Patch from Dirk Clemens so df doesn't run numbers together on drives >100gig.

diff --git a/coreutils/df.c b/coreutils/df.c
index 51d3cc632fb9300c5a405b8d3915f8142475639b..ef08f91d476b5393da8b60e19c929c3307bafb82 100644 (file)
--- a/coreutils/df.c
+++ b/coreutils/df.c
@@ -136,7 +136,7 @@ extern int df_main(int argc, char **argv)
                        }
 
 #ifdef CONFIG_FEATURE_HUMAN_READABLE
-                       bb_printf("%-21s%9s ", device,
+                       bb_printf("%-20s %9s ", device,
                                          make_human_readable_str(s.f_blocks, s.f_bsize, df_disp_hr));
 
                        bb_printf("%9s ",
@@ -147,7 +147,7 @@ extern int df_main(int argc, char **argv)
                                          make_human_readable_str(s.f_bavail, s.f_bsize, df_disp_hr),
                                          blocks_percent_used, mount_point);
 #else
-                       bb_printf("%-21s%9ld %9ld %9ld %3ld%% %s\n",
+                       bb_printf("%-20s %9ld %9ld %9ld %3ld%% %s\n",
                                          device,
                                          kscale(s.f_blocks, s.f_bsize),
                                          kscale(s.f_blocks-s.f_bfree, s.f_bsize),