Re: RFC: ld: Add --text-section-ordering-file (version 2)

[Hans-Peter Nilsson <hp@bitrange.com>]

On Fri, 26 Apr 2024, Nick Clifton wrote:
> Hi Hans-Peter,
> 
> > Sounds great! ...except:
> > 
> > >      match.  So ".t*t" would match any section starting with ".t" and
> > >      ending with "t" and would put it into the .text section.  In this
> > >      version however the output section is selected based upon matching
> > >      the fixed part at the start of the pattern with the output section.
> > >      So ".t*t" would only work if the output section was called ".t".
> > 
> > Do you mean '...if the output section was called ".tt"'?  Else,
> > having a pattern .t*t matching an instance .t sounds confusing.
> 
> No.  Basically the code compares each entry in the section ordering file
> with the name of the output section that is currently being processed.
> So if the output section is called ".text" then any entry in the ordering
> file that starts with ".text" will be considered as valid for this section.
> But this is a straight string comparison operation, not a regular expression
> match.  So an entry for ".t*t" would not match an output section called
> ".text".
> 
> Now if the output section was called ".t" then this would match the ".t*t"
> entry (and all of the .text.<whatever> entries that might be in the ordering
> file).  Which would probably cause the ordering file not to work the way that
> the user wants.  But since this only works if the linker script has allowed
> ordering in the .t section, the user would have to deliberately create a
> strange linker script, eg:
> 
>   SECTIONS
>   {
>      .t : { INCLUDE section-ordering-file ; *(.t) }
>      .text { INCLUDE section-ordering-file ; *(.text) }
>   }
> 
> I hope that this makes sense. :-)

Honestly, no.  Sorry.  But feel free to ignore my ignorance.
I just hope my confusion makes sense to the wider audience.

I don't think ".t*t" should match ".t" but that it should match 
".text" and ".tt".  I'm certainly not going to insist.

brgds, H-P