Projects / ATLAS / Releases / Bugfixes

RSS All releases tagged Bugfixes

  •  10 Nov 2013 20:03

Release Notes: Major refactoring of code. Removal of blocks for better threading and superior performance. The new GEMM tuning framework has been extended to all types and precisions. Support for newer chips, such as Intel's Haswell (Corei3), has been added. Assorted bugfixes. Partial migration of the tuning defs to the new architecture.

Release Notes: This release fixes rare bugs in the ATL_syr, ATL_her, ATL_syr2, and ATL_her2 L2 BLAS functions, adds GEMV accelerators for TRMV, TRSV, SYMV, and HEMV for large problems, adds code to recognize IvyBridge as Corei2, applies fixes for ARM GEMM kernels, updates documentation, and tweaks the Windows DLL build.

Release Notes: This release builds under mingw for the Windows 32bit platform, adds assorted bugfixes, and removes unsupported archdefs.

Release Notes: This release added the ability to specify which processors to use when tuning and fixes issue with virtual processors. Bugs in POWER7 support were fixed, and all compliance checks now pass. GCC 4.7.0 support was improved. ATLAS can now be compiled with MSVC++. ARM assembly was cleaned and ARM NEON support was improved. Basic support was provided for Intel's compilers. Threads and 64-bit support in Windows are now both native and longer need Cygwin.

Release Notes: This release fixes a bug where non-x86 architectures couldn't build threaded libraries. It fixes optimization flags for gfortran. It adds archdefs for HAMMER32SSE3, Corei264SSE3, and POWER764VSX. It updates power7 gcc flags to work with gcc 4.6.2. It fixes a compile problem with newer versions of OS X due to changes in the toolchain.

Release Notes: This release improves ranking of possible gccs during configure. It fixes a buffer overrun in config.c which caused seg fault on Windows and fixes stack overwrites in several modules. It adds archdefs for PPCG432AltiVec, USIII64, Corei232AVX, Corei132SSE3, Core232SSE3, PPCG532AltiVec, AMDDozer, P4E64SSE3, and HAMMER64SSE3, adds code to enforce in-order writes or not use PCA for weakly-ordered memory systems like IA64, PPC, POWER, and ARM, and adds a probe and kernels for the FMA4 ISA extension.

  •  31 Jan 2012 22:46

Release Notes: This release updates the API of the test harness to work with current LAPACK. It reworks LAPACK bindings accordingly (which breaks compatibility with versions of LAPACK prior to 3.4.0). There are assorted bugfixes.

Release Notes: This release fixes errors in TRSM for non-SSE/AVX kernels. It adds a BETA=0 case to the AVX cgemvT kernel. It fixes a divide-by-zero error for small threaded SYRK. New archdefs for AMD64K10h64SSE3. It gets rid of obsolete PowerPC archdefs. It changes archinfo so it recognizes model 46 or Xeon X7560 as Corei1. There are other minor fixes.

Release Notes: Fixed to work with LAPack 3.4.0. Fixes for buffer overruns in configuration. The addition of the AVX cgemvT kernel. New Corei264AVX architecture definitions for gcc 4.6.2.

Release Notes: This release improves automatic selection of compilers and libraries, to produce the best results. It adds further development and bugfixes to the multithreading support.

Screenshot

Project Spotlight

PPCG

A polyhedral parallel code generator.

Screenshot

Project Spotlight

gpgpwd

A commandline password manager based around GnuPG.