The hard X-ray detector (HXD) onboard {it Suzaku} measured soft $gamma$-rays from the Type Ia supernova SN2014J at $77pm2$ days after the explosion. Although the confidence level of the signal is about 90% (i.e., $2 sigma$), the $3 sigma$ upper limit has been derived at $< 2.2 times10^{-4}$ ph s$^{-1}$ cm$^{-2}$ in the 170 -- 250 keV band as the first independent measurement of soft $gamma$-rays with an instrument other than {it INTEGRAL}. For this analysis, we have examined the reproducibility of the NXB model of HXD/GSO using blank sky data. We find that the residual count rate in the 90 -- 500 keV band is distributed around an average of 0.19% with a standard deviation of 0.42% relative to the NXB rate. The averaged residual signals are consistent with that expected from the cosmic X-ray background. The flux of SN2014J derived from {it Suzaku} measurements taken in one snapshot at $t=77pm2$ days after the explosion is consistent with the {it INTEGRAL} values averaged over the period between $t=$50 and 100 days and also with explosion models of single or double degenerate scenarios. Being sensitive to the total ejecta mass surrounding the radioactive material, the ratio between continuum and line flux in the soft gamma-ray regime might distinguish different progenitor models. The {it Suzaku} data have been examined with this relation at $t=77pm2$ days, but could not distinguish models between single and double degenerate-progenitors. We disfavor explosion models with larger $^{56}$Ni masses than 1 $M_odot$, from our $1 sigma$ error on the 170-250 keV X-ray flux of $(1.2pm0.7) times10^{-4}$ ph s$^{-1}$ cm$^{-2}$.