On 05/07/2013 10:19 AM, Waiman Long wrote:
The slowdown that I was trying to fix was in the "perf record" part of the profiling process, not the "perf report" part. I didn't try perf-record on perf-record as the performance counters are limited resources and I don't want resource conflicts to affect the results. As the slow-down was entirely in the user space, I decided to use gprof to do the profiling.
Just to clarify that the post-processing time that I referred to in the patch is the time spent after the workload finishes in the perf-record session and the perf command itself quits. I could take a really long time for perf to quit without the patch.
Regards, Longman -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
