more on tame

2015-10-03 Thread Theo de Raadt
42 tame calls have been commited to 28 userland programs so far. For instance gzip, md5, ping, traceroute, tcpdump, script, arp, whois, ntpd, sshd... Below is a tree of roughly a hundred more programs. Not all are fully verified yet, but they being placed in snapshots. Some of these I did myself

Re: tame userland diff

2015-10-03 Thread Sebastien Marie
On Sat, Oct 03, 2015 at 09:52:13PM +0200, Mike Burns wrote: > On 2015-10-03 09.53.54 -0600, Theo de Raadt wrote: > > > > I don't know why you added "proc". I don't see a need for it. Do > > you have a seperate test cases that prompts this? > > Yes, here is a simple test for it: > > $ echo "!ec

Re: tame(2) nologin(8)

2015-10-03 Thread Sebastien Marie
On Sun, Oct 04, 2015 at 01:33:12AM +0200, Mike Burns wrote: > - Use stdio instead of rw because of mprotect(2) in atexit. > - Pass the path to the file that can be opened. with "rpath" request, you can already open the file you want. The `path' argument of tame(2) is used to *restrict* the files y

tame(2) nologin(8)

2015-10-03 Thread Mike Burns
- Use stdio instead of rw because of mprotect(2) in atexit. - Pass the path to the file that can be opened. - Switch to return instead of exit(3) in the nearby code. I suspect that the paths argument is unused or not yet ready, but I include in here regardless merely so that I can ask about it. T

Re: Removing GCC 2 support from cdefs.h

2015-10-03 Thread Philip Guenther
On Sat, Oct 3, 2015 at 2:33 PM, Michael McConville wrote: > Philip Guenther wrote: >> On Thu, Oct 1, 2015 at 10:52 PM, Jonathan Gray wrote: >> > This isn't just removing support for gcc 2 it is removing support >> > for anything that doesn't define __GNUC__ and implement all the gnu >> > attribut

Re: tame userland diff

2015-10-03 Thread Mike Burns
I see that I am too late: http://marc.info/?l=openbsd-cvs&m=144388023505289&w=2 http://marc.info/?l=openbsd-cvs&m=144388037405304&w=2 On 2015-10-03 22.44.22 +0200, Mike Burns wrote: > Fix tame(2) for patch(1). To recreate: > > /usr/src/regress/usr.bin/diff$ cat t2.1 > Below is an example license

Re: Removing GCC 2 support from cdefs.h

2015-10-03 Thread Michael McConville
Philip Guenther wrote: > On Thu, Oct 1, 2015 at 10:52 PM, Jonathan Gray wrote: > > This isn't just removing support for gcc 2 it is removing support > > for anything that doesn't define __GNUC__ and implement all the gnu > > attributes which isn't what we want. > > Yep. We've now seen a couple d

Re: tame userland diff

2015-10-03 Thread Mike Burns
Fix tame(2) for patch(1). To recreate: /usr/src/regress/usr.bin/diff$ cat t2.1 Below is an example license to be used for new code in OpenBSD, modeled after the ISC license. It is important to specify the year of the copyright. Additional years should be separated by a comma, e.g. Copyright

Re: Removing GCC 2 support from cdefs.h

2015-10-03 Thread Philip Guenther
On Thu, Oct 1, 2015 at 10:52 PM, Jonathan Gray wrote: > This isn't just removing support for gcc 2 it is removing support for > anything that doesn't define __GNUC__ and implement all the gnu > attributes which isn't what we want. Yep. We've now seen a couple diffs labeled "remove gcc2 support"

Re: iwm(4) newstate task (again)

2015-10-03 Thread Stefan Sperling
On Sun, Sep 27, 2015 at 08:00:19PM +0200, Stefan Sperling wrote: > This is yet another attempt at improving the iwm(4) newstate task. This diff has been working nicely for me, with many suspend/resume cycles. Never had a problem connecting to several wifis. Any objections? Any Oks? > The goal is

Re: tame userland diff

2015-10-03 Thread Mike Burns
On 2015-10-03 09.53.54 -0600, Theo de Raadt wrote: > > Before, from the tame patch in snapshots: > > > > $ dc /dev/null > > Killed > > $ > > > > After the below patch: > > > > $ dc /dev/null > > $ > > > > Found via the regress tests. > > I don't know why you added "proc". I don't see a need

Re: CVS: cvs.openbsd.org: src

2015-10-03 Thread Philip Guenther
On Sat, Oct 3, 2015 at 2:12 AM, Vadim Zhukov wrote: > CVSROOT:/cvs > Module name:src > Changes by: z...@cvs.openbsd.org2015/10/03 03:12:39 > > Modified files: > usr.bin/kdump : kdump.c > > Log message: > Fix wrong cast. > > This one should be an unsigned long in theory

Re: rcctl action multiple daemons

2015-10-03 Thread Antoine Jacoutot
On Wed, Sep 30, 2015 at 01:40:44PM +0200, Antoine Jacoutot wrote: > On Wed, Sep 30, 2015 at 12:30:52PM +0200, Martijn van Duren wrote: > > ping > > I am still pondering the pros and cons... I'll probably commit your diff tomorrow, thanks. -- Antoine

Re: tame userland diff

2015-10-03 Thread Theo de Raadt
> Before, from the tame patch in snapshots: > > $ dc /dev/null > Killed > $ > > After the below patch: > > $ dc /dev/null > $ > > Found via the regress tests. I don't know why you added "proc". I don't see a need for it. Do you have a seperate test cases that prompts this? So we can do bet

Re: tame userland diff

2015-10-03 Thread Mike Burns
Before, from the tame patch in snapshots: $ dc /dev/null Killed $ After the below patch: $ dc /dev/null $ Found via the regress tests. Index: dc.c === RCS file: /cvs/src/usr.bin/dc/dc.c,v retrieving revision 1.13 diff -u -p -r1.1

Re: login_radius(8)

2015-10-03 Thread Claudio Jeker
On Fri, Oct 02, 2015 at 11:09:06AM +0100, Stuart Henderson wrote: > On 2015/10/01 14:35, Jason McIntyre wrote: > > CVSROOT:/cvs > > Module name:src > > Changes by: j...@cvs.openbsd.org2015/10/01 14:35:55 > > > > Modified files: > > libexec/login_radius: login_radius.8 > >