Re: PR82943 - Suggested patch to fix

[Jerry D <jvdelisle2@gmail.com>]

On 1/20/24 11:08 AM, Jerry D wrote:
> On 1/20/24 10:46 AM, Alexander Westbrooks wrote:
>> Hello and Happy New Year!
>>
>> I wanted to follow up on this patch I made to address PR82943 for 
>> GFortran. Has anyone had a chance to review it?
>>
>> Thanks,
>>
>> Alexander Westbrooks
>>
> 
> Inserting myself in here just a little bit.  I will apply and test today 
> if I can. Paul is unavailable for a few weeks. Harald can chime in.
> 
> Do you have commit rights for gcc?
> 
> Your efforts are appreciated.
> 
> Regards,
> 
> Jerry
> 
> 
> 

I did send you an invite to our Mattermost workspace.

I did apply your patch. Some comments.

The ChangeLog files are auto generated so do not get manually changed 
with a patch.  The push process to the gcc git repository will generate 
those from the git commit log.  If the git commit log is not formatted 
correctly the push will be rejected.

The patch attempts to create a PDT_33.f03 test case. There is one 
already there so probably rename that one.

In resolve.cc There was a compile error that looked like an extra '&&' 
in the conditional.  I deleted that and all compiled ok and all 
regression tested OK, including your new test cases and the existing 
PDT_33.f03 case.  I did not try your new test case yet for PDT_33.

So next steps are walk you through using the git scripts to make commits 
with the logs formatted correctly. (assuming no one else chimes in with 
any other comment about code changes themselves..

Regards,

Jerry