[patch] doas(1): use a #define for the environment variable name length limit

2017-04-04 Thread bytevolcano
No functional or user-visible changes here. On that note, where did the "1024" (1023-char) come from? Is there anyone who has environment variables whose name goes near 1023 chars? Index: usr.bin/doas/env.c === RCS file: /cvs/src/usr.

Re: [patch 1/2] doas(1): Moved some parsing from env.c into parse.y

2017-04-04 Thread bytevolcano
Hello Philippe, On Thu, 16 Mar 2017 10:19:12 -0400 Philippe Meunier wrote: > Ted Unangst wrote: > >Did I get it backwards? If you have setenv { HERE= there }, your diff > >changes behavior. > > Speaking from the peanut gallery here, but I find this syntax rather > confusing and error-prone, e

Re: __pure __dead

2017-04-04 Thread Philip Guenther
On Wed, 5 Apr 2017, Mark Kettenis wrote: > Our only defines these if when __STRICT_ANSI__ isn't > defined. As a result, functions like exit(3) are not properly marked > as "noreturn". This in turn makes compilers complain that other > "noreturn" functions actually may return. ok guenther@ on yo

Re: __pure __dead

2017-04-04 Thread Todd C. Miller
OK millert@ We should probably consider pruning out all the gcc < 3.x bits from sys/cdefs.h. - todd

Re: syslogd log_debug

2017-04-04 Thread Alexander Bluhm
On Fri, Mar 17, 2017 at 02:09:35AM +0100, Alexander Bluhm wrote: > This is the next step for refactoring internal syslogd(8) logging. > > Replace logdebug() with generic log_debug() from log.c. Implement > log_debugadd() to construct debug message incrementally. Updated diff after unlock, ok? b

Re: Better error output for readlink(1)

2017-04-04 Thread Scott Cheloha
> On Apr 4, 2017, at 4:46 PM, Nicholas Marriott > wrote: > > readlink is explicitly documented to silently exit 1 if run without -f, > and GNU readlink behaves the same way. I doubt that should change. Yeah, I saw that. I (incorrectly, I guess?) interpreted it to mean "does not print anything

__pure __dead

2017-04-04 Thread Mark Kettenis
Our only defines these if when __STRICT_ANSI__ isn't defined. As a result, functions like exit(3) are not properly marked as "noreturn". This in turn makes compilers complain that other "noreturn" functions actually may return. This is most visible in various llvm projects that tend to be compi

Re: Better error output for readlink(1)

2017-04-04 Thread Nicholas Marriott
Hi readlink is explicitly documented to silently exit 1 if run without -f, and GNU readlink behaves the same way. I doubt that should change. On Tue, Apr 04, 2017 at 04:40:19PM -0500, Scott Cheloha wrote: > This patch replaces a custom error message in readlink.c with err(3). > The custom messa

Better error output for readlink(1)

2017-04-04 Thread Scott Cheloha
This patch replaces a custom error message in readlink.c with err(3). The custom message and call to strlen(3) aren't needed because readlink(2) checks the length of the argument string and sets errno to ENAMETOOLONG if it is too long. Dropping strlen(3) lets us drop string.h. Using err(3) also l

Re: rtadvd: asprintf

2017-04-04 Thread Theo de Raadt
> > The last commit to this file fixed two (void)asprintf calls. Fix the > > two remaining calls: depend on the ret buffer being set to NULL on > > failure. > > Personally, I would leave vltimexpire and pltimexpire initialized > to NULL and avoid those else clauses. Anything which makes the next

Re: rtadvd: asprintf

2017-04-04 Thread Todd C. Miller
On Tue, 04 Apr 2017 19:08:13 +0200, Jeremie Courreges-Anglas wrote: > The last commit to this file fixed two (void)asprintf calls. Fix the > two remaining calls: depend on the ret buffer being set to NULL on > failure. Personally, I would leave vltimexpire and pltimexpire initialized to NULL and

rtadvd: asprintf

2017-04-04 Thread Jeremie Courreges-Anglas
The last commit to this file fixed two (void)asprintf calls. Fix the two remaining calls: depend on the ret buffer being set to NULL on failure. ok? Index: dump.c === RCS file: /d/cvs/src/usr.sbin/rtadvd/dump.c,v retrieving revisi

rtadvd: recvmsg/sendmsg return value

2017-04-04 Thread Jeremie Courreges-Anglas
Said functions return an ssize_t. Comparing (size_t < 0) isn't going to give good results. The type change in ra_output() deserves its own commit. Also, use the same type and the same name for both return values; "i" is too generic. ra_input()/rs_input() take an int, but I don't think we need

Re: ifconfig.8: Small improvement, typofix

2017-04-04 Thread Jason McIntyre
On Tue, Apr 04, 2017 at 04:12:16PM +0200, Mike Belopuhov wrote: > On Tue, Apr 04, 2017 at 15:32 +0200, Klemens Nanni wrote: > > Hey, > > > > blocknonip's description is a tad clearer that way, the rest is mostly > > cosmetical but still. > > > Index: sbin/ifconfig/ifconfig.8 > > =

Re: ifconfig.8: Small improvement, typofix

2017-04-04 Thread Denis Fondras
> -If the master does not advertise within three times this interval, this host > +If the master does not advertise this interval within three times, this host > I think the old phrasing is easier to understand.

Re: ifconfig.8: Small improvement, typofix

2017-04-04 Thread Mike Belopuhov
On Tue, Apr 04, 2017 at 15:32 +0200, Klemens Nanni wrote: > Hey, > > blocknonip's description is a tad clearer that way, the rest is mostly > cosmetical but still. > Index: sbin/ifconfig/ifconfig.8 > === > RCS file: /cvs/src/sbin/ifc

ifconfig.8: Small improvement, typofix

2017-04-04 Thread Klemens Nanni
Hey, blocknonip's description is a tad clearer that way, the rest is mostly cosmetical but still. Index: sbin/ifconfig/ifconfig.8 === RCS file: /cvs/src/sbin/ifconfig/ifconfig.8,v retrieving revision 1.280 diff -u -p -r1.280 ifconfig

Re: httpd/libtls: TLS client certificate revocation checking

2017-04-04 Thread William Ahern
On Sat, Apr 01, 2017 at 07:10:35PM +1030, Jack Burton wrote: > On Fri, 31 Mar 2017 13:03:44 -0700 > William Ahern wrote: > > Basically, anything short of passing through the entire certificate > > is going to be severely limiting and frustrating, to the point of > > uselessness. And as a practica