OpenBLAS/kernel
Jerome Robert e9d9a8eae3 Allow to do gemv and ger buffer allocation on the stack
ger and gemv call blas_memory_alloc/free which in their turn
call blas_lock. blas_lock create thread contention when matrices
are small and the number of thread is high enough. We avoid
call blas_memory_alloc by replacing it with stack allocation.
This can be enabled with:
make -DMAX_STACK_ALLOC=2048
The given size (in byte) must be high enough to avoid thread contention
and small enough to avoid stack overflow.

Fix #478
2014-12-27 14:33:12 +01:00
..
alpha Remove all trailing whitespace except lapack-netlib 2014-06-27 12:05:18 -07:00
arm add reference in C for symv_U 2014-08-16 13:52:50 +02:00
arm64 # The first commit's message is: 2014-11-11 22:19:23 +08:00
generic Update organization info. 2014-11-25 15:28:58 +08:00
ia64 Remove all trailing whitespace except lapack-netlib 2014-06-27 12:05:18 -07:00
mips64 Update organization info. 2014-11-25 15:28:58 +08:00
power Remove all trailing whitespace except lapack-netlib 2014-06-27 12:05:18 -07:00
sparc Remove all trailing whitespace except lapack-netlib 2014-06-27 12:05:18 -07:00
x86 Lapack-test Windows 32bit now error free 2014-07-10 11:01:47 +02:00
x86_64 Allow to do gemv and ger buffer allocation on the stack 2014-12-27 14:33:12 +01:00
Makefile Remove all trailing whitespace except lapack-netlib 2014-06-27 12:05:18 -07:00
Makefile.L1 Fixed #395. Enable optimized cgemm for Sandybridge. Added optimized sdot kernel. 2014-06-29 10:34:51 +08:00
Makefile.L2 Remove all trailing whitespace except lapack-netlib 2014-06-27 12:05:18 -07:00
Makefile.L3 Refs #467. Added generic kernel file for x86_64. 2014-11-24 15:34:48 +08:00
Makefile.LA Support NO_LAPACK=1 to build the lib without LAPACK functions. 2011-03-04 11:51:32 +08:00
setparam-ref.c enabled use of GEMM3M functions 2014-09-20 14:27:10 +02:00