Here's a quick summary of the code review comments I've gotten. At
this point I consider the reviews done, unless someone else chimes in soon.
get_cpu_mstate()
move 2nd membar_consumer() higher
always assume NCMSTATES as nstates (remove nstates)
cpuvar.h
#include <sys/msacct.h> change not needed
interrupt.s
1110 stx %l1, [%o3 + %o4] ! delay - store new counter
~~~~~
cpu.c
comment before get_cpu_mstate() needs to explain how it's possible
to have decreasing values read from get_cpu_mstate()
usr/src/uts/i86pc/os/intr.c
intr_thread_prolog():294
How come intrtime gets set to "t->t_intr_start - tsc_read()"
and not "tsc_read() - t->t_intr_start" like everywhere else?
Is it intentional?
In machdep.c* we could make intracct[] as hrtime_t ?
uint64_t intracct[NCMSTATES];
-- Ethan
_______________________________________________
perf-discuss mailing list
[email protected]