On Thu, May 18, 2006 at 05:10:34PM +0200, Phil Carns wrote:
> Index: pvfs2_src/src/server/perf-mon.sm
> ===================================================================
> --- pvfs2_src/src/server/perf-mon.sm (revision 1541)
> +++ pvfs2_src/src/server/perf-mon.sm (revision 1542)
> @@ -86,6 +86,7 @@
> int valid_count = 0;
> uint32_t tmp_next_id;
> int ret = -1;
> + int idx;
>
> PINT_STATE_DEBUG("do_work");
>
> @@ -177,23 +178,24 @@
> {
> if(i<valid_count && static_start_time_array_ms[i] != 0)
> {
> - s_op->resp.u.mgmt_perf_mon.perf_array[i].valid_flag = 1;
> - s_op->resp.u.mgmt_perf_mon.perf_array[i].id = 0;
> - s_op->resp.u.mgmt_perf_mon.perf_array[i].id +=
> + idx = valid_count - i - 1;
> + s_op->resp.u.mgmt_perf_mon.perf_array[idx].valid_flag = 1;
> + s_op->resp.u.mgmt_perf_mon.perf_array[idx].id = 0;
> + s_op->resp.u.mgmt_perf_mon.perf_array[idx].id +=
> (uint32_t)(static_start_time_array_ms[i] % 1000000000);
Am i reading it right that you are now filling in
mgmt_perf_mon.perf_array backwards? Nothing in the code cares which
way you fill it in?
==rob
--
Rob Latham
Mathematics and Computer Science Division A215 0178 EA2D B059 8CDF
Argonne National Labs, IL USA B29D F333 664A 4280 315B
_______________________________________________
Pvfs2-developers mailing list
[email protected]
http://www.beowulf-underground.org/mailman/listinfo/pvfs2-developers
