's actuallty the KAME guys doing
a lot of work to integrate into _us_.
> 5) Will the v6 code by multi-threaded?
Hopefully, there's nothing inherently bad with it that will make it
too difficult.
> Well, that's a start.
Why the sudden interest?
--
-Alfred Perlstein [[EMAIL PROT
r 2002) the book won't be too far behind it.
Interesting! I would target 5.0 as that's where a bunch of activity
is going to be, I imagine a lot of changes may occur, but there's
just about zero possibility of a rewrite, that would just be absurd.
--
-Alfred Perlstein [[EMAIL PROTECTED
russ(1) output might help, along with the code in question, along
with what type of descriptors it's polling.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring 30 years of a
e I really don't feel much anyone has the skills necessary
to duplicate the equivelant of several hundered man years of
developement in just a couple of months no matter how highly they
think of themselves.
In summary, I completely agree with Jonathan and all the others
opposing such a motion. Yo
* George V. Neville-Neil <[EMAIL PROTECTED]> [011130 15:15] wrote:
> > I don't think a rewrite is ever going to be much of a good idea,
> > a restructuring might, meaning that fixing up all the layering
> > and making it more flat like Van Jacobson suggested (and Linux
> > implemented in one of th
* Alfred Perlstein <[EMAIL PROTECTED]> [011213 12:08] wrote:
> * Bruce Evans <[EMAIL PROTECTED]> [011213 05:56] wrote:
> >
> > >From POSIX.1-200x-draft7 (this has not changed since at least the 1990
> > version):
> >
> > ! 36609 Wh
* Alfred Perlstein <[EMAIL PROTECTED]> [011215 15:10] wrote:
>
> Can people take a look at this fix? It seems to dtrt, but I need feedback
> here.
>
> It basically backs out my last two revisions and changes the hacks the
> poll call to seemingly do the right thing.
I
* Bruce Evans <[EMAIL PROTECTED]> [011217 07:10] wrote:
> On Sat, 15 Dec 2001, Alfred Perlstein wrote:
>
> > Can people take a look at this fix? It seems to dtrt, but I need feedback
> > here.
> >
> > It basically backs out my last two revisions and ch
* Alfred Perlstein <[EMAIL PROTECTED]> [011217 12:47] wrote:
> * Bruce Evans <[EMAIL PROTECTED]> [011217 07:10] wrote:
>
> > > +
> > > filetmp.f_data = (caddr_t)ap->a_vp->v_fifoinfo->fi_readsock;
> > > + so = (stru
Memory address is
> wrong.
>
> Can any one tell me if socket call can be used in kernel level? If not, how
> can I accomplish socket communication in the kernel level?
This is nowhere near enough information for anyone to be able to
help you. Please be more specific, show us some example
* Bruce Evans <[EMAIL PROTECTED]> [011221 08:27] wrote:
>
> I think it should be more like:
>
> if (ap->a_events & POLLIN)
> ap->a_events = (ap->a_events & ~POLLIN) | POLLIN_IGNORE_EOF;
>
> We can't pass in flags that we don't really care about (POLLIN in
> this case) since
amp;sinsiz)) == -1)
fatal ("Error in accept(): %s", strerror (errno));
OOPS!
You're missing:
sinsiz = sizeof(sin);
before the call to accept.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
st
* Harti Brandt <[EMAIL PROTECTED]> [020124 00:54] wrote:
> On Wed, 23 Jan 2002, Terry Lambert wrote:
>
> TL>Ruslan Ermilov wrote:
> TL>> [Redirected to -net with -hackers Bcc:ed]
> TL>> > is there any way to get at the if_multiaddrs list from user space (except
> TL>> > for digging through the ke
* Vinod Namboodiri <[EMAIL PROTECTED]> [020206 11:51] wrote:
> can anyone tell me the url where i can find the TCP/IP
> stack source code.is the following url the right one?
> http://www.FreeBSD.org/cgi/cvsweb.cgi/src/sys/netinet/
Yes, also look in the sys, net and kern directories.
-Alfred
To
* Ari Suutari <[EMAIL PROTECTED]> [020211 22:50] wrote:
> Hi,
>
>
> On Monday 11 February 2002 16:15, Mike Silbersack wrote:
> > On Mon, 11 Feb 2002, Alfred Perlstein wrote:
> >
> > > another way would be to loop doing recvfrom's until EAGAIN is retu
* Nuno Miguel Fernandes Sucena Almeida <[EMAIL PROTECTED]> [020216 20:29] wrote:
> Hello,
> i'm using getifaddrs to get the IP and HW address of an ethernet
> interface but it seems that the ifap->ifa_next pointer is never NULL so I
> end up with an infinite loop! If i uncomment the // print
kernel after inbound processing. (And the machine works fine on a
> network; I found this by accident.)
>
> Any clues?
I'm not 100% on this, but bpf may be queueing the packet without
doing a deep copy of it? Another thing may be that the card performs
the checksum but uses the
* Marco Molteni <[EMAIL PROTECTED]> [020225 00:45] wrote:
> Hi all,
>
> I have a small patch to add a comment to struct inpcb, explaining the
> role of member "vflag" (IP version flag, v4/v6). It took me a while to
> figure it out so I think it might be better to explicitly explain it.
Your patc
* Alfred Perlstein <[EMAIL PROTECTED]> [020225 01:43] wrote:
> * Marco Molteni <[EMAIL PROTECTED]> [020225 00:45] wrote:
> > Hi all,
> >
> > I have a small patch to add a comment to struct inpcb, explaining the
> > role of member "vflag" (IP versi
o on. 80K worked but for all others, the system halted
> during the boot saying 'pager_swap_zone = null'. I am guessing that the
> kernel zone allocator cannot allocate enough map space for the pager swap?
> I am wondering if anyone knows how to get around this problem. Thanks.
am wondering if anyone knows how to get around this problem. Thanks.
> >
> > 4-stable has a fix for this.
>
> If the problem doesn't involve many files, is there anyway to get a diff
> patch against 4.3-RELEASE? Thanks.
Sure. Look at src/sys/vm/swap_pager.c near revisio
error 13'.
Now give it two or three shots and you may get the server
to lock up as well! (seems to run out of mbufs)
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring 30 ye
* Bruce M Simpson <[EMAIL PROTECTED]> [020301 01:58] wrote:
>
> o Would anybody be interested in my adding support for this beast's crypto
> features to the fxp driver?
Yes. :)
--
-Alfred Perlstein [[EMAIL PROTECTED]]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with
eeBSD too? I know the concepts are going to be
> similar like ip networking.But when it comes to system
> calls and stuff is it much different?
> Wouldappreciate the help very much.
> Vinod
The Stevens TCP/IP book Vol II is what you want for a FreeBSD
kernel networking internals walkthroug
like
the latest 4.5 version has this chipset... You aren't
running an older release or a custom kernel are you?
It should work with the 'rl' driver if you've got 4.5.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970
close to the same thing that
the zone allocator does except in a more flexible manner?
Sorry if the question is niave, I'm not extremely familiar with the
previous and current code.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s tec
* Jeff Roberson <[EMAIL PROTECTED]> [020320 12:29] wrote:
>
>
> On Wed, 20 Mar 2002, Alfred Perlstein wrote:
>
> >
> > That depends on what this implies. :)
> >
> > Does it mean that when giving M_NOWAIT there's a chance it may fail
> >
#x27;d be interested in
seeing patches to do this safely.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring 30 years of accumulated wisdom.'
Tax deductible donations for Fre
che flush.
However, If you were to make this a tunable (defaults to off) I
would commit it.
If you can show that Solaris caches open()s as well I'd be ok with
turning it on by default. :)
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using &quo
> Do you mean this is what you'd like or that it exists? I get:
>
> # sysctl net.inet.tcp.deleteconn="209.157.133.226.4296:216.136.204.21.80"
> sysctl: unknown oid 'net.inet.tcp.deleteconn'
> #
>
> when I try it.
It's a yahoo specific os hack. :)
--
-Alfred P
, otherwise store the allocated object into the
global and continue.
you may not call malloc(9) with M_WAITOK while holding a mutex.
---
entry = jumbo_kmap_inuse.slh_first;
I'm sure that should use a list macro.
---
That's all I see off the bat. :) Looks cool though.
--
-Alfred Per
t.
>
...
> Suggestions?
*slaps forhead*
Probably a SYSINIT?
> Cool, thanks for the feedback!
np, this is promising work!
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring
Something bizzaro with the 'struct ng_parse_struct_info' declarations,
please suggest or make a fix:
cc -c -O -pipe -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes
-Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual -Wno-format -ansi -g
-nostdinc -I- -I. -I../../.. -I../.
* Julian Elischer <[EMAIL PROTECTED]> [020603 12:41] wrote:
> this is YET ANOTHER case for the Atomic reference counting ABI that
> jhb has been talking about...
I was the initial person to request an atomic_t API.
-Alfred
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd
t; evolve relatively quickly.)
I submitted some patches to use sendfile(2) that weren't accepted
for some reason. It's not too hard, you just have to adjust the code
not to close(2) the descriptors and make the mmap() function a stub
type thing.
really out of date...
http://people.free
duce usage.
True, but it looks like he could raise nmbclusters in his config intead
of throttling.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring 30 years of accumulated wisdom.&
* Patrick Thomas <[EMAIL PROTECTED]> [020627 09:11] wrote:
>
> two followups:
>
> 1) is the tcp.recvspace an immediate tunable, or to get best results
> should I set it in rc.local ?
Why not look and find out? :)
> 2) when you say raise nmbclusters "in his config", may I assume you men my
> ke
and MCLGET are rather long macros, but in -CURRENT they
> are functions so there is not really much added code cache pollution.
Heh... :)
If the pool is per-device-softc then it doesn't need locks and will
be a lot more efficient than even grabbing from the per cpu pool.
Also, (/me does a
not my native language.
Your english is fine and this is an ok list to ask technical network
related questions.
Ok, now that you're re-assured, check out the manpage for 'getifaddrs'
that sounds like what you might need. The getifaddrs(3) function returns
a list of all the network
http://people.freebsd.org/~alfred/deltas/sockunion.diff.gz
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring 30 years of accumulated wisdom.'
Tax deductible donations for Fre
* Mike Silbersack <[EMAIL PROTECTED]> [020707 12:36] wrote:
>
> On Sun, 7 Jul 2002, Alfred Perlstein wrote:
>
> > Some time ago I noticed that there appeared to be several members
> > of struct socket that were either only used by listen sockets or
> > only u
eral case.
>
> I think Alfred was just trying to get his changes done first before I got
> around to committing what I have. :-) :-) :-) (It's currently in p4)
pfft, actually your return motivated me enough to try something I've
been meaning to try. :)
--
-Alfred Perlstein [[
, as it
should already be in network order.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring 30 years of accumulated wisdom.'
Tax deductible donations for FreeBSD: http://www.freebsdfounda
* Andrew Gallatin <[EMAIL PROTECTED]> [020711 11:32] wrote:
>
> Alfred Perlstein writes:
> > Some time ago I noticed that there appeared to be several members
> > of struct socket that were either only used by listen sockets or
> > only used by data sockets.
&
;t do per-cluster locking, or anything ridiculous like
> that), so unless Alfred has the reason again, I'll consider that method
> again instead. Thanks for the constructive feedback.
Yes it was NetBSD that did this.
How do you plan on manipulating a linked list without switching
is easy to tell by examining the ring pointer.
> I.e., you never have to iterate through the entire ring.
That's true, but could someone explain how one can safely and
effeciently manipulate such a structure in an SMP environment?
I'm not saying it's impossible, I'm just saying it didn't seem
intuative to me back then, as well as now.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-net" in the body of the message
* Julian Elischer <[EMAIL PROTECTED]> [020712 00:00] wrote:
>
>
> On Thu, 11 Jul 2002, Alfred Perlstein wrote:
> >
> > That's true, but could someone explain how one can safely and
> > effeciently manipulate such a structure in an SMP environment?
>
&g
*/
#define CSUM_FRAGMENT 0x0010 /* will do IP fragmentation */
Just use the first 3, have a look at the if_bge.c driver for
an example.
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technology,"
start asking why
It appears that we expect the ip_len and ip_off feilds to be sent
in host byte order as the stack will fix it to network byte order
in ip_output.
Is this a bug or feature? :)
--
-Alfred Perlstein [[EMAIL PROTECTED]]
'Instead of asking why a piece of software is using "1970s technolog
Any suggestion of the kind of cable one should look for at Frys
to run between two gigE card (intel em0) to function as a crossover?
--
- Alfred Perlstein
- Research Engineering Development Inc.
- email: [EMAIL PROTECTED] cell: 408-480-4684
___
[EMAIL
I had a cat5e cable, but either:
a) the box needed to reboot
b) 4.9 has a problem whereas 4-stable post 4.9 is ok with em0
I dunno, but it's working with a standard cat5e cable now after
the upgrade.
* Michael Sierchio <[EMAIL PROTECTED]> [031220 14:13] wrote:
> Alfred Perlstein
201 - 251 of 251 matches
Mail list logo
