Re: [PATCH] Work around incorrect/broken pathnames in NT_FILE note

[Kevin Buettner <kevinb@redhat.com>]

On Mon, 31 Aug 2020 15:50:07 +0100
Pedro Alves <pedro@palves.net> wrote:

> Hi Kevin,
> 
> On 8/29/20 4:25 AM, Kevin Buettner wrote:
> > Hi Pedro,
> > 
> > I've made all of the changes you suggested except for this one...
> >   
> >>> @@ -227,6 +244,7 @@ core_target::build_file_mappings ()
> >>>  
> >>>  	    if (bfd == nullptr || !bfd_check_format (bfd, bfd_object))
> >>>  	      {
> >>> +		m_core_unavailable_mappings.emplace_back (start, end - start);
> >>>  		/* If we get here, there's a good chance that it's due to
> >>>  		   an internal error.  We issue a warning instead of an
> >>>  		   internal error because of the possibility that the    
> >>
> >> Now that we're expecting to reach here, the comment about an internal
> >> error seems stale.  
> > 
> > IMO, we'll (still) only reach the code in question due either to an
> > internal error or due to the condition mentioned later in the comment. 
> > Bear in mind that we've already checked for the existence of the file
> > and have even opened the file (though we're discarding the result) in
> > the call to exec_file_find().  The test with accompanying warning message
> > following the call to exec_file_find() should handle most if not all cases
> > where the path name is bad / broken.
> > 
> > The calls to bfd_openr() and bfd_check_format() _should_ be successful,
> > but we still need to check the result, hence second block containing the
> > comment and the warning, etc.
> >   
> 
> I see, thanks.
> 
> This version LGTM.
> 
> Pedro Alves
> 

Thanks for the reviews.  I've pushed this version.

Kevin