Le 08/09/2023 à 23:22, Harald Anlauf via Fortran a écrit :
> Am 08.09.23 um 12:04 schrieb Mikael Morin via Gcc-patches:
>> Hello,
>>
>> this avoids some redundant work in the symbol deletion code, which is
>> used a lot by the parser to cancel statements that fail to match in the
>> end.
>> I haven't tried to measure the performance effect, if any, on a 
>> pathological example;
>> just passed the fortran testsuite on x86_64-pc-linux-gnu.
>> OK for master?
> 
> This is OK.
> 
Thanks.
I had forgotten function comments.
This is what I have pushed.