Moving to email since that CL may not be the right place to have a back and
forth conversation.

Sorry about that, my understanding was that this code wouldn't be executed
during the build, it would just be read, compiled into byte code, and then
stuffed into the gem5 binary. Could you send the error message that you're
getting? That might give us a better idea of what's going on which could
help fix it.

It also seems odd that m5.defines exists, USE_SYSTEMC has a setting and
that setting is True. I would expect there to be some sort of problem with
the import of defines failing, or buildEnv not being available, or
USE_SYSTEMC not being present.

Are you hitting this problem on a Mac? I could believe the checks in
src/systemc/SConscript would affect what other SConscripts see, but not
what gets built into defines.buildEnv. I'll look to see if there's a better
place for those checks which fits with other mechanisms like buildEnv more
completely.

Gabe

On Wed, Feb 27, 2019 at 3:36 PM Andrea Mondelli (Gerrit) <
[email protected]> wrote:

> View Change <https://gem5-review.googlesource.com/c/public/gem5/+/16563>
>
> 1 comment:
>
>    -
>
>    File src/python/m5/__init__.py:
>    
> <https://gem5-review.googlesource.com/#/c/16563/4/src/python/m5/__init__.py>
>    -
>
>       Patch Set #4, Line 54:
>       
> <https://gem5-review.googlesource.com/#/c/16563/4/src/python/m5/__init__.py@54>
>
>           if defines.buildEnv['USE_SYSTEMC']:
>               from . import systemc
>               from . import tlm
>
>       Hi Gabe.
>       This patch ignores the env['USE_SYSTEMC'] status set during the
>       compilation.
>       Even if the SYSTEMC was disabled, gem5 tries to do:
>       from . import systemc
>       when __init__.py is executed.
>
>       I was trying to find a clean way to keep the USE_SYSTEMC setting
>       persist (using the generated defines.py file) but I haven't found a good
>       solution (my python is very limited).
>
>       Do you have an idea how we can fix this bug?
>
> To view, visit change 16563
> <https://gem5-review.googlesource.com/c/public/gem5/+/16563>. To
> unsubscribe, or for help writing mail filters, visit settings
> <https://gem5-review.googlesource.com/settings>.
> Gerrit-Project: public/gem5
> Gerrit-Branch: master
> Gerrit-Change-Id: Iecb218daec5e15772152b5ad22b51f43b86c3d4b
> Gerrit-Change-Number: 16563
> Gerrit-PatchSet: 4
> Gerrit-Owner: Gabe Black <[email protected]>
> Gerrit-Reviewer: Andreas Sandberg <[email protected]>
> Gerrit-Reviewer: Gabe Black <[email protected]>
> Gerrit-Reviewer: Giacomo Travaglini <[email protected]>
> Gerrit-Reviewer: Jason Lowe-Power <[email protected]>
> Gerrit-Reviewer: Matthias Jung <[email protected]>
> Gerrit-Reviewer: Weiping Liao <[email protected]>
> Gerrit-CC: Andrea Mondelli <[email protected]>
> Gerrit-Comment-Date: Wed, 27 Feb 2019 23:36:09 +0000
> Gerrit-HasComments: Yes
> Gerrit-Has-Labels: No
> Gerrit-MessageType: comment
>
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev

Reply via email to