From mboxrd@z Thu Jan 1 00:00:00 1970 From: Clark Williams Subject: Re: rt file i/o Date: Wed, 4 Nov 2009 16:44:13 -0600 Message-ID: <20091104164413.3aff479c@torg> References: <5bdc1c8b0911040743h7e4eaa08w97ac8287b806b54f@mail.gmail.com> <4AF1A327.3040103@klingt.org> <20091104171412.GB1702@opentech.at> <4AF1BAB9.5080104@klingt.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=PGP-SHA1; boundary="Sig_/LOfj972bbxRHDbPpM4zAR8l"; protocol="application/pgp-signature" Cc: Nicholas Mc Guire , y@opentech.at, Mark Knecht , linux-rt-users@vger.kernel.org To: Tim Blechmann Return-path: Received: from mx1.redhat.com ([209.132.183.28]:22975 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932730AbZKDWo3 (ORCPT ); Wed, 4 Nov 2009 17:44:29 -0500 In-Reply-To: <4AF1BAB9.5080104@klingt.org> Sender: linux-rt-users-owner@vger.kernel.org List-ID: --Sig_/LOfj972bbxRHDbPpM4zAR8l Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Wed, 04 Nov 2009 18:32:41 +0100 Tim Blechmann wrote: > >> well, if i understand the rt howto correctly, _no_ disc access is > >> allowed, neither from rt nor from non-rt threads, since it may produce > >> page faults, which introduce latencies ... > >=20 > > I would be supprised if the rt howto states that page-faults in non-rt > > threads is a critical problem - that would not significantly impact RT > > performance - atelast not the worst case - it will (as every other syst= em > > load) impact the average case. so having a non-rt thread reading disk-f= iles > > to a buffer and a rt-thread processing this buffer should be perfectly = fine. >=20 > that is what i thought for years ... but according to [1] a page fault > in an rt process freezes the entire process with both rt and non-rt > threads until the page fault is handled ... >=20 No, we need to get that fixed in the wiki. It may have been the case that the entire process was frozen while a fault was being handled, but I don't believe that is the case now. If you're in a thread on core0 and a sibling thread on core1 faults, you shouldn't be impacted unless you're trying to reference the same memory.=20 Clark --Sig_/LOfj972bbxRHDbPpM4zAR8l Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.12 (GNU/Linux) iEYEARECAAYFAkryA8AACgkQHyuj/+TTEp0rEgCcD8EXObK22smMhwF2fTJCWckW cMgAnA1lzVEkL4jgZYf5XoF9v49IRJ/M =REBI -----END PGP SIGNATURE----- --Sig_/LOfj972bbxRHDbPpM4zAR8l--