I think we can relax the path restriction if there's no restriction on
command.
Index: doas.c
===
RCS file: /cvs/src/usr.bin/doas/doas.c,v
retrieving revision 1.39
diff -u -p -r1.39 doas.c
--- doas.c 27 Aug 2015 16:31:02 -
Michael McConville wrote:
> Ted Unangst wrote:
> > Michael McConville wrote:
> > > Also, why is fs on line 390 cast to char* when afree() takes void*?
> >
> > this code is older than void. and NULL apparently. please remove the casts
> > too.
>
> Do
Martijn van Duren wrote:
> Hello tech@,
>
> I took a quick glance at ksh and one of the first things I noticed was
> that it uses some sanatizing code on argv. When looking at execve(2) I
> see that EINVAL or EFAULT are returned when argv isn't properly
> formatted. I've also verified this quic
Walter Neto wrote:
> Hi,
>
> Here is the first patch to bring Journaling to OpenBSD based on the McKusick
> paper: https://www.bsdcan.org/2010/schedule/attachments/141_suj-slides.pd and
> FreeBSD 10
>
> This first patch is just for structures and definitions.
>
> I know is my first patch sugge
Walter Neto wrote:
> Hi guys,
>
> Studying more about wapbl, I saw It is a little faster than SU+J and so much
> more easy to implement, correct me if I’m wrong, please.
>
> About a diff with working code, my next would be one with it, but not with
> the whole thing working, It was to be homeop
Martin Pieuchot wrote:
> On 02/09/15(Wed) 08:45, trondd wrote:
> > I just noticed on my -current systems, memory reporting from netstat -m
> > seems to show that memory is overcommited.
> >
> > $ netstat -m
> > 535 mbufs in use:
> > 289 mbufs allocated to data
> > 8 mbufs allocated
David CARLIER wrote:
> Hi all,
>
> Both internal state variables, lclptr and gmtptr are allocated once but
> never freed so I was wondering if this little diff which avoids the
> unnecessary dynamic allocations would have any use ?
Looks about right. There's little cause to optimize for static sp
Mike Larkin wrote:
> > Is this hypervisor more similar to "micro"-hypervisor or to monolithic
> > hypervisor?
> >
>
> I don't know what those terms mean.
It's a milli-hypervisor!
> - * NOTE: INT32 may end up being more than 32 bits.
> /* Table flag type - needs > 16 and < 32 bits */
> -typedef INT32 Tflag;
awkward...
Stuart Henderson wrote:
> Currently NSD's compat functions for b64_pton and b64_ntop are always
> used. Worse, they're the old versions which abort(). This is because
> the library symbols are prefixed by __ (not just in OpenBSD) and are
> only accessible as b64_xx when resolv.h is included.
>
> F
Michael McConville wrote:
> Alexey Suslikov wrote:
> > Michael McConville sccs.swarthmore.edu> writes:
> >
> > > RCS file: /cvs/src/bin/ksh/c_ksh.c,v
> >
> >
> >
> > > - shprintf(newline);
> > > + shprintf("\n");
> >
> > In terms of portability, are you sure ne
Michael McConville wrote:
> There will probably be more similar patches to come if this is
> acceptable. The legwork was done with the following Coccinelle script:
>
I think it should be split up, but I'm happy to see diffs like this.
Nicholas Marriott wrote:
> Here is a diff, also make %p better
that looks better. sigh. why does this even exist?
Michael McConville wrote:
> Index: mv.c
> ===
> RCS file: /cvs/src/bin/mv/mv.c,v
> retrieving revision 1.40
> diff -u -p -r1.40 mv.c
> --- mv.c 24 Aug 2015 00:10:59 - 1.40
> +++ mv.c 14 Sep 2015 13:38:13 -
> @@ -
Nicholas Marriott wrote:
> Hi
>
> As far as I can tell there haven't been any new bits added for almost 20
> years, so I expect we can do without the Tflag typedef.
>
> ok?
yup
Miod Vallat wrote:
> > Hi,
> >
> > Here a first sets of "if(x) free(x)" cleanup in sys/arch/
> >
> > This patch contains only trivial if(x) removal. The size argument in
> > free is keep untouched (because it is already setted, or because it
> > makes sens to keep it to 0).
> >
> > Comments ? OK
doas doesn't need any other open files and should probably shut them all.
Index: doas.c
===
RCS file: /cvs/src/usr.bin/doas/doas.c,v
retrieving revision 1.41
diff -u -p -r1.41 doas.c
--- doas.c 3 Sep 2015 20:05:58 - 1.
Sebastian Benoit wrote:
> ok, but in other places we have closefrom(STDERR_FILENO + 1)
is that really more clear? it only makes sense if you know stderr is 2.
if you sometimes forget which is 1 and which is 2, then the macro only makes
it more confusing because now you have to decide what comes a
I don't think there's anything special about init that requires mode 0500. if
a regular user wants to run it, let them. i mean, if running init would do
something evil, i've got bad news. i can build my own...
also, the makefile contains (thankfully) ignored install flags. delete them
too.
Index
Martijn van Duren wrote:
> Hello tech@,
>
> The following patch adds $PATH resolving to doas. This has two advantages:
> 1) People are forced to configure doas with a full path, which makes
> sure that applications can't be spoofed by users changing their $PATH to
> a writeable location and maki
/bin/sleep does a lot more work than you'd expect (just ktrace it).
this is because it calls setlocale(). apparently so that isdigit() doesn't get
confused by wacky foreigners with funny numbers.
there is another solution, given that the problem of identifying digits is not
particularly challengi
Michael McConville wrote:
I'm late to the party, but we could tidy things up a bit by moving the
function into the c file it's used in and using errx() internally.
> Index: locate/fastfind.c
> ===
> RCS file: /cvs/src/usr.bin/locate/
The frequency of the cpu is likely to be of interest (in fact, it is to me).
We can include it in the apm info.
Index: arch/i386/i386/apm.c
===
RCS file: /cvs/src/sys/arch/i386/i386/apm.c,v
retrieving revision 1.113
diff -u -p -r1.11
Stefan Sperling wrote:
>
> Just remove the setlocale() call to leave sleep(1) in the default "C" locale.
here's a diff that does that. you're right, there's no need to rewrite ctype
if it hasn't been altered. I also checked that there are no calls to err()
or strerror() or whatever, and so there
I think this will help pflogd sleep better at night...
My understanding of the pcap/bpf timeout is that it will always wait that
long, even if packets are received, so that you don't get one read() per
packet. But using this timeout doesn't mean wait forever until you get
something.
We can put a
Todd C. Miller wrote:
> On Mon, 21 Sep 2015 20:13:05 -0400, "Ted Unangst" wrote:
>
> > We can put a "long" poll() in front of pcap to wait until there are packets
> > (maybe never if you aren't using pf logging), and then let the timeout work
> > i
Stefan Sperling wrote:
> Properly reset the RX ring by clearing RX buffer status exposed to hardware.
> Found by Matthew Dillon while porting FreeBSD's iwm(4) to Dragonfly.
>
> ok?
>
> Index: if_iwm.c
> ===
> RCS file: /cvs/src/sys/d
Tobias Ulmer wrote:
> On Tue, Sep 22, 2015 at 11:46:08AM -0400, Ted Unangst wrote:
> > Todd C. Miller wrote:
> > > On Mon, 21 Sep 2015 20:13:05 -0400, "Ted Unangst" wrote:
> > >
> > > > We can put a "long" poll() in front of pcap to wa
Dimitris Papastamos wrote:
> I did some very crude benchmarking on my machine and I could push
> about 100k messages per second through a single message queue.
>
> At this point, I can start working on manpages + the necessary
> userspace glue.
>
> Is this patch going in the right direction?
I d
Jason McIntyre wrote:
> On Mon, Sep 21, 2015 at 10:40:29PM -0400, Michael McConville wrote:
> > Could also say "Use the err(3) and warn(3) family of functions." That's
> > what the err(3) man page intro refers to it as.
> >
>
> please some developer commit or reject this. thanks,
i liked the wor
Michael McConville wrote:
> Am I interpreting this correctly?
>
> This is the least invasive fix, but it's unfortunate that this function
> allows the supplied buffer to be NULL. If we made it unconditionally
> allocate a new buffer, we would have to change some program logic
> because uses pass s
Michael Reed wrote:
> Hi all,
>
> I test it out as both `arch' and `machine' and didn't observe
> any differences. Also, I figured I might as well convert exit(3)
> to return while touching this code, as was done in [1].
this is cleaner i think.
Index: arch.c
==
Fritjof Bornebusch wrote:
> Hi,
>
> this diff changes the following:
>
> - exit(3) to return at the end of main functions
> - use /* NOTREACHED */ were it belongs according to style(9)
I think that's obsolete. We no longer use lint. It's just noise now.
Anthony J. Bentley wrote:
> Kevin Zhang writes:
> > Using >, & in raw text is invalid HTML.
>
> You're right about &, but > does not need to be escaped.
Even & only needs escaping if the following letters could be interpreted as an
entity... Since & is only followed by & and space, that's not the
Earlier this month, kettenis identified a bug in uvm from an incompletely
reverted change. Clearing the wrong bit on a page would sometimes result in
the page queues being corrupted, and then eventually the system would crash or
panic.
This issue affects 5.7 and the forthcoming 5.8 release.
Patch
there's nothing random about this file, and it upsets me whenever i see it's
name go by, because i panic thinking mg has some stupid random number
generator that hasn't been fixed yet. util.c is a better name.
(and cvs rm/add)
Index: Makefile
==
Several utilities have command line parsing that predates getopt. Several of
these are the old kmem grovellers, which allowed specifying the custom nlist,
etc.
I think this is kind of unusual and think that generally, the fewer
undocumented features the better. This is as likely to trap people who
Sebastien Marie wrote:
> - if an exec'ed program starts with herited TAME flags: the
> initialisation of the program would be difficult as it would be
> already tamed.
i've been thinking about this some more. true in some cases, but i think in
many cases, what we are banning should be ba
as seen in cron, make the popen replacement nicer. this also repairs two
abuses of comma operators and an unnecessary function pointer.
Index: extern.h
===
RCS file: /cvs/src/libexec/ftpd/extern.h,v
retrieving revision 1.18
diff -u -
Rob Pierce wrote:
> On Mon, Oct 05, 2015 at 05:38:34PM +0059, Jason McIntyre wrote:
> > On Mon, Oct 05, 2015 at 11:50:49AM -0400, Rob Pierce wrote:
> > > There are some offending braces. I just added leading tabs in the right
> > > places to correct indentation.
> > >
> > > Rob
> > >
> >
> > why
Benny Lofgren wrote:
> On 2015-10-05 22:21, Rob Pierce wrote:
> > On Mon, Oct 05, 2015 at 05:38:34PM +0059, Jason McIntyre wrote:
> >> On Mon, Oct 05, 2015 at 11:50:49AM -0400, Rob Pierce wrote:
> > Is it KNF compliant to have an exit() or return() at the same indentation as
> > the closing functio
Without mucking about in the internals, here are some toplevel tame calls.
Index: signify.c
===
RCS file: /cvs/src/usr.bin/signify/signify.c,v
retrieving revision 1.100
diff -u -p -r1.100 signify.c
--- signify.c 16 Jan 2015 06:16:1
Ted Unangst wrote:
> Without mucking about in the internals, here are some toplevel tame calls.
check return values. ok, ok.
in the fairly common verify case of piping msgfile to - (as in patching), we
can cut things down a bit more as well.
Index: signif
ksh does a little dance to try and gift history files to their original owner
if it's somehow running as a different user. this of course only works as
root, and is probably a terrible idea.
ksh should simply refuse to open a history file that's owned by somebody else.
Index: history.c
=
i think it's idiotic for a program to offer users the choice between stdio and
mmap modes. nobody knows which to pick, and it reveals some terrible
incompetence somewhere else in the code.
mmap is the default. it's effectively the only tested codepath. so burn down
the stdio code path.
pros: ever
Philip Guenther wrote:
>
> As a general rule, programs should not use errno values as an exit status.
>
> Compare "sleep 10001" w/ and w/o this diff.
>
> ok?
agreed, but why not return 1? i don't want to have to slap 2>/dev/null around
all my sleep calls now.
>
> Index: sleep.c
>
Stefan Sperling wrote:
> The iwm(4) driver pre-allocates fw command payload buffers of 320 bytes.
>
> For some firmware commands, particularly those used when configuring
> the PHY (iwm_send_phy_db_cmd) and running scans (iwm_mvm_scan_request),
> the payload exceeds 320 bytes. I've seen somewhere
Stefan Sperling wrote:
> On Sun, Oct 11, 2015 at 11:31:35AM -0400, Ted Unangst wrote:
> > huh? Why an mbuf? Is dma_alloc not a better choice?
>
> The mbuf pointer already exists to keep track of packets on the
> otheor Tx queues. I guess that's why iwn (where this came f
Philip Guenther wrote:
> On Sun, Oct 11, 2015 at 8:19 AM, Ted Unangst wrote:
> > agreed, but why not return 1? i don't want to have to slap 2>/dev/null
> > around
> > all my sleep calls now.
>
> So don't. Why would you need to?
doesn't err() print a message?
Sevan Janiyan wrote:
>
>
> On 12/10/2015 16:30, Theo de Raadt wrote:
> > You have failed to explain the usage case; I don't believe there is
> > any scenario where people benefit from more architecture divergence.
>
> As an example, in pkgsrc, CPUs architectures are grouped under a common
> alia
Remove some more of the crypt() calls in login_chpass.
Index: login_chpass/Makefile
===
RCS file: /cvs/src/libexec/login_chpass/Makefile,v
retrieving revision 1.7
diff -u -p -r1.7 Makefile
--- login_chpass/Makefile 22 Apr 2014
The OBJ_obj2txt function in libcrypto contains a one byte buffer overrun
and memory leak, as reported by Qualys Security. This can be abused by an
attacker to cause a denial of service in some cases.
Patches are now available for OpenBSD as well as new releases of LibreSSL
portable. 5.6, 5.7, and
Stuart Henderson wrote:
> It does not build as-is, there's an issue with "c_get_text" (a renamed
> yyget_text) with our version of flex. I have two possible workarounds
> so far but neither is particularly nice;
>
> - add a configlexer.c file generated with newer flex to the tree (e.g.
> the one i
André Schneider wrote:
> Hi,
>
> maybe i'm overlooking something, but in the rebuild and install
> instructions,
> shouldn't it say
>cd /usr/src/lib/libcrypto
> instead of
>cd src/lib/libcrypto
either or. maybe somebody has their src somewhere else.
ul appears somewhat useless for its intended purpose.
echo _xxx_ | ul does not result in underlined text in an xterm, so I doubt
many people are using this.
Unlike, say, mandoc, it can't output Greek letters. I also imagine most people
have moved on to some form of markdown for their other text m
Nicholas Marriott wrote:
> Well, it does work:
>
> printf 'A\bA_\bB'|ul
>
> I still think it is not useful, I say kill it.
Oh! Is that how you use it? The man page doesn't explain, apparently expecting
that everybody just knows there's only one true way to mark up text.
Christian Weisgerber wrote:
> On 2015-10-23, "Ted Unangst" wrote:
>
> > ul appears somewhat useless for its intended purpose.
>
> mandoc /usr/share/man/man1/ls.1 | ul
>
> Works fine. Of course that functionality has been incorporated
> into more/less de
Stefan Sperling wrote:
> On Fri, Oct 23, 2015 at 05:50:53AM -0400, Ted Unangst wrote:
> > well, it doesn't work with utf-8 because it tries to underline only half the
> > character. i'm aiming for the "quick fix"...
>
> Why not at least try a kind of bett
rs doesn't print nicely aligned columns with utf-8 inputs.
there's a few ways to handle this; here's just one.
note that the source is riddled with lots of code like:
if (maxlen < p - *ep) /* update maxlen */
maxlen = p - *ep;
I'm very scared to try count
So, third diff to ponder as we evaluate this approach. This one also uses a
u8len() function to help get the column widths correct.
(Still not dealing with combining or otherwise not 1 width glyphs.)
Index: ls.c
===
RCS file: /cvs/sr
Stefan Sperling wrote:
> This removes the isprint() check entirely. Do we really want that?
>
> To what degree should tools like ls protect terminals from escape codes?
I think this is beyond the scope of what ls should care about. du doesn't have
such a check. Does the shell perform a check befo
Peter Hessler wrote:
> As a different approach to ls, I wrote this a while ago. This uses the
> wchar_t functions, but only in putname().
This will correct the alignment of columns, but if you have a filename like
pöp
the columns will be super wide instead of nicely sized.
Nicholas Marriott wrote:
> Hi
>
> This doesn't account for UTF-8 double width characters, so they will
> still throw the column widths off?
right. maybe we will steal some code from tmux for that :). but getting
u8len() into the right places is the first step.
i don't think we want a isu8cont()
Christian Weisgerber wrote:
> Ted Unangst:
>
> > --- ul.c10 Oct 2015 16:15:03 - 1.19
> > +++ ul.c23 Oct 2015 10:29:43 -
> > @@ -241,6 +241,8 @@ mfilter(FILE *f)
> > obuf[col].c_mode |= BOLD|mode;
> > e
Christian Weisgerber wrote:
> On 2015-10-23, "Ted Unangst" wrote:
>
> >> To what degree should tools like ls protect terminals from escape codes?
> >
> > I think this is beyond the scope of what ls should care about. du doesn't
> > have
> &g
Pablo Méndez Hernández wrote:
> Hi,
>
> I would say that some information about the authors is missing...
I would assume that would be included at the top of the new files, but that
does point out they're missing.
cvs add, then diff. (with -N if they still don't show up.)
this burns the tmpdir out of diff, so as to let pledge tmppath work a little
better.
Index: diff.1
===
RCS file: /cvs/src/usr.bin/diff/diff.1,v
retrieving revision 1.44
diff -u -p -r1.44 diff.1
--- diff.1 5 Oct 2015 20:15:00 -00
Walter Neto wrote:
> Hi guys, sorry for the inconvenience.
>
> Once you only have use git at console, cvs is nightmare :(
>
> But I am learning :)
this indeed looks a lot better. although it's quite the diff to review.
some brave soul testing of this might be appreciated.
Reyk Floeter wrote:
> Hi,
>
> as requested by Theo and discussed with many, the following diff moves
> it into a new driver. This also allowed to improve the logic of link
> states related to the connection (as discussed with Claudio).
>
> The new driver is called pair(4).
>
> # ifconfig
Alexander Bluhm wrote:
> Hi,
>
> Before removig nls, I would like to unify all error messages in
> - include comment
> - libc comment
> - nls C
> - man page
>
> The EPROGUNAVAIL change is just beautification. Any objections
> against that?
ok
Jan Stary wrote:
> Any comments/OK?
>
> On Oct 22 13:49:28, h...@stare.cz wrote:
> > games/quiz.c popen()s a PAGER to display the help message.
> > Throw that away and let the use pipe into a PAGER if needed.
> > Then we can just pledge "stdio rpath".
yeah, works for me.
Ted Unangst wrote:
> Christian Weisgerber wrote:
> > On 2015-10-23, "Ted Unangst" wrote:
> >
> > >> To what degree should tools like ls protect terminals from escape codes?
> > >
> > > I think this is beyond the scope of what ls should care
Raf Czlonka wrote:
> This is unrelated to your diff but what I do instead is to check
> the BUILDINFO file - it's tiny and all the information you need,
> is already there.
While we're on the subject, I'll point out that sometimes only the base sets
change, but not the x sets. If you download SHA2
Otto Moerbeek wrote:
> Hi,
>
> Malloc maintains a list if 16 slots of chunks to be freed. On free a
> chunk is put in a random slot and the existing chunk in that slot is
> actually freed. Currently, the code only checks the slot selected for
> a double free.
>
> This diff adds code to check all
Stuart Henderson wrote:
> I have been bumping into maxproc limits a lot recently on my
> workstation. While I do have quite a lot of processes at times, I don't
> think I have quite been tripping into the limits of number of actual
> processes.
>
> It seems that perhaps threads are counted towards
Ingo Schwarze wrote:
> [EILSEQ]
> A wide-character code that does not correspond to a valid
> character has been detected.
>
> That means that the functions are *required* to fail ("shall fail")
> if encoding errors can be detected, that -1 must be returned, and
> that errno must b
Stefan Sperling wrote:
> Or is modifying ifconfig sufficient?
> We are more concerned about textual display rather than the
> kernel/userland ioctl boundary, correct?
>
> The option list for ifconfig is [-AaC]. Plenty of letters available.
> We could add:
>
>-P Show authentication details su
Jeremie Courreges-Anglas wrote:
>
> I don't think anything has been committed regarding this issue, right?
this looks ok to me.
Alexander Kuleshov wrote:
> pipe()/pipe2() system calls may fail now with ENOMEM error during creation in
> pipe_create() -> pipespace(). The bellow diff replaces ENOMEM with ENFILE
> errno which is already documented for pipe()/pipe2() to decrease amount of
> possible errno(s) with similar behavio
Theo Buehler wrote:
> The diff below only skips 6 letters if they actually are '--- a/' or
> '+++ b/' and seems to do the trick for me in various scenarios.
asymptotically approaching correct, i like it.
unifdef some foolishness. note that BSD is not currently defined when this
file is built.
Index: ftpcmd.y
===
RCS file: /cvs/src/libexec/ftpd/ftpcmd.y,v
retrieving revision 1.60
diff -u -p -r1.60 ftpcmd.y
--- ftpcmd.y18 Aug 2015
Daniel Micay wrote:
> This patch adds an opt-in malloc configuration option placing canaries after
> small allocations to detect heap overflows on free(...). It's intended to be
> used alongside guard pages for large allocations. Since it's essentially
> adding extra padding to all small allocation
it only gets deeper and thicker...
this decodes chars and prints ? for bytes it doesn't like, as well as
codepoints (128-159) it doesn't like.
(this is extracted from some old utf8 code i had laying around. it's a bit
simpler than the stringprep stuff but it seems to handle the case of some
incor
Damien Miller wrote:
> rather than scattering hacks in each program that needs to
> output utf8 to the console, how about making something
> for libutil that they all can use?
Yes, that is certainly the plan, but I think it's easier to see what's needed
if we convert a few programs first to identi
when i run top, i think uptime would be interesting info to include. there's
some space on the second line under the hostname and time. i tried to squeeze
it on the first line, but there isn't much room with 80 columns.
Index: display.c
Old bug in pwcache functions. Calling setpassent(1) to keep the passwd
database open is a surprising abstraction violation for the caller of
user_from_uid. Now it has a file descriptor it must close before exec by
calling endpwent(), but this fact is not mentioned. (find is affected by this,
for ex
This improves the cache. Basically, it's kind of like four way associative
now, with LRU replacement. Also we can cache nameless entries instead of going
back to getpwuid every time.
Index: gen/pwcache.c
===
RCS file: /cvs/src/lib/lib
Michael McConville wrote:
> We have a pretty strong guarantee that it can only happen once per
> process...
I don't think this really matters. What does it do to the assmembly, and how
does that make things faster?
Todd C. Miller wrote:
> On Mon, 26 Oct 2015 11:39:48 -0400, "Ted Unangst" wrote:
>
> > This improves the cache. Basically, it's kind of like four way
> > associative now, with LRU replacement. Also we can cache nameless
> > entries instead of going back
Philip Guenther wrote:
> On Mon, Oct 26, 2015 at 6:47 AM, Ted Unangst wrote:
> > Old bug in pwcache functions. Calling setpassent(1) to keep the passwd
> > database open is a surprising abstraction violation for the caller of
> > user_from_uid. Now it has a file descriptor
Philip Guenther wrote:
> On Mon, Oct 26, 2015 at 9:05 AM, Ted Unangst wrote:
> > Philip Guenther wrote:
> >> On Mon, Oct 26, 2015 at 6:47 AM, Ted Unangst wrote:
> >> > Old bug in pwcache functions. Calling setpassent(1) to keep the passwd
> >> >
Michael McConville wrote:
> Ted Unangst wrote:
> > Michael McConville wrote:
> > > We have a pretty strong guarantee that it can only happen once per
> > > process...
> >
> > I don't think this really matters. What does it do to the assmembly,
>
Constantine Aleksandrovich Murenin wrote:
> As reported elsewhere (http://seclists.org/oss-sec/2015/q4/87 via
> http://www.opennet.ru/43146), both of these errors were introduced as
> part of the refactoring.
>
> Quick glance through
> http://bxr.su/o/lib/libssl/src/crypto/objects/obj_dat.c#OBJ_ob
Constantine Aleksandrovich Murenin wrote:
> So you confirm that using block scope in such scenario is the new best
> practice now?
I don't think it was ever not best practice; though whether people do it or
not is a separate matter.
Anthony J. Bentley wrote:
> Stefan Sperling writes:
> > On Mon, Oct 26, 2015 at 03:58:58PM -0600, Anthony J. Bentley wrote:
> > > "Ted Unangst" writes:
> > > > it only gets deeper and thicker...
> > >
> > > Indeed.
> > >
&
Anthony J. Bentley wrote:
> "Ted Unangst" writes:
> > Fixing citrus is a pretty massive effort in itself. I'd prefer to see the
> > replacement code prove itself as a separate API first, then we can remove
> > citrus and change the wchar functions to use t
This adds a quite limited understanding of utf-8 to hexdump. I've found it
helpful trying to see exactly what's coming out of some utilities instead of
trying to decode utf-8 by hand.
Index: display.c
===
RCS file: /cvs/src/usr.bin/he
Stuart Henderson wrote:
> On 2015/10/27 06:51, Ted Unangst wrote:
> > This adds a quite limited understanding of utf-8 to hexdump. I've found it
> > helpful trying to see exactly what's coming out of some utilities instead of
> > trying to decode utf-8 by hand.
&g
Dimitris Papastamos wrote:
> There was a comment in the code that indicated that it might be worth
> investigating the use of trees. I have not currently done any kind of
> serious benchmarking on this but I am looking into it.
nice.
> +static int
> +reqcmp(struct request *r1, struct request *r2
Dimitris Papastamos wrote:
> Hi,
>
> I thought it would be cool for rebound(8) to load balance on a number of
> DNS servers.
>
> While I was working on this, I did not manage to convince myself as to
> whether this should be the default behaviour.
>
> An alternative default would be to use the m
701 - 800 of 1831 matches
Mail list logo
