HGH LDA pseudopotential fails at ps_filter for Ti in the main branch

The debug trace says there is an error with the gsl library. This did not happen in previous versions of the code (for example commit df45245c was working, didn't try with the latest hotfix yet). I tried with other elements (C, O) and it works.

Here is the input file:

CalculationMode = gs
Dimensions = 3
Debug = trace
DebugTrapSignals = no

PseudopotentialSet = hgh_lda
%Coordinates
'Ti' | 0.0 | 0.0 | 0.0
%

Spacing = 0.1*angstrom
Radius = 6.0*angstrom

And this is the the last part of the output:

Input: [FilterPotentials = filter_TS]
gsl: interp.c:83: ERROR: x values must be strictly increasing
Default GSL error handler invoked.

Program received signal SIGABRT: Process abort signal.

Backtrace for this error:
#0  0x7fca5e264d5f in ???
	at ./signal/../sysdeps/unix/sysv/linux/x86_64/sigaction.c:0
#1  0x7fca5e264ce1 in __GI_raise
	at ../sysdeps/unix/sysv/linux/raise.c:51
#2  0x7fca5e24e536 in __GI_abort
	at ./stdlib/abort.c:79
#3  0x7fca5e49111c in ???
#4  0x7fca5e4bc4c2 in ???
#5  0x15558f2 in oct_spline_fit_
	at ../../src/math/spline_low.cc:57
#6  0x145f8c0 in __splines_oct_m_MOD_spline_fit
	at ../../src/math/splines.F90:522
#7  0x145d1f8 in __spline_filter_oct_m_MOD_spline_filter_mask
	at ../../src/math/spline_filter.F90:142
#8  0x1478c32 in __ps_oct_m_MOD_ps_filter
	at ../../src/species/ps.F90:718
#9  0xa0c880 in __species_oct_m_MOD_species_pot_init
	at ../../src/species/species.F90:970
#10  0xdd3f0d in __epot_oct_m_MOD_epot_init
	at ../../src/hamiltonian/epot.F90:162
#11  0xe81ad8 in __hamiltonian_elec_oct_m_MOD_hamiltonian_elec_init
	at ../../src/hamiltonian/hamiltonian_elec.F90:332
#12  0x55104f in __electrons_oct_m_MOD_electrons_init_parallelization
	at ../../src/electrons/electrons.F90:435
#13  0x40fb81 in __run_oct_m_MOD_run
	at ../../src/main/run.F90:175
#14  0x40eef9 in octopus
	at ../../src/main/main.F90:137
#15  0x40e88c in main
	at ../../src/main/main.F90:23