This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Update pass manager to handle generate_summary/transform functions
>>>>> "Rafael" == Rafael Espindola <espindola@google.com> writes:
>> Why is it necessary to set input_location before calling this warning?
>> The warning uses %H anyhow.
Rafael> Because with the patch it uses input_location to find out if we are in
Rafael> a system header or not.
I have been thinking about this some more.
It seems very weird to me that the diagnostic machinery will use one
location to test the "in system header" property but then emit the
actual diagnostic using a different location.
I tried to think of a situation where this would be the right thing to
do, but I can't think of one.
I guess this does not help much, though, since the point of the
input_location test is to short-circuit the remaining work when
possible.
Tom