Re: CVS commit: src/usr.bin/ldd

[Valery Ushakov <uwe%stderr.spb.ru@localhost>]

On Sat, Feb 01, 2025 at 14:08:09 -0000, Christos Zoulas wrote:

> In article <Z54fWRlrQq-4nefU%snips.stderr.spb.ru@localhost>,
> Valery Ushakov  <uwe%stderr.spb.ru@localhost> wrote:
> >On Fri, Jan 31, 2025 at 13:52:37 -0500, Christos Zoulas wrote:
> >
> >> Module Name:	src
> >> Committed By:	christos
> >> Date:		Fri Jan 31 18:52:37 UTC 2025
> >> 
> >> Modified Files:
> >> 	src/usr.bin/ldd: ldd.1
> >> 
> >> Log Message:
> >> update %a description (it was wrong before; it was never __progname).
> >
> >+The main program's name, either set from
> >+.Dv AT_SUN_EXECNAME
> >+or from the first element of the argument vector.
> >
> >What does AT_SUN_EXECNAME even mean here, in the context of ldd(1)?
> >The program(s) are not being exec'ed.  Please, can you clarify?  TIA.
> 
> Is the description better now?

Not really.  What I am confused about is (as I said in the original
mail) how does AT_SUN_EXECNAME get into the picture.  When a program
is executed - the kernel passes the name used to execute the program
in the aux vector.  But here, in ldd case, the program is _not_
executed, so there is no aux vector, so there's no AT_SUN_EXECNAME
entry in this non-existent aux vector.

Consider, e.g.:

    ldd -o /bin/*

where -o uses the format that includes %a that we are talking about.

-uwe