[lldb-dev] [PATCH] make dosep.ty test driver multi-threadable

Todd Fiala tfiala at google.com
Mon Mar 17 14:22:48 PDT 2014


>  I am not sure why the environment variable was ever used instead of
using an option.

I think the reason we went this way was that these are the common ways we
run tests:

(cmake/ninja):
ninja check-lldb

(configure/make):
make -C tools/lldb/test

I don't know that it's entirely clear what the best way is to pass thread
arguments to dosep.ty via either of those.  Do we want to add --threads to
the default instantiations of both of those so it picks up a good default?


On Mon, Mar 17, 2014 at 2:03 PM, Todd Fiala <tfiala at google.com> wrote:

> I'll have a run on my end in a bit here.
>
>
> On Mon, Mar 17, 2014 at 11:54 AM, Ed Maste <emaste at freebsd.org> wrote:
>
>> On 17 March 2014 14:31, Greg Clayton <gclayton at apple.com> wrote:
>> > The patch below adds a new "--threads" option whose default value is
>> "multiprocessing.cpu_count()". And only if "multiprocessing.cpu_count()"
>> returns 0 will it fall back to using the LLDB_TEST_THREADS environment
>> variable. I am not sure why the environment variable was ever used instead
>> of using an option.
>> >
>> > If everyone can try out this patch and make sure it does things as
>> expected:
>>
>> It gets the correct number of threads by default for me.  Looks like
>> you'll want to clean up the messages about the env variable though:
>>
>> default:
>>
>> Running tests with 8 threads
>> Running multithreaded with 8 threads (from LLDB_TEST_THREADS)
>>
>> or w/ --threads 1:
>>
>> Running tests with 1 threads
>> NOT running multithreaded.  Consider setting LLDB_TEST_THREADS
>> environment variable.
>>
>
>
>
> --
> Todd Fiala | Software Engineer |  tfiala at google.com |  650-943-3180
>



-- 
Todd Fiala | Software Engineer | tfiala at google.com | 650-943-3180
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-dev/attachments/20140317/78e0c687/attachment.html>


More information about the lldb-dev mailing list