch64_simd_nonimmediate_operand") (match_operand:SI 2 "immediate_operand")] "TARGET_SIMD" @@ -3550,7 +3518,7 @@ ;; gimple_fold'd to the IFN_REDUC_(MAX|MIN) function. (This is FP smax/smin). (define_expand "reduc__scal_" [(match_operand: 0 "register_operand") - (unspec: [(match_operand:VHSDF_P 1 "register_operand")] + (unspec: [(match_operand:VHSDF 1 "register_operand")] FMAXMINV)] "TARGET_SIMD" { @@ -3565,7 +3533,7 @@ (define_expand "reduc__scal_" [(match_operand: 0 "register_operand") - (unspec: [(match_operand:VHSDF_P 1 "register_operand")] + (unspec: [(match_operand:VHSDF 1 "register_operand")] FMAXMINNMV)] "TARGET_SIMD" { @@ -3609,8 +3577,8 @@ ) (define_insn "aarch64_reduc__internal" - [(set (match_operand:VHSDF_P 0 "register_operand" "=w") - (unspec:VHSDF_P [(match_operand:VHSDF_P 1 "register_operand" "w")] + [(set (match_operand:VHSDF 0 "register_operand" "=w") + (unspec:VHSDF [(match_operand:VHSDF 1 "register_operand" "w")] FMAXMINV))] "TARGET_SIMD" "\\t%0, %1." @@ -4255,7 +4223,7 @@ (define_insn_and_split "aarch64_get_lane" [(set (match_operand: 0 "aarch64_simd_nonimmediate_operand" "=?r, w, Utv") (vec_select: - (match_operand:VMOVE 1 "register_operand" "w, w, w") + (match_operand:VALL_F16 1 "register_operand" "w, w, w") (parallel [(match_operand:SI 2 "immediate_operand" "i, i, i")])))] "TARGET_SIMD" { @@ -8060,7 +8028,7 @@ ;; Standard pattern name vec_init. (define_expand "vec_init" - [(match_operand:VMOVE 0 "register_operand") + [(match_operand:VALL_F16 0 "register_operand") (match_operand 1 "" "")] "TARGET_SIMD" { @@ -8139,7 +8107,7 @@ (define_expand "vec_extract" [(match_operand: 0 "aarch64_simd_nonimmediate_operand") - (match_operand:VMOVE 1 "register_operand") + (match_operand:VALL_F16 1 "register_operand") (match_operand:SI 2 "immediate_operand")] "TARGET_SIMD" { diff --git a/gcc/config/aarch64/aarch64.cc b/gcc/config/aarch64/aarch64.cc index 9a79a9e7928..7591a66e008 100644 --- a/gcc/config/aarch64/aarch64.cc +++ b/gcc/config/aarch64/aarch64.cc @@ -3634,7 +3634,6 @@ aarch64_classify_vector_mode (machine_mode mode) case E_V8BFmode: case E_V4SFmode: case E_V2DFmode: - case E_V2HFmode: return TARGET_FLOAT ? VEC_ADVSIMD : 0; default: diff --git a/gcc/config/aarch64/iterators.md b/gcc/config/aarch64/iterators.md index a521dbde1ec..5b26443e5b6 100644 --- a/gcc/config/aarch64/iterators.md +++ b/gcc/config/aarch64/iterators.md @@ -160,10 +160,6 @@ (define_mode_iterator VHSDF [(V4HF "TARGET_SIMD_F16INST") (V8HF "TARGET_SIMD_F16INST") V2SF V4SF V2DF]) -;; Advanced SIMD Float modes suitable for reduction or pairwise operations -(define_mode_iterator VHSDF_P [(V4HF "TARGET_SIMD_F16INST") - (V8HF "TARGET_SIMD_F16INST") - V2SF V4SF V2DF (V2HF "TARGET_SIMD_F16INST")]) ;; Advanced SIMD Float modes, and DF. (define_mode_iterator VDQF_DF [V2SF V4SF V2DF DF]) @@ -192,22 +188,15 @@ (define_mode_iterator VALLF [V2SF V4SF V2DF SF DF]) ;; Advanced SIMD Float modes with 2 elements. -(define_mode_iterator V2F [V2SF V2DF V2HF]) +(define_mode_iterator V2F [V2SF V2DF]) ;; All Advanced SIMD modes on which we support any arithmetic operations. (define_mode_iterator VALL [V8QI V16QI V4HI V8HI V2SI V4SI V2DI V2SF V4SF V2DF]) -;; The set of all modes for which vld1 intrinsics are provided. +;; All Advanced SIMD modes suitable for moving, loading, and storing. (define_mode_iterator VALL_F16 [V8QI V16QI V4HI V8HI V2SI V4SI V2DI V4HF V8HF V4BF V8BF V2SF V4SF V2DF]) -;; All Advanced SIMD modes suitable for moving, loading, and storing -;; including V2HF -(define_mode_iterator VMOVE [V8QI V16QI V4HI V8HI V2SI V4SI V2DI - V4HF V8HF V4BF V8BF V2SF V4SF V2DF - (V2HF "TARGET_SIMD_F16INST")]) - - ;; The VALL_F16 modes except the 128-bit 2-element ones. (define_mode_iterator VALL_F16_NO_V2Q [V8QI V16QI V4HI V8HI V2SI V4SI V4HF V8HF V2SF V4SF]) @@ -1090,7 +1079,7 @@ (V2SF "2") (V4SF "4") (V1DF "1") (V2DF "2") (DI "1") (DF "1") - (V8DI "8") (V2HF "2")]) + (V8DI "8")]) ;; Map a mode to the number of bits in it, if the size of the mode ;; is constant. @@ -1209,7 +1198,7 @@ (define_mode_attr Vetype [(V8QI "b") (V16QI "b") (V4HI "h") (V8HI "h") (V2SI "s") (V4SI "s") - (V2DI "d") (V2HF "h") + (V2DI "d") (V4HF "h") (V8HF "h") (V2SF "s") (V4SF "s") (V2DF "d") @@ -1301,7 +1290,7 @@ ;; more accurately. (define_mode_attr stype [(V8QI "b") (V16QI "b") (V4HI "s") (V8HI "s") (V2SI "s") (V4SI "s") (V2DI "d") (V4HF "s") - (V8HF "s") (V2SF "s") (V4SF "s") (V2DF "d") (V2HF "s") + (V8HF "s") (V2SF "s") (V4SF "s") (V2DF "d") (HF "s") (SF "s") (DF "d") (QI "b") (HI "s") (SI "s") (DI "d")]) @@ -1376,8 +1365,8 @@ (V4HF "HF") (V8HF "HF") (V2SF "SF") (V4SF "SF") (DF "DF") (V2DF "DF") - (SI "SI") (V2HF "HF") - (QI "QI") (HI "HI") + (SI "SI") (HI "HI") + (QI "QI") (V4BF "BF") (V8BF "BF") (VNx16QI "QI") (VNx8QI "QI") (VNx4QI "QI") (VNx2QI "QI") (VNx8HI "HI") (VNx4HI "HI") (VNx2HI "HI") @@ -1397,7 +1386,7 @@ (V2SF "sf") (V4SF "sf") (V2DF "df") (DF "df") (SI "si") (HI "hi") - (QI "qi") (V2HF "hf") + (QI "qi") (V4BF "bf") (V8BF "bf") (VNx16QI "qi") (VNx8QI "qi") (VNx4QI "qi") (VNx2QI "qi") (VNx8HI "hi") (VNx4HI "hi") (VNx2HI "hi") @@ -1882,7 +1871,7 @@ (V4HF "") (V8HF "_q") (V4BF "") (V8BF "_q") (V2SF "") (V4SF "_q") - (V2HF "") (V2DF "_q") + (V2DF "_q") (QI "") (HI "") (SI "") (DI "") (HF "") (SF "") (DF "") (V2x8QI "") (V2x16QI "_q") (V2x4HI "") (V2x8HI "_q") @@ -1921,7 +1910,6 @@ (V2SI "p") (V4SI "v") (V2DI "p") (V2DF "p") (V2SF "p") (V4SF "v") - (V2HF "p") (V4HF "v") (V8HF "v")]) (define_mode_attr vsi2qi [(V2SI "v8qi") (V4SI "v16qi") diff --git a/gcc/config/arm/types.md b/gcc/config/arm/types.md index 0a656bd792d..83e29563c8e 100644 --- a/gcc/config/arm/types.md +++ b/gcc/config/arm/types.md @@ -484,7 +484,6 @@ ; neon_fp_minmax_s_q ; neon_fp_minmax_d ; neon_fp_minmax_d_q -; neon_fp_reduc_add_h ; neon_fp_reduc_add_s ; neon_fp_reduc_add_s_q ; neon_fp_reduc_add_d @@ -1035,7 +1034,6 @@ neon_fp_minmax_d,\ neon_fp_minmax_d_q,\ \ - neon_fp_reduc_add_h,\ neon_fp_reduc_add_s,\ neon_fp_reduc_add_s_q,\ neon_fp_reduc_add_d,\ @@ -1260,8 +1258,8 @@ neon_fp_compare_d, neon_fp_compare_d_q, neon_fp_minmax_s,\ neon_fp_minmax_s_q, neon_fp_minmax_d, neon_fp_minmax_d_q,\ neon_fp_neg_s, neon_fp_neg_s_q, neon_fp_neg_d, neon_fp_neg_d_q,\ - neon_fp_reduc_add_h, neon_fp_reduc_add_s, neon_fp_reduc_add_s_q,\ - neon_fp_reduc_add_d, neon_fp_reduc_add_d_q, neon_fp_reduc_minmax_s,\ + neon_fp_reduc_add_s, neon_fp_reduc_add_s_q, neon_fp_reduc_add_d,\ + neon_fp_reduc_add_d_q, neon_fp_reduc_minmax_s, neon_fp_reduc_minmax_s_q, neon_fp_reduc_minmax_d,\ neon_fp_reduc_minmax_d_q,\ neon_fp_cvt_narrow_s_q, neon_fp_cvt_narrow_d_q,\ diff --git a/gcc/testsuite/gcc.target/aarch64/sve/slp_1.c b/gcc/testsuite/gcc.target/aarch64/sve/slp_1.c index e6021c5a427..07d71a63414 100644 --- a/gcc/testsuite/gcc.target/aarch64/sve/slp_1.c +++ b/gcc/testsuite/gcc.target/aarch64/sve/slp_1.c @@ -30,9 +30,11 @@ vec_slp_##TYPE (TYPE *restrict a, TYPE b, TYPE c, int n) \ TEST_ALL (VEC_PERM) /* We should use one DUP for each of the 8-, 16- and 32-bit types, - We should use two DUPs for each of the three 64-bit types. */ + although we currently use LD1RW for _Float16. We should use two + DUPs for each of the three 64-bit types. */ /* { dg-final { scan-assembler-times {\tmov\tz[0-9]+\.h, [hw]} 2 } } */ -/* { dg-final { scan-assembler-times {\tmov\tz[0-9]+\.s, [sw]} 3 } } */ +/* { dg-final { scan-assembler-times {\tmov\tz[0-9]+\.s, [sw]} 2 } } */ +/* { dg-final { scan-assembler-times {\tld1rw\tz[0-9]+\.s, } 1 } } */ /* { dg-final { scan-assembler-times {\tmov\tz[0-9]+\.d, [dx]} 9 } } */ /* { dg-final { scan-assembler-times {\tzip1\tz[0-9]+\.d, z[0-9]+\.d, z[0-9]+\.d\n} 3 } } */ /* { dg-final { scan-assembler-not {\tzip2\t} } } */ @@ -51,7 +53,7 @@ TEST_ALL (VEC_PERM) /* { dg-final { scan-assembler-times {\twhilelo\tp[0-7]\.s} 6 } } */ /* { dg-final { scan-assembler-times {\twhilelo\tp[0-7]\.d} 6 } } */ /* { dg-final { scan-assembler-not {\tldr} } } */ -/* { dg-final { scan-assembler-not {\tstr} } } */ -/* { dg-final { scan-assembler-not {\tstr\th[0-9]+} } } */ +/* { dg-final { scan-assembler-times {\tstr} 2 } } */ +/* { dg-final { scan-assembler-times {\tstr\th[0-9]+} 2 } } */ /* { dg-final { scan-assembler-not {\tuqdec} } } */