svn commit: r354694 - in head: lib/libc/gen sys/sys

Ed Maste emaste at freebsd.org
Wed Nov 13 23:46:49 UTC 2019


On Wed, 13 Nov 2019 at 18:17, Konstantin Belousov <kostikbel at gmail.com> wrote:
>
> On Wed, Nov 13, 2019 at 10:55:35PM +0000, Brooks Davis wrote:
> > On Thu, Nov 14, 2019 at 12:46:40AM +0200, Konstantin Belousov wrote:
> > > On Wed, Nov 13, 2019 at 09:51:55PM +0000, Brooks Davis wrote:
> > > > Author: brooks
> > > > Date: Wed Nov 13 21:51:55 2019
> > > > New Revision: 354694
> > > > URL: https://svnweb.freebsd.org/changeset/base/354694
> > > >
> > > > Log:
> > > >   elf_aux_info: Add support for AT_EXECPATH.
> > > >
> > > >   Reviewed by:    emaste, sef
> > > >   MFC after:      3 days
> > > >   Sponsored by:   DARPA, AFRL
> > > >   Differential Revision:  https://reviews.freebsd.org/D22353
> > > >
> > > > Modified:
> > > >   head/lib/libc/gen/auxv.3
> > > >   head/lib/libc/gen/auxv.c
> > > >   head/sys/sys/param.h
> > > >
> > > > Modified: head/lib/libc/gen/auxv.3
> > > > ==============================================================================
> > > > --- head/lib/libc/gen/auxv.3      Wed Nov 13 21:49:46 2019        (r354693)
> > > > +++ head/lib/libc/gen/auxv.3      Wed Nov 13 21:51:55 2019        (r354694)
> > > > @@ -24,7 +24,7 @@
> > > >  .\"
> > > >  .\" $FreeBSD$
> > > >  .\"
> > > > -.Dd April 25, 2019
> > > > +.Dd November 13, 2019
> > > >  .Dt ELF_AUX_INFO 3
> > > >  .Os
> > > >  .Sh NAME
> > > > @@ -48,6 +48,10 @@ can be requested:
> > > >  .Bl -tag -width AT_OSRELDATE
> > > >  .It AT_CANARY
> > > >  The canary value for SSP.
> > > > +.It AT_EXECPATH
> > > > +The path of executed program.
> > > > +This will not be present if the process was initialized by
> > > This is too strong statement.  The auxv element might be not present
> > > if vn_fullpath(9) failed, otherwise we do provide the path.
> > >
> > > > +.Xr fexecve 2 .
> >
> > Would it be more accurate to change it from "will" to "may"?
> > Referencing vn_fullpath(9) seems unhelpful in userspace programmer
> > documentation.
>
> I do not suggest to reference vn_fullpath(9).  'May' is enough, you might
> also mention namecache.

Perhaps also mention that in that case it's not guaranteed to be the
pathname that was actually used, if there are multiple links.


More information about the svn-src-head mailing list