[Profiling Background] Now we are profiling the performance of ext4 and f2fs on an eMMC card with iozone, we find a case that ext4 is better than f2fs in random write under the test of "iozone -s 262144 -r 64 -i 0 -i 2". We want to analyze the I/O delay of the two file systems. We have got a conclusion that 1% of sys_write takes up 60% time of the overall sys_write (262144/64=4096). We want to find out the call stack during this specific 1% sys_write. Our idea is to record the stack in a certain time period and since the specific 1% case takes up 60% time, the total number of records of its stack should also takes up 60% of the total records, then we can recognize those stacks and figure out what the I/O stack of f2fs is doing in the 1% case.
[Profiling Problem] Although perf can record the events (with call stack) of a specified pid, e.g. using "perf record -g iozone -s 262144 -r 64 -i 0 -i 2". But we find iozone is interrupted and the CPU is scheduled to other process. As a result, perf will not record any events of iozone until iozone's context is recovered and the CPU is scheduled to continue processing the sys_write of iozone. This obeys our initial idea which is described in [Profiling Background], since we cannot recognize the call stacks of the specific 1% case by using the ratio of the record number. [Alternative Solution without Perf] We can use /proc/#pid/stack to record the stack in a certain period (e.g. 1ms) of iozone, no matter whether iozone is interrupted or not. However, we have not taken a deep sight into this, since we want to use perf to do this kind of thing. [Question about Perf] So we have a question that "How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process?" -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
