From: ebiederm@xmission.com (Eric W. Biederman)
To: benh@kernel.crashing.org
Cc: Suresh Siddha <suresh.b.siddha@intel.com>,
Matthew Wilcox <matthew@wil.cx>,
"linux-pci@vger.kernel.org" <linux-pci@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"grundler@parisc-linux.org" <grundler@parisc-linux.org>,
"mingo@elte.hu" <mingo@elte.hu>,
"tglx@linutronix.de" <tglx@linutronix.de>,
"jgarzik@pobox.com" <jgarzik@pobox.com>,
"linux-ide@vger.kernel.org" <linux-ide@vger.kernel.org>,
"jbarnes@virtuousgeek.org" <jbarnes@virtuousgeek.org>,
"rdunlap@xenotime.net" <rdunlap@xenotime.net>,
"mtk.manpages@gmail.com" <mtk.manpages@gmail.com>
Subject: Re: Multiple MSI, take 3
Date: Sun, 13 Jul 2008 16:29:45 -0700 [thread overview]
Message-ID: <m1tzet2wie.fsf@frodo.ebiederm.org> (raw)
In-Reply-To: <1215989044.7549.219.camel@pasglop> (Benjamin Herrenschmidt's message of "Mon, 14 Jul 2008 08:44:04 +1000")
Ben. Multi-MSI is a crap hardware design. Why do you think we have
MSI-X? MSI-X as specced is a properly operating irq controller that
we don't need kludges to support. Multi-MSI with a full set of
kludges almost work but not quite fits the linux irq model.
Any hardware designer who choose to implement Multi-MSI instead of
MSI-X was not really concerned about having a high performance device.
If we can find a way to model the portable capabilities of Multi-MSI
cleanly then we can support it, and our drivers and our users and our
intermediate layers won't get surprised.
So far we have too close fits but neither model really works.
Further this is all about driver optimization, so none of this is
necessary to have working hardware. Which makes kludges much less
appropriate. Modelling Multi-MSI irqs as normal irqs requires a lot
of nasty kludges.
One of the kludges is allocating a continuous chunk of irq targets,
and the resulting fragmentation issues that you get when you start
allowing different sized allocations.
Overall if Multi-MSI was to become common I think we would really
regret it.
Eric
next prev parent reply other threads:[~2008-07-13 23:29 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-07-11 0:57 Multiple MSI, take 3 Matthew Wilcox
2008-07-11 0:59 ` [PATCH] PCI MSI: Replace 'type' with 'is_msix' Matthew Wilcox
2008-07-11 0:59 ` [PATCH] PCI: Add support for multiple MSI Matthew Wilcox
2008-07-11 8:28 ` Hidetoshi Seto
2008-07-11 9:45 ` Matthew Wilcox
2008-07-12 3:45 ` Benjamin Herrenschmidt
2008-07-11 1:00 ` [PATCH] Rewrite MSI-HOWTO Matthew Wilcox
2008-09-26 6:42 ` Grant Grundler
2008-07-11 1:00 ` [PATCH] AHCI: Request multiple MSIs Matthew Wilcox
2008-07-11 1:00 ` [PATCH] x86-64: Support for " Matthew Wilcox
2008-07-11 4:50 ` Kenji Kaneshige
2008-07-11 8:50 ` Matthew Wilcox
2008-07-14 1:08 ` Kenji Kaneshige
2008-07-11 10:06 ` Multiple MSI, take 3 Eric W. Biederman
2008-07-11 10:23 ` Matthew Wilcox
2008-07-11 10:32 ` David Miller
2008-07-11 10:41 ` Matthew Wilcox
2008-07-11 11:05 ` Eric W. Biederman
2008-07-11 11:34 ` Eric W. Biederman
2008-07-11 12:17 ` Matthew Wilcox
2008-07-11 15:10 ` Matthew Wilcox
2008-07-11 21:59 ` Suresh Siddha
2008-07-11 22:59 ` Eric W. Biederman
2008-07-11 23:15 ` Suresh Siddha
2008-07-11 23:59 ` Eric W. Biederman
2008-07-12 3:52 ` Benjamin Herrenschmidt
2008-07-12 4:41 ` Eric W. Biederman
2008-07-12 7:36 ` Benjamin Herrenschmidt
2008-07-13 22:30 ` Eric W. Biederman
2008-07-13 22:44 ` Benjamin Herrenschmidt
2008-07-13 23:29 ` Eric W. Biederman [this message]
2008-07-14 0:17 ` Benjamin Herrenschmidt
2008-07-14 0:44 ` David Miller
2008-07-14 2:03 ` Eric W. Biederman
2008-07-14 3:19 ` David Miller
2008-09-26 5:30 ` Jike Song
2008-09-27 19:04 ` Matthew Wilcox
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=m1tzet2wie.fsf@frodo.ebiederm.org \
--to=ebiederm@xmission.com \
--cc=benh@kernel.crashing.org \
--cc=grundler@parisc-linux.org \
--cc=jbarnes@virtuousgeek.org \
--cc=jgarzik@pobox.com \
--cc=linux-ide@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=matthew@wil.cx \
--cc=mingo@elte.hu \
--cc=mtk.manpages@gmail.com \
--cc=rdunlap@xenotime.net \
--cc=suresh.b.siddha@intel.com \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).