From 3e38f80ba5e42cd66d98692edd4cef7075661a5b Mon Sep 17 00:00:00 2001 From: Weiqun Zhang Date: Sat, 2 Jan 2021 09:25:05 -0800 Subject: [PATCH] Update CHANGES (#1693) --- CHANGES | 75 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 75 insertions(+) diff --git a/CHANGES b/CHANGES index 5c57036a255..c70fb68eb33 100644 --- a/CHANGES +++ b/CHANGES @@ -1,3 +1,78 @@ +# 21.01 + + -- Optimization of MLCellLinOp::applyBC (#1690) + + -- Optimization of iMultiFab sum (#1684) + + -- Sanitize the input of mkconfig (#1685) + + -- New versions of Reduce::Min, Max and MinMax taking callable (#1679) + + -- Only issue prefetch if arena uses managed memory (#1678) + + -- GNU Make: Fix the library flags (#1641) + + -- PolymorphicArray4, a thin layer around Array4 allowing us to access + both AoS and SoA data with operator() (#1643) + + -- CMake: dowstream projects must enable Fortran if AMReX_FORTRAN=ON + (#1669) + + -- Optimization of FaceLinear interpolater (#1663) + + -- Allow alternate form of "check_pair" function (#1667) + + -- ParallelFor for Reduction (#1658) + + -- Add features to fextract (#1657) + + -- Fix parallel add bug when both USE_GPU and USE_OMP are defined (#1654) + + -- Write a plotfile to ASCII (#1638) + + -- Fix NodalProjector::getGradPhi for multilevel case (#1642) + + -- Remove GPU pragma support (#1650) (#1649) + + -- Remove custom MFIter reduction code (#1648) + + -- Remove support for old versions of DPCPP (#1644) + + -- Fused FillBoundary (#1570) + + -- Divergence Free FillPatch (#1483) + + -- Optimization of constant coefficient MacProjector (#1636) + + -- Optimization of nodal solver with constant coefficient (#1629) + + -- Expose the index of the neighboring particle to users of the + iterator. (#1628) + + -- Implement overset mask for MLPoisson too (#1604) + + -- Fuse nodal jacobi smoother for GPU (#1625) + + -- Support for SoA data for neighbor particles. (#1590) + + -- Fix a bug that causes errors when writing HDF5 plotfile and particle + (#1621) + + -- Fortran Inteface: Porting more MultiFab functions (#1609) + + -- Remove alternative operators from AMReX (#1599) + + -- EB from STL (#1592) + + -- AMReX_Config.H generated at built time is included in all + headers. (#1566) + + -- MacProjector: Update setDomainBC API (#1589) + + -- Reimplement NumParticlesAtLevel (#1581) + + -- Add msvc cuda support (#1573) + # 20.12 -- EB2::PolyIF (#1567) The existing PolynomialIF contains a dynamic size