The cosmic evolution of gamma-ray burst (GRB) luminosity is essential for revealing the GRB physics and for using GRBs as cosmological probes. We investigate the luminosity evolution of long GRBs with a large sample of 258 Swift/BAT GRBs. By describing the peak luminosity evolution of individual GRBs as L_p_{prop.to}(1+z)^k^, we get k=1.49+/-0.19 using the nonparametric {tau} statistics method without considering observational biases of GRB trigger and redshift measurement. By modeling these biases with the observed peak flux and characterizing the peak luminosity function of long GRBs as a smoothly broken power law with a break that evolves as L_b_{prop.to}(1+z)^kb^, we obtain k_b_=1.14^+0.99^_-0.47_ through simulations based on the assumption that the long GRB rate follows the star formation rate incorporating the cosmic metallicity history. The derived k and k_b_ values are systematically smaller than those reported in previous papers. By removing the observational biases of the GRB trigger and redshift measurement based on our simulation analysis, we generate mock complete samples of 258 and 1000 GRBs to examine how these biases affect the {tau} statistics method. We get k=0.94+/-0.14 and k=0.80+/-0.09 for the two samples, indicating that these observational biases may lead to overestimating the k value. With the large uncertainty of k_b_ derived from our simulation analysis, one cannot even convincingly argue for a robust evolution feature of the GRB luminosity.