TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that sexy?)

TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that sexy?)

[Dan Kegel]

Tony Finch wrote:
> 
> Tony Finch <dot@dotat.at> wrote:
> >Linus Torvalds <torvalds@transmeta.com> wrote:
> >>
> >>Without proper uncorking (and it really shouldn't be that hard to
> >>add), TCP_NOPUSH simply can't be used in the generic sense.
> >
> >It was easy :-) I've put up a patch for FreeBSD that adds proper
> >uncorking on my homepage <http://apache.org/~fanf/> in the "Stuff"
> >section.
> 
> ... and it has been committed to -CURRENT, too.

How very cool.

How close is TCP_NOPUSH to behaving identically to TCP_CORK now?
If it does behave identically, it might be time to standardize
the symbolic name for this option, to make apps more portable
between the two OS's.  (It'd be nice to also standardize the
numeric value, in the interest of making the ABI's more compatible, too.)

Here are the definitions in the two OS's at the moment:

FreeBSD: netinet/tcp.h (from 
http://minnie.cs.adfa.oz.au/FreeBSD-srctree/newsrc/netinet/tcp.h.html#TCP_NOPUSH )

/*
 * User-settable options (used with setsockopt).
 */
#define TCP_NODELAY     0x01    /* don't delay send to coalesce packets */
#define TCP_MAXSEG      0x02    /* set maximum segment size */
#define TCP_NOPUSH      0x04    /* don't push last block of write */

Linux: netinet/tcp.h:

/*
 * User-settable options (used with setsockopt).
 */
#define TCP_NODELAY 0x01    /* don't delay send to coalesce packets */
#define TCP_MAXSEG  0x02    /* set maximum segment size */
#define TCP_CORK    0x03    /* control sending of partial frames */      

- Dan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that

[Alan Cox]

> How close is TCP_NOPUSH to behaving identically to TCP_CORK now?
> If it does behave identically, it might be time to standardize
> the symbolic name for this option, to make apps more portable
> between the two OS's.  (It'd be nice to also standardize the
> numeric value, in the interest of making the ABI's more compatible, too.)

That one isnt practical because of the way the implementations handle 
boolean options. BSD uses bitmask based option setting for the basic
options and Linus uses switch statements
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that

[Dan Kegel]

Alan Cox wrote:
> 
> > How close is TCP_NOPUSH to behaving identically to TCP_CORK now?
> > If it does behave identically, it might be time to standardize
> > the symbolic name for this option, to make apps more portable
> > between the two OS's.  (It'd be nice to also standardize the
> > numeric value, in the interest of making the ABI's more compatible, too.)
> 
> That one isnt practical because of the way the implementations handle
> boolean options. BSD uses bitmask based option setting for the basic
> options and Linus uses switch statements

OK, well, at least a common symbolic name could be chosen.
- Dan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that sexy?)

[Tony Finch]

Dan Kegel <dank@alumni.caltech.edu> wrote:
>
>How close is TCP_NOPUSH to behaving identically to TCP_CORK now?

They are exactly the same.

>If it does behave identically, it might be time to standardize
>the symbolic name for this option, to make apps more portable
>between the two OS's.  (It'd be nice to also standardize the
>numeric value, in the interest of making the ABI's more compatible, too.)

I wonder if it's a bit late for that now...

This reminds me to make sure that FreeBSD's Linux emulation supports
TCP_CORK properly.

Tony.
-- 
f.a.n.finch    fanf@covalent.net    dot@dotat.at
" ``Well, let's go down and find out who's grave it is.''
``How?''  ``By going down and finding out!'' "
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that

[Tony Finch]

Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
>> How close is TCP_NOPUSH to behaving identically to TCP_CORK now?
>> If it does behave identically, it might be time to standardize
>> the symbolic name for this option, to make apps more portable
>> between the two OS's.  (It'd be nice to also standardize the
>> numeric value, in the interest of making the ABI's more compatible, too.)
>
>That one isnt practical because of the way the implementations handle 
>boolean options. BSD uses bitmask based option setting for the basic
>options and Linus uses switch statements

No, that's only true for some of the socket-level options. For the TCP
options there isn't a direct correspondance between the option number
and the number of the flag in the PCB.

Tony.
-- 
f.a.n.finch    fanf@covalent.net    dot@dotat.at
"Dead! And yet there he stands!"
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that

[kuznet]

Hello!

> > How close is TCP_NOPUSH to behaving identically to TCP_CORK now?

They have not so much of common.

TCP_NOPUSH enables T/TCP and its presense used to mean that
T/TCP is possible on this system. Linux headers cannot
even contain TCP_NOPUSH.

Alexey
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that

[Dan Kegel]

Alexy wrote:

> > > How close is TCP_NOPUSH to behaving identically to TCP_CORK now? 
> 
> They have not so much of common. 
> 
> TCP_NOPUSH enables T/TCP and its presense used to mean that 
> T/TCP is possible on this system. Linux headers cannot 
> even contain TCP_NOPUSH.

But Tony Finch wrote:
> They are exactly the same. 

Alexy, Tony just checked in a change to FreeBSD to make TCP_NOPUSH behave the
same as TCP_CORK.

Tony, are people using the TCP_NOPUSH define as a way to detect
the presence of T/TCP support?  In that case, perhaps the right
thing to do to achieve source compatibility would be for FreeBSD
to also define TCP_CORK (and give it TCP_NOPUSH as a value, perhaps).

- Dan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that

[Tony Finch]

Dan Kegel <dank@alumni.caltech.edu> wrote:
>
>Tony, are people using the TCP_NOPUSH define as a way to detect
>the presence of T/TCP support?

No, MSG_EOF is the right way to do that.

Tony.
-- 
f.a.n.finch    fanf@covalent.net    dot@dotat.at
FAIR ISLE: WESTERLY VEERING NORTHERLY 4 OR 5. WINTRY SHOWERS. MAINLY GOOD.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: TCP_NOPUSH on FreeBSD, TCP_CORK on Linux (was: Is sendfile all that

[Linus Torvalds]

In article <20010208035803.L74296@hand.dotat.at>,
Tony Finch  <dot@dotat.at> wrote:
>Dan Kegel <dank@alumni.caltech.edu> wrote:
>>
>>Tony, are people using the TCP_NOPUSH define as a way to detect
>>the presence of T/TCP support?
>
>No, MSG_EOF is the right way to do that.

However, I think ank is at least partially correct: TCP_NOPUSH has some
magic behaviour for sockets in listen state, and turns on at least some
T/TCP semantics, if I remember correctly. Tony?

If I remember correctly, may I suggest something: make a new BSD option
called (ehh, just random name ;) TCP_CORK, and make the old BSD
TCP_NOPUSH option be a superset of TCP_CORK that also turns on T/TCP on
listen sockets.

Linux TCP_CORK doesn't have anything to do with T/TCP (not surprisingly,
as T/TCP is considered a broken protocol in Linux and other circles).

And Linux TCP_CORK _is_ used on listen sockets: it makes sockets that
are accepted from the listen socket have the corking semantics. In
contrast, BSD TCP_NOPUSH, I think, has this overloading issue..

		Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/