On 21/06/2018 14:42, Greg Troxel wrote:
I'm not following this but something seems strange.
Are you really saying that a program that is in /sbin will not work with
/usr not mounted? That seems quite broken; the whole point of /sbin
being separate from /usr/sbin was always that things would wor
On 21/06/2018 16:42, Emmanuel Dreyfus wrote:
On Thu, Jun 21, 2018 at 04:12:32PM +0100, Roy Marples wrote:
So if anyone has any really good reasons to keep dhclient around, now is the
time to discuss them. I've added and supported every request in dhcpcd that
anyone here as ever asked.
On 21/06/2018 16:19, Julian Coleman wrote:
The release of NetBSD-8 is probably imminent, now is probably the best time
to do this.
Will it suprise people that dhclient has gone from base? Maybe deprecate it
in the 8.0 release notes and remove it from current? Or, is that what you
meant?
How
On 21/06/2018 18:22, Emmanuel Dreyfus wrote:
Roy Marples wrote:
Could dhcpcd behave as dhclient when argv{0] is dhclient? That way
we could hardlink dhclient to dhcpcd and remove it while no user
would get surprised by the change.
That wouldn't be great. At lot of installations woul
On 21/06/2018 19:42, Martin Husemann wrote:
On Thu, Jun 21, 2018 at 05:19:32PM +0200, Julian Coleman wrote:
Hi,
The release of NetBSD-8 is probably imminent, now is probably the best time
to do this.
Will it suprise people that dhclient has gone from base? Maybe deprecate it
in the 8.0 r
Hi
So I just fixed an issue where RTM_ONEWADDR was recieved by route(8) and
it tried to parse it as a route message rather than an address message.
My question is this - should the `route monitor` command discard old
message types or are they somehow important to someone?
For reference the
On 16/08/2018 13:08, Christos Zoulas wrote:
I think that it is time to re-think how the routing socket works. Having
each program and the kernel deal with sending and receing all versions of
all messages is wasteful and painful to maintain. In the short term we
should fix things to use a message
On 17/08/2018 09:36, Christos Zoulas wrote:
| If there is anything better we can do long term, I'm sure out of ideas
| right now.
Looks like the way that we are evolving the routing socket API is very
expensive and error prone. If we designed the individual messages differently
(so that they cou
Hi Ian
On 30/08/2018 19:11, Ian Bruene wrote:
I am one of the developers on NTPsec and we are getting warnings on
NetBSD 8.0 in the form:
../../ntpd/ntp_io.c:1962:14: warning: %m is only allowed in syslog(3)
like functions [-Wformat=]
This feature appears to have been added here:
https://g
On 31/08/2018 13:29, Martin Husemann wrote:
On Fri, Aug 31, 2018 at 01:23:51PM +0100, Roy Marples wrote:
This allows dhcpcd to be compiled with full warnings, but I've disabled the
check for specific functions. This is actually self defeating as it's just
the kind of function the wa
On 31/08/2018 14:15, Martin Husemann wrote:
On Fri, Aug 31, 2018 at 02:10:12PM +0100, Roy Marples wrote:
The function is currently marked as __printflike.
If I re-enable the warning gcc warns me that it should be marked as
__sysloglike because it calls syslog. If I change it to
On 16/02/2019 03:11, John Nemeth wrote:
I often get difficult to read paint factory
explosions and basic things like "ifconfig" have disappeared (every
freaking unix-like system I've touched in the last 30 years has
had "ifconfig", but now a few NIHs are trying to make it go away).
One of my la
Hi List
tree(3) states that it's deprecated and rbtree(3) should be used instead.
There seems to be no equivalent of the _FOREACH_SAFE macros.
I'm really new to rbtree stuff, but to convert from my existing TAILQ
implementation, I need this.
Attached is a patch (not even compile tested yet) to
On 03/03/2019 12:49, Roy Marples wrote:
Attached is a patch (not even compile tested yet) to add the
_FOREACH_SAFE macros and also _NEXT and _PREV as well.
Does this look good or have I made a schoolboy error somewhere?
This one might even compile!
Index: rbtree.h
On 03/03/2019 16:47, a...@100acres.us wrote:
Attached is a patch (not even compile tested yet) to add the _FOREACH_SAFE
macros and also _NEXT and _PREV as well.
Does this look good or have I made a schoolboy error somewhere?
This one might even compile!
Do your _SAFE variants work for rbtrees
On 11/03/2019 06:21, Martin Husemann wrote:
On Mon, Mar 11, 2019 at 02:01:44PM +0900, Rin Okuyama wrote:
(3) Put "state" struct __window (suggested by uwe)
Not __window _cursesi_screen which is __screen.
I prefer this one too.
I prefer this also.
Roy
On 25/08/2019 15:39, Joerg Sonnenberger wrote:
There is no technical reason really for prefering void *
Having to write code to exclusively deal with NetBSD's different API to
compile without warnings sounds like a pretty good technical reason to me.
And I'm just talking about the C interface.
On 25/08/2019 16:48, Joerg Sonnenberger wrote:
On Sun, Aug 25, 2019 at 04:43:51PM +0100, Roy Marples wrote:
On 25/08/2019 15:39, Joerg Sonnenberger wrote:
There is no technical reason really for prefering void *
Having to write code to exclusively deal with NetBSD's different API to
co
On 27/08/2019 17:24, Kamil Rytarowski wrote:
Last year, I wrote this patch to add curses_version() for curses(3).
http://netbsd.org/~kamil/patch-00073-curses-version.txt
The only purpose of this function is to get better compat with ncurses
software.
I needed it originally in qemu. It's someti
Using Blymns correct email :)
On 27/08/2019 18:28, Roy Marples wrote:
On 27/08/2019 17:24, Kamil Rytarowski wrote:
Last year, I wrote this patch to add curses_version() for curses(3).
http://netbsd.org/~kamil/patch-00073-curses-version.txt
The only purpose of this function is to get better
On 27/08/2019 21:14, Kamil Rytarowski wrote:
On 27.08.2019 19:29, Roy Marples wrote:
Using Blymns correct email :)
On 27/08/2019 18:28, Roy Marples wrote:
On 27/08/2019 17:24, Kamil Rytarowski wrote:
Last year, I wrote this patch to add curses_version() for curses(3).
http://netbsd.org
On 29/08/2019 03:45, Thor Lancelot Simon wrote:
On Wed, Aug 28, 2019 at 10:42:13PM -0400, Thor Lancelot Simon wrote:
On Wed, Aug 28, 2019 at 01:54:54AM +0100, Roy Marples wrote:
Where 8.2 is taken from the .so version?
Roy
I think that it would be confused with NetBSD release and it could
On 29/08/2019 14:19, Kamil Rytarowski wrote:
In my opinion artificial versioning (1.0.0) of native code adds no
interesting information and adds burden on us for superfluous versioning
model, orthogonal to __NetBSD_Version__.
Adding extra versioning would make sense if we could maintain curses
On 29/08/2019 14:27, Kamil Rytarowski wrote:
On 29.08.2019 15:23, Roy Marples wrote:
On 29/08/2019 14:19, Kamil Rytarowski wrote:
In my opinion artificial versioning (1.0.0) of native code adds no
interesting information and adds burden on us for superfluous versioning
model, orthogonal to
On 30/08/2019 17:45, Roy Marples wrote:
On 29/08/2019 14:27, Kamil Rytarowski wrote:
On 29.08.2019 15:23, Roy Marples wrote:
On 29/08/2019 14:19, Kamil Rytarowski wrote:
In my opinion artificial versioning (1.0.0) of native code adds no
interesting information and adds burden on us for
On 29/08/2019 14:27, Kamil Rytarowski wrote:
On 29.08.2019 15:23, Roy Marples wrote:
On 29/08/2019 14:19, Kamil Rytarowski wrote:
In my opinion artificial versioning (1.0.0) of native code adds no
interesting information and adds burden on us for superfluous versioning
model, orthogonal to
On 30/08/2019 18:09, Kamil Rytarowski wrote:
On 30.08.2019 18:55, Roy Marples wrote:
return "NetBSD-Curses " CURSES_VERSION
I propose to go for:
return "NetBSD-" CURSES_VERSION " Curses";
With removed __DATE__ that is not MKREPRO friendly.
The rest looks
Hi List!
dhcpcd-9 is almost ready for release and the big feature for it is privilege
separation. The question is which system user should it use on NetBSD? Other
BSD's have the _dhcp user - which shares DHCP serving on OpenBSD at least.
Is this to generic and we should have a _dhcpcd user or
syslogd is a powerful syslog implementation.
It supports authenticated and encrypted TLS connections and signing messages.
Because of this it lives in /usr due to the libraries it needs.
/usr traditionally depends on mountcritremote which in turn relies on networking
being up.
The irony being t
BSD Foundation, Inc.
+.\" All rights reserved.
+.\"
+.\" This code is derived from software contributed to The NetBSD Foundation
+.\" by Roy Marples.
+.\"
+.\" Redistribution and use in source and binary forms, with or without
+.\" modification, are permitted provid
On 11/03/2020 15:52, Christos Zoulas wrote:
In article <6d301f35-05ea-25b4-fb19-c473d68fd...@marples.name>,
Roy Marples wrote:
-=-=-=-=-=-
Hi
So I have an IPv6 capable interface on my router.
When I run route monitor, I get a lot of RTM_MISS for non-existant nodes on my
prefix. Loggin
On 27/03/2020 02:06, Christos Zoulas wrote:
@@ -494,6 +530,7 @@
if (tic == NULL)
return NULL;
+ tic->rtype = _ti_find_rtype(cap);
buf.buf = NULL;
buf.buflen = 0;
If I'm reading this patch correctly, it means that the old terminfo library
cannot read
On 27/03/2020 13:15, Christos Zoulas wrote:
On Mar 27, 2020, at 4:05 AM, Roy Marples wrote:
On 27/03/2020 02:06, Christos Zoulas wrote:
@@ -494,6 +530,7 @@
if (tic == NULL)
return NULL;
+ tic->rtype = _ti_find_rtype(cap);
buf.buf = N
On 27/03/2020 14:32, Christos Zoulas wrote:
In article <431d1631-ef82-d1ec-98ff-e39b71418...@marples.name>,
Roy Marples wrote:
On 27/03/2020 13:15, Christos Zoulas wrote:
On Mar 27, 2020, at 4:05 AM, Roy Marples wrote:
On 27/03/2020 02:06, Christos Zoulas wrote:
@@ -494,6
On 30/03/2020 02:19, Christos Zoulas wrote:
Recently you changed blacklistd without prior discussion and
you introduced a security bug. The terminfo2 changes were not discussed
either - I would have objected to creating yet another file to hold the
compiled definitions.
You don't discuss any dh
On 30/03/2020 02:19, Christos Zoulas wrote:
The terminfo database before I cleaned it up was a mess. There was
no build process, or explanation how to import it, or what the local
changes where.
When I imported it there were no local changes, as such nothing to document
other than whence it ca
On 30/03/2020 02:15, Brett Lymn wrote:
On Mon, Mar 30, 2020 at 01:59:35AM +0100, Roy Marples wrote:
Your import of newer terminfo descriptions broke numerics larger than int16_t.
I take equal blame for this because there should have been a check for this
in the initial code.
But still you
On 30/03/2020 03:07, Christos Zoulas wrote:
On Mar 29, 2020, at 9:42 PM, Roy Marples wrote:
On 30/03/2020 02:19, Christos Zoulas wrote:
The terminfo database before I cleaned it up was a mess. There was
no build process, or explanation how to import it, or what the local
changes where
tos
On Mar 29, 2020, at 9:25 PM, Roy Marples wrote:
On 30/03/2020 02:19, Christos Zoulas wrote:
Recently you changed blacklistd without prior discussion and
you introduced a security bug. The terminfo2 changes were not discussed
either - I would have objected to creating yet another file to ho
On 30/03/2020 04:05, Christos Zoulas wrote:
On Mar 29, 2020, at 10:37 PM, Roy Marples wrote:
blacklistd was not working for me and the ACL check you mention was certainly
not described anywhere I saw. After reading the Fine Man Page, I came to the
conclusion that passing a sockaddr with a
On 02/02/2016 22:11, Constantine A. Murenin wrote:
> On 2016-02-01 1:07, Roy Marples wrote:
>> On 30/01/2016 19:39, Constantine A. Murenin wrote:
>>> In general, I personally don't think it ever makes sense to shutdown
>>> by default when the temperature
pidfile(3) is pretty crap - it just writes to the file without any locking.
Also, you can only write the pidfile once - subsequent calls to the same file
are a non-op so it's useless after forking.
pidlock(3) is a bit better, but the "lock" is just does the file exist or not
which isn't great if
On Sunday 20 March 2016 09:26:23 Michael van Elst wrote:
> r...@marples.name (Roy Marples) writes:
> >So I've created pidfile_lock (patch attached) to address these problems.
>
> Does it work on NFS root?
I've not tested it especially, but I would assume so as flock(2)
Updated patch with man page changes, including an example of it's use.
Added pidfile_remove() which fixes the existing BUGS section.
Added pidfile_close() so a forked process can close it safely.
Made pidfile_read() visible so it's easy to obtain the PID to send a signal to.
RpyIndex: include/util
On Sunday 20 March 2016 14:04:52 Roy Marples wrote:
> Updated patch with man page changes, including an example of it's use.
> Added pidfile_remove() which fixes the existing BUGS section.
> Added pidfile_close() so a forked process can close it safely.
> Made pidfile_read() vis
On Sunday 20 March 2016 14:53:46 you wrote:
> On Sat, 19 Mar 2016 23:29:37 +
>
> Roy Marples wrote:
> > pidfile(3) is pretty crap - it just writes to the file without any
> > locking.
>
> I don't understand why you think any of that matters. Lo
On Sunday 20 March 2016 23:05:12 Christos Zoulas wrote:
> In article <1547312.w6y7ml9...@uberpc.marples.name>,
> Roy Marples wrote:
>
> There is no need for pidfile_lock(), just fix pid_file() to return pid_t.
> I've audited the code in the tree and the code that ch
On 21/03/2016 00:46, Thor Lancelot Simon wrote:
> On Sun, Mar 20, 2016 at 08:43:31PM -0400, Thor Lancelot Simon wrote:
>> On Sun, Mar 20, 2016 at 08:53:03PM +0000, Roy Marples wrote:
>>> On Sunday 20 March 2016 14:53:46 you wrote:
>>>> On Sat, 19 Mar 2016 23:29:
On 21/03/2016 14:55, Thor Lancelot Simon wrote:
> On Mon, Mar 21, 2016 at 02:30:43PM +0000, Roy Marples wrote:
>>>
>>> In fact, what you really want is just the guts of shlock(1), specifically
>>> the shlock -p behavior. I can't see any reason why pidfile
On Thursday 24 March 2016 10:25:20 Brett Lymn wrote:
> On Tue, Mar 22, 2016 at 02:13:01PM +0100, Martin Husemann wrote:
> > None of my (plenty) diskless machines does that - /var/run is not
> > performance critical, so why bother?
>
> Because it automagically gets cleared out when the machine rebo
Lets go back to the original question.
On 20/03/2016 09:26, Michael van Elst wrote:
> r...@marples.name (Roy Marples) writes:
>
>> So I've created pidfile_lock (patch attached) to address these problems.
>
> Does it work on NFS root?
After taking on board what everyone
On 24/03/2016 12:04, Jonathan A. Kollasch wrote:
> On Thu, Mar 24, 2016 at 12:49:31PM +0100, Kamil Rytarowski wrote:
>> POSIX extension to ISO C says:
>>
>> [CX] [Option Start] If the value of errnum is a valid error number,
>> the message string shall indicate what error occurred; if the value of
On 24/03/2016 12:47, Michael van Elst wrote:
> r...@marples.name (Roy Marples) writes:
>
>> If it doesn't work on NFS and you care so much, please spend your time
>> fixing NFS
>
> If you want to change things you should not try to pass your responsibility
> to so
On 20/03/2016 18:53, James K. Lowden wrote:
> The biggest problem with pidfiles IMO is stale files left over when the
> process terminates abnormally. That could be remedied with a separate
> daemon, modelled on syslogd, that would create, update and delete
> them. Modify pidfile(3) to use the da
On 22/05/2016 15:22, co...@sdf.org wrote:
> hi,
>
> I'd like to write code that works with the input or output typically
> done with ifconfig.
> as an example, I would like to know, say, which interfaces are wireless
> and which are wired, as they tend to be controlled differently, and
> whether t
On 10/06/2016 10:03, Paul Goyette wrote:
> Does anyone have suggestions and web-sites for a couple of linux
> distros? I'm not going to check them all... :)
http://man7.org/linux/man-pages/man1/find.1.html
Quite a few time related options.
Roy
Hi List
I use recvmsg(2) a fair bit in code.
However, a lot of the time I can't be certain how large a buffer I would
need, so they all ship with a handy function called recvmsg_realloc,
which as you can guess, will realloc the last iovec so that the whole
message fits. Think of it as a convenienc
On 07/09/2016 13:44, Christos Zoulas wrote:
> Are you planning to change our kernel so that it handles MSG_TRUNC like linux
> does (return the actual length of the buffer instead of ignoring the flag)?
Yes, because it saves on looping.
But I can do that separately from this.
> How about handling
Hi List!
There seems to be a fair bit of code in our tree which outputs to both
syslog and stdout/stderr via printf/perror/warnx, etc.
I'll make the assumption that is because syslog output via LOG_PERROR
isn't pretty. It also only goes to stderr.
Also, there are cases where you want to run a daem
On 2016-10-06 18:25, Timo Buhrmester wrote:
LOG_POUT - output only the formattted message to stdout
if pri > LOG_ERR, otherwise stderr
Writing diagnostics to stdout sounds very wrong, I wouldn't want
any program to do that. What for?
Well, some programs write informative ou
On 07/10/2016 00:36, Roy Marples wrote:
> On 2016-10-06 18:25, Timo Buhrmester wrote:
>>>LOG_POUT - output only the formattted message to stdout
>>>if pri > LOG_ERR, otherwise stderr
>> Writing diagnostics to stdout sounds very wrong, I woul
On 07/10/2016 21:33, Timo Buhrmester wrote:
>> what is stdout actually for then.
> I guess the usual distinction is that stdout gets the "proper" output
> of a non-sink program (e.g. for ls(1) the actual directory listing,
> while diagnostic information (ls: foo: Permission denied) goes
> to stderr
On 30/11/2016 22:11, Carsten Kunze wrote:
> Brett Lymn wrote:
>
>> But here we see:
>>
>>> kbeg=\E[1~
>>> khome=\E[1~
>>
>> So, we have two keys that map to ESC [1~ so it would depend entirely on
>> how the termcap is processed as to which leaf symbol is assigned to the
>> escape
Hi List
It seems our beloved curses library has been made portable, to work on
Linux + musl libc.
https://github.com/sabotage-linux/netbsd-curses
I'm in discussions with one of their devs about pushing some of their
patches into NetBSD so it's easier for them to sync our changes back
down.
Hi List!
While setting up a new server for FTP access, I found that:
* We ship a FTP server in the base system.
* We supply /var/spool/ftp as a recommened skeletion for anon FTP.
* We DO NOT supply a ftp user in master.passwd (or group).
Is there a reason for this? Surely we just need t
Hi Martin
On 26/01/2017 00:12, Martin Neitzel wrote:
* We ship a FTP server in the base system.
* We supply /var/spool/ftp as a recommened skeletion for anon FTP.
* We DO NOT supply a ftp user in master.passwd (or group).
Is there a reason for this? Surely we just need to supply a FTP
On 30/01/2017 04:22, co...@sdf.org wrote:
> I've applied a patch to audio/moc. It works on my end now.
> It was attempting set_escdelay (25); very early, before setting
> up a window.
>
> On nbcurses it was:_cursesi_screen->ESCDELAY = escdelay;
> which was a null deref.
>
> for ncurses, it was re
Hi List!
tic(8) is currently growing some changes.
The big one is the ability to write to a directory/file database rather
than a singular database. The general idea is to use the singular
database for the base system and the directory/file database for pkgsrc.
So ncurses (for the vt220 terminfo
On 05/05/2017 20:54, Kamil Rytarowski wrote:
> I have a pending task to write from scratch in C# for CoreFX (.NET
> platform assemblies) support for extracting registered entries from our
> native terminfo cdb file. (This is currently enforced to be this way,
> the same happens for ncurses)
Intere
On 09/05/2017 12:51, Kamil Rytarowski wrote:
> I don't have a preference, but I'm surprised that our single .cdb
> approach is going to be hybrid with support for ncurses layout.
Other ideas welcome on how to allow pkgsrc to integrate terminfo
descriptions.
Roy
signature.asc
Description: OpenP
On 25/06/2017 14:28, Kamil Rytarowski wrote:
On 25.06.2017 14:06, Martin Husemann wrote:
You can also read random bytes from /dev/urandom.
Martin
Right this is the most portable solution right now. I don't think
RANDOM_BITS or similar is currently useful, it will grow incompatible
extension -
On 25/06/2017 15:19, Robert Elz wrote:
Date:Sun, 25 Jun 2017 13:53:02 +0200
From:Kamil Rytarowski
Message-ID: <93fba221-3f6c-07be-8dc8-efd744c83...@gmx.com>
| There is a libc function for this kind of tasks: arc4random_uniform(3).
Sorry, I had missed this mes
On 29/06/2017 02:06, Taylor R Campbell wrote:
>> we've been able to run ntpd as non-root for a while. this is not the
>> default if you innocently ntpd=yes in rc.conf. it requires
>> /dev/clockctl, and most things have it, even one of the sun2 kernels.
>>
>> can I change this to become the default,
On 05/10/2017 05:34, David Holland wrote:
> After some discussion on irc today the following tentative plan's been
> floated for addressing hmac(3):
>
> 1. Give it its own hmac.h; this is consistent with the way each hash
> function currently has its own header file. Remove it from stdlib.h,
> whe
On 05/10/2017 06:12, Taylor R Campbell wrote:
>> Date: Thu, 5 Oct 2017 04:34:08 +
>> From: David Holland
>>
>> 3. Going forward, create a new crypthash.h, and have it include all
>> the hash function headers (hmac.h, md5.h, sha1.h, rmd160.h, etc etc)
>> and document all the latter as deprecate
On 11/8/2017 11:18 AM, Valery Ushakov wrote:
On Wed, Nov 08, 2017 at 07:47:17 +1030, Brett Lymn wrote:
On Mon, Nov 06, 2017 at 07:09:06PM +0300, Valery Ushakov wrote:
This does leave us in the invidious position of our correctness
being interpreted as bugs because we don't do what ncurses
does
On 30/04/2018 02:48, Robert Elz wrote:
Back in May 2005. dsl@ added this in the BUGS section of sh(1):
The characters generated by filename completion should probably be quoted
to ensure that the filename is still valid after the input line has been
processed.
Does anyone (inc
77 matches
Mail list logo
