From mboxrd@z Thu Jan 1 00:00:00 1970 From: ebiederm@xmission.com (Eric W. Biederman) Subject: Re: Multiple MSI, take 3 Date: Sun, 13 Jul 2008 19:03:52 -0700 Message-ID: References: <1215989044.7549.219.camel@pasglop> <1215994659.7549.227.camel@pasglop> <20080713.174457.82768245.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: In-Reply-To: <20080713.174457.82768245.davem@davemloft.net> (David Miller's message of "Sun, 13 Jul 2008 17:44:57 -0700 (PDT)") Sender: linux-pci-owner@vger.kernel.org To: David Miller Cc: benh@kernel.crashing.org, suresh.b.siddha@intel.com, matthew@wil.cx, linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org, grundler@parisc-linux.org, mingo@elte.hu, tglx@linutronix.de, jgarzik@pobox.com, linux-ide@vger.kernel.org, jbarnes@virtuousgeek.org, rdunlap@xenotime.net, mtk.manpages@gmail.com List-Id: linux-ide@vger.kernel.org David Miller writes: > The x86 system designers decided to implement multi-MSI in an > inconvenient way, it is not a "crap hardware design", merely > some (unfortunately common) implementations of it happen to be. To be clear I was referring to the PCI spec that describes multi-MSI as a crap hardware design. At the very least you are left with the problem of allocating multiple contiguous destinations. Which has the potential to create fragmentation on all supported platforms. Optional mask bits are also nasty. My honest opinion is that the should have deprecated multi-msi after the introduction of the msi-x specification. Eric