svn commit: r215178 - in head: lib/libc/sys sys/kern sys/sys

[Ivan Voras]

On 16 November 2010 02:48, Luigi Rizzo <rizzo at iet.unipi.it> wrote:
> On Tue, Nov 16, 2010 at 01:33:53AM +0100, Ivan Voras wrote:
>> On 15 November 2010 18:10, Luigi Rizzo <rizzo at iet.unipi.it> wrote:
>>
>> > 2. [generic] passing pointers between userland and kernel
>> > requires remapping the pointer when going up or down.
>> > As the mapping would be application specific, i don't
>> > see much use in allowing room for a pointer without kernel code
>> > to map userland <-> kernel pointers.
>>
>> I'm not thinking of passing a *working* pointer into the kernel but
>> used as a cookie, similar to how it's used in kqueue: the intention
>> being the application can send and get a pointer which means something
>> to the application, not something usable to the kernel.
>
> oh, but then you are thinking of something completely different.
> The SO_USER_COOKIE is never returned to the application; it is
> only passed to another kernel subsystem, so it must be significant
> there, not for the application.

Ah, ok then. I was thinking you are maybe trying to do something else,
like tagging a socket with additional information from userland.