Re: bc makefile tweak

2017-07-03 Thread David Gwynne
> On 4 Jul 2017, at 12:54, Michael W. Bombardieri wrote: > > Hi, > > When building bc, yacc can directly write bc.c instead of renaming > the file with mv. Does this look any better? there's some debate at the moment about the least worst or best way of handling this kind of

Re: armv7 alignment faults

2017-07-03 Thread Theo de Raadt
> > i think i've noted about this before, around 13months ago freebsd > > first disabled alignment faults, and they haven't enabled them since. > > deja vu, or not, i don't recall if the last diff like below did go > > anywhere, nor if it got discussed about, so i'm sorry in advance, > > if i'm

Re: armv7 alignment faults

2017-07-03 Thread Jeremie Courreges-Anglas
Artturi Alm writes: > Hi, > > i think i've noted about this before, around 13months ago freebsd > first disabled alignment faults, and they haven't enabled them since. > deja vu, or not, i don't recall if the last diff like below did go > anywhere, nor if it got discussed

bc makefile tweak

2017-07-03 Thread Michael W. Bombardieri
Hi, When building bc, yacc can directly write bc.c instead of renaming the file with mv. Does this look any better? - Michael Index: Makefile === RCS file: /cvs/src/usr.bin/bc/Makefile,v retrieving revision 1.9 diff -u -p -u -r1.9

armv7 alignment faults

2017-07-03 Thread Artturi Alm
Hi, i think i've noted about this before, around 13months ago freebsd first disabled alignment faults, and they haven't enabled them since. deja vu, or not, i don't recall if the last diff like below did go anywhere, nor if it got discussed about, so i'm sorry in advance, if i'm banging my head

install.sub: Typo/whitespace nit

2017-07-03 Thread Klemens Nanni
Remove duplicate full stop and add space after function name. Feedback/OK? Index: install.sub === RCS file: /cvs/src/distrib/miniroot/install.sub,v retrieving revision 1.1019 diff -u -p -r1.1019 install.sub --- install.sub 2 Jul

Re: install.sub: Clean v[46]_info() ouput

2017-07-03 Thread Klemens Nanni
On Mon, Jul 03, 2017 at 10:47:31PM +, Robert Peichaer wrote: Dokument explicitely possible outputs and tweak the sed expressions to remove the superfluous whitespaces. I guess that does the trick. Index: install.sub === RCS

Re: install.sub: Clean v[46]_info() ouput

2017-07-03 Thread Robert Peichaer
On Wed, Jun 14, 2017 at 05:37:07PM +0200, Klemens Nanni wrote: > With this patch, v[46]_info() both output exactly what their description > says. > > As of now, these functions are only used through > set -- $(v4_info $_if) > which gracefully handles any constellation of whitespaces in the

Re: serial console and ddb

2017-07-03 Thread Hrvoje Popovski
On 3.7.2017. 23:42, Stuart Henderson wrote: > The phrase "break sequence" is often used, but it's a bit of a misnomer. > When a serial port is connected but not actively transmitting data the tx > line is usually held high. A "break" is when that line is low for more > than a frame duration (the

Re: serial console and ddb

2017-07-03 Thread Stuart Henderson
On 2017/07/03 21:05, Hrvoje Popovski wrote: > Hi all, > > i'm having two firewalls fw1 and fw2 and on fw1 i'm sending console > output to com0. > > root@fw1:~ > # cat /etc/boot.conf > stty com0 115200 > set tty com0 > > root@fw1:~ > # cat /etc/ttys | grep tty00 > tty00 "/usr/libexec/getty

ifstated diff rename variables to avoid state confusion

2017-07-03 Thread Rob Pierce
ifstated monitors interface state and the return state of invoked commands, and takes action accordingly, all of which is managed with the help of a finite state machine. That makes for a lot of "state" references in the code. The following diff renames variables to make a distinction between

Re: magic.5: Add missing types

2017-07-03 Thread Klemens Nanni
On Mon, Jul 03, 2017 at 05:36:52PM +0100, Nicholas Marriott wrote: Hi On Thu, Jun 29, 2017 at 09:29:57PM +0200, Klemens Nanni wrote: [...] What about the current version being 4.21? We're clearly ahead of this, it seems magic(5) wasn't updated when nicm@ reimplemented things. This patch

ifstated whitespace diff

2017-07-03 Thread Rob Pierce
Fix some variable alignment whitespace. Rob Index: ifstated.h === RCS file: /cvs/src/usr.sbin/ifstated/ifstated.h,v retrieving revision 1.15 diff -u -p -r1.15 ifstated.h --- ifstated.h 2 Jul 2017 15:28:26 - 1.15 +++

Re: serial console and ddb

2017-07-03 Thread Mark Kettenis
> From: Hrvoje Popovski > Date: Mon, 3 Jul 2017 21:05:01 +0200 > > Hi all, > > i'm having two firewalls fw1 and fw2 and on fw1 i'm sending console > output to com0. > > root@fw1:~ > # cat /etc/boot.conf > stty com0 115200 > set tty com0 > > root@fw1:~ > # cat /etc/ttys | grep

serial console and ddb

2017-07-03 Thread Hrvoje Popovski
Hi all, i'm having two firewalls fw1 and fw2 and on fw1 i'm sending console output to com0. root@fw1:~ # cat /etc/boot.conf stty com0 115200 set tty com0 root@fw1:~ # cat /etc/ttys | grep tty00 tty00 "/usr/libexec/getty std.115200" vt220 on secure on fw2 i'm using "cu -s 115200" to play

Re: ifstated readability diff

2017-07-03 Thread Sebastian Benoit
commited, thanks. Rob Pierce(r...@2keys.ca) on 2017.07.03 09:45:35 -0400: > On Sun, Jul 02, 2017 at 11:50:56PM -0400, Rob Pierce wrote: > > Remove obvious clear_config() comments and misleading state_change() > > comments. > > > > Also relocate do_action() calls for the init block from

elf.h

2017-07-03 Thread Karel Gardas
Hello, I'm curious if it's possible to provide /usr/include/elf.h file on OpenBSD to improve its niceness to software porting from other Unixes. Following patch adds this for me and is tested with GHC where I'd like to kill code like: #if !defined(openbsd_HOST_OS) # include #else /*

Re: magic.5: Add missing types

2017-07-03 Thread Nicholas Marriott
Hi On Thu, Jun 29, 2017 at 09:29:57PM +0200, Klemens Nanni wrote: > While reading file(1)'s code in #openbsd-daily mulander noted that the > 'name' and 'use' types were missing from magic(5). > > I'm not entirely sure yet whether this is complete, so here's what I > did: > > magic(5) provided

Re: sblock() & solock() ordering

2017-07-03 Thread Alexander Bluhm
On Mon, Jul 03, 2017 at 11:42:19AM +0200, Martin Pieuchot wrote: > Updated diff that fixes some issues reported by visa@: > > - prevents relocking the netlock in the 'restart' case. > - always call solock() after sbunlock() in sosplice(). > > Alexander is there an easy way to trigger the

Re: ifstated readability diff

2017-07-03 Thread Rob Pierce
On Sun, Jul 02, 2017 at 11:50:56PM -0400, Rob Pierce wrote: > Remove obvious clear_config() comments and misleading state_change() comments. > > Also relocate do_action() calls for the init block from change_state() to > occur with the corresponding do_action() calls for the body block within >

Re: fo_ioctl & solock

2017-07-03 Thread Alexander Bluhm
On Mon, Jul 03, 2017 at 02:48:49PM +0200, Martin Pieuchot wrote: > soo_ioctl() will need to grab the socket lock since it modifies its > states. Sadly this function is sometimes called from socket-only > syscalls which already held the corresponding socket lock. > > So the diff below simply

Re: so_state & solock

2017-07-03 Thread Alexander Bluhm
On Mon, Jul 03, 2017 at 02:42:15PM +0200, Martin Pieuchot wrote: > I'd like to assert that the socket lock is held when modifying > `so_sate'. > > ok? OK bluhm@ > > Index: kern/uipc_socket2.c > === > RCS file:

Re: so_qlen & solock

2017-07-03 Thread Alexander Bluhm
On Mon, Jul 03, 2017 at 02:41:15PM +0200, Martin Pieuchot wrote: > I'd like to assert the socket lock is held when `so_qlen' is modified > or when it is accessed as part of a sequence that needs atomicity. > > ok? OK bluhm@ > > Index: kern/uipc_socket2.c >

Re: isakmpd(8) use-after-free

2017-07-03 Thread Martin Pieuchot
On 03/07/17(Mon) 11:18, Martin Pieuchot wrote: > On 08/06/17(Thu) 15:23, Martin Pieuchot wrote: > > Michał Koc reported a crash on misc@, turns out it's a use-after-free: > > http://marc.info/?l=openbsd-misc=149597472223216=2 > > > > The trace indicates that argument given to

Re: CVS: cvs.openbsd.org: src

2017-07-03 Thread Martijn van Duren
On 07/01/17 18:14, Mark Kettenis wrote: > CVSROOT: /cvs > Module name: src > Changes by: kette...@cvs.openbsd.org2017/07/01 10:14:10 > > Modified files: > sys/dev/pci/drm: drm_irq.c drm_linux.c drm_linux.h >drm_linux_list.h drm_mm.c drm_mm.h

fo_ioctl & solock

2017-07-03 Thread Martin Pieuchot
soo_ioctl() will need to grab the socket lock since it modifies its states. Sadly this function is sometimes called from socket-only syscalls which already held the corresponding socket lock. So the diff below simply set/remove SS_NBIO directly in places where we are dealing with sockets and

so_state & solock

2017-07-03 Thread Martin Pieuchot
I'd like to assert that the socket lock is held when modifying `so_sate'. ok? Index: kern/uipc_socket2.c === RCS file: /cvs/src/sys/kern/uipc_socket2.c,v retrieving revision 1.80 diff -u -p -r1.80 uipc_socket2.c ---

so_qlen & solock

2017-07-03 Thread Martin Pieuchot
I'd like to assert the socket lock is held when `so_qlen' is modified or when it is accessed as part of a sequence that needs atomicity. ok? Index: kern/uipc_socket2.c === RCS file: /cvs/src/sys/kern/uipc_socket2.c,v retrieving

Re: netstat(1) print PID for sockets.

2017-07-03 Thread Alexander Bluhm
On Sat, Jul 01, 2017 at 08:50:36PM +0200, Sebastian Benoit wrote: > @@ -149,7 +149,7 @@ protopr(kvm_t *kvmd, u_long pcbaddr, u_int tableid, int > proto) > struct kinfo_file *kf; > int i, fcnt; > > - kf = kvm_getfiles(kvmd, KERN_FILE_BYFILE, DTYPE_SOCKET, > + kf =

Re: netstat(1) show only listening sockets

2017-07-03 Thread Alexander Bluhm
On Sat, Jul 01, 2017 at 04:44:14PM +0200, Sebastian Benoit wrote: > This makes netstat show only listening sockets for tcp sockets > when invoked as netstat -l. > @@ -294,9 +294,14 @@ netdomainpr(struct kinfo_file *kf, int proto) > } > > /* filter listening sockets out unless -a is

Re: netstat(1) show only listening sockets

2017-07-03 Thread Craig Skinner
Hi Sebastian, On Sat, 1 Jul 2017 16:44:14 +0200 Sebastian Benoit wrote: > This makes netstat show only listening sockets for tcp sockets > when invoked as netstat -l. > > With it "netstat -l -finet -p tcp" is equivalent to > "netstat -a -finet | grep LISTEN" This shows listening UDP ports too:

NET_LOCK() w/o SPL

2017-07-03 Thread Martin Pieuchot
All network processing contexts, with the exception of hardware interrupt handlers, are now process contexts. That means the SPL protection is no longer needed inside the NET_LOCK(). So the diff below removes the splsofnet()/splx() dance from the NET_LOCK(). I'm not changing the NET_LOCK()

Re: sblock() & solock() ordering

2017-07-03 Thread Martin Pieuchot
On 26/06/17(Mon) 16:15, Martin Pieuchot wrote: > I'd like to enforce the following "lock" ordering: always hold the > socket lock when calling sblock(). > > This would allow me to protect `so_state' in sosend() when setting the > SS_ISSENDING bit. > > Diff below implements that. It also gets

isakmpd NULL dereference

2017-07-03 Thread Martin Pieuchot
Michał Koc reported another isakmpd(8) crash, this time related to a NULL dereference: #0 0x076e6ff12959 in ipsec_sa_check_flow_any (sa=Variable "sa" is not available.) at /usr/src/sbin/isakmpd/ipsec.c:275 #1 0x076e6ff1c215 in sa_find (check=0x76e6ff128d0 , arg=0x7708ab93a00) at

Re: isakmpd(8) use-after-free

2017-07-03 Thread Martin Pieuchot
On 08/06/17(Thu) 15:23, Martin Pieuchot wrote: > Michał Koc reported a crash on misc@, turns out it's a use-after-free: > http://marc.info/?l=openbsd-misc=149597472223216=2 > > The trace indicates that argument given to pf_key_v2_stayalive() is no > longer valid: > > #0 conf_get_str

Re: systemd compat for doas

2017-07-03 Thread Damien Miller
On Mon, 3 Jul 2017, Franco Fichtner wrote: > > > On 2. Jul 2017, at 8:59 PM, Ted Unangst wrote: > > > > If the username starts with a digit, but isn't a number, treat it like root. > > I question the simplicity of this patch due to the fact that it leaves > no head room

Re: systemd compat for doas

2017-07-03 Thread Franco Fichtner
> On 2. Jul 2017, at 8:59 PM, Ted Unangst wrote: > > If the username starts with a digit, but isn't a number, treat it like root. I question the simplicity of this patch due to the fact that it leaves no head room for further security-related regressions. Maybe more

Re: dont use the malloc ks_limit in softdeps

2017-07-03 Thread David Gwynne
hrm. ignore this. On Mon, Jul 03, 2017 at 04:30:11PM +1000, David Gwynne wrote: > the only thing using ks_limit is the ffs softdep code. > > this cleans that up. > > ok? > > Index: ffs/ffs_softdep.c > === > RCS file:

dont use the malloc ks_limit in softdeps

2017-07-03 Thread David Gwynne
the only thing using ks_limit is the ffs softdep code. this cleans that up. ok? Index: ffs/ffs_softdep.c === RCS file: /cvs/src/sys/ufs/ffs/ffs_softdep.c,v retrieving revision 1.135 diff -u -p -r1.135 ffs_softdep.c ---