Re: [RFC 3/3] Test on solib load and unload

[Doug Evans <dje@google.com>]

On Mon, Sep 23, 2013 at 7:30 PM, Yao Qi <yao@codesourcery.com> wrote:
> On 09/23/2013 08:14 AM, Doug Evans wrote:
>>
>> I think it is the test framework's responsibility to provide the utilities
>> to.
>> Large tests (of the size we need to collect data for) are best not
>> written by hand, and if we're going to machine generate source, I
>> would rather such generators come from the framework than always be
>> hardcoded into every such test.  [Obviously some tests may have unique
>> needs though.]
>>
>
> Doug,
> Generating source is easy in this test case.  However, I am not sure it is
> easy to generate source for other perf test cases, like symbols and types.
> Supposing we want to generate source files have 1 million classes, with some
> hierarchies, the generation script can't be simple, IMO.  On the other hand,
> I don't know how representative the generated program is, compared with the
> real large applications, such as openoffice, clang, etc.

Hi.
Found this in my inbox and realized I hadn't replied.
[At least I can't find a reply.]

It's easy enough to generate programs with a million symbols.
A quick hack that used bash did it in a reasonable amount of time.
[I'm not suggesting we use bash, I just used it as a quick hack to see
how long it would take.]

As for being representative, gdb doesn't care what the program does,
the program just has to look representative.
E.g., 10K DWARF CUs, 500K DWARF TUs, 4M ELF symbols, 5000 shared libs,
and so on.

I don't envision the scripts to generate this being too complex.