Date: Tue, 6 Nov 2012 07:38:07 -0800 From: Garrett Cooper <yanegomi@gmail.com> To: Chuck Burns <break19@gmail.com> Cc: "freebsd-current@freebsd.org" <freebsd-current@freebsd.org> Subject: Re: polling's future [was: Re: Dynamic Ticks/HZ] Message-ID: <94F3E4AA-E51F-48F2-8191-659994A209D8@gmail.com> In-Reply-To: <32992731.3rb8pPz1Mv@blackbeast.local> References: <509758B8.1000409@rewt.org.uk> <20121106113014.GA26313@onelab2.iet.unipi.it> <5098F64E.1070704@freebsd.org> <32992731.3rb8pPz1Mv@blackbeast.local>
next in thread | previous in thread | raw e-mail | index | archive | help
On Nov 6, 2012, at 4:31 AM, Chuck Burns <break19@gmail.com> wrote: > On Tuesday, November 06, 2012 12:36:46 PM Andre Oppermann wrote: >> On 06.11.2012 12:30, Luigi Rizzo wrote: >>> On Tue, Nov 06, 2012 at 11:23:34AM +0100, Andre Oppermann > wrote: >>> ... >>>=20 >>>> Hi Luigi, >>>>=20 >>>> do you agree on polling having outlived its usefulness in the light >>>> of interrupt moderating NIC's and SMP > complications/disadvantages? >>>=20 >>> yes, we should let it rest in peace. >>=20 >> Thank you for this non-complicated answer. :-) >=20 > I worry about what happens for those people who would be running=20 > FreeBSD on older equipment where polling might still make sense. >=20 > Do we throw them under the bus? I think that adding a big fat warning at boot if the driver supports proper h= ybrid polling (think: witness) might dissuade some folks. Thanks! -Garrett=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?94F3E4AA-E51F-48F2-8191-659994A209D8>