Re: CVS commit: src/sys/kern

2019-09-14 Thread Michael van Elst
On Sun, Sep 15, 2019 at 03:47:32AM +0200, Emmanuel Dreyfus wrote: > Michael van Elst wrote: > > > A real solution would just support the NAME=* syntax in getwedgename(). > > It might also allow for a case-insensitive match like userland. Then > > it would just work for config, for boot parameters

Re: more fexecve questions

2019-09-14 Thread maya
Please add a comment explaining that this is for pathnames we can't resolve in chroots. > @@ -224,9 +225,11 @@ elf_populate_auxv(struct lwp *l, struct > a->a_v = l->l_proc->p_stackbase; > a++; > > - execname = a; > - a->a_type = AT_SUN_EXECNAM

Re: CVS commit: src/sys/kern

2019-09-14 Thread Emmanuel Dreyfus
Michael van Elst wrote: > A real solution would just support the NAME=* syntax in getwedgename(). > It might also allow for a case-insensitive match like userland. Then > it would just work for config, for boot parameters and for interactive > entries. You mean this change? --- sys/kern/kern_s

Re: more fexecve questions

2019-09-14 Thread Christos Zoulas
In article <20190910145247.c52cc17f...@rebar.astron.com>, Christos Zoulas wrote: 1. So the consensus is to leave the file descriptor alone. 2. I've added the reverse cache lookup, and now script execution works. 3. Nobody suggested anything to improve security. Here's the latest patch; if I don'

Re: fcntl(F_GETPATH) support

2019-09-14 Thread Christos Zoulas
In article <2f29ca9a-0ae1-48d3-b3f4-1556912d4...@me.com>, Jason Thorpe wrote: > > >> On Sep 14, 2019, at 2:52 PM, Kamil Rytarowski wrote: >> >> On 14.09.2019 23:34, Christos Zoulas wrote: >>> Comments? >>> >> >> Question. How does it handle symbolic/hardlinks? > >Symbolic links, of course, ar

Re: fcntl(F_GETPATH) support

2019-09-14 Thread Jason Thorpe
> On Sep 14, 2019, at 2:52 PM, Kamil Rytarowski wrote: > > On 14.09.2019 23:34, Christos Zoulas wrote: >> Comments? >> > > Question. How does it handle symbolic/hardlinks? Symbolic links, of course, are simply continuations of the lookup, so there's "nothing to see here" with a symbolic li

Re: fcntl(F_GETPATH) support

2019-09-14 Thread Kamil Rytarowski
On 14.09.2019 23:34, Christos Zoulas wrote: > Comments? > Question. How does it handle symbolic/hardlinks? > christos > > > Index: kern/sys_descrip.c > === > RCS file: /cvsroot/src/sys/kern/sys_descrip.c,v > retrieving revision 1.

fcntl(F_GETPATH) support

2019-09-14 Thread Christos Zoulas
Comments? christos Index: kern/sys_descrip.c === RCS file: /cvsroot/src/sys/kern/sys_descrip.c,v retrieving revision 1.34 diff -u -p -u -r1.34 sys_descrip.c --- kern/sys_descrip.c 26 Aug 2019 10:19:08 - 1.34 +++ kern/sys

Re: NCHNAMLEN vnode cache limitation removal

2019-09-14 Thread Jason Thorpe
> On Sep 14, 2019, at 1:20 AM, David Holland wrote: > >> Given the list of smart people I've seen discussing this, I'm >> presumably just missing something, but what? > > Loonix. Linux isn't the only system that has such a capability. -- thorpej

Re: NCHNAMLEN vnode cache limitation removal

2019-09-14 Thread David Holland
On Fri, Sep 13, 2019 at 07:49:54PM -0400, Mouse wrote: > > [...], because fexecve is causing rumbles about doing significantly > > more reverse lookups. > > Why is everyone so concerned about finding "the" name of an inode, or > indeed any name for an inode? As far as I can tell there has ne

Re: CVS commit: src/sys/kern

2019-09-14 Thread Emmanuel Dreyfus
Martin Husemann wrote: > Oh, *B*ootdev vs. *R*ootdev - I see. > Not sure why bootdev is important, however we should use the same syntax! The change handles the parameter passed from bootloader when booting Xen. In boot.cfg, NAME=label is now supported everywhere, and this was the only missing

Re: CVS commit: src/sys/kern

2019-09-14 Thread Michael van Elst
mar...@duskware.de (Martin Husemann) writes: >> config netbsd root on "wedge:Guru-Root" type ffs >> for ages (in the netbsd-7 branch even). >Oh, *B*ootdev vs. *R*ootdev - I see. >Not sure why bootdev is important, however we should use the same syntax! bootdev already act