Thanks a lot!
>From the ANL bug reproducer side, there were 33 bug fixes and 15 regressions on PVC (Sunspot). The details are below. Please take a look if you submitted one of these bugs!
For applications, 3 tests fixed (OpenMC, HYPRE, ELPA/BGW-kernels, 5 tests regressed (GAMESS, E3SM/YAKL, QMCPACK, pFUnit,VTK-m). Details and workarounds are below,
Thanks,
Colleen
# Applications:
# Regressions:
GAMESS: compile hang and compile ICE (WA: -O0)
E3SM/YAKL: Compile error for specifying sub-group size (WA: remove -fsycl-default-sub-group)
QMCPACK: some unit tests produce incorrect value due to the CPU vectorizer (WA: -fp-precise or adding explicitly #pragma omp simd reduction(:+) where "required"
pFUnit: compile fail for "check -all" (WA: build Release)
VTK-m: compile fail, still under debug
# Bug reproducers:
# Fixes:
source/reproducers/dpcpp/VirtualFunction:VirtualFunction_exw-virtuals-app_ath_compile [CMPLRLLVM-35295,XDEPS-6157] P0
source/reproducers/dpcpp/VirtualFunction:VirtualFunction_minimal_ath_compile [CMPLRLLVM-35295,XDEPS-6157] P0
source/reproducers/dpcpp/VirtualFunction:VirtualFunction_minimal_ath_run [CMPLRLLVM-35295,XDEPS-6157] P0
source/reproducers/dpcpp/copy_race:copy_race_ath_run
source/reproducers/dpcpp/event_command_submit [CMPLRLLVM-35289,XDEPS-3796]
source/reproducers/dpcpp/hypre_onedpl [ONEDPL-729]
source/reproducers/dpcpp/oneDPL_incl_excl_scan_inplace:oneDPL_incl_excl_scan_inplace_larger [ONEDPL-490] P0
source/reproducers/dpcpp/sycl_free_memory [CMPLRLLVM-41689] P0
source/reproducers/dpcpp_ct/dpct_constant_iterator [CTST-3984]
source/reproducers/ifx/CMPLRLLVM-35621
source/reproducers/ifx/XDEPS-4611
source/reproducers/ifx/XDEPS-5141
source/reproducers/ifx/XDEPS-5191
source/reproducers/ifx/array_reduction_ice
source/reproducers/ifx/phasta_target_simd [CMPLRLLVM-35621,CMPLRLLVM-40195] P0
source/reproducers/ifx/red_enter_data:red_enter_data_ath_run [CMPLRLLVM-42763]
source/reproducers/ifx/red_implicit_scaling:red_implicit_scaling_ath_run
source/reproducers/ifx/threadprivate_ice [CMPLRLLVM-45498]
source/reproducers/mkl/bbfft-repro-e3sm [MKLD-15002]
source/reproducers/mkl/blas-sparse_usm:blas-sparse_usm_ath_run
source/reproducers/mkl/getri_strided_group_compare:getri_strided_group_compare_ath_run [MKLD-13828]
source/reproducers/openmp/atomics_testing:atomics_testing_cpp [CMPLRLLVM-46117,CMPLRLLVM-44907]
source/reproducers/openmp/dsps_gamess_mini_wrong_answer:dsps_gamess_mini_wrong_answer_offload_spd_ath_run
source/reproducers/openmp/gamess_mini_all_kernels:gamess_mini_all_kernels_j13 [XDEPS-5609,IGC-6246,CMPLRLLVM-45082]
source/reproducers/openmp/ifx_common_blocks:ifx_common_blocks_ath_run [CMPLRLLVM-42010]
source/reproducers/openmp/long_double [CMPLRLLVM-10445]
source/reproducers/openmp/multi_gpu_ms67:multi_gpu_ms67_ath_run
source/reproducers/openmp/openmc_segfault_jira:openmc_segfault_jira_ath_run [CMPLRLIBS-34379]
source/reproducers/openmp/simd_fortran_ice:simd_fortran_ice-simple [CMPLRLLVM-41961]
source/reproducers/openmp/simd_private_array [CMPLRLLVM-21649] P0
source/reproducers/openmp/simple_local_allocate:simple_local_allocate_omp_new [CMPLRLIBS-34061,CMPLRLLVM-37421,CMPLRLLVM-39430,CMPLRLLVM-42572] P0
source/reproducers/openmp/simple_local_allocate:simple_local_allocate_omp_new2 [CMPLRLIBS-34061,CMPLRLLVM-37421,CMPLRLLVM-39430,CMPLRLLVM-42572] P0
source/reproducers/openmp/split-parallel-for-reduction:split-parallel-for-reduction_gemv-case-4-1 [CMPLRLLVM-10985,CMPLRLLVM-43840,CMPLRLLVM-44357]
# Regressions:
source/reproducers/mkl/sparse_gemvdot:sparse_gemvdot_ath_run
source/reproducers/mkl/zgetrs_batch_slowdown [MKLD-15212]
source/reproducers/mpi/hdf5_paralle_incorrect:hdf5_paralle_incorrect_autoconf [HPCS-8110]
source/reproducers/openmp/ddpp_gamess_mini_wrong_answer_fp_precise
source/reproducers/openmp/ddpp_gamess_mini_wrong_answer_fp_precise:ddpp_gamess_mini_wrong_answer_fp_precise_no_fp_precise
source/reproducers/openmp/ddpp_gamess_mini_wrong_answer_fp_precise:ddpp_gamess_mini_wrong_answer_fp_precise_offload_disabled
source/reproducers/openmp/gamess_mini_all_kernels:gamess_mini_all_kernels_j18 [XDEPS-5609,IGC-6246,CMPLRLLVM-45082]
source/reproducers/openmp/gamess_mini_all_kernels:gamess_mini_all_kernels_j19 [XDEPS-5609,IGC-6246,CMPLRLLVM-45082]
source/reproducers/openmp/gamess_mini_wrong_answer:gamess_mini_wrong_answer_offload_ath_compile [CMPLRLLVM-45082]
source/reproducers/openmp/gamess_minispd_nan:gamess_minispd_nan_w2_ath_run [XDEPS-5607,XDEPS-6215,GSD-3402] P0
source/reproducers/openmp/hanging_gamess:hanging_gamess_ath_compile [CMPLRLLVM-39824,GSD-2102,XDEPS-5609]
source/reproducers/openmp/ifx-O2:ifx-O2_ath_compile [CMPLRLLVM-43398] P0
source/reproducers/openmp/ifx_flag_O2:ifx_flag_O2_ath_compile
source/reproducers/openmp/ifx_flag_O2:ifx_flag_O2_ath_run
source/reproducers/openmp/omp_offload_isnan:omp_offload_isnan_ath_compile [CMPLRLLVM-25014,CMPLRLLVM-25019,CMPLRLIBS-33270]
________________________________
From: Ecp-aurora-sdk-discuss <ecp-aurora-sdk-discuss-bounces(a)lists.jlse.anl.gov> on behalf of Chan-nui, Christopher via Ecp-aurora-sdk-discuss <ecp-aurora-sdk-discuss(a)lists.jlse.anl.gov>
Sent: Wednesday, April 12, 2023 12:10 PM
To: Ecp-aurora-sdk-discuss <ecp-aurora-sdk-discuss(a)lists.jlse.anl.gov>
Subject: [Ecp-aurora-sdk-discuss] Updating default engineering oneAPI module on JLSE/Sunspot
We are pleased to provide a new drop of engineering components to JLSE/Sunspot.
The new modulefile is currently available as:
module add oneapi/eng-compiler/2022.12.30.005
This is not the default environment.
Provided are updates to the compilers, dpcpp-ct, MKL, advisor, DAL and DPL.
--
Christopher Chan-Nui
christopher.chan-nui(a)intel.com
