[Libva] [PATCH 3/9] configure: use ${libdir} over the hardcoded /usr/lib

[Emil Velikov]

On 18/09/14 21:54, Emil Velikov wrote:
> On 18/09/14 21:17, Gwenole Beauchesne wrote:
>> Hi,
>>
>> 2014-09-18 18:45 GMT+02:00 Emil Velikov <emil.l.velikov at gmail.com>:
>>> Signed-off-by: Emil Velikov <emil.l.velikov at gmail.com>
>>> ---
>>>  configure.ac | 2 +-
>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/configure.ac b/configure.ac
>>> index f8ce348..65fe0e9 100644
>>> --- a/configure.ac
>>> +++ b/configure.ac
>>> @@ -127,7 +127,7 @@ dnl Check for VA-API drivers path
>>>  AC_MSG_CHECKING([for VA drivers path])
>>>  LIBVA_DRIVERS_PATH=`$PKG_CONFIG libva --variable driverdir`
>>>  if test -z "$LIBVA_DRIVERS_PATH"; then
>>> -    LIBVA_DRIVERS_PATH="/usr/lib/xorg/modules/drivers"
>>> +    LIBVA_DRIVERS_PATH="${libdir}/xorg/modules/drivers"
>>>  fi
>>>  AC_MSG_RESULT([$LIBVA_DRIVERS_PATH])
>>>  AC_SUBST(LIBVA_DRIVERS_PATH)
>>> --
>>> 2.1.0
>>
>> Did you really hit an issue with that?
>>
>> The only ways this could fail are (i) you are trying to build against
>> an ancient version of libva < 1.0.x where x is to be determined, or
>> (ii) the pkg-config check failed somehow. (i) cannot occur because you
>> cannot even build the current VA intel-driver git master branch
>> against libva 1.3 I'd guess. So you are left with (ii) and another fix
>> might be needed then. i.e. why did the pkg-config check command
>> failed?
>>
> Or (iii) I've hacked up the latest libva.pc :P
> So as you can see I've gone through and made 'make distcheck' work, which
> currently fails without commenting out the driverdir variable in the pc (as
> mentioned in bug https://bugs.freedesktop.org/show_bug.cgi?id=84065)
> 
> We should fix the current case (not sure how yet) although I think this is a
> good idea as well :)
> 
So it seems that there is a feature [1] in (auto)make with no way to
disable. One can resolve the whole thing by exporting DESTDIR prior to
running make distcheck.

Imho it still would be nice to have a sensible fallback rather than a
hardcoded one :)


-Emil

[1] The feature:
_All_ the files have to be installed in --prefix. This is one of the
tests that "make distcheck" does and currently fails on. Reported in
2009 but no-one has stepped to implement a switch to disable it.

> Speaking of bugs, should I follow up to the patches with their respective
> bugzilla entry ?
> 
> Cheers,
> Emil
> 
> 
>> Regards,
>>
>