Re: Kqueue write event position?
- From: "Vlad Galu" <dudu@xxxxxxx>
- Date: Thu, 23 Nov 2006 12:22:09 +0200
On 11/23/06, Ivan Voras <ivoras@xxxxxx> wrote:
Vlad Galu wrote:
> It seems to me you would have to propagate that info along the
> VOP_WRITE_POST->VFS_KNOTE_LOCKED->VN_KNOTE->knote() chain. Since
> knote() is generic and is used for all types of notifications, you can
> probably roll down your own replacement and call it from VN_KNOTE. Of
> course, there probably is a better way :)
Before I get any further with this, can anyone predict what would be the
impact of monitoring a large number of files (50,000 or more) with
kqueue? For kqueue to work, each file would have to be opened (and take
up one file descriptor).
My guess is that it won't be remarcably high. However, you can
create those files, add them to your notification list and randomly
write bytes to them, to see how your system performs. One more
suggestion, I think it would be better if, in case you extend the
vnode API, you only send the size of the write in the data field of
the kevent structure and you store the offset in your program and
adding the size to it everytime an event occurs.
_______________________________________________
freebsd-hackers@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@xxxxxxxxxxx"
--
If it's there, and you can see it, it's real.
If it's not there, and you can see it, it's virtual.
If it's there, and you can't see it, it's transparent.
If it's not there, and you can't see it, you erased it.
_______________________________________________
freebsd-hackers@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@xxxxxxxxxxx"
- Follow-Ups:
- Re: Kqueue write event position?
- From: Ivan Voras
- Re: Kqueue write event position?
- References:
- Kqueue write event position?
- From: Ivan Voras
- Re: Kqueue write event position?
- From: Vlad Galu
- Re: Kqueue write event position?
- From: Ivan Voras
- Kqueue write event position?
- Prev by Date: Re: Kqueue write event position?
- Next by Date: Re: Kqueue write event position?
- Previous by thread: Re: Kqueue write event position?
- Next by thread: Re: Kqueue write event position?
- Index(es):
Relevant Pages
- Re: mtx_lock_recurse/mtx_unlock_recurse functions (proof-of-concept).
... >> The headache of kqueue is that KNOTEultimately calls back the caller ...
maybe we can attach an event queue to a knote list. ... > The MAC Framework is also
an incestuous subsystem, but it does a lot less ... If you treat the KNOTE lock
as a leaf lock, ... (freebsd-arch) - kqueue giant-locking (&kq_Giant, locking)
... with no ill effect:) The locking thus far is one global kqueue lock, ... KNOTE()
can't end up calling another KNOTEunless ... Kqueue does the locking for you when it comes
to the non-object lists. ... KNOTEwith locks held, but you should reference your object.
... (freebsd-arch) - once again, MP-safe kqueue...
... EVFILT_PROC+NOTE_TRACK and allows recursing KNOTE() ) calls. ... There is one
subsystem lock still, ... There can be one writer which may sleep and drop the kqueue
subsystem ... (freebsd-arch) - Re: lio_listio fixes and adding kqueue notification patch
... > Here's a patch to fix some lio_listio bugs and add kqueue support to it. ...
> during the locking changes. ... there could be a knote added... ...
[aio stuff deleted] ... (freebsd-current)