2013/1/14, Jaegeuk Kim <jaegeuk.kim@xxxxxxxxxxx>: > 2013-01-12 (토), 14:41 +0900, Namjae Jeon: >> From: Namjae Jeon <namjae.jeon@xxxxxxxxxxx> >> >> #> cat /sys/kernel/debug/tracing/events/f2fs/enable >> 0 >> #> echo 1 > /sys/kernel/debug/tracing/events/f2fs/enable >> #> cat /sys/kernel/debug/tracing/events/f2fs/enable >> 1 >> >> #> mount -t f2fs /dev/sdb3 /mnt/ >> #> >> #> cat /sys/kernel/debug/tracing/trace >> # tracer: nop >> # >> #TASK-PID CPU# TIMESTAMP FUNCTION >> # | | | | | >> mount-423 [001] 49.879188: f2fs_iget: dev 8,19 ino 2 >> mount-423 [001] 49.879215: f2fs_get_page: dev 8,19 page_index 81, >> type 2 >> mount-423 [001] 49.879224: f2fs_readpage: dev 8,19 ino 2 page_index >> 81 >> mount-423 [001] 49.879819: f2fs_get_page: dev 8,19 page_index 83, >> type 2 >> mount-423 [001] 49.879830: f2fs_readpage: dev 8,19 ino 2 page_index >> 83 >> mount-423 [001] 49.880434: f2fs_get_page: dev 8,19 page_index 593, >> type 2 >> mount-423 [001] 49.880442: f2fs_readpage: dev 8,19 ino 2 page_index >> 593 >> mount-423 [001] 49.889931: f2fs_get_page: dev 8,19 page_index 595, >> type 2 >> mount-423 [001] 49.889939: f2fs_readpage: dev 8,19 ino 2 page_index >> 595 >> mount-423 [001] 49.892169: f2fs_get_page: dev 8,19 page_index 594, >> type 2 >> mount-423 [001] 49.892186: f2fs_readpage: dev 8,19 ino 2 page_index >> 594 >> ... >> .. >> >> Signed-off-by: Namjae Jeon <namjae.jeon@xxxxxxxxxxx> >> Signed-off-by: Pankaj Kumar <pankaj.km@xxxxxxxxxxx> >> --- >> fs/f2fs/checkpoint.c | 17 + >> fs/f2fs/data.c | 17 + >> fs/f2fs/file.c | 14 + >> fs/f2fs/inode.c | 10 + >> fs/f2fs/namei.c | 7 +- >> fs/f2fs/node.c | 23 +- >> fs/f2fs/segment.c | 5 + >> fs/f2fs/super.c | 4 + >> include/trace/events/f2fs.h | 1220 >> +++++++++++++++++++++++++++++++++++++++++++ > > IMO, it needs to split this patch into two independent patches like: > 1. adding a header file to the event subsystem, and > 2. introducing tracepoints in f2fs. > > Then, we need to get acks from all the related maintainers. > > Moreover, it is obvious that the patch-set should describe each of > tracepoints precisely and concretely. > > Anyway, it must be a valuable job. :) > Thanks, Hi Jaegeuk. Okay, I see. I will split the patch as your suggestion. And I will check more performance side and the reason why we should add trace point here. Thanks. > > -- > Jaegeuk Kim > Samsung > -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
