Tuning the levmar implementation

There are probably many ways to tune the levmar program to the system you are running it on. I highlight three below that I have come across.

On line 54 of the file "misc.h" we have:
#define __BLOCKSZ__ 32 


The properties of the system I am using are shown below via the CPU-Z program.

Given my larger L1 cache I bumped up this value from 32 to 64 (64 * 64 * 8 = 32,768 bytes).

On line 683 of the file "lmbc_core.c" there are seven choices for different linear solvers. I selected the SVD option for its robustness.
issolved=AX_EQ_B_SVD(jacTjac, jacTe, Dp, m); ++nlss; linsolver=AX_EQ_B_SVD;

On line 885 of file "Axb_core.c" I opted for the GESDD LAPACK routine given that my system has lots of memory.

GESDD("A", (int *)&m, (int *)&m, NULL, (int *)&m, NULL, NULL, (int *)&m, NULL, (int *)&m, (LM_REAL *)&thresh, (int *)&worksz, NULL, &info);

If you know of more options that can fine tune the routine, please mention in the comments below.