On Sun, Mar 17, 2013 at 6:05 PM, Luca Barbato <[email protected]> wrote:
> On 17/03/13 17:43, Diego Biurrun wrote:
>> On Sun, Mar 17, 2013 at 07:59:35AM +0100, Reinhard Tartler wrote:
>>>
>>> --- a/configure
>>> +++ b/configure
>>> @@ -2155,6 +2155,10 @@ case "$toolchain" in
>>> add_cflags -fthread-sanitizer
>>> add_ldflags -fthread-sanitizer
>>> ;;
>>> + valgrind-memcheck)
>>
>> That valgrind is a memory checker is well-known. I would prefer a name
>> that suggests more of a combination with FATE, which is what the option
>> does, such as "fate-valgrind" or "valgrind-fate" or just plain "valgrind".
>
> memcheck is the name of the tool in use.
Indeed, I think valgrind-memcheck is a good choice. We might want to
add other tools, such as valgrind-cachegrind, valgrind-callgrind, or
valgrind-hellgrind at some point in the future.
>
>> Why not just add the args to target_exec?
>
> so you can override the valgrind and keep the command line.
>
>> It would seem that fate.texi is a better location for this note.
>
> if we have a toolchain section it should stay there, an hint about it
> would do.
AFAIUI, fate.texi describes how fate works. This option can is
independent of that; developers might want to run arbitrary, possibly
fuzzed or otherwise broken, samples as well. That's why I do not think
that fate.texi would be a good place for that.
--
regards,
Reinhard
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
