Re: kernel - Regarding tools/testing/ktests or selftests

[Greg KH]

On Tue, Oct 10, 2017 at 01:15:11PM +0530, Pintu Kumar wrote:
> Hi All,
> 
> Is anybody familiar with ktests of selftests framework under: tool/testing/ ?
> I have some queries.
> 
> 1) What is the difference between ktests and selftests ? How to choose ?

It depends on what you want to do, but usually new things go in the
selftest section.

> 2) How to invoke a test under ktests?
> For example:
> I have some user space C program, that tests some kernel driver IOCTL 
> calls.
> I have a Makefile to build 2 executable for the tests.
> I invoke those executable from a bash shell script and execute all
> the tests.
> How can I do it automatically from ktests ?

Have you read the documentation the kernel has for this:
$ make help | grep test
Kernel selftest:
  kselftest   - Build and run kernel selftest (run as root)
running kselftest on it
  kselftest-clean - Remove all generated kselftest files
  kselftest-merge - Merge all the config dependencies of kselftest to existed

> 3) Similarly, how can I do it using selftests?
> Using selftests looks much better option for me.
> But the problem is I have some sub directories to execute the tests.
> Example:
> Under, tools/testing/selftests , I have created 2 sub folders.
> selftests/folders/folder1/
> My test setup is under: folder1
> How to invoke selftests, for folder1 ?
> 
> If anybody have added new tests under ktests of selftests, please
> provide some reference.

Have you looked at the many patches in the kernel source tree that show
how the existing tests are added?  That's the best thing to use as a
template.

Hope this helps,

greg k-h

___
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

Re: kernel - Regarding tools/testing/ktests or selftests

[Pintu Kumar]

Hi,

I am actually interested in kselftests.
My requirement is:
I have a sub folders for test. Example: tools/testing/selftests//
I have my own Makefile, and some set of C source files under folder1
I also have a test script under folder1, that tests 2 executable
generated out of 7 C files.
I wanted to know about to make use of selftest framework to generate
these 2 executable, and also execute by script to run the test.
I have already includes  in top level Makefile under selftests.
TARGETS = 

Do, I need to create another Makefile under  and invoke
folder1/Makefile from there?
Or is there another way?

If anybody have already done a new addition in selftests, please let me know.


Thanks,
Pintu



On Tue, Oct 10, 2017 at 1:15 PM, Pintu Kumar  wrote:
> Hi All,
>
> Is anybody familiar with ktests of selftests framework under: tool/testing/ ?
> I have some queries.
>
> 1) What is the difference between ktests and selftests ? How to choose ?
> 2) How to invoke a test under ktests?
> For example:
> I have some user space C program, that tests some kernel driver IOCTL 
> calls.
> I have a Makefile to build 2 executable for the tests.
> I invoke those executable from a bash shell script and execute all
> the tests.
> How can I do it automatically from ktests ?
> 3) Similarly, how can I do it using selftests?
> Using selftests looks much better option for me.
> But the problem is I have some sub directories to execute the tests.
> Example:
> Under, tools/testing/selftests , I have created 2 sub folders.
> selftests/folders/folder1/
> My test setup is under: folder1
> How to invoke selftests, for folder1 ?
>
> If anybody have added new tests under ktests of selftests, please
> provide some reference.
>
>
> Thanks,
> Pintu

___
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

Re: boot time variable

[Tobin C. Harding]

On Tue, Oct 10, 2017 at 09:08:30AM +0200, Greg KH wrote:
> On Tue, Oct 10, 2017 at 10:17:09AM +1100, Tobin C. Harding wrote:
> > Hi,
> > 
> > I would like to create a boot time variable i.e a variable that is set once 
> > at boot time. Variable
> > does not need to be globally accessible. (actually I am using two 
> > variables).
> 
> static foo = 42;
> 
> should be all you need, right?
> 
> If not, what exactly do you mean by "boot time variable"?
> 
> > Could any one point me to examples of this already intree please?
> > 
> > I have tried the following but it has a race condition on the zero check 
> > and assignment of randval/oddval.
> > 
> > /* Maps a pointer to a unique identifier. */
> > static char *ptr_to_id(char *buf, char *end, void *ptr, struct printf_spec 
> > spec)
> > {
> > long hashval;
> > static long randval = 0;
> > static long oddval = 0;
> > 
> > if (oddval == 0 && randval == 0) {
> > randval = get_random_long();
> > oddval = get_random_odd_long();
> > }
> > 
> > hashval = ptr_obfuscate((unsigned long)ptr, randval, oddval);
> > spec.base = 16;
> > 
> > return number(buf, end, hashval, spec);
> > }
> 
> What's wrong with this code?

Maybe I just have race conditions on the brain.

> > And the compiler doesn't like
> > 
> > static long randval = get_random_long();
> > static long oddval = get_random_odd_long();
> 
> Yeah, that will not work, static initializers are at link/load time, not
> runtime.
> 
> > I thought of wrapping oddval/randval in a struct and protecting it with a 
> > lock but I don't know
> > how/where to initialize the lock in a race free manner?
> 
> Put a local lock in the function when testing if the variables are == 0,
> if you are worried that two different callers will enter it at the same
> time.

Okay.

> hope this helps,
> 
> greg k-h

Cheers Greg

___
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

Re: boot time variable

[valdis . kletnieks]

On Tue, 10 Oct 2017 10:17:09 +1100, "Tobin C. Harding" said:
> I would like to create a boot time variable i.e a variable that is set once 
> at boot time. Variable
> does not need to be globally accessible. (actually I am using two variables).

The canonical way to have stuff happen at boot is usually declaring
an initcall.

> how/where to initialize the lock in a race free manner?

Step 0: Figure out how once-at-boot code could *possibly* be called
a second time.  What are you worried about racing here?


pgpXkbCtNa2Jt.pgp
Description: PGP signature
___
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

Hook Functions - Dynamic Instrumentation and Monitoring

[Milad Kahsari]

Hi guys.

I want to write a code (or library) which it used AFL and QEMU in order to
monitor functions which called by an executable.

I have worked with AFL for binary instrumentation and fuzzing. Also, I
could find out multiple vulnerabilities in third-party libraries but I
can't realize how is it possible I do the task - hooking functions and
monitoring the behavior of a program with AFL and Qemu.

Any ideas?
___
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

kernel - Regarding tools/testing/ktests or selftests

[Pintu Kumar]

Hi All,

Is anybody familiar with ktests of selftests framework under: tool/testing/ ?
I have some queries.

1) What is the difference between ktests and selftests ? How to choose ?
2) How to invoke a test under ktests?
For example:
I have some user space C program, that tests some kernel driver IOCTL calls.
I have a Makefile to build 2 executable for the tests.
I invoke those executable from a bash shell script and execute all
the tests.
How can I do it automatically from ktests ?
3) Similarly, how can I do it using selftests?
Using selftests looks much better option for me.
But the problem is I have some sub directories to execute the tests.
Example:
Under, tools/testing/selftests , I have created 2 sub folders.
selftests/folders/folder1/
My test setup is under: folder1
How to invoke selftests, for folder1 ?

If anybody have added new tests under ktests of selftests, please
provide some reference.


Thanks,
Pintu

___
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

Re: boot time variable

[Greg KH]

On Tue, Oct 10, 2017 at 10:17:09AM +1100, Tobin C. Harding wrote:
> Hi,
> 
> I would like to create a boot time variable i.e a variable that is set once 
> at boot time. Variable
> does not need to be globally accessible. (actually I am using two variables).

static foo = 42;

should be all you need, right?

If not, what exactly do you mean by "boot time variable"?

> Could any one point me to examples of this already intree please?
> 
> I have tried the following but it has a race condition on the zero check and 
> assignment of randval/oddval.
> 
> /* Maps a pointer to a unique identifier. */
> static char *ptr_to_id(char *buf, char *end, void *ptr, struct printf_spec 
> spec)
> {
>   long hashval;
>   static long randval = 0;
>   static long oddval = 0;
> 
>   if (oddval == 0 && randval == 0) {
>   randval = get_random_long();
>   oddval = get_random_odd_long();
>   }
> 
>   hashval = ptr_obfuscate((unsigned long)ptr, randval, oddval);
>   spec.base = 16;
> 
>   return number(buf, end, hashval, spec);
> }

What's wrong with this code?

> And the compiler doesn't like
> 
> static long randval = get_random_long();
>   static long oddval = get_random_odd_long();

Yeah, that will not work, static initializers are at link/load time, not
runtime.

> I thought of wrapping oddval/randval in a struct and protecting it with a 
> lock but I don't know
> how/where to initialize the lock in a race free manner?

Put a local lock in the function when testing if the variables are == 0,
if you are worried that two different callers will enter it at the same
time.

hope this helps,

greg k-h

___
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies