[PATCH macros 1/2] Fix cflag test compiler message and cache ids

[Jon TURNEY]

On 12/03/2012 22:39, Chase Douglas wrote:
> On 03/12/2012 02:28 PM, Gaetan Nadon wrote:
>> On 12-03-12 02:57 PM, Chase Douglas wrote:
>>> When the language is C++, the flag checking message references $CC
>>> instead of $CXX. The cache id is also xorg_cv_cc_* instead of
>>> xorg_cv_cxx_*. This change fixes both issues.
>>>
>>> Signed-off-by: Chase Douglas <chase.douglas-Z7WLFzj8eWMS+FvcfC7Uqw at public.gmane.org>
>>> ---
>>>  xorg-macros.m4.in |    8 ++++++--
>>>  1 files changed, 6 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/xorg-macros.m4.in b/xorg-macros.m4.in
>>> index ee356e1..2da57c2 100644
>>> --- a/xorg-macros.m4.in
>>> +++ b/xorg-macros.m4.in
>>> @@ -1503,9 +1503,13 @@ AC_LANG_CASE(
>>>  	[C], [
>>>  		AC_REQUIRE([AC_PROG_CC_C99])
>>>  		define([PREFIX], [C])
>>> +		define([CACHE_PREFIX], [cc])
>>> +		define([COMPILER], [$CC])
>>>  	],
>>>  	[C++], [
>>>  		define([PREFIX], [CXX])
>>> +		define([CACHE_PREFIX], [cxx])
>>> +		define([COMPILER], [$CXX])
>>>  	]
>>>  )
>>>  
>>> @@ -1550,8 +1554,8 @@ m4_foreach([flag], m4_cdr($@), [
>>>  		PREFIX[FLAGS]="$PREFIX[FLAGS] ]flag["
>>>  
>>>  dnl Some hackery here since AC_CACHE_VAL can't handle a non-literal varname
>>> -		AC_MSG_CHECKING([if $CC supports ]flag[])
>>> -		cacheid=`AS_ECHO([xorg_cv_cc_flag_]flag[])`
>>> +		AC_MSG_CHECKING([if ]COMPILER[ supports]flag[])
>>> +		cacheid=`AS_ECHO([xorg_cv_]CACHE_PREFIX[_flag_]flag[])`
>>>  		AC_CACHE_VAL(AS_TR_SH($cacheid),
>>>  			     [AC_LINK_IFELSE([AC_LANG_PROGRAM([int i;])],
>>>  					     [eval AS_TR_SH($cacheid)=yes],
>>
>> Either this patch or Jon Turney's patch will need to be rebased due to
>> the removal of AS_ECHO.
>> http://lists.x.org/archives/xorg-devel/2012-March/029738.html
>>
>>
>> Reviewed-by: Gaetan Nadon<memsize-XzQKRVe1yT0V+D8aMU/kSg at public.gmane.org>
> 
> Since Jon's patch has your reviewed-by, I added my reviewed-by and
> pushed it. Then I rebased my two patches and pushed them.
> 
> We should be good for a release now.

Can we get some confirmation from someone who actually uses autoconf 2.61 that
I managed to get this right on the third attempt, before making a release? :-)