All releases tagged Bugfixes
Release Notes: Performance improvements to Real TRSM. Architectural defaults for Atom64SSE3. A bug in gerk assembly has been fixed.
Release Notes: QR-related routines were cleaned up. Improved performance for parallel LU and QR and for serial LQ and RQ. Numerous new platform defaults. Many bugfixes.
Release Notes: This release fixes errors in threaded GEMV and GER. It fixes building for experimental Win64 code. It has improved detection of clock speed under Cygwin. Minor fixes to sanity checks and test kit. Improved sanity check support for multi-threaded environments. Improved charting functions. A fix for a bug in threaded s/dGEMM that happens anytime the NB is not a multiple of 8.
Release Notes: This release adds the ability to autobuild performance charts in results. It adds all-alignment testing for GER and GEMV. It adds a results/ directory that can autobuild performance charts. Numerous fixes to qrtest and some fixes for the QR fact routines. A chapter in atlas_install explaining how to use mmflagsearch. A fix for an uninitialized memory read caused by copying data ATLAS doesn't reference in parallel GEMM. A fix for an uninitialized memory read in gemvT. This release changes extendedmodel=2, model=5 from Corei2 to Corei1 in archinfo_x86.
Release Notes: ARMv7 support and tunings have been added. Assorted bugs were fixed.
Release Notes: This release adds beta versions of simple threaded GEMV & GER. It adds threaded L2 testing to the tester. It fixes a bug in axpby where it called SCAL with alpha=0, which fixes GEMM error for BETA=0 case. It fixes several simple buffer overruns in the full tester. It adds dynamically scheduled tgemm, which is used whenever all dimensions are large. It adds support for complex types for both dynamic cases (rank-K, large). It fixes several errors in GEMM that occurred when K dim was cut.
Release Notes: Improved affinity setting. Fixes for numerous bugs.
Release Notes: Extensive bugfixes were made. Preliminary support was added for threaded/parallel lapack. The architecture definitions were updated for P4ESSE3, PPCG564AltiVec, Core264SSE, and AMD64K10h64SSE3. PCA codes were added for LU and QR.
Release Notes: Improved timers for Core264SSE3, AMD64K10h64SSE3, and Corei764SSE3. Bugfixes for symbols, unnecessary tests, and build problems on some platforms.
Release Notes: Repeated GER/GER2 symbols that prevented shared library building were fixed. A bug causing x_tfindCE/txover to use CPU time rather than wall time was fixed. R2K.sum files were added to archdef Makefile and to Core264 and k10h64 arch defs. Configure support was added for UltraSPARC T2. Affinity was turned off for T2, where it decreases parallel performance. Architectural defaults for UST264 and UST232 were added. Errors in GER2 handling were fixed. Other minor bugs were fixed.
