Re: [Call For Wartectomy] CRLF conversion out of kernel

Khimenko Victor (khim@sch57.msk.ru)
Fri, 16 Jul 1999 03:08:42 +0400 (MSD)

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: Kanoj Sarcar: "gcc/as question"
Previous message: Richard Gooch: "Re: kernel thread support - LWP's"
Maybe in reply to: Alexander Viro: "[Call For Wartectomy] CRLF conversion out of kernel"
Next in thread: Brian Gerst: "Re: [Call For Wartectomy] CRLF conversion out of kernel"

In <19990715222441.J25966@mencheca.ch.genedata.com> Matthew Wilcox (Matthew.Wilcox@genedata.com) wrote:
> On Thu, Jul 15, 1999 at 09:29:45PM +0200, Jamie Lokier wrote:
>> Marc Mutz wrote:
>> > No. IP masq. changes IP _headers_; he was talking of data.
>>
>> IP masq changes the data too for specific apps. Simply rewriting the
>> headers is not enough for applications which pass addresses & port
>> numbers in their protocol.

> I guess the important difference is that IP masquerading only substitutes
> bytes; it neither inserts nor deletes bytes. Not to mention that you
> can't mmap() a network stream. It's not a good comparison to CRLF
> conversion. My vote is to kill that feature in 2.3 and retroactively
> add a warning to 2.2.11.

IP masq. just can not be effectively and reliable done in userspace. That's
main difference IMO...

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

Next message: Kanoj Sarcar: "gcc/as question"
Previous message: Richard Gooch: "Re: kernel thread support - LWP's"
Maybe in reply to: Alexander Viro: "[Call For Wartectomy] CRLF conversion out of kernel"
Next in thread: Brian Gerst: "Re: [Call For Wartectomy] CRLF conversion out of kernel"