Thanks for fixing my typos, Steve! I moved some of the info into the first
section like you asked. I agree that it works better there. The page isn't
perfect, but at least next time *I* go to add something to the regression
tester I'll know how :). Hopefully it will help others too.

Happy 4th!
Jason

On Thu, Jul 2, 2015 at 2:26 PM Steve Reinhardt <[email protected]> wrote:

> Thanks, Jason!  This looks like a valuable addition.
>
> My main comment is that the first two steps you have in your new "execution
> description" section basically overlap what's already there in the "running
> regressions" section.  I'd prefer to see you expand the "running
> regressions" section with additional detail rather than duplicating
> information across the two sections. Does that make sense?
>
> I also did some minor cleanup on the bottom part, you may want to check
> that too.
>
> Steve
>
>
> On Thu, Jul 2, 2015 at 9:23 AM Jason Power <[email protected]> wrote:
>
> > Thanks, Andreas. This was exactly what I was missing.
> >
> > I've updated the wiki page on regressions with everything that I learned.
> > If someone that knows more about the regression tester could check it out
> > and make sure I didn't miss anything and I'm not getting anything wrong,
> > that would be great.
> > http://gem5.org/Regression_Tests
> >
> > Cheers,
> > Jason
> >
> > On Fri, Jun 26, 2015 at 4:53 PM Andreas Hansson <[email protected]
> >
> > wrote:
> >
> > > Hi Jason,
> > >
> > > Have a look at tests/SConscript. At the end there is a list of configs
> > > created, and then a call to glob to actually find the regression
> targets.
> > > You need to add your new test here as well.
> > >
> > > I usually end up removing the build/ARM/tests directory to force a
> > re-run.
> > > If the binary is up to date scons should not rebuild anything, and only
> > > run the tests.
> > >
> > > Andreas
> > >
> > > On 26/06/2015 22:35, "gem5-dev on behalf of Jason Power"
> > > <[email protected] on behalf of [email protected]> wrote:
> > >
> > > >Ok, thanks for confirming that it's not possible :).
> > > >
> > > >Now, I'm trying to add a new script to tests/configs, but I'm having
> > > >trouble getting scons to recognize it. Here's what I've done.
> > > >
> > > >1) created a directory
> > tests/quick/se/00.hello/ref/x86/linux/learning_gem5
> > > >2) added blank config.ini  simerr*  simout*  stats.txt files to the
> > > >directory
> > > >3) created a script learning_gem5.py in tests/config
> > > >
> > > >Then I run:
> > > >scons build/X86/tests/opt/quick/se/00.hello/x86/linux/learning_gem5
> > > >
> > > >The result is:
> > > >scons: *** [all-targets] Explicit dependency
> > > >`tests/quick/se/00.hello/x86/linux/learning_gem5' not found, needed by
> > > >target `all-targets'.
> > > >scons: building terminated because of errors.
> > > >
> > > >Any idea what I'm missing?
> > > >
> > > >Note: this works as expected:
> > > >scons
> build/X86/tests/opt/quick/se/00.hello/x86/linux/simple-timing-ruby
> > > >
> > > >One other question. Does anyone know how to force the regressions to
> > > >re-run? Is there an option to pass scons that forces it to only re-run
> > the
> > > >regressions and not rebuild gem5?
> > > >
> > > >Thanks again!
> > > >Jason
> > > >
> > > >On Fri, Jun 26, 2015 at 3:13 PM Steve Reinhardt <[email protected]>
> > wrote:
> > > >
> > > >> Unfortunately one of the biggest limitations of the current
> regression
> > > >> tester is that it uses separate config files, which makes it harder
> > > >>than it
> > > >> should be to take a "normal" execution and turn it into a regression
> > > >>test,
> > > >> and also makes it impossible to test config scripts themselves
> > directly.
> > > >>
> > > >> If you look in tests/configs, though, several of the test scripts do
> > > >> include files from "../configs", so there might be a way to get it
> to
> > > >>work
> > > >> indirectly.
> > > >>
> > > >> Steve
> > > >>
> > > >>
> > > >> On Fri, Jun 26, 2015 at 12:54 PM Jason Power <[email protected]>
> > > wrote:
> > > >>
> > > >> > Hi all,
> > > >> >
> > > >> > I'm trying to do something that I thought would be very simple,
> but
> > I
> > > >> can't
> > > >> > figure out how to do it. What I would like to do is add a
> regression
> > > >>step
> > > >> > that runs a simple script in configs/.
> > > >> >
> > > >> > I would like to run the following test as a regression test:
> > > >> > build/X86/gem5.opt configs/learning_gem5/part1/two_level.py
> > > >> >
> > > >> > Or, without needing my script, another example would be to test
> > > >>something
> > > >> > like this:
> > > >> > build/X86/gem5.opt configs/example/se.py -c
> > > >> > tests/test-progs/hello/bin/x86/linux/hello
> > > >> >
> > > >> > Is there any way to do this with our current regression tester?
> From
> > > >> what I
> > > >> > can tell, there isn't any way to run a script in configs/.
> > > >> >
> > > >> > Also, does anyone have a pointer to documentation on the
> regression
> > > >> tester
> > > >> > other than this page on the wiki (
> http://gem5.org/Regression_Tests
> > )?
> > > >> It's
> > > >> > pretty out-of-date.
> > > >> >
> > > >> > As a meta point, I'm working on committing the scripts that are
> used
> > > >> part 1
> > > >> > of the Learning gem5 document I'm working on. I would like to
> > commit a
> > > >> > regression at the same time that ensures these scripts will be
> > tested
> > > >> > during future commits.
> > > >> >
> > > >> > Thanks,
> > > >> > Jason
> > > >> > _______________________________________________
> > > >> > gem5-dev mailing list
> > > >> > [email protected]
> > > >> > http://m5sim.org/mailman/listinfo/gem5-dev
> > > >> >
> > > >> _______________________________________________
> > > >> gem5-dev mailing list
> > > >> [email protected]
> > > >> http://m5sim.org/mailman/listinfo/gem5-dev
> > > >>
> > > >_______________________________________________
> > > >gem5-dev mailing list
> > > >[email protected]
> > > >http://m5sim.org/mailman/listinfo/gem5-dev
> > >
> > >
> > > -- IMPORTANT NOTICE: The contents of this email and any attachments are
> > > confidential and may also be privileged. If you are not the intended
> > > recipient, please notify the sender immediately and do not disclose the
> > > contents to any other person, use it for any purpose, or store or copy
> > the
> > > information in any medium.  Thank you.
> > >
> > > ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ,
> > > Registered in England & Wales, Company No:  2557590
> > > ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1
> 9NJ,
> > > Registered in England & Wales, Company No:  2548782
> > > _______________________________________________
> > > gem5-dev mailing list
> > > [email protected]
> > > http://m5sim.org/mailman/listinfo/gem5-dev
> > >
> > _______________________________________________
> > gem5-dev mailing list
> > [email protected]
> > http://m5sim.org/mailman/listinfo/gem5-dev
> >
> _______________________________________________
> gem5-dev mailing list
> [email protected]
> http://m5sim.org/mailman/listinfo/gem5-dev
>
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev

Reply via email to