Date: Tue, 10 Feb 2015 18:30:27 +0000 (GMT) From: Rui Paulo <rpaulo@me.com> To: John Baldwin <jhb@freebsd.org> Cc: "svn-src-head@freebsd.org" <svn-src-head@freebsd.org>, Adrian Chadd <adrian@freebsd.org>, "src-committers@freebsd.org" <src-committers@freebsd.org>, "svn-src-all@freebsd.org" <svn-src-all@freebsd.org> Subject: Re: svn commit: r278479 - in head: etc sys/kern Message-ID: <8e5503e1-755c-49e4-ab4d-a0ad1ae91f97@me.com>
next in thread | raw e-mail | index | archive | help
On Feb 10, 2015, at 07:37 AM, John Baldwin <jhb@freebsd.org> wrote:=0AThat= wasn't really my question. My question was if we want distinct streams =0A= or if we want want unified stream. Having a unified stream might very well= =0Amake sense (and if so we could rename devd to make that more obvious).= =0A=0AI'm fine with renaming devd to eventd or something else, but Ian was= saying that he's worried about the number of notifications that devd has = to process. =C2=A0I'm not sure that's a real problem at this point, though= . =C2=A0On freefall, devd used 0.07 seconds of CPU time and has been runni= ng for a 1 day and a half. =C2=A0On my BeagleBone, devd used 0.61 seconds = of CPU time and it has been up for 5 days and a half. =C2=A0 On my VM that= has been up for 5 days and a half, it used 4 seconds of CPU time. =C2=A0R= enaming sounds like a good idea and it looks like we could leave the optim= isations to a later time.=0A=0AAnother thing I had in mind (which is more = work) was to abstract the devctl kernel code in an API which could make it= easy to fan out the notifications to multiple /dev devices. =C2=A0However= , that may be overkill.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?8e5503e1-755c-49e4-ab4d-a0ad1ae91f97>