* can we measure the traffic between user and kernel mode
@ 2011-01-27 3:42 loody
2011-01-27 4:41 ` Alexandre Courbot
` (2 more replies)
0 siblings, 3 replies; 14+ messages in thread
From: loody @ 2011-01-27 3:42 UTC (permalink / raw)
To: kernelnewbies
Dear all:
Is there any directory under proc/sys or tools that can help us to
measure the traffic between user and kernel mode?
(such as ioctl, open, close, etc.)
--
Regards,
miloody
^ permalink raw reply [flat|nested] 14+ messages in thread* can we measure the traffic between user and kernel mode 2011-01-27 3:42 can we measure the traffic between user and kernel mode loody @ 2011-01-27 4:41 ` Alexandre Courbot 2011-01-27 5:02 ` Mulyadi Santosa 2011-01-27 5:56 ` Michael Blizek 2 siblings, 0 replies; 14+ messages in thread From: Alexandre Courbot @ 2011-01-27 4:41 UTC (permalink / raw) To: kernelnewbies > Is there any directory under proc/sys or tools that can help us to > measure the traffic between user and kernel mode? > (such as ioctl, open, close, etc.) Not that I know of, but you can use strace in order to monitor system calls performed by a particular application. Alex. ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 3:42 can we measure the traffic between user and kernel mode loody 2011-01-27 4:41 ` Alexandre Courbot @ 2011-01-27 5:02 ` Mulyadi Santosa 2011-01-27 5:56 ` loody 2011-01-27 5:56 ` Michael Blizek 2 siblings, 1 reply; 14+ messages in thread From: Mulyadi Santosa @ 2011-01-27 5:02 UTC (permalink / raw) To: kernelnewbies On Thu, Jan 27, 2011 at 10:42, loody <miloody@gmail.com> wrote: > Dear all: > Is there any directory under proc/sys or tools that can help us to > measure the traffic between user and kernel mode? > (such as ioctl, open, close, etc.) By "traffic", you mean context switch latency? Perhaps you mean a tool like lmbench? -- regards, Mulyadi Santosa Freelance Linux trainer and consultant blog: the-hydra.blogspot.com training: mulyaditraining.blogspot.com ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 5:02 ` Mulyadi Santosa @ 2011-01-27 5:56 ` loody 2011-01-27 6:20 ` Mulyadi Santosa 0 siblings, 1 reply; 14+ messages in thread From: loody @ 2011-01-27 5:56 UTC (permalink / raw) To: kernelnewbies hi all: thanks for your kind reply :-) 2011/1/27 Mulyadi Santosa <mulyadi.santosa@gmail.com>: > On Thu, Jan 27, 2011 at 10:42, loody <miloody@gmail.com> wrote: >> Dear all: >> Is there any directory under proc/sys or tools that can help us to >> measure the traffic between user and kernel mode? >> (such as ioctl, open, close, etc.) > > By "traffic", you mean context switch latency? Perhaps you mean a tool > like lmbench? I got a user-mode application without symbol and source, and it keeps console no response about 2 mins. I just want to know whether it is busy on processing itself or do the context switch frequently for calling kernel drivers at that time. thanks a lot, miloody > > -- > regards, > > Mulyadi Santosa > Freelance Linux trainer and consultant > > blog: the-hydra.blogspot.com > training: mulyaditraining.blogspot.com > -- Regards, ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 5:56 ` loody @ 2011-01-27 6:20 ` Mulyadi Santosa 2011-01-27 11:56 ` Pravin Shedage 2011-01-27 13:48 ` loody 0 siblings, 2 replies; 14+ messages in thread From: Mulyadi Santosa @ 2011-01-27 6:20 UTC (permalink / raw) To: kernelnewbies On Thu, Jan 27, 2011 at 12:56, loody <miloody@gmail.com> wrote: > I got a user-mode application without symbol and source, and it keeps > console no response about 2 mins. > I just want to know whether it is busy on processing itself or do the > context switch frequently for calling kernel drivers at that time. well then, strace/ltrace is probably the right tool to do the job :) or just use /proc/<pid>/stack :) -- regards, Mulyadi Santosa Freelance Linux trainer and consultant blog: the-hydra.blogspot.com training: mulyaditraining.blogspot.com ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 6:20 ` Mulyadi Santosa @ 2011-01-27 11:56 ` Pravin Shedage 2011-01-27 13:48 ` loody 1 sibling, 0 replies; 14+ messages in thread From: Pravin Shedage @ 2011-01-27 11:56 UTC (permalink / raw) To: kernelnewbies Hi, You can use *top* command to know traffic of processes in your cpu. another if you want to debug that running application then you can also try gdb # gdb --pid <pid_of_application> Thanks & Regards, -------PraviN------- On Thu, Jan 27, 2011 at 11:50 AM, Mulyadi Santosa <mulyadi.santosa@gmail.com > wrote: > On Thu, Jan 27, 2011 at 12:56, loody <miloody@gmail.com> wrote: > > I got a user-mode application without symbol and source, and it keeps > > console no response about 2 mins. > > I just want to know whether it is busy on processing itself or do the > > context switch frequently for calling kernel drivers at that time. > > well then, strace/ltrace is probably the right tool to do the job :) > or just use /proc/<pid>/stack :) > > -- > regards, > > Mulyadi Santosa > Freelance Linux trainer and consultant > > blog: the-hydra.blogspot.com > training: mulyaditraining.blogspot.com > > _______________________________________________ > Kernelnewbies mailing list > Kernelnewbies at kernelnewbies.org > http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies > -- Thanks & Regards, ---------PraviN--------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20110127/a8ce56e6/attachment.html ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 6:20 ` Mulyadi Santosa 2011-01-27 11:56 ` Pravin Shedage @ 2011-01-27 13:48 ` loody 2011-01-27 17:45 ` Mulyadi Santosa 2011-01-28 1:07 ` Alexandre Courbot 1 sibling, 2 replies; 14+ messages in thread From: loody @ 2011-01-27 13:48 UTC (permalink / raw) To: kernelnewbies hi all: 2011/1/27 Mulyadi Santosa <mulyadi.santosa@gmail.com>: > On Thu, Jan 27, 2011 at 12:56, loody <miloody@gmail.com> wrote: >> I got a user-mode application without symbol and source, and it keeps >> console no response about 2 mins. >> I just want to know whether it is busy on processing itself or do the >> context switch frequently for calling kernel drivers at that time. > > well then, strace/ltrace is probably the right tool to do the job :) > or just use /proc/<pid>/stack :) I have tried the suggestions you gave. :) Can I combine all the message together? Supposedly, my console shows: A begins A finish B begins B finish and I measure 2mins from '#A finish' to "#B begins". I can "strace -o -tt 123 a.out" before "A begin" but how could I combine console output with the content of strace? such as A begins strace content ...... A finish strace content ...... B begins strace content ...... B finish appreciate your kind help :-) miloody ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 13:48 ` loody @ 2011-01-27 17:45 ` Mulyadi Santosa 2011-01-30 7:50 ` loody 2011-01-28 1:07 ` Alexandre Courbot 1 sibling, 1 reply; 14+ messages in thread From: Mulyadi Santosa @ 2011-01-27 17:45 UTC (permalink / raw) To: kernelnewbies Hi :) On Thu, Jan 27, 2011 at 20:48, loody <miloody@gmail.com> wrote: > I can "strace -o -tt 123 a.out" before "A begin" but how could I > combine console output with the content of strace? > such as > A begins > strace content ...... > A finish > strace content ...... > B begins > strace content ...... > B finish Not sure if I understand you correctly...what are A and B here? function calls? -- regards, Mulyadi Santosa Freelance Linux trainer and consultant blog: the-hydra.blogspot.com training: mulyaditraining.blogspot.com ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 17:45 ` Mulyadi Santosa @ 2011-01-30 7:50 ` loody 0 siblings, 0 replies; 14+ messages in thread From: loody @ 2011-01-30 7:50 UTC (permalink / raw) To: kernelnewbies hi: 2011/1/28 Mulyadi Santosa <mulyadi.santosa@gmail.com>: > Hi :) > > On Thu, Jan 27, 2011 at 20:48, loody <miloody@gmail.com> wrote: >> I can "strace -o -tt 123 a.out" before "A begin" but how could I >> combine console output with the content of strace? >> such as >> A begins >> strace content ...... >> A finish >> strace content ...... >> B begins >> strace content ...... >> B finish > > Not sure if I understand you correctly...what are A and B here? function calls? Yes, the A and B here I mean is the function call I wrote in C, such as the "open("/home/miloody", O_RDONLY) " etc. Thank you ~ miloody ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 13:48 ` loody 2011-01-27 17:45 ` Mulyadi Santosa @ 2011-01-28 1:07 ` Alexandre Courbot 2011-01-30 8:07 ` loody 1 sibling, 1 reply; 14+ messages in thread From: Alexandre Courbot @ 2011-01-28 1:07 UTC (permalink / raw) To: kernelnewbies > I can "strace -o -tt 123 a.out" before "A begin" but how could I > combine console output with the content of strace? strace does output to stderr unless you specify the -o option. So by using something like $ strace -tt your_program or $ strace -tt your_program &> log.txt you will have both outputs intermingled. Note that the console output of your program will appear in the middle of the corresponding write() system call. Alex. ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-28 1:07 ` Alexandre Courbot @ 2011-01-30 8:07 ` loody 2011-01-31 1:38 ` Alexandre Courbot 0 siblings, 1 reply; 14+ messages in thread From: loody @ 2011-01-30 8:07 UTC (permalink / raw) To: kernelnewbies hi: 2011/1/28 Alexandre Courbot <gnurou@gmail.com>: >> I can "strace -o -tt 123 a.out" before "A begin" but how could I >> combine console output with the content of strace? > > strace does output to stderr unless you specify the -o option. So by > using something like > > $ strace -tt your_program > > or > > $ strace -tt your_program &> log.txt > > you will have both outputs intermingled. Note that the console output > of your program will appear in the middle of the corresponding write() > system call. I have tried both ways you mentioned. in case 1: it did interleave the output of standard output with standard error. in case 2: the standard output seems will be collect for flush at later, combine.strace.txt. (I attach the testing result for your reference.) theoretically, file re-direction should not be effect the content of output, right? appreciate your help :-) miloody. -------------- next part -------------- A non-text attachment was scrubbed... Name: strace.output.tar.bz2 Type: application/x-bzip2 Size: 1882 bytes Desc: not available Url : http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20110130/5b02b45e/attachment.bz2 ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-30 8:07 ` loody @ 2011-01-31 1:38 ` Alexandre Courbot 2011-01-31 1:40 ` Alexandre Courbot 0 siblings, 1 reply; 14+ messages in thread From: Alexandre Courbot @ 2011-01-31 1:38 UTC (permalink / raw) To: kernelnewbies > I have tried both ways you mentioned. > in case 1: > it did interleave the output of standard output with standard error. > in case 2: > the standard output seems will be collect for flush at later, > combine.strace.txt. > (I attach the testing result for your reference.) > theoretically, file re-direction should not be effect the content of > output, right? That's what I thought too, but it turns out the libc is smarter than we expected. File redirection *does* affect the order of output, as testified by your logs. See the non-redirected behavior: 15:55:35.205409 write(1, "\n", 1) = 1 15:55:35.205517 write(1, "open fail\n", 10) = 10 15:55:35.205628 open("/home/miloody", O_RDONLY) = -1 ENOENT (No such file or directory) 15:55:35.205745 write(1, "\n", 1) = 1 15:55:35.205847 write(1, "open fail\n", 10) = 10 15:55:35.205957 open("/home/miloody", O_RDONLY) = -1 ENOENT (No such file or directory) 15:55:35.206073 write(1, "\n", 1) = 1 15:55:35.206174 write(1, "open fail\n", 10) = 10 15:55:35.206284 open("/home/miloody", O_RDONLY) = -1 ENOENT (No such file or directory) ... and compare it to what happens when we use redirection: ... 15:56:21.105236 open("/home/miloody", O_RDONLY) = -1 ENOENT (No such file or directory) 15:56:21.105294 open("/home/miloody", O_RDONLY) = -1 ENOENT (No such file or directory) 15:56:21.105353 open("/home/miloody", O_RDONLY) = -1 ENOENT (No such file or directory) 15:56:21.105417 write(1, "\nopen fail\n\nopen fail\n\nopen fail"..., 110 Here the libc has grouped all your printfs into one single system call because of user-space buffering. Fortunately you can prevent this behavior using unbuffer: http://linuxcommand.org/man_pages/unbuffer1.html So your strace invokation would be like: $ strace -tt stdbuf -o0 -e0 ./foo &>output Which produces the expected output on my system. Alex. ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-31 1:38 ` Alexandre Courbot @ 2011-01-31 1:40 ` Alexandre Courbot 0 siblings, 0 replies; 14+ messages in thread From: Alexandre Courbot @ 2011-01-31 1:40 UTC (permalink / raw) To: kernelnewbies > Here the libc has grouped all your printfs into one single system call > because of user-space buffering. Fortunately you can prevent this > behavior using unbuffer: > > http://linuxcommand.org/man_pages/unbuffer1.html Sorry, that should have read stdbuf, and the correct man page is here: http://www.pixelbeat.org/programming/stdio_buffering/stdbuf-man.html Alex. ^ permalink raw reply [flat|nested] 14+ messages in thread
* can we measure the traffic between user and kernel mode 2011-01-27 3:42 can we measure the traffic between user and kernel mode loody 2011-01-27 4:41 ` Alexandre Courbot 2011-01-27 5:02 ` Mulyadi Santosa @ 2011-01-27 5:56 ` Michael Blizek 2 siblings, 0 replies; 14+ messages in thread From: Michael Blizek @ 2011-01-27 5:56 UTC (permalink / raw) To: kernelnewbies Hi! On 11:42 Thu 27 Jan , loody wrote: > Dear all: > Is there any directory under proc/sys or tools that can help us to > measure the traffic between user and kernel mode? > (such as ioctl, open, close, etc.) If you are interested in number of system calls per second, you can run dstat. However I do not know a way to measure the traffic in bytes/sec. -Michi -- programing a layer 3+4 network protocol for mesh networks see http://michaelblizek.twilightparadox.com ^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2011-01-31 1:40 UTC | newest] Thread overview: 14+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2011-01-27 3:42 can we measure the traffic between user and kernel mode loody 2011-01-27 4:41 ` Alexandre Courbot 2011-01-27 5:02 ` Mulyadi Santosa 2011-01-27 5:56 ` loody 2011-01-27 6:20 ` Mulyadi Santosa 2011-01-27 11:56 ` Pravin Shedage 2011-01-27 13:48 ` loody 2011-01-27 17:45 ` Mulyadi Santosa 2011-01-30 7:50 ` loody 2011-01-28 1:07 ` Alexandre Courbot 2011-01-30 8:07 ` loody 2011-01-31 1:38 ` Alexandre Courbot 2011-01-31 1:40 ` Alexandre Courbot 2011-01-27 5:56 ` Michael Blizek
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.