Li, Aubrey wrote: >> --- Comment #1 from Rafael Vanoni <rafael.vanoni at sun.com> 2009-04-14 >> 21:05:40 --- My putback for >> >> 6828678 PowerTOP has erroneous use of memset(3C) >> >> had an error. One of the calls to memset remained with the incorrect >> value: >> >> 266 (void) memset(g_event_info, 0, >> 267 EVENT_NUM_MAX * sizeof (event_info_t)); >> 268 (void) memset(g_cstate_info, NSTATES >> 269 * sizeof (state_info_t), 0); >> >> But the same putback removed the code in cpu_idle.c that was manually >> zeroing the g_cstate_info array item by item. >> >> Apologies for the mistake, I'll putback a fix first thing tomorrow. >> > Hi Rafael, > > Manually patch modification is not a good idea, I also noticed > the string g_msg_idle_state is
The c-state error was a mismerge, I should've double checked the diffs before pushing. > "C-state (idle power state)" > > in the powertop gate while is > > "C-state (idle power states)" > > in the ONNV gate. There are also other difference. What other differences ? I just diffed the code in nevada from a clone of the project gate and didn't find anything aside from the c-state issue and the idle and frequency messages. Rafael
