-
Notifications
You must be signed in to change notification settings - Fork 92
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Checking status for data recording #259
Comments
there is no "record" command, if you wanna record events then execute commands as below.
recording features use ftrace so it is very accurate but it requires big trace buffer (memory).
or
please check varius command examples and output in README.md. |
Are you going to delete references to the other command name?
Would you like to add any more information for such functionality to the software documentation? |
What do you mean? :)
There are many command examples including such functionality in README.md. You can use help command to list all commands.
And you can check options and examples for each command.
|
I suggest to reconsider the following specification once more. 🤔
Partly, yes. I am looking for more helpful background information.
👀 Would you like to compare any analyses with approaches which were published also for a tool like uftrace (for example)? |
You are right!
Do you want to compare features of Guider with other tools? |
Occasionally also for my needs, yes.
Would you like to improve any details in the published PDF file? 🤔
|
Can execution durations be recorded for selected functions in the call hierarchy? |
Do you want to trace function calls for a specific task? |
Yes, of course. 👉 I published analysis approaches according to special software application concerns for further development considerations. |
Guider has a function tracing feature for specific tasks. first, you can analyze performance at function-level with sampling technique such a below command.
then you can know which function takes long time.
finally, if you want to trace specific functions then use a below command.
check [RET]=RETURN_VALUE/ELAPSED_TIME for each line. |
in addition,
|
Thanks for such a warning.
Will such background information become better represented in the evolving software documentation?
Would you like to extend any help descriptions accordingly? Which data structures are provided by a file like “guider.out”? I became especially interested in further clarification for repeated calls of the function “statvfs64” (which gets called also by the function “QStorageInfoPrivate::retrieveVolumeInfo” for example). |
I will try! :)
first line describes call info of the target function "getdents64".
the total test program execution durations that you mentioned are very nice and accurate. |
Does this feedback express a target conflict for desirable precise software run time analyses? |
yes, measuring some kind of function calls is good! |
I tried it out already to measure run time characteristics for selected function call hierarchies (including calls for Linux kernel functions). |
The biggest cause of the overhead is not about python code. |
I have tried this tool out with the version 3.9.88.
How does such information fit to the usage description?
The text was updated successfully, but these errors were encountered: