Age | Commit message (Collapse) | Author | Files | Lines |
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h:
(_mm512_mask_blend_ph): New intrinsic.
(_mm512_permutex2var_ph): Ditto.
(_mm512_permutexvar_ph): Ditto.
* config/i386/avx512fp16vlintrin.h:
(_mm256_mask_blend_ph): New intrinsic.
(_mm256_permutex2var_ph): Ditto.
(_mm256_permutexvar_ph): Ditto.
(_mm_mask_blend_ph): Ditto.
(_mm_permutex2var_ph): Ditto.
(_mm_permutexvar_ph): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-14.c: New test.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h: Add new intrinsics.
(_mm512_conj_pch): New intrinsic.
(_mm512_mask_conj_pch): Ditto.
(_mm512_maskz_conj_pch): Ditto.
* config/i386/avx512fp16vlintrin.h: Add new intrinsics.
(_mm256_conj_pch): New intrinsic.
(_mm256_mask_conj_pch): Ditto.
(_mm256_maskz_conj_pch): Ditto.
(_mm_conj_pch): Ditto.
(_mm_mask_conj_pch): Ditto.
(_mm_maskz_conj_pch): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-conjugation-1.c: New test.
* gcc.target/i386/avx512fp16vl-conjugation-1.c: New test.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_MM512_REDUCE_OP): New macro
(_mm512_reduce_add_ph): New intrinsic.
(_mm512_reduce_mul_ph): Ditto.
(_mm512_reduce_min_ph): Ditto.
(_mm512_reduce_max_ph): Ditto.
* config/i386/avx512fp16vlintrin.h
(_MM256_REDUCE_OP/_MM_REDUCE_OP): New macro.
(_mm256_reduce_add_ph): New intrinsic.
(_mm256_reduce_mul_ph): Ditto.
(_mm256_reduce_min_ph): Ditto.
(_mm256_reduce_max_ph): Ditto.
(_mm_reduce_add_ph): Ditto.
(_mm_reduce_mul_ph): Ditto.
(_mm_reduce_min_ph): Ditto.
(_mm_reduce_max_ph): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-reduce-op-1.c: New test.
* gcc.target/i386/avx512fp16vl-reduce-op-1.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (__m512h_u, __m256h_u,
__m128h_u): New typedef.
(_mm512_load_ph): New intrinsic.
(_mm256_load_ph): Ditto.
(_mm_load_ph): Ditto.
(_mm512_loadu_ph): Ditto.
(_mm256_loadu_ph): Ditto.
(_mm_loadu_ph): Ditto.
(_mm512_store_ph): Ditto.
(_mm256_store_ph): Ditto.
(_mm_store_ph): Ditto.
(_mm512_storeu_ph): Ditto.
(_mm256_storeu_ph): Ditto.
(_mm_storeu_ph): Ditto.
(_mm512_abs_ph): Ditto.
* config/i386/avx512fp16vlintrin.h
(_mm_abs_ph): Ditto.
(_mm256_abs_ph): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-13.c: New test.
|
|
The code sequence emitted uses CC internally.
gcc/ChangeLog:
* config/s390/tpf.md (prologue_tpf, epilogue_tpf): Add cc clobber.
|
|
Avoid emitting a strict low part move if the insv target actually
affects the whole target reg.
gcc/ChangeLog:
PR target/102222
* config/s390/s390.c (s390_expand_insv): Emit a normal move if it
is actually a full copy of the source operand into the target.
Don't emit a strict low part move if source and target mode match.
gcc/testsuite/ChangeLog:
* gcc.target/s390/pr102222.c: New test.
|
|
gcc/ChangeLog:
* config/i386/i386.md (define_attr "isa"): Add
fma_or_avx512vl.
(define_attr "enabled"): Correspond fma_or_avx512vl to
TARGET_FMA || TARGET_AVX512VL.
* config/i386/mmx.md (fmav2sf4): Extend to AVX512 fma.
(fmsv2sf4): Ditto.
(fnmav2sf4): Ditto.
(fnmsv2sf4): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512vl-pr95046.c: New test.
|
|
gcc/ChangeLog:
* config/i386/i386.md (cstorehf3): New define_expand.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-builtin-fpcompare-1.c: New test.
* gcc.target/i386/avx512fp16-builtin-fpcompare-2.c: New test.
|
|
gcc/ChangeLog:
* config/i386/i386.md (<rounding_insn>hf2): New expander.
(sse4_1_round<mode>2): Extend from MODEF to MODEFH.
* config/i386/sse.md (*sse4_1_round<ssescalarmodesuffix>):
Extend from VF_128 to VFH_128.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-builtin-round-1.c: New test.
|
|
gcc/ChangeLog:
* config/i386/i386-features.c (i386-features.c): Handle
E_HFmode.
* config/i386/i386.md (sqrthf2): New expander.
(*sqrthf2): New define_insn.
* config/i386/sse.md
(*<sse>_vmsqrt<mode>2<mask_scalar_name><round_scalar_name>):
Extend to VFH_128.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-builtin-sqrt-1.c: New test.
* gcc.target/i386/avx512fp16vl-builtin-sqrt-1.c: New test.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm_mask_fcmadd_sch):
New intrinsic.
(_mm_mask3_fcmadd_sch): Likewise.
(_mm_maskz_fcmadd_sch): Likewise.
(_mm_fcmadd_sch): Likewise.
(_mm_mask_fmadd_sch): Likewise.
(_mm_mask3_fmadd_sch): Likewise.
(_mm_maskz_fmadd_sch): Likewise.
(_mm_fmadd_sch): Likewise.
(_mm_mask_fcmadd_round_sch): Likewise.
(_mm_mask3_fcmadd_round_sch): Likewise.
(_mm_maskz_fcmadd_round_sch): Likewise.
(_mm_fcmadd_round_sch): Likewise.
(_mm_mask_fmadd_round_sch): Likewise.
(_mm_mask3_fmadd_round_sch): Likewise.
(_mm_maskz_fmadd_round_sch): Likewise.
(_mm_fmadd_round_sch): Likewise.
(_mm_fcmul_sch): Likewise.
(_mm_mask_fcmul_sch): Likewise.
(_mm_maskz_fcmul_sch): Likewise.
(_mm_fmul_sch): Likewise.
(_mm_mask_fmul_sch): Likewise.
(_mm_maskz_fmul_sch): Likewise.
(_mm_fcmul_round_sch): Likewise.
(_mm_mask_fcmul_round_sch): Likewise.
(_mm_maskz_fcmul_round_sch): Likewise.
(_mm_fmul_round_sch): Likewise.
(_mm_mask_fmul_round_sch): Likewise.
(_mm_maskz_fmul_round_sch): Likewise.
* config/i386/i386-builtin.def: Add corresponding new builtins.
* config/i386/sse.md
(avx512fp16_fmaddcsh_v8hf_maskz<round_expand_name>): New expander.
(avx512fp16_fcmaddcsh_v8hf_maskz<round_expand_name>): Ditto.
(avx512fp16_fma_<complexopname>sh_v8hf<mask_scalarcz_name><round_scalarcz_name>):
New define insn.
(avx512fp16_<complexopname>sh_v8hf_mask<round_name>): Ditto.
(avx512fp16_<complexopname>sh_v8hf<mask_scalarc_name><round_scalarcz_name>):
Ditto.
* config/i386/subst.md (mask_scalarcz_name): New.
(mask_scalarc_name): Ditto.
(mask_scalarc_operand3): Ditto.
(mask_scalarcz_operand4): Ditto.
(round_scalarcz_name): Ditto.
(round_scalarc_mask_operand3): Ditto.
(round_scalarcz_mask_operand4): Ditto.
(round_scalarc_mask_op3): Ditto.
(round_scalarcz_mask_op4): Ditto.
(round_scalarcz_constraint): Ditto.
(round_scalarcz_nimm_predicate): Ditto.
(mask_scalarcz): Ditto.
(mask_scalarc): Ditto.
(round_scalarcz): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm512_fcmadd_pch):
New intrinsic.
(_mm512_mask_fcmadd_pch): Likewise.
(_mm512_mask3_fcmadd_pch): Likewise.
(_mm512_maskz_fcmadd_pch): Likewise.
(_mm512_fmadd_pch): Likewise.
(_mm512_mask_fmadd_pch): Likewise.
(_mm512_mask3_fmadd_pch): Likewise.
(_mm512_maskz_fmadd_pch): Likewise.
(_mm512_fcmadd_round_pch): Likewise.
(_mm512_mask_fcmadd_round_pch): Likewise.
(_mm512_mask3_fcmadd_round_pch): Likewise.
(_mm512_maskz_fcmadd_round_pch): Likewise.
(_mm512_fmadd_round_pch): Likewise.
(_mm512_mask_fmadd_round_pch): Likewise.
(_mm512_mask3_fmadd_round_pch): Likewise.
(_mm512_maskz_fmadd_round_pch): Likewise.
(_mm512_fcmul_pch): Likewise.
(_mm512_mask_fcmul_pch): Likewise.
(_mm512_maskz_fcmul_pch): Likewise.
(_mm512_fmul_pch): Likewise.
(_mm512_mask_fmul_pch): Likewise.
(_mm512_maskz_fmul_pch): Likewise.
(_mm512_fcmul_round_pch): Likewise.
(_mm512_mask_fcmul_round_pch): Likewise.
(_mm512_maskz_fcmul_round_pch): Likewise.
(_mm512_fmul_round_pch): Likewise.
(_mm512_mask_fmul_round_pch): Likewise.
(_mm512_maskz_fmul_round_pch): Likewise.
* config/i386/avx512fp16vlintrin.h (_mm_fmadd_pch):
New intrinsic.
(_mm_mask_fmadd_pch): Likewise.
(_mm_mask3_fmadd_pch): Likewise.
(_mm_maskz_fmadd_pch): Likewise.
(_mm256_fmadd_pch): Likewise.
(_mm256_mask_fmadd_pch): Likewise.
(_mm256_mask3_fmadd_pch): Likewise.
(_mm256_maskz_fmadd_pch): Likewise.
(_mm_fcmadd_pch): Likewise.
(_mm_mask_fcmadd_pch): Likewise.
(_mm_mask3_fcmadd_pch): Likewise.
(_mm_maskz_fcmadd_pch): Likewise.
(_mm256_fcmadd_pch): Likewise.
(_mm256_mask_fcmadd_pch): Likewise.
(_mm256_mask3_fcmadd_pch): Likewise.
(_mm256_maskz_fcmadd_pch): Likewise.
(_mm_fmul_pch): Likewise.
(_mm_mask_fmul_pch): Likewise.
(_mm_maskz_fmul_pch): Likewise.
(_mm256_fmul_pch): Likewise.
(_mm256_mask_fmul_pch): Likewise.
(_mm256_maskz_fmul_pch): Likewise.
(_mm_fcmul_pch): Likewise.
(_mm_mask_fcmul_pch): Likewise.
(_mm_maskz_fcmul_pch): Likewise.
(_mm256_fcmul_pch): Likewise.
(_mm256_mask_fcmul_pch): Likewise.
(_mm256_maskz_fcmul_pch): Likewise.
* config/i386/i386-builtin-types.def (V8HF_FTYPE_V8HF_V8HF_V8HF,
V8HF_FTYPE_V16HF_V16HF_V16HF, V16HF_FTYPE_V16HF_V16HF_V16HF_UQI,
V32HF_FTYPE_V32HF_V32HF_V32HF_INT,
V32HF_FTYPE_V32HF_V32HF_V32HF_UHI_INT): Add new builtin types.
* config/i386/i386-builtin.def: Add new builtins.
* config/i386/i386-expand.c: Handle new builtin types.
* config/i386/subst.md (SUBST_CV): New.
(maskc_name): Ditto.
(maskc_operand3): Ditto.
(maskc): Ditto.
(sdc_maskz_name): Ditto.
(sdc_mask_op4): Ditto.
(sdc_mask_op5): Ditto.
(sdc_mask_mode512bit_condition): Ditto.
(sdc): Ditto.
(round_maskc_operand3): Ditto.
(round_sdc_mask_operand4): Ditto.
(round_maskc_op3): Ditto.
(round_sdc_mask_op4): Ditto.
(round_saeonly_sdc_mask_operand5): Ditto.
* config/i386/sse.md (unspec): Add complex fma unspecs.
(avx512fmaskcmode): New.
(UNSPEC_COMPLEX_F_C_MA): Ditto.
(UNSPEC_COMPLEX_F_C_MUL): Ditto.
(complexopname): Ditto.
(<avx512>_fmaddc_<mode>_maskz<round_expand_name>): New expander.
(<avx512>_fcmaddc_<mode>_maskz<round_expand_name>): Ditto.
(fma_<complexopname>_<mode><sdc_maskz_name><round_name>): New
define insn.
(<avx512>_<complexopname>_<mode>_mask<round_name>): Ditto.
(<avx512>_<complexopname>_<mode><maskc_name><round_name>): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
This patch follows the discussion here[1], where Segher suggested
parameterizing those exact magic constants for density heuristics,
to make it easier to tweak if need.
The change here should be "No Functional Change". But I verified
it with SPEC2017 at option sets O2-vect and Ofast-unroll on Power8,
the result is neutral as expected.
[1]https://gcc.gnu.org/pipermail/gcc-patches/2021-September/579121.html
gcc/ChangeLog:
* config/rs6000/rs6000.opt (rs6000-density-pct-threshold,
rs6000-density-size-threshold, rs6000-density-penalty,
rs6000-density-load-pct-threshold,
rs6000-density-load-num-threshold): New parameter.
* config/rs6000/rs6000.c (rs6000_density_test): Adjust with
corresponding parameters.
|
|
We have a small unwinder shim that is only used for Darwin10
(and only then in quite specific cases). To avoid linking
this code for every executable or DSO, we can present the crt
as a convenience library (rather than a .o file).
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/ChangeLog:
* config/darwin.h (LINK_COMMAND_SPEC_A): Use Darwin10
unwinder shim as a convenience library.
libgcc/ChangeLog:
* config.host: Use convenience library for Darwin10
unwinder shim.
* config/t-darwin: Build Darwin10 unwinder shim as a
convenience library.
|
|
gcc/
* config/avr/avr-mcus.def: Add atmega324pb.
* doc/avr-mmcu.texi: Corresponding changes.
|
|
Besides conversion instructions, pass_rpad also handles scalar
sqrt/rsqrt/rcp/round instructions, while r12-3614 should only want to
handle conversion instructions, so fix it.
gcc/ChangeLog:
* config/i386/i386-features.c (remove_partial_avx_dependency):
Restrict TARGET_USE_VECTOR_FP_CONVERTS and
TARGET_USE_VECTOR_CONVERTS to conversion instructions only.
|
|
Add vfmadd[132,213,231]sh/vfnmadd[132,213,231]sh/
vfmsub[132,213,231]sh/vfnmsub[132,213,231]sh.
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm_fmadd_sh):
New intrinsic.
(_mm_mask_fmadd_sh): Likewise.
(_mm_mask3_fmadd_sh): Likewise.
(_mm_maskz_fmadd_sh): Likewise.
(_mm_fmadd_round_sh): Likewise.
(_mm_mask_fmadd_round_sh): Likewise.
(_mm_mask3_fmadd_round_sh): Likewise.
(_mm_maskz_fmadd_round_sh): Likewise.
(_mm_fnmadd_sh): Likewise.
(_mm_mask_fnmadd_sh): Likewise.
(_mm_mask3_fnmadd_sh): Likewise.
(_mm_maskz_fnmadd_sh): Likewise.
(_mm_fnmadd_round_sh): Likewise.
(_mm_mask_fnmadd_round_sh): Likewise.
(_mm_mask3_fnmadd_round_sh): Likewise.
(_mm_maskz_fnmadd_round_sh): Likewise.
(_mm_fmsub_sh): Likewise.
(_mm_mask_fmsub_sh): Likewise.
(_mm_mask3_fmsub_sh): Likewise.
(_mm_maskz_fmsub_sh): Likewise.
(_mm_fmsub_round_sh): Likewise.
(_mm_mask_fmsub_round_sh): Likewise.
(_mm_mask3_fmsub_round_sh): Likewise.
(_mm_maskz_fmsub_round_sh): Likewise.
(_mm_fnmsub_sh): Likewise.
(_mm_mask_fnmsub_sh): Likewise.
(_mm_mask3_fnmsub_sh): Likewise.
(_mm_maskz_fnmsub_sh): Likewise.
(_mm_fnmsub_round_sh): Likewise.
(_mm_mask_fnmsub_round_sh): Likewise.
(_mm_mask3_fnmsub_round_sh): Likewise.
(_mm_maskz_fnmsub_round_sh): Likewise.
* config/i386/i386-builtin-types.def
(V8HF_FTYPE_V8HF_V8HF_V8HF_UQI_INT): New builtin type.
* config/i386/i386-builtin.def: Add new builtins.
* config/i386/i386-expand.c: Handle new builtin type.
* config/i386/sse.md (fmai_vmfmadd_<mode><round_name>):
Ajdust to support FP16.
(fmai_vmfmsub_<mode><round_name>): Ditto.
(fmai_vmfnmadd_<mode><round_name>): Ditto.
(fmai_vmfnmsub_<mode><round_name>): Ditto.
(*fmai_fmadd_<mode>): Ditto.
(*fmai_fmsub_<mode>): Ditto.
(*fmai_fnmadd_<mode><round_name>): Ditto.
(*fmai_fnmsub_<mode><round_name>): Ditto.
(avx512f_vmfmadd_<mode>_mask<round_name>): Ditto.
(avx512f_vmfmadd_<mode>_mask3<round_name>): Ditto.
(avx512f_vmfmadd_<mode>_maskz<round_expand_name>): Ditto.
(avx512f_vmfmadd_<mode>_maskz_1<round_name>): Ditto.
(*avx512f_vmfmsub_<mode>_mask<round_name>): Ditto.
(avx512f_vmfmsub_<mode>_mask3<round_name>): Ditto.
(*avx512f_vmfmsub_<mode>_maskz_1<round_name>): Ditto.
(*avx512f_vmfnmsub_<mode>_mask<round_name>): Ditto.
(*avx512f_vmfnmsub_<mode>_mask3<round_name>): Ditto.
(*avx512f_vmfnmsub_<mode>_mask<round_name>): Ditto.
(*avx512f_vmfnmadd_<mode>_mask<round_name>): Renamed to ...
(avx512f_vmfnmadd_<mode>_mask<round_name>) ... this, and
adjust to support FP16.
(avx512f_vmfnmadd_<mode>_mask3<round_name>): Ditto.
(avx512f_vmfnmadd_<mode>_maskz_1<round_name>): Ditto.
(avx512f_vmfnmadd_<mode>_maskz<round_expand_name>): New
expander.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/sse.md (avx512fmaskmodelower): Extend to support
HF modes.
(maskload<mode><avx512fmaskmodelower>): Ditto.
(maskstore<mode><avx512fmaskmodelower>): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-xorsign-1.c: New test.
|
|
1. FP16 vector xor/ior/and/andnot/abs/neg
2. FP16 scalar abs/neg/copysign/xorsign
gcc/ChangeLog:
* config/i386/i386-expand.c (ix86_expand_fp_absneg_operator):
Handle HFmode.
(ix86_expand_copysign): Ditto.
(ix86_expand_xorsign): Ditto.
* config/i386/i386.c (ix86_build_const_vector): Handle HF vector
modes.
(ix86_build_signbit_mask): Ditto.
(ix86_can_change_mode_class): Ditto.
* config/i386/i386.md
(SSEMODEF): Add HFmode.
(ssevecmodef): Ditto.
(<code>hf2): New define_expand.
(*<code>hf2_1): New define_insn_and_split.
(copysign<mode>): Extend to support HFmode under AVX512FP16.
(xorsign<mode>): Ditto.
* config/i386/sse.md (VFB): New mode iterator.
(VFB_128_256): Ditto.
(VFB_512): Ditto.
(sseintvecmode2): Support HF vector mode.
(<code><mode>2): Use new mode iterator.
(*<code><mode>2): Ditto.
(copysign<mode>3): Ditto.
(xorsign<mode>3): Ditto.
(<code><mode>3<mask_name>): Ditto.
(<code><mode>3<mask_name>): Ditto.
(<sse>_andnot<mode>3<mask_name>): Adjust for HF vector mode.
(<sse>_andnot<mode>3<mask_name>): Ditto.
(*<code><mode>3<mask_name>): Ditto.
(*<code><mode>3<mask_name>): Ditto.
|
|
Add vfmadd[132,213,231]ph/vfnmadd[132,213,231]ph/vfmsub[132,213,231]ph/
vfnmsub[132,213,231]ph.
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm512_mask_fmadd_ph):
New intrinsic.
(_mm512_mask3_fmadd_ph): Likewise.
(_mm512_maskz_fmadd_ph): Likewise.
(_mm512_fmadd_round_ph): Likewise.
(_mm512_mask_fmadd_round_ph): Likewise.
(_mm512_mask3_fmadd_round_ph): Likewise.
(_mm512_maskz_fmadd_round_ph): Likewise.
(_mm512_fnmadd_ph): Likewise.
(_mm512_mask_fnmadd_ph): Likewise.
(_mm512_mask3_fnmadd_ph): Likewise.
(_mm512_maskz_fnmadd_ph): Likewise.
(_mm512_fnmadd_round_ph): Likewise.
(_mm512_mask_fnmadd_round_ph): Likewise.
(_mm512_mask3_fnmadd_round_ph): Likewise.
(_mm512_maskz_fnmadd_round_ph): Likewise.
(_mm512_fmsub_ph): Likewise.
(_mm512_mask_fmsub_ph): Likewise.
(_mm512_mask3_fmsub_ph): Likewise.
(_mm512_maskz_fmsub_ph): Likewise.
(_mm512_fmsub_round_ph): Likewise.
(_mm512_mask_fmsub_round_ph): Likewise.
(_mm512_mask3_fmsub_round_ph): Likewise.
(_mm512_maskz_fmsub_round_ph): Likewise.
(_mm512_fnmsub_ph): Likewise.
(_mm512_mask_fnmsub_ph): Likewise.
(_mm512_mask3_fnmsub_ph): Likewise.
(_mm512_maskz_fnmsub_ph): Likewise.
(_mm512_fnmsub_round_ph): Likewise.
(_mm512_mask_fnmsub_round_ph): Likewise.
(_mm512_mask3_fnmsub_round_ph): Likewise.
(_mm512_maskz_fnmsub_round_ph): Likewise.
* config/i386/avx512fp16vlintrin.h (_mm256_fmadd_ph):
New intrinsic.
(_mm256_mask_fmadd_ph): Likewise.
(_mm256_mask3_fmadd_ph): Likewise.
(_mm256_maskz_fmadd_ph): Likewise.
(_mm_fmadd_ph): Likewise.
(_mm_mask_fmadd_ph): Likewise.
(_mm_mask3_fmadd_ph): Likewise.
(_mm_maskz_fmadd_ph): Likewise.
(_mm256_fnmadd_ph): Likewise.
(_mm256_mask_fnmadd_ph): Likewise.
(_mm256_mask3_fnmadd_ph): Likewise.
(_mm256_maskz_fnmadd_ph): Likewise.
(_mm_fnmadd_ph): Likewise.
(_mm_mask_fnmadd_ph): Likewise.
(_mm_mask3_fnmadd_ph): Likewise.
(_mm_maskz_fnmadd_ph): Likewise.
(_mm256_fmsub_ph): Likewise.
(_mm256_mask_fmsub_ph): Likewise.
(_mm256_mask3_fmsub_ph): Likewise.
(_mm256_maskz_fmsub_ph): Likewise.
(_mm_fmsub_ph): Likewise.
(_mm_mask_fmsub_ph): Likewise.
(_mm_mask3_fmsub_ph): Likewise.
(_mm_maskz_fmsub_ph): Likewise.
(_mm256_fnmsub_ph): Likewise.
(_mm256_mask_fnmsub_ph): Likewise.
(_mm256_mask3_fnmsub_ph): Likewise.
(_mm256_maskz_fnmsub_ph): Likewise.
(_mm_fnmsub_ph): Likewise.
(_mm_mask_fnmsub_ph): Likewise.
(_mm_mask3_fnmsub_ph): Likewise.
(_mm_maskz_fnmsub_ph): Likewise.
* config/i386/i386-builtin.def: Add corresponding new builtins.
* config/i386/sse.md
(<avx512>_fmadd_<mode>_maskz<round_expand_name>): Adjust to
support HF vector modes.
(<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name><round_name>):
Ditto.
(*<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name>_bcst_1): Ditto.
(*<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name>_bcst_2): Ditto.
(*<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name>_bcst_3): Ditto.
(<avx512>_fmadd_<mode>_mask<round_name>): Ditto.
(<avx512>_fmadd_<mode>_mask3<round_name>): Ditto.
(<avx512>_fmsub_<mode>_maskz<round_expand_name>): Ditto.
(<sd_mask_codefor>fma_fmsub_<mode><sd_maskz_name><round_name>):
Ditto.
(*<sd_mask_codefor>fma_fmsub_<mode><sd_maskz_name>_bcst_1): Ditto.
(*<sd_mask_codefor>fma_fmsub_<mode><sd_maskz_name>_bcst_2): Ditto.
(*<sd_mask_codefor>fma_fmsub_<mode><sd_maskz_name>_bcst_3): Ditto.
(<avx512>_fmsub_<mode>_mask<round_name>): Ditto.
(<avx512>_fmsub_<mode>_mask3<round_name>): Ditto.
(<sd_mask_codefor>fma_fnmadd_<mode><sd_maskz_name><round_name>):
Ditto.
(*<sd_mask_codefor>fma_fnmadd_<mode><sd_maskz_name>_bcst_1): Ditto.
(*<sd_mask_codefor>fma_fnmadd_<mode><sd_maskz_name>_bcst_2): Ditto.
(*<sd_mask_codefor>fma_fnmadd_<mode><sd_maskz_name>_bcst_3): Ditto.
(<avx512>_fnmadd_<mode>_mask<round_name>): Ditto.
(<avx512>_fnmadd_<mode>_mask3<round_name>): Ditto.
(<avx512>_fnmsub_<mode>_maskz<round_expand_name>): Ditto.
(<sd_mask_codefor>fma_fnmsub_<mode><sd_maskz_name><round_name>):
Ditto.
(*<sd_mask_codefor>fma_fnmsub_<mode><sd_maskz_name>_bcst_1): Ditto.
(*<sd_mask_codefor>fma_fnmsub_<mode><sd_maskz_name>_bcst_2): Ditto.
(*<sd_mask_codefor>fma_fnmsub_<mode><sd_maskz_name>_bcst_3): Ditto.
(<avx512>_fnmsub_<mode>_mask<round_name>): Ditto.
(<avx512>_fnmsub_<mode>_mask3<round_name>): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test fot new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm512_fmaddsub_ph):
New intrinsic.
(_mm512_mask_fmaddsub_ph): Likewise.
(_mm512_mask3_fmaddsub_ph): Likewise.
(_mm512_maskz_fmaddsub_ph): Likewise.
(_mm512_fmaddsub_round_ph): Likewise.
(_mm512_mask_fmaddsub_round_ph): Likewise.
(_mm512_mask3_fmaddsub_round_ph): Likewise.
(_mm512_maskz_fmaddsub_round_ph): Likewise.
(_mm512_mask_fmsubadd_ph): Likewise.
(_mm512_mask3_fmsubadd_ph): Likewise.
(_mm512_maskz_fmsubadd_ph): Likewise.
(_mm512_fmsubadd_round_ph): Likewise.
(_mm512_mask_fmsubadd_round_ph): Likewise.
(_mm512_mask3_fmsubadd_round_ph): Likewise.
(_mm512_maskz_fmsubadd_round_ph): Likewise.
* config/i386/avx512fp16vlintrin.h (_mm256_fmaddsub_ph):
New intrinsic.
(_mm256_mask_fmaddsub_ph): Likewise.
(_mm256_mask3_fmaddsub_ph): Likewise.
(_mm256_maskz_fmaddsub_ph): Likewise.
(_mm_fmaddsub_ph): Likewise.
(_mm_mask_fmaddsub_ph): Likewise.
(_mm_mask3_fmaddsub_ph): Likewise.
(_mm_maskz_fmaddsub_ph): Likewise.
(_mm256_fmsubadd_ph): Likewise.
(_mm256_mask_fmsubadd_ph): Likewise.
(_mm256_mask3_fmsubadd_ph): Likewise.
(_mm256_maskz_fmsubadd_ph): Likewise.
(_mm_fmsubadd_ph): Likewise.
(_mm_mask_fmsubadd_ph): Likewise.
(_mm_mask3_fmsubadd_ph): Likewise.
(_mm_maskz_fmsubadd_ph): Likewise.
* config/i386/i386-builtin.def: Add corresponding new builtins.
* config/i386/sse.md (VFH_SF_AVX512VL): New mode iterator.
* (<avx512>_fmsubadd_<mode>_maskz<round_expand_name>): New expander.
* (<avx512>_fmaddsub_<mode>_maskz<round_expand_name>): Use
VFH_SF_AVX512VL.
* (<sd_mask_codefor>fma_fmaddsub_<mode><sd_maskz_name><round_name>):
Ditto.
* (<avx512>_fmaddsub_<mode>_mask<round_name>): Ditto.
* (<avx512>_fmaddsub_<mode>_mask3<round_name>): Ditto.
* (<sd_mask_codefor>fma_fmsubadd_<mode><sd_maskz_name><round_name>):
Ditto.
* (<avx512>_fmsubadd_<mode>_mask<round_name>): Ditto.
* (<avx512>_fmsubadd_<mode>_mask3<round_name>): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
gcc/ChangeLog:
PR target/87767
* config/i386/i386.c (ix86_print_operand): Handle
V8HF/V16HF/V32HFmode.
* config/i386/i386.h (VALID_BCST_MODE_P): Add HFmode.
* config/i386/sse.md (avx512bcst): Remove.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-broadcast-1.c: New test.
* gcc.target/i386/avx512fp16-broadcast-2.c: New test.
|
|
This is an optimisation for middle-end worker-partitioning support (used
to support multiple workers on AMD GCN). At present, barriers may be
emitted in cases where they aren't needed and cannot be optimised away.
This patch stops the extraneous barriers from being emitted in the
first place.
One exception to the above (where the barrier is still needed) is for
predicated blocks of code that perform a write to gang-private shared
memory from one worker. We must execute a barrier before other workers
read that shared memory location.
gcc/
* config/gcn/gcn.c (gimple.h): Include.
(gcn_fork_join): Emit barrier for worker-level joins.
* omp-oacc-neuter-broadcast.cc (find_local_vars_to_propagate): Add
writes_gang_private bitmap parameter. Set bit for blocks
containing gang-private variable writes.
(worker_single_simple): Don't emit barrier after predicated block.
(worker_single_copy): Don't emit barrier if we're not broadcasting
anything and the block contains no gang-private writes.
(neuter_worker_single): Don't predicate blocks that only contain
NOPs or internal marker functions. Pass has_gang_private_write
argument to worker_single_copy.
(oacc_do_neutering): Add writes_gang_private bitmap handling.
|
|
This patch implements an algorithm to lay out local data-share (LDS)
space. It currently works for AMD GCN. At the moment, LDS is used for
three things:
1. Gang-private variables
2. Reduction temporaries (accumulators)
3. Broadcasting for worker partitioning
After the patch is applied, (2) and (3) are placed at preallocated
locations in LDS, and (1) continues to be handled by the backend (as it
is at present prior to this patch being applied). LDS now looks like this:
+--------------+ (gang-private size + 1024, = 1536)
| free space |
| ... |
| - - - - - - -|
| worker bcast |
+--------------+
| reductions |
+--------------+ <<< -mgang-private-size=<number> (def. 512)
| gang-private |
| vars |
+--------------+ (32)
| low LDS vars |
+--------------+ LDS base
So, gang-private space is fixed at a constant amount at compile time
(which can be increased with a command-line switch if necessary
for some given code). The layout algorithm takes out a slice of the
remainder of usable space for reduction vars, and uses the rest for
worker partitioning.
The partitioning algorithm works as follows.
1. An "adjacency" set is built up for each basic block that might
do a broadcast. This is calculated by starting at each such block,
and doing a recursive DFS walk over successors to find the next
block (or blocks) that *also* does a broadcast
(dfs_broadcast_reachable_1).
2. The adjacency set is inverted to get adjacent predecessor blocks also.
3. Blocks that will perform a broadcast are sorted by size of that
broadcast: the biggest blocks are handled first.
4. A splay tree structure is used to calculate the spans of LDS memory
that are already allocated by the blocks adjacent to this one
(merge_ranges{,_1}.
5. The current block's broadcast space is allocated from the first free
span not allocated in the splay tree structure calculated above
(first_fit_range). This seems to work quite nicely and efficiently
with the splay tree structure.
6. Continue with the next-biggest broadcast block until we're done.
In this way, "adjacent" broadcasts will not use the same piece of
LDS memory.
PR96334 "openacc: Unshare reduction temporaries for GCN" got merged in:
The GCN backend uses tree nodes like MEM((__lds TYPE *) <constant>)
for reduction temporaries. Unlike e.g. var decls and SSA names, these
nodes cannot be shared during gimplification, but are so in some
circumstances. This is detected when appropriate --enable-checking
options are used. This patch unshares such nodes when they are reused
more than once.
gcc/
* config/gcn/gcn-protos.h
(gcn_goacc_create_worker_broadcast_record): Update prototype.
* config/gcn/gcn-tree.c (gcn_goacc_get_worker_red_decl): Use
preallocated block of LDS memory. Do not cache/share decls for
reduction temporaries between invocations.
(gcn_goacc_reduction_teardown): Unshare VAR on second use.
(gcn_goacc_create_worker_broadcast_record): Add OFFSET parameter
and return temporary LDS space at that offset. Return pointer in
"sender" case.
* config/gcn/gcn.c (acc_lds_size, gang_private_hwm, lds_allocs):
New global vars.
(ACC_LDS_SIZE): Define as acc_lds_size.
(gcn_init_machine_status): Don't initialise lds_allocated,
lds_allocs, reduc_decls fields of machine function struct.
(gcn_option_override): Handle default size for gang-private
variables and -mgang-private-size option.
(gcn_expand_prologue): Use LDS_SIZE instead of LDS_SIZE-1 when
initialising M0_REG.
(gcn_shared_mem_layout): New function.
(gcn_print_lds_decl): Update comment. Use global lds_allocs map and
gang_private_hwm variable.
(TARGET_GOACC_SHARED_MEM_LAYOUT): Define target hook.
* config/gcn/gcn.h (machine_function): Remove lds_allocated,
lds_allocs, reduc_decls. Add reduction_base, reduction_limit.
* config/gcn/gcn.opt (gang_private_size_opt): New global.
(mgang-private-size=): New option.
* doc/tm.texi.in (TARGET_GOACC_SHARED_MEM_LAYOUT): Place
documentation hook.
* doc/tm.texi: Regenerate.
* omp-oacc-neuter-broadcast.cc (targhooks.h, diagnostic-core.h):
Add includes.
(build_sender_ref): Handle sender_decl being pointer.
(worker_single_copy): Add PLACEMENT and ISOLATE_BROADCASTS
parameters. Pass placement argument to
create_worker_broadcast_record hook invocations. Handle
sender_decl being pointer and isolate_broadcasts inserting extra
barriers.
(blk_offset_map_t): Add typedef.
(neuter_worker_single): Add BLK_OFFSET_MAP parameter. Pass
preallocated range to worker_single_copy call.
(dfs_broadcast_reachable_1): New function.
(idx_decl_pair_t, used_range_vec_t): New typedefs.
(sort_size_descending): New function.
(addr_range): New class.
(splay_tree_compare_addr_range, splay_tree_free_key)
(first_fit_range, merge_ranges_1, merge_ranges): New functions.
(execute_omp_oacc_neuter_broadcast): Rename to...
(oacc_do_neutering): ... this. Add BOUNDS_LO, BOUNDS_HI
parameters. Arrange layout of shared memory for broadcast
operations.
(execute_omp_oacc_neuter_broadcast): New function.
(pass_omp_oacc_neuter_broadcast::gate): Remove num_workers==1
handling from here. Enable pass for all OpenACC routines in order
to call shared memory-layout hook.
* target.def (create_worker_broadcast_record): Add OFFSET
parameter.
(shared_mem_layout): New hook.
libgomp/
* testsuite/libgomp.oacc-c-c++-common/broadcast-many.c: Update.
|
|
This patch just duplicates a couple of functions and adjusts them to use the
new builtin names. There's no logical change otherwise.
2021-09-17 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000.c (rs6000-builtins.h): New include.
(rs6000_new_builtin_vectorized_function): New function.
(rs6000_new_builtin_md_vectorized_function): Likewise.
(rs6000_builtin_vectorized_function): Call
rs6000_new_builtin_vectorized_function.
(rs6000_builtin_md_vectorized_function): Call
rs6000_new_builtin_md_vectorized_function.
|
|
Peter Bergner recently added two new builtins __builtin_vsx_lxvp and
__builtin_vsx_stxvp. These happened to break a pattern in MMA builtins that
I had been using to automate gimple folding of MMA builtins. Previously,
every MMA function that could be folded had an associated internal function
that it was folded into. The LXVP/STXVP builtins are just folded directly
into memory operations.
Instead of relying on this pattern, this patch adds a new attribute to
builtins called "mmaint," which is set for all MMA builtins that have an
associated internal builtin. The naming convention that adds _INTERNAL to
the builtin index name remains.
The rest of the patch is just duplicating Peter's patch, using the new
builtin infrastructure.
2021-09-17 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000-builtin-new.def (ASSEMBLE_ACC): Add mmaint flag.
(ASSEMBLE_PAIR): Likewise.
(BUILD_ACC): Likewise.
(DISASSEMBLE_ACC): Likewise.
(DISASSEMBLE_PAIR): Likewise.
(PMXVBF16GER2): Likewise.
(PMXVBF16GER2NN): Likewise.
(PMXVBF16GER2NP): Likewise.
(PMXVBF16GER2PN): Likewise.
(PMXVBF16GER2PP): Likewise.
(PMXVF16GER2): Likewise.
(PMXVF16GER2NN): Likewise.
(PMXVF16GER2NP): Likewise.
(PMXVF16GER2PN): Likewise.
(PMXVF16GER2PP): Likewise.
(PMXVF32GER): Likewise.
(PMXVF32GERNN): Likewise.
(PMXVF32GERNP): Likewise.
(PMXVF32GERPN): Likewise.
(PMXVF32GERPP): Likewise.
(PMXVF64GER): Likewise.
(PMXVF64GERNN): Likewise.
(PMXVF64GERNP): Likewise.
(PMXVF64GERPN): Likewise.
(PMXVF64GERPP): Likewise.
(PMXVI16GER2): Likewise.
(PMXVI16GER2PP): Likewise.
(PMXVI16GER2S): Likewise.
(PMXVI16GER2SPP): Likewise.
(PMXVI4GER8): Likewise.
(PMXVI4GER8PP): Likewise.
(PMXVI8GER4): Likewise.
(PMXVI8GER4PP): Likewise.
(PMXVI8GER4SPP): Likewise.
(XVBF16GER2): Likewise.
(XVBF16GER2NN): Likewise.
(XVBF16GER2NP): Likewise.
(XVBF16GER2PN): Likewise.
(XVBF16GER2PP): Likewise.
(XVF16GER2): Likewise.
(XVF16GER2NN): Likewise.
(XVF16GER2NP): Likewise.
(XVF16GER2PN): Likewise.
(XVF16GER2PP): Likewise.
(XVF32GER): Likewise.
(XVF32GERNN): Likewise.
(XVF32GERNP): Likewise.
(XVF32GERPN): Likewise.
(XVF32GERPP): Likewise.
(XVF64GER): Likewise.
(XVF64GERNN): Likewise.
(XVF64GERNP): Likewise.
(XVF64GERPN): Likewise.
(XVF64GERPP): Likewise.
(XVI16GER2): Likewise.
(XVI16GER2PP): Likewise.
(XVI16GER2S): Likewise.
(XVI16GER2SPP): Likewise.
(XVI4GER8): Likewise.
(XVI4GER8PP): Likewise.
(XVI8GER4): Likewise.
(XVI8GER4PP): Likewise.
(XVI8GER4SPP): Likewise.
(XXMFACC): Likewise.
(XXMTACC): Likewise.
(XXSETACCZ): Likewise.
(ASSEMBLE_PAIR_V): Likewise.
(BUILD_PAIR): Likewise.
(DISASSEMBLE_PAIR_V): Likewise.
(LXVP): New.
(STXVP): New.
* config/rs6000/rs6000-call.c (rs6000_gimple_fold_new_mma_builtin):
Handle RS6000_BIF_LXVP and RS6000_BIF_STXVP.
* config/rs6000/rs6000-gen-builtins.c (attrinfo): Add ismmaint.
(parse_bif_attrs): Handle ismmaint.
(write_decls): Add bif_mmaint_bit and bif_is_mmaint.
(write_bif_static_init): Handle ismmaint.
|
|
This is another patch that looks bigger than it really is. Because we
have a new namespace for the builtins, allowing us to have both the old
and new builtin infrastructure supported at once, we need versions of
these functions that use the new builtin namespace. Otherwise the code is
unchanged.
2021-09-17 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000-call.c (rs6000_gimple_fold_new_builtin): New
forward decl.
(rs6000_gimple_fold_builtin): Call rs6000_gimple_fold_new_builtin.
(rs6000_new_builtin_valid_without_lhs): New function.
(rs6000_gimple_fold_new_mma_builtin): Likewise.
(rs6000_gimple_fold_new_builtin): Likewise.
|
|
I over-restricted use of __builtin_mffsl, since I was unaware that it
automatically uses mffs when mffsl is not available. Paul Clarke pointed
this out in discussion of his SSE 4.1 compatibility patches.
2021-08-31 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000-builtin-new.def (__builtin_mffsl): Move from
[power9] to [always].
|
|
1. Replace TARGET_SSE_PARTIAL_REG_DEPENDENCY with
TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY in SSE FP to FP splitters.
2. Replace TARGET_SSE_PARTIAL_REG_DEPENDENCY with
TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY in SSE INT to FP splitters.
3. Also check TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY and
TARGET_SSE_PARTIAL_REG_DEPENDENCY when handling avx_partial_xmm_update
attribute. Don't convert AVX partial XMM register update if there is no
partial SSE register dependency for SSE conversion.
gcc/
* config/i386/i386-features.c (remove_partial_avx_dependency):
Also check TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY and
and TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY before generating
vxorps.
* config/i386/i386.h (TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY):
New.
(TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY): Likewise.
* config/i386/i386.md (SSE FP to FP splitters): Replace
TARGET_SSE_PARTIAL_REG_DEPENDENCY with
TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY.
(SSE INT to FP splitter): Replace TARGET_SSE_PARTIAL_REG_DEPENDENCY
with TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY.
* config/i386/x86-tune.def
(X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY): New.
(X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY): Likewise.
gcc/testsuite/
* gcc.target/i386/avx-covert-1.c: New file.
* gcc.target/i386/avx-fp-covert-1.c: Likewise.
* gcc.target/i386/avx-int-covert-1.c: Likewise.
* gcc.target/i386/sse-covert-1.c: Likewise.
* gcc.target/i386/sse-fp-covert-1.c: Likewise.
* gcc.target/i386/sse-int-covert-1.c: Likewise.
|
|
Check TARGET_USE_VECTOR_FP_CONVERTS or TARGET_USE_VECTOR_CONVERTS when
handling avx_partial_xmm_update attribute. Don't convert AVX partial
XMM register update if vector packed SSE conversion should be used.
gcc/
PR target/101900
* config/i386/i386-features.c (remove_partial_avx_dependency):
Check TARGET_USE_VECTOR_FP_CONVERTS and TARGET_USE_VECTOR_CONVERTS
before generating vxorps.
gcc/testsuite
PR target/101900
* gcc.target/i386/pr101900-1.c: New test.
* gcc.target/i386/pr101900-2.c: Likewise.
* gcc.target/i386/pr101900-3.c: Likewise.
|
|
Simply memcpy and memset inline strategies to avoid branches for
-mtune=tremont:
1. Create Tremont cost model from generic cost model.
2. With MOVE_RATIO and CLEAR_RATIO == 17, GCC will use integer/vector
load and store for up to 16 * 16 (256) bytes when the data size is
fixed and known.
3. Inline only if data size is known to be <= 256.
a. Use "rep movsb/stosb" with simple code sequence if the data size
is a constant.
b. Use loop if data size is not a constant.
4. Use memcpy/memset libray function if data size is unknown or > 256.
* config/i386/i386-options.c (processor_cost_table): Use
tremont_cost for Tremont.
* config/i386/x86-tune-costs.h (tremont_memcpy): New.
(tremont_memset): Likewise.
(tremont_cost): Likewise.
* config/i386/x86-tune.def (X86_TUNE_PREFER_KNOWN_REP_MOVSB_STOSB):
Enable for Tremont.
|
|
Initial -mtune=tremont update
1. Use Haswell scheduling model.
2. Assume that stack engine allows to execute push&pop instructions in
parall.
3. Prepare for scheduling pass as -mtune=generic.
4. Use the same issue rate as -mtune=generic.
5. Enable partial_reg_dependency.
6. Disable accumulate_outgoing_args
7. Enable use_leave
8. Enable push_memory
9. Disable four_jump_limit
10. Disable opt_agu
11. Disable avoid_lea_for_addr
12. Disable avoid_mem_opnd_for_cmove
13. Enable misaligned_move_string_pro_epilogues
14. Enable use_cltd
16. Enable avoid_false_dep_for_bmi
17. Enable avoid_mfence
18. Disable expand_abs
19. Enable sse_typeless_stores
20. Enable sse_load0_by_pxor
21. Disable split_mem_opnd_for_fp_converts
22. Disable slow_pshufb
23. Enable partial_reg_dependency
This is the first patch to tune for Tremont. With all patches applied,
performance impacts on SPEC CPU 2017 are:
500.perlbench_r 1.81%
502.gcc_r 0.57%
505.mcf_r 1.16%
520.omnetpp_r 0.00%
523.xalancbmk_r 0.00%
525.x264_r 4.55%
531.deepsjeng_r 0.00%
541.leela_r 0.39%
548.exchange2_r 1.13%
557.xz_r 0.00%
geomean for intrate 0.95%
503.bwaves_r 0.00%
507.cactuBSSN_r 6.94%
508.namd_r 12.37%
510.parest_r 1.01%
511.povray_r 3.70%
519.lbm_r 36.61%
521.wrf_r 8.79%
526.blender_r 2.91%
527.cam4_r 6.23%
538.imagick_r 0.28%
544.nab_r 21.99%
549.fotonik3d_r 3.63%
554.roms_r -1.20%
geomean for fprate 7.50%
gcc/ChangeLog
* common/config/i386/i386-common.c: Use Haswell scheduling model
for Tremont.
* config/i386/i386.c (ix86_sched_init_global): Prepare for Tremont
scheduling pass.
* config/i386/x86-tune-sched.c (ix86_issue_rate): Change Tremont
issue rate to 4.
(ix86_adjust_cost): Handle Tremont.
* config/i386/x86-tune.def (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY):
Enable for Tremont.
(X86_TUNE_USE_LEAVE): Likewise.
(X86_TUNE_PUSH_MEMORY): Likewise.
(X86_TUNE_MISALIGNED_MOVE_STRING_PRO_EPILOGUES): Likewise.
(X86_TUNE_USE_CLTD): Likewise.
(X86_TUNE_AVOID_FALSE_DEP_FOR_BMI): Likewise.
(X86_TUNE_AVOID_MFENCE): Likewise.
(X86_TUNE_SSE_TYPELESS_STORES): Likewise.
(X86_TUNE_SSE_LOAD0_BY_PXOR): Likewise.
(X86_TUNE_ACCUMULATE_OUTGOING_ARGS): Disable for Tremont.
(X86_TUNE_FOUR_JUMP_LIMIT): Likewise.
(X86_TUNE_OPT_AGU): Likewise.
(X86_TUNE_AVOID_LEA_FOR_ADDR): Likewise.
(X86_TUNE_AVOID_MEM_OPND_FOR_CMOVE): Likewise.
(X86_TUNE_EXPAND_ABS): Likewise.
(X86_TUNE_SPLIT_MEM_OPND_FOR_FP_CONVERTS): Likewise.
(X86_TUNE_SLOW_PSHUFB): Likewise.
|
|
float32/float64/integer.
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm_undefined_ph):
New intrinsic.
(_mm256_undefined_ph): Likewise.
(_mm512_undefined_ph): Likewise.
(_mm_cvtsh_h): Likewise.
(_mm256_cvtsh_h): Likewise.
(_mm512_cvtsh_h): Likewise.
(_mm512_castph_ps): Likewise.
(_mm512_castph_pd): Likewise.
(_mm512_castph_si512): Likewise.
(_mm512_castph512_ph128): Likewise.
(_mm512_castph512_ph256): Likewise.
(_mm512_castph128_ph512): Likewise.
(_mm512_castph256_ph512): Likewise.
(_mm512_zextph128_ph512): Likewise.
(_mm512_zextph256_ph512): Likewise.
(_mm512_castps_ph): Likewise.
(_mm512_castpd_ph): Likewise.
(_mm512_castsi512_ph): Likewise.
* config/i386/avx512fp16vlintrin.h (_mm_castph_ps):
New intrinsic.
(_mm256_castph_ps): Likewise.
(_mm_castph_pd): Likewise.
(_mm256_castph_pd): Likewise.
(_mm_castph_si128): Likewise.
(_mm256_castph_si256): Likewise.
(_mm_castps_ph): Likewise.
(_mm256_castps_ph): Likewise.
(_mm_castpd_ph): Likewise.
(_mm256_castpd_ph): Likewise.
(_mm_castsi128_ph): Likewise.
(_mm256_castsi256_ph): Likewise.
(_mm256_castph256_ph128): Likewise.
(_mm256_castph128_ph256): Likewise.
(_mm256_zextph128_ph256): Likewise.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-typecast-1.c: New test.
* gcc.target/i386/avx512fp16-typecast-2.c: Ditto.
* gcc.target/i386/avx512fp16vl-typecast-1.c: Ditto.
* gcc.target/i386/avx512fp16vl-typecast-2.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm_cvtsh_ss):
New intrinsic.
(_mm_mask_cvtsh_ss): Likewise.
(_mm_maskz_cvtsh_ss): Likewise.
(_mm_cvtsh_sd): Likewise.
(_mm_mask_cvtsh_sd): Likewise.
(_mm_maskz_cvtsh_sd): Likewise.
(_mm_cvt_roundsh_ss): Likewise.
(_mm_mask_cvt_roundsh_ss): Likewise.
(_mm_maskz_cvt_roundsh_ss): Likewise.
(_mm_cvt_roundsh_sd): Likewise.
(_mm_mask_cvt_roundsh_sd): Likewise.
(_mm_maskz_cvt_roundsh_sd): Likewise.
(_mm_cvtss_sh): Likewise.
(_mm_mask_cvtss_sh): Likewise.
(_mm_maskz_cvtss_sh): Likewise.
(_mm_cvtsd_sh): Likewise.
(_mm_mask_cvtsd_sh): Likewise.
(_mm_maskz_cvtsd_sh): Likewise.
(_mm_cvt_roundss_sh): Likewise.
(_mm_mask_cvt_roundss_sh): Likewise.
(_mm_maskz_cvt_roundss_sh): Likewise.
(_mm_cvt_roundsd_sh): Likewise.
(_mm_mask_cvt_roundsd_sh): Likewise.
(_mm_maskz_cvt_roundsd_sh): Likewise.
* config/i386/i386-builtin-types.def
(V8HF_FTYPE_V2DF_V8HF_V8HF_UQI_INT,
V8HF_FTYPE_V4SF_V8HF_V8HF_UQI_INT,
V2DF_FTYPE_V8HF_V2DF_V2DF_UQI_INT,
V4SF_FTYPE_V8HF_V4SF_V4SF_UQI_INT): Add new builtin types.
* config/i386/i386-builtin.def: Add corrresponding new builtins.
* config/i386/i386-expand.c: Handle new builtin types.
* config/i386/sse.md (VF48_128): New mode iterator.
(avx512fp16_vcvtsh2<ssescalarmodesuffix><mask_scalar_name><round_saeonly_scalar_name>):
New.
(avx512fp16_vcvt<ssescalarmodesuffix>2sh<mask_scalar_name><round_scalar_name>):
Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm512_cvtph_pd):
New intrinsic.
(_mm512_mask_cvtph_pd): Likewise.
(_mm512_maskz_cvtph_pd): Likewise.
(_mm512_cvt_roundph_pd): Likewise.
(_mm512_mask_cvt_roundph_pd): Likewise.
(_mm512_maskz_cvt_roundph_pd): Likewise.
(_mm512_cvtxph_ps): Likewise.
(_mm512_mask_cvtxph_ps): Likewise.
(_mm512_maskz_cvtxph_ps): Likewise.
(_mm512_cvtx_roundph_ps): Likewise.
(_mm512_mask_cvtx_roundph_ps): Likewise.
(_mm512_maskz_cvtx_roundph_ps): Likewise.
(_mm512_cvtxps_ph): Likewise.
(_mm512_mask_cvtxps_ph): Likewise.
(_mm512_maskz_cvtxps_ph): Likewise.
(_mm512_cvtx_roundps_ph): Likewise.
(_mm512_mask_cvtx_roundps_ph): Likewise.
(_mm512_maskz_cvtx_roundps_ph): Likewise.
(_mm512_cvtpd_ph): Likewise.
(_mm512_mask_cvtpd_ph): Likewise.
(_mm512_maskz_cvtpd_ph): Likewise.
(_mm512_cvt_roundpd_ph): Likewise.
(_mm512_mask_cvt_roundpd_ph): Likewise.
(_mm512_maskz_cvt_roundpd_ph): Likewise.
* config/i386/avx512fp16vlintrin.h (_mm_cvtph_pd):
New intrinsic.
(_mm_mask_cvtph_pd): Likewise.
(_mm_maskz_cvtph_pd): Likewise.
(_mm256_cvtph_pd): Likewise.
(_mm256_mask_cvtph_pd): Likewise.
(_mm256_maskz_cvtph_pd): Likewise.
(_mm_cvtxph_ps): Likewise.
(_mm_mask_cvtxph_ps): Likewise.
(_mm_maskz_cvtxph_ps): Likewise.
(_mm256_cvtxph_ps): Likewise.
(_mm256_mask_cvtxph_ps): Likewise.
(_mm256_maskz_cvtxph_ps): Likewise.
(_mm_cvtxps_ph): Likewise.
(_mm_mask_cvtxps_ph): Likewise.
(_mm_maskz_cvtxps_ph): Likewise.
(_mm256_cvtxps_ph): Likewise.
(_mm256_mask_cvtxps_ph): Likewise.
(_mm256_maskz_cvtxps_ph): Likewise.
(_mm_cvtpd_ph): Likewise.
(_mm_mask_cvtpd_ph): Likewise.
(_mm_maskz_cvtpd_ph): Likewise.
(_mm256_cvtpd_ph): Likewise.
(_mm256_mask_cvtpd_ph): Likewise.
(_mm256_maskz_cvtpd_ph): Likewise.
* config/i386/i386-builtin.def: Add corresponding new builtins.
* config/i386/i386-builtin-types.def: Add corresponding builtin types.
* config/i386/i386-expand.c: Handle new builtin types.
* config/i386/sse.md
(VF4_128_8_256): New.
(VF48H_AVX512VL): Ditto.
(ssePHmode): Add HF vector modes.
(castmode): Add new convertable modes.
(qq2phsuff): Ditto.
(ph2pssuffix): New.
(avx512fp16_vcvt<castmode>2ph_<mode><mask_name><round_name>): Ditto.
(avx512fp16_vcvt<castmode>2ph_<mode>): Ditto.
(*avx512fp16_vcvt<castmode>2ph_<mode>): Ditto.
(avx512fp16_vcvt<castmode>2ph_<mode>_mask): Ditto.
(*avx512fp16_vcvt<castmode>2ph_<mode>_mask): Ditto.
(*avx512fp16_vcvt<castmode>2ph_<mode>_mask_1): Ditto.
(avx512fp16_float_extend_ph<mode>2<mask_name><round_saeonly_name>):
Ditto.
(avx512fp16_float_extend_ph<mode>2<mask_name>): Ditto.
(*avx512fp16_float_extend_ph<mode>2_load<mask_name>): Ditto.
(avx512fp16_float_extend_phv2df2<mask_name>): Ditto.
(*avx512fp16_float_extend_phv2df2_load<mask_name>): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm_cvttsh_i32):
New intrinsic.
(_mm_cvttsh_u32): Likewise.
(_mm_cvtt_roundsh_i32): Likewise.
(_mm_cvtt_roundsh_u32): Likewise.
(_mm_cvttsh_i64): Likewise.
(_mm_cvttsh_u64): Likewise.
(_mm_cvtt_roundsh_i64): Likewise.
(_mm_cvtt_roundsh_u64): Likewise.
* config/i386/i386-builtin.def: Add corresponding new builtins.
* config/i386/sse.md
(avx512fp16_fix<fixunssuffix>_trunc<mode>2<round_saeonly_name>):
New.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx512fp16-vcvttsh2si-1a.c: New test.
* gcc.target/i386/avx512fp16-vcvttsh2si-1b.c: Ditto.
* gcc.target/i386/avx512fp16-vcvttsh2si64-1a.c: Ditto.
* gcc.target/i386/avx512fp16-vcvttsh2si64-1b.c: Ditto.
* gcc.target/i386/avx512fp16-vcvttsh2usi-1a.c: Ditto.
* gcc.target/i386/avx512fp16-vcvttsh2usi-1b.c: Ditto.
* gcc.target/i386/avx512fp16-vcvttsh2usi64-1a.c: Ditto.
* gcc.target/i386/avx512fp16-vcvttsh2usi64-1b.c: Ditto.
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
vcvttph2w/vcvttph2uw/vcvttph2dq/vcvttph2qq/vcvttph2udq/vcvttph2uqq
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm512_cvttph_epi32):
New intrinsic.
(_mm512_mask_cvttph_epi32): Likewise.
(_mm512_maskz_cvttph_epi32): Likewise.
(_mm512_cvtt_roundph_epi32): Likewise.
(_mm512_mask_cvtt_roundph_epi32): Likewise.
(_mm512_maskz_cvtt_roundph_epi32): Likewise.
(_mm512_cvttph_epu32): Likewise.
(_mm512_mask_cvttph_epu32): Likewise.
(_mm512_maskz_cvttph_epu32): Likewise.
(_mm512_cvtt_roundph_epu32): Likewise.
(_mm512_mask_cvtt_roundph_epu32): Likewise.
(_mm512_maskz_cvtt_roundph_epu32): Likewise.
(_mm512_cvttph_epi64): Likewise.
(_mm512_mask_cvttph_epi64): Likewise.
(_mm512_maskz_cvttph_epi64): Likewise.
(_mm512_cvtt_roundph_epi64): Likewise.
(_mm512_mask_cvtt_roundph_epi64): Likewise.
(_mm512_maskz_cvtt_roundph_epi64): Likewise.
(_mm512_cvttph_epu64): Likewise.
(_mm512_mask_cvttph_epu64): Likewise.
(_mm512_maskz_cvttph_epu64): Likewise.
(_mm512_cvtt_roundph_epu64): Likewise.
(_mm512_mask_cvtt_roundph_epu64): Likewise.
(_mm512_maskz_cvtt_roundph_epu64): Likewise.
(_mm512_cvttph_epi16): Likewise.
(_mm512_mask_cvttph_epi16): Likewise.
(_mm512_maskz_cvttph_epi16): Likewise.
(_mm512_cvtt_roundph_epi16): Likewise.
(_mm512_mask_cvtt_roundph_epi16): Likewise.
(_mm512_maskz_cvtt_roundph_epi16): Likewise.
(_mm512_cvttph_epu16): Likewise.
(_mm512_mask_cvttph_epu16): Likewise.
(_mm512_maskz_cvttph_epu16): Likewise.
(_mm512_cvtt_roundph_epu16): Likewise.
(_mm512_mask_cvtt_roundph_epu16): Likewise.
(_mm512_maskz_cvtt_roundph_epu16): Likewise.
* config/i386/avx512fp16vlintrin.h (_mm_cvttph_epi32):
New intirnsic.
(_mm_mask_cvttph_epi32): Likewise.
(_mm_maskz_cvttph_epi32): Likewise.
(_mm256_cvttph_epi32): Likewise.
(_mm256_mask_cvttph_epi32): Likewise.
(_mm256_maskz_cvttph_epi32): Likewise.
(_mm_cvttph_epu32): Likewise.
(_mm_mask_cvttph_epu32): Likewise.
(_mm_maskz_cvttph_epu32): Likewise.
(_mm256_cvttph_epu32): Likewise.
(_mm256_mask_cvttph_epu32): Likewise.
(_mm256_maskz_cvttph_epu32): Likewise.
(_mm_cvttph_epi64): Likewise.
(_mm_mask_cvttph_epi64): Likewise.
(_mm_maskz_cvttph_epi64): Likewise.
(_mm256_cvttph_epi64): Likewise.
(_mm256_mask_cvttph_epi64): Likewise.
(_mm256_maskz_cvttph_epi64): Likewise.
(_mm_cvttph_epu64): Likewise.
(_mm_mask_cvttph_epu64): Likewise.
(_mm_maskz_cvttph_epu64): Likewise.
(_mm256_cvttph_epu64): Likewise.
(_mm256_mask_cvttph_epu64): Likewise.
(_mm256_maskz_cvttph_epu64): Likewise.
(_mm_cvttph_epi16): Likewise.
(_mm_mask_cvttph_epi16): Likewise.
(_mm_maskz_cvttph_epi16): Likewise.
(_mm256_cvttph_epi16): Likewise.
(_mm256_mask_cvttph_epi16): Likewise.
(_mm256_maskz_cvttph_epi16): Likewise.
(_mm_cvttph_epu16): Likewise.
(_mm_mask_cvttph_epu16): Likewise.
(_mm_maskz_cvttph_epu16): Likewise.
(_mm256_cvttph_epu16): Likewise.
(_mm256_mask_cvttph_epu16): Likewise.
(_mm256_maskz_cvttph_epu16): Likewise.
* config/i386/i386-builtin.def: Add new builtins.
* config/i386/sse.md
(avx512fp16_fix<fixunssuffix>_trunc<mode>2<mask_name><round_saeonly_name>):
New.
(avx512fp16_fix<fixunssuffix>_trunc<mode>2<mask_name>): Ditto.
(*avx512fp16_fix<fixunssuffix>_trunc<mode>2_load<mask_name>): Ditto.
(avx512fp16_fix<fixunssuffix>_truncv2di2<mask_name>): Ditto.
(avx512fp16_fix<fixunssuffix>_truncv2di2_load<mask_name>): Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm_cvtsh_i32): New intrinsic.
(_mm_cvtsh_u32): Likewise.
(_mm_cvt_roundsh_i32): Likewise.
(_mm_cvt_roundsh_u32): Likewise.
(_mm_cvtsh_i64): Likewise.
(_mm_cvtsh_u64): Likewise.
(_mm_cvt_roundsh_i64): Likewise.
(_mm_cvt_roundsh_u64): Likewise.
(_mm_cvti32_sh): Likewise.
(_mm_cvtu32_sh): Likewise.
(_mm_cvt_roundi32_sh): Likewise.
(_mm_cvt_roundu32_sh): Likewise.
(_mm_cvti64_sh): Likewise.
(_mm_cvtu64_sh): Likewise.
(_mm_cvt_roundi64_sh): Likewise.
(_mm_cvt_roundu64_sh): Likewise.
* config/i386/i386-builtin-types.def: Add corresponding builtin types.
* config/i386/i386-builtin.def: Add corresponding new builtins.
* config/i386/i386-expand.c (ix86_expand_round_builtin):
Handle new builtin types.
* config/i386/sse.md
(avx512fp16_vcvtsh2<sseintconvertsignprefix>si<rex64namesuffix><round_name>):
New define_insn.
(avx512fp16_vcvtsh2<sseintconvertsignprefix>si<rex64namesuffix>_2): Likewise.
(avx512fp16_vcvt<floatsuffix>si2sh<rex64namesuffix><round_name>): Likewise.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
Although this patch looks quite large, the changes are fairly minimal.
Most of it is duplicating the large function that does the overload
resolution using the automatically generated data structures instead of
the old hand-generated ones. This doesn't make the patch terribly easy to
review, unfortunately. Just be aware that generally we aren't changing
the logic and functionality of overload handling.
2021-09-16 Bill Schmidt <wschmidt@linux.ibm.com>
gcc/
* config/rs6000/rs6000-c.c (rs6000-builtins.h): New include.
(altivec_resolve_new_overloaded_builtin): New forward decl.
(rs6000_new_builtin_type_compatible): New function.
(altivec_resolve_overloaded_builtin): Call
altivec_resolve_new_overloaded_builtin.
(altivec_build_new_resolved_builtin): New function.
(altivec_resolve_new_overloaded_builtin): Likewise.
* config/rs6000/rs6000-call.c (rs6000_new_builtin_is_supported):
Likewise.
* config/rs6000/rs6000-gen-builtins.c (write_decls): Remove _p from
name of rs6000_new_builtin_is_supported.
|
|
After a recent change only a boolean value is returned.
2021-09-16 Uroš Bizjak <ubizjak@gmail.com>
gcc/
* config/i386/i386-protos.h (ix86_decompose_address):
Change return type to bool.
* config/i386/i386.c (ix86_decompose_address): Ditto.
|
|
This mimics what the main Makefile.in does: compile the generator
files under build (with Makefile.in's 'build/%.o' rule for compilation).
It also adds $(RUN_GEN) to optionally run it with valgrind and
the $(build_exeext) suffix.
Before, the .o files were compiled with $(COMPILE), causing link
error with $(LINKER_FOR_BUILD) for build != host.
gcc/
PR target/102353
* config/rs6000/t-rs6000 (build/rs6000-gen-builtins.o, build/rbtree.o):
Added 'build/' to target, use build/%.o rule.
(build/rs6000-gen-builtins$(build_exeext)): Add 'build/' and
'$(build_exeext)' to target and 'build/' for the *.o files.
(rs6000-builtins.c): Update for those changes; run rs6000-gen-builtins
with $(RUN_GEN).
|
|
The LEON5 can often dual issue instructions from the same 64-bit aligned
double word if there are no data dependencies. Add scheduling information
to avoid scheduling unpairable instructions back-to-back.
gcc/ChangeLog:
* config/sparc/sparc-opts.h (enum sparc_processor_type): Add LEON5
* config/sparc/sparc.c (struct processor_costs): Add LEON5 costs
(leon5_adjust_cost): Increase cost of store with data dependency
on ALU instruction and FPU anti-dependencies.
(sparc_option_override): Add LEON5 costs
(sparc_adjust_cost): Add LEON5 cost adjustments
* config/sparc/sparc.h: Add LEON5
* config/sparc/sparc.md: Include LEON5 scheduling information
* config/sparc/sparc.opt: Add LEON5
* doc/invoke.texi: Add LEON5
* config/sparc/leon5.md: New file.
|
|
This is needed to prevent the Store -> (Non-store or load) -> Store
sequence.
gcc/ChangeLog:
* config/sparc/sparc.md (stack_protect_set32): Add NOP to prevent
sensitive sequence for B2BST errata workaround.
|
|
A call to the function might have a load instruction in the delay slot
and a load followed by an atomic function could cause a deadlock.
gcc/ChangeLog:
* config/sparc/sparc.c (sparc_do_work_around_errata): Do not begin
functions with atomic instruction in the UT700 errata workaround.
|
|
This version detects multiple empty assembly statements in a row and also
detects non-memory barrier empty assembly statements (__asm__("")). It
can be used instead of next_active_insn().
gcc/ChangeLog:
* config/sparc/sparc.c (next_active_non_empty_insn): New function
that returns next active non empty assembly instruction.
(sparc_do_work_around_errata): Use new function.
|
|
Check the attribute of instruction to determine if it performs a store
or load operation. This more generic approach sees the last instruction
in the GOTdata_op model as a potential load and treats the memory barrier
as a potential store instruction.
gcc/ChangeLog:
* config/sparc/sparc.c (store_insn_p): Add predicate for store
attributes.
(load_insn_p): Add predicate for load attributes.
(sparc_do_work_around_errata): Use new predicates.
|
|
gcc/ChangeLog:
* config/sparc/sparc.c (dump_target_flag_bits): Print bit names for
LEON and LEON3.
|
|
gcc/ChangeLog:
* config/mips/netbsd.h: Fix typo in name of a macro.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm512_cvtepi32_ph): New
intrinsic.
(_mm512_mask_cvtepi32_ph): Likewise.
(_mm512_maskz_cvtepi32_ph): Likewise.
(_mm512_cvt_roundepi32_ph): Likewise.
(_mm512_mask_cvt_roundepi32_ph): Likewise.
(_mm512_maskz_cvt_roundepi32_ph): Likewise.
(_mm512_cvtepu32_ph): Likewise.
(_mm512_mask_cvtepu32_ph): Likewise.
(_mm512_maskz_cvtepu32_ph): Likewise.
(_mm512_cvt_roundepu32_ph): Likewise.
(_mm512_mask_cvt_roundepu32_ph): Likewise.
(_mm512_maskz_cvt_roundepu32_ph): Likewise.
(_mm512_cvtepi64_ph): Likewise.
(_mm512_mask_cvtepi64_ph): Likewise.
(_mm512_maskz_cvtepi64_ph): Likewise.
(_mm512_cvt_roundepi64_ph): Likewise.
(_mm512_mask_cvt_roundepi64_ph): Likewise.
(_mm512_maskz_cvt_roundepi64_ph): Likewise.
(_mm512_cvtepu64_ph): Likewise.
(_mm512_mask_cvtepu64_ph): Likewise.
(_mm512_maskz_cvtepu64_ph): Likewise.
(_mm512_cvt_roundepu64_ph): Likewise.
(_mm512_mask_cvt_roundepu64_ph): Likewise.
(_mm512_maskz_cvt_roundepu64_ph): Likewise.
(_mm512_cvtepi16_ph): Likewise.
(_mm512_mask_cvtepi16_ph): Likewise.
(_mm512_maskz_cvtepi16_ph): Likewise.
(_mm512_cvt_roundepi16_ph): Likewise.
(_mm512_mask_cvt_roundepi16_ph): Likewise.
(_mm512_maskz_cvt_roundepi16_ph): Likewise.
(_mm512_cvtepu16_ph): Likewise.
(_mm512_mask_cvtepu16_ph): Likewise.
(_mm512_maskz_cvtepu16_ph): Likewise.
(_mm512_cvt_roundepu16_ph): Likewise.
(_mm512_mask_cvt_roundepu16_ph): Likewise.
(_mm512_maskz_cvt_roundepu16_ph): Likewise.
* config/i386/avx512fp16vlintrin.h (_mm_cvtepi32_ph): New
intrinsic.
(_mm_mask_cvtepi32_ph): Likewise.
(_mm_maskz_cvtepi32_ph): Likewise.
(_mm256_cvtepi32_ph): Likewise.
(_mm256_mask_cvtepi32_ph): Likewise.
(_mm256_maskz_cvtepi32_ph): Likewise.
(_mm_cvtepu32_ph): Likewise.
(_mm_mask_cvtepu32_ph): Likewise.
(_mm_maskz_cvtepu32_ph): Likewise.
(_mm256_cvtepu32_ph): Likewise.
(_mm256_mask_cvtepu32_ph): Likewise.
(_mm256_maskz_cvtepu32_ph): Likewise.
(_mm_cvtepi64_ph): Likewise.
(_mm_mask_cvtepi64_ph): Likewise.
(_mm_maskz_cvtepi64_ph): Likewise.
(_mm256_cvtepi64_ph): Likewise.
(_mm256_mask_cvtepi64_ph): Likewise.
(_mm256_maskz_cvtepi64_ph): Likewise.
(_mm_cvtepu64_ph): Likewise.
(_mm_mask_cvtepu64_ph): Likewise.
(_mm_maskz_cvtepu64_ph): Likewise.
(_mm256_cvtepu64_ph): Likewise.
(_mm256_mask_cvtepu64_ph): Likewise.
(_mm256_maskz_cvtepu64_ph): Likewise.
(_mm_cvtepi16_ph): Likewise.
(_mm_mask_cvtepi16_ph): Likewise.
(_mm_maskz_cvtepi16_ph): Likewise.
(_mm256_cvtepi16_ph): Likewise.
(_mm256_mask_cvtepi16_ph): Likewise.
(_mm256_maskz_cvtepi16_ph): Likewise.
(_mm_cvtepu16_ph): Likewise.
(_mm_mask_cvtepu16_ph): Likewise.
(_mm_maskz_cvtepu16_ph): Likewise.
(_mm256_cvtepu16_ph): Likewise.
(_mm256_mask_cvtepu16_ph): Likewise.
(_mm256_maskz_cvtepu16_ph): Likewise.
* config/i386/i386-builtin-types.def: Add corresponding builtin types.
* config/i386/i386-builtin.def: Add corresponding new builtins.
* config/i386/i386-expand.c
(ix86_expand_args_builtin): Handle new builtin types.
(ix86_expand_round_builtin): Ditto.
* config/i386/i386-modes.def: Declare V2HF and V6HF.
* config/i386/sse.md (VI2H_AVX512VL): New.
(qq2phsuff): Ditto.
(sseintvecmode): Add HF vector modes.
(avx512fp16_vcvt<floatsuffix><sseintconvert>2ph_<mode><mask_name><round_name>):
New.
(avx512fp16_vcvt<floatsuffix><sseintconvert>2ph_<mode>): Ditto.
(*avx512fp16_vcvt<floatsuffix><sseintconvert>2ph_<mode>): Ditto.
(avx512fp16_vcvt<floatsuffix><sseintconvert>2ph_<mode>_mask): Ditto.
(*avx512fp16_vcvt<floatsuffix><sseintconvert>2ph_<mode>_mask): Ditto.
(*avx512fp16_vcvt<floatsuffix><sseintconvert>2ph_<mode>_mask_1): Ditto.
(avx512fp16_vcvt<floatsuffix>qq2ph_v2di): Ditto.
(*avx512fp16_vcvt<floatsuffix>qq2ph_v2di): Ditto.
(avx512fp16_vcvt<floatsuffix>qq2ph_v2di_mask): Ditto.
(*avx512fp16_vcvt<floatsuffix>qq2ph_v2di_mask): Ditto.
(*avx512fp16_vcvt<floatsuffix>qq2ph_v2di_mask_1): Ditto.
* config/i386/subst.md (round_qq2phsuff): New subst_attr.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|
|
gcc/ChangeLog:
* config/i386/avx512fp16intrin.h (_mm512_cvtph_epi32):
New intrinsic/
(_mm512_mask_cvtph_epi32): Likewise.
(_mm512_maskz_cvtph_epi32): Likewise.
(_mm512_cvt_roundph_epi32): Likewise.
(_mm512_mask_cvt_roundph_epi32): Likewise.
(_mm512_maskz_cvt_roundph_epi32): Likewise.
(_mm512_cvtph_epu32): Likewise.
(_mm512_mask_cvtph_epu32): Likewise.
(_mm512_maskz_cvtph_epu32): Likewise.
(_mm512_cvt_roundph_epu32): Likewise.
(_mm512_mask_cvt_roundph_epu32): Likewise.
(_mm512_maskz_cvt_roundph_epu32): Likewise.
(_mm512_cvtph_epi64): Likewise.
(_mm512_mask_cvtph_epi64): Likewise.
(_mm512_maskz_cvtph_epi64): Likewise.
(_mm512_cvt_roundph_epi64): Likewise.
(_mm512_mask_cvt_roundph_epi64): Likewise.
(_mm512_maskz_cvt_roundph_epi64): Likewise.
(_mm512_cvtph_epu64): Likewise.
(_mm512_mask_cvtph_epu64): Likewise.
(_mm512_maskz_cvtph_epu64): Likewise.
(_mm512_cvt_roundph_epu64): Likewise.
(_mm512_mask_cvt_roundph_epu64): Likewise.
(_mm512_maskz_cvt_roundph_epu64): Likewise.
(_mm512_cvtph_epi16): Likewise.
(_mm512_mask_cvtph_epi16): Likewise.
(_mm512_maskz_cvtph_epi16): Likewise.
(_mm512_cvt_roundph_epi16): Likewise.
(_mm512_mask_cvt_roundph_epi16): Likewise.
(_mm512_maskz_cvt_roundph_epi16): Likewise.
(_mm512_cvtph_epu16): Likewise.
(_mm512_mask_cvtph_epu16): Likewise.
(_mm512_maskz_cvtph_epu16): Likewise.
(_mm512_cvt_roundph_epu16): Likewise.
(_mm512_mask_cvt_roundph_epu16): Likewise.
(_mm512_maskz_cvt_roundph_epu16): Likewise.
* config/i386/avx512fp16vlintrin.h (_mm_cvtph_epi32):
New intrinsic.
(_mm_mask_cvtph_epi32): Likewise.
(_mm_maskz_cvtph_epi32): Likewise.
(_mm256_cvtph_epi32): Likewise.
(_mm256_mask_cvtph_epi32): Likewise.
(_mm256_maskz_cvtph_epi32): Likewise.
(_mm_cvtph_epu32): Likewise.
(_mm_mask_cvtph_epu32): Likewise.
(_mm_maskz_cvtph_epu32): Likewise.
(_mm256_cvtph_epu32): Likewise.
(_mm256_mask_cvtph_epu32): Likewise.
(_mm256_maskz_cvtph_epu32): Likewise.
(_mm_cvtph_epi64): Likewise.
(_mm_mask_cvtph_epi64): Likewise.
(_mm_maskz_cvtph_epi64): Likewise.
(_mm256_cvtph_epi64): Likewise.
(_mm256_mask_cvtph_epi64): Likewise.
(_mm256_maskz_cvtph_epi64): Likewise.
(_mm_cvtph_epu64): Likewise.
(_mm_mask_cvtph_epu64): Likewise.
(_mm_maskz_cvtph_epu64): Likewise.
(_mm256_cvtph_epu64): Likewise.
(_mm256_mask_cvtph_epu64): Likewise.
(_mm256_maskz_cvtph_epu64): Likewise.
(_mm_cvtph_epi16): Likewise.
(_mm_mask_cvtph_epi16): Likewise.
(_mm_maskz_cvtph_epi16): Likewise.
(_mm256_cvtph_epi16): Likewise.
(_mm256_mask_cvtph_epi16): Likewise.
(_mm256_maskz_cvtph_epi16): Likewise.
(_mm_cvtph_epu16): Likewise.
(_mm_mask_cvtph_epu16): Likewise.
(_mm_maskz_cvtph_epu16): Likewise.
(_mm256_cvtph_epu16): Likewise.
(_mm256_mask_cvtph_epu16): Likewise.
(_mm256_maskz_cvtph_epu16): Likewise.
* config/i386/i386-builtin-types.def: Add new builtin types.
* config/i386/i386-builtin.def: Add new builtins.
* config/i386/i386-expand.c
(ix86_expand_args_builtin): Handle new builtin types.
(ix86_expand_round_builtin): Ditto.
* config/i386/sse.md (sseintconvert): New.
(ssePHmode): Ditto.
(UNSPEC_US_FIX_NOTRUNC): Ditto.
(sseintconvertsignprefix): Ditto.
(avx512fp16_vcvtph2<sseintconvertsignprefix><sseintconvert>_<mode><mask_name><round_name>):
Ditto.
gcc/testsuite/ChangeLog:
* gcc.target/i386/avx-1.c: Add test for new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/sse-14.c: Add test for new intrinsics.
* gcc.target/i386/sse-22.c: Ditto.
|