From: Nathan Zimmer Date: Mon, 15 Apr 2013 14:53:33 +0000 (-0500) Subject: btreplay: Machines are now large enough that holes need to be dealt with X-Git-Tag: blktrace-1.1.0~9 X-Git-Url: https://git.kernel.dk/?a=commitdiff_plain;h=0c2df13d11dd91a1c3933dcc8faa98f2b59a5793;p=blktrace.git btreplay: Machines are now large enough that holes need to be dealt with The current method fails if once we hit the first offlined cpu. This will correct that case. However this still underreports the number cpus if the last cpu are offlined. Cc: Jens Axboe Signed-off-by: Nathan Zimmer Signed-off-by: Jens Axboe --- diff --git a/btreplay/btreplay.c b/btreplay/btreplay.c index 20494e0..6f0ba63 100644 --- a/btreplay/btreplay.c +++ b/btreplay/btreplay.c @@ -502,6 +502,7 @@ static inline void start_iter(void) */ static void get_ncpus(void) { + long last_cpu; cpu_set_t cpus; if (sched_getaffinity(getpid(), sizeof(cpus), &cpus)) { @@ -509,12 +510,11 @@ static void get_ncpus(void) /*NOTREACHED*/ } - /* - * XXX This assumes (perhaps wrongly) that there are no /holes/ - * XXX in the mask. - */ - for (ncpus = 0; ncpus < CPU_SETSIZE && CPU_ISSET(ncpus, &cpus); ncpus++) - ; + ncpus = -1; + for (last_cpu = 0; last_cpu < CPU_SETSIZE && CPU_ISSET(last_cpu, &cpus); last_cpu++) + if (CPU_ISSET( last_cpu, &cpus) ) + ncpus = last_cpu; + ncpus++; if (ncpus == 0) { fatal(NULL, ERR_SYSCALL, "Insufficient number of CPUs\n"); /*NOTREACHED*/