Does subject labelling affect p-values of a lmer model in R?

Question

I encounter a problem while running an R script. The study is a within-subject study. I run a linear mixed-effect model:

lmer(var1 ~ cond1 * cond2 + (1|subj)