Before this commit, the "position < NUM_BUFFERS" loop condition from
blas_memory_free will be completely optimized away by GCC. This is
because the condition can only be false after undefined behavior has
already been invoked (reading past the end of an array). As a
consequence of this bug, GCC also removes the subsequent if statement
and all the code after the error label because all of it is dead.

This commit switches the loop condition around so it works as intended.

Ref. #995, may possibly help solve issues seen in 660,883

USE NPROCESSOR_CONF instaed of NPORCESSOR_ONLN

to determine the number of CPU. In ARM platform,
online CPU will increasing when there is more workload.
while configure cpu is the max number of CPU.

I cannot reproduce SEGFAULT of lapack-test with default stack size
on ARM Linux.

Let openblas_get_num_threads return the number of active threads

illumos fixes

... not the number of allocated threads.

Close #760

Conflicts:
	driver/others/memory.c

Conflicts:
	driver/others/memory.c

should fix issue #597

e.g.
make HOSTCC=gcc CC=arm-linux-androideabi-gcc NO_LAPACK=1 TARGET=ARMV7

In Android NDK, it uses armeabi-v7a-hard ABI.
TARGET_CFLAGS += -mhard-float -D_NDK_MATH_NO_SOFTFP=1
TARGET_LDFLAGS += -Wl,--no-warn-mismatch -lm_hard
For more information, please check hard-float example at
android_ndk/tests/device/hard-float/jni/.

Refactor such that you can build with ALLOC_SHM=0. HughTLB
implicity depends on ALLOC_SHM=1. This patch allows
building for Android NDK r10d.

For gemv_t, directly use malloc to create the buffer.

Fix race condition during shutdown causing a crash in
gotoblas_set_affinity().

get_nprocs() is a GNU convenience function equivalent to POSIX2008
sysconf(_SC_NPROCESSORS_ONLN); the latter should be available in unistd.h
on any current *nix. (OS X supports this call since 10.5, and FreeBSD
currently supports it. But this commit does not change FreeBSD or OS X
versions.)

Conflicts:
	driver/others/memory.c

Signed-off-by: Timothy Gu <timothygu99@gmail.com>

This reverts commit 3617c22a56.

The problem is the mutex we used in blas_server. Thus, we must clear
the mutex before the fork and re-init them at parent and child process.

If you used OpenMP, GOMP has the same problem by now. Please try other OpenMP
implemantation.