On April 4, 2015 6:53:52 PM EDT, Valdis.Kletnieks@vt.edu wrote:
On Sat, 04 Apr 2015 15:28:23 -0400, Nicholas Krause said:
After reading this it seems that the issue is my lack of effort not my abilities.
Being able to realize that the find|xargs grep solution was actually less work, with quicker results, is *itself* an ability. I agree my strength is not in finding the most effective way to debug or find an answer to a question. I will slowly start working on it. In the meantime I am having issues with the rc 6 candidate after enabling debugging for locks as my builds for kernels are now slower by up to 3 times, I am assuming this is normal however please know if otherwise. Thanks, Nick
-- Sent from my Android device with K-9 Mail. Please excuse my brevity.
On Sat, 04 Apr 2015 19:00:42 -0400, Nicholas Krause said:
are now slower by up to 3 times,
So what debugging did you already do to try to narrow this down? Did you even run 'top' or 'vmstat' or anything else like that? Did you check to make sure you didn't replace a stripped down custom config with a 'make allmodconfig'? Did you compare the current .config with the previous one to make sure enabling lock debugging was the only change? Did you reboot back into a known-good kernel and test build a known-good kernel tree with a known-good .config (if that was still slow, I'd start checking the system hardware to make sure you haven't lost a 4G memory DIMM or other similar issue). (Replying mostly for the archives, and the edification of others)
On 2015-04-05 01:15:31 (-0400), Valdis.Kletnieks@vt.edu <Valdis.Kletnieks@vt.edu> wrote:
On Sat, 04 Apr 2015 19:00:42 -0400, Nicholas Krause said:
are now slower by up to 3 times,
So what debugging did you already do to try to narrow this down?
While we're on the subject, let me indulge a pet peeve: How did you measure this? Can you reproduce the measurement? Is the difference statistically significant? What aspect of performance is worse? Task switching time? IO throughput? Benchmarking is *hard*, and 'up to 3 times' is not the result of a careful, well thought out benchmark. Regards, Kristof
On April 5, 2015 8:29:47 AM EDT, Kristof Provost <kristof@sigsegv.be> wrote:
On 2015-04-05 01:15:31 (-0400), Valdis.Kletnieks@vt.edu <Valdis.Kletnieks@vt.edu> wrote:
On Sat, 04 Apr 2015 19:00:42 -0400, Nicholas Krause said:
are now slower by up to 3 times,
So what debugging did you already do to try to narrow this down?
While we're on the subject, let me indulge a pet peeve: How did you measure this? Can you reproduce the measurement? Is the difference statistically significant? What aspect of performance is worse? Task switching time? IO throughput?
Benchmarking is *hard*, and 'up to 3 times' is not the result of a careful, well thought out benchmark.
Regards, Kristof I ran the make command on all of my CPU cores and timed it. Furthermore it seems to be slower up to 3 times then my distribution kernel based off 3.16. Nick -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
On 2015-04-05 14:45:28 (-0400), Nicholas Krause <xerofoify@gmail.com> wrote:
On April 5, 2015 8:29:47 AM EDT, Kristof Provost <kristof@sigsegv.be> wrote:
On 2015-04-05 01:15:31 (-0400), Valdis.Kletnieks@vt.edu <Valdis.Kletnieks@vt.edu> wrote:
On Sat, 04 Apr 2015 19:00:42 -0400, Nicholas Krause said:
are now slower by up to 3 times,
So what debugging did you already do to try to narrow this down?
While we're on the subject, let me indulge a pet peeve: How did you measure this? Can you reproduce the measurement? Is the difference statistically significant? What aspect of performance is worse? Task switching time? IO throughput?
Benchmarking is *hard*, and 'up to 3 times' is not the result of a careful, well thought out benchmark.
I ran the make command on all of my CPU cores and timed it. Furthermore it seems to be slower up to 3 times then my distribution kernel based off 3.16.
So, in other words, no, you've not done a careful benchmark at all. What's the standard deviation on both measurements? How confident are you (your answer should be a number) that there is a difference? How have you accounted for VFS cache effects? Have you ensured that both builds were from a completely clean tree? How have you ensured that no other jobs interfered with the build (i.e. cron, user interaction, ...)? Once you've established that there really is a difference you need to start isolating contributing factors and figure out what's causing the difference. Alternatively, you can start trying to figure out which part of overall system performance has been degraded. That requires actually understanding how the system works and using inspection tools (top, iostat, perf, ...) to find out where the performance bottleneck is. None of that is easy or quick. None of these questions can be answered in one quick sentence. Certainly your overall answer should include more information, and demonstrate more thought and effort, than 'I did a build.'. Regards, Kristof
On April 5, 2015 3:05:18 PM EDT, Kristof Provost <kristof@sigsegv.be> wrote:
On 2015-04-05 14:45:28 (-0400), Nicholas Krause <xerofoify@gmail.com> wrote:
On April 5, 2015 8:29:47 AM EDT, Kristof Provost <kristof@sigsegv.be>
wrote:
On 2015-04-05 01:15:31 (-0400), Valdis.Kletnieks@vt.edu <Valdis.Kletnieks@vt.edu> wrote:
On Sat, 04 Apr 2015 19:00:42 -0400, Nicholas Krause said:
are now slower by up to 3 times,
So what debugging did you already do to try to narrow this down?
While we're on the subject, let me indulge a pet peeve: How did you measure this? Can you reproduce the measurement? Is the difference statistically significant? What aspect of performance is worse? Task switching time? IO throughput?
Benchmarking is *hard*, and 'up to 3 times' is not the result of a careful, well thought out benchmark.
I ran the make command on all of my CPU cores and timed it. Furthermore it seems to be slower up to 3 times then my distribution kernel based off 3.16.
So, in other words, no, you've not done a careful benchmark at all.
What's the standard deviation on both measurements? How confident are you (your answer should be a number) that there is a difference? How have you accounted for VFS cache effects? Have you ensured that both builds were from a completely clean tree? How have you ensured that no other jobs interfered with the build (i.e. cron, user interaction, ...)?
Once you've established that there really is a difference you need to start isolating contributing factors and figure out what's causing the difference. Alternatively, you can start trying to figure out which part of overall system performance has been degraded. That requires actually understanding how the system works and using inspection tools (top, iostat, perf, ...) to find out where the performance bottleneck is.
None of that is easy or quick. None of these questions can be answered in one quick sentence. Certainly your overall answer should include more information, and demonstrate more thought and effort, than 'I did a build.'.
Regards, Kristof The only difference I can tell is the kernel version. I was wondering however if anyone who has more experience in the kernel can give me tips on how to address my benchmarking. Thanks, Nick
-- Sent from my Android device with K-9 Mail. Please excuse my brevity.
On 2015-04-05 15:15:12 (-0400), Nicholas Krause <xerofoify@gmail.com> wrote:
The only difference I can tell is the kernel version. I was wondering however if anyone who has more experience in the kernel can give me tips on how to address my benchmarking.
Nick, as you've been told again and again: *you* need to do the work. No one is going to divide up the work in bite-sized chunks for you. You have to do research, you have to think about what the logical next step is, you have to actually *work* at this. You keep asking what you can do to improve your reputation or get unbanned from the kernel mailing list. It's actually simple enough: Do the work, then show us that you've done it. Next time you ask a question explain what you've tried and researched to answer your own question. When people ask you questions, *answer* *all* of them. You've essentially ignored all of the questions gave you (hint: those are not just questions, they are parts of the work you have to do). Regards, Kristof
On April 5, 2015 3:26:09 PM EDT, Kristof Provost <kristof@sigsegv.be> wrote:
On 2015-04-05 15:15:12 (-0400), Nicholas Krause <xerofoify@gmail.com> wrote:
The only difference I can tell is the kernel version. I was wondering however if anyone who has more experience in the kernel can give me tips on how to address my benchmarking.
Nick, as you've been told again and again: *you* need to do the work. No one is going to divide up the work in bite-sized chunks for you. You have to do research, you have to think about what the logical next step is, you have to actually *work* at this.
You keep asking what you can do to improve your reputation or get unbanned from the kernel mailing list. It's actually simple enough: Do the work, then show us that you've done it. Next time you ask a question explain what you've tried and researched to answer your own question. When people ask you questions, *answer* *all* of them.
You've essentially ignored all of the questions gave you (hint: those are not just questions, they are parts of the work you have to do).
Regards, Kristofer. I am starting to understand. I am busy today but willing to start looking into this tomorrow. Nick
-- Sent from my Android device with K-9 Mail. Please excuse my brevity.
participants (3)
-
Kristof Provost -
Nicholas Krause -
Valdis.Kletnieks＠vt.edu