Date: Thu, 27 Feb 2014 23:16:19 -0800 From: John-Mark Gurney <jmg@funkthat.com> To: Kohji Okuno <okuno.kohji@jp.panasonic.com> Cc: freebsd-current@freebsd.org Subject: Re: About kevent Message-ID: <20140228071619.GO47921@funkthat.com> In-Reply-To: <20140228.111358.1991189164034042401.okuno.kohji@jp.panasonic.com> References: <20140228.111358.1991189164034042401.okuno.kohji@jp.panasonic.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Kohji Okuno wrote this message on Fri, Feb 28, 2014 at 11:13 +0900:
> I have a question about kevent.
>
> How should the userland judge knote which is cleared from knlist by
> knlist_clear() or knlist_delete()?
It looks like I need to read the code better... knlist_clear (killkn=0)
and knlist_delete (killkn=1) are wrappers around knlist_cleardel...
Looking at the code of knlist_cleardel, if killkn is set
(knlist_delete) the knote will be dropped (free'd)... if it is not set,
the flags _EOF and _ONESHOT will be set such that it'll be returned
soon..
Now that I look at the code, KNOTE_ACTIVATE is never called to be put
on the list to be delivered, so now I'm not sure if it works the way
it's suppose to... I have a feeling that the notes might hang around
forever until the kq fd is closed...
I'm also puzzled as to why _DETACHED isn't set, which would seem to
mean that we'll call f_detach when we close the kq, which I assume
could cause a panic...
This needs to be investigated/tested...
--
John-Mark Gurney Voice: +1 415 225 5579
"All that I will do, has been done, All that I have, has not."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20140228071619.GO47921>