Home » Mailing list archives » linux-kernel » 2008 » July » 11

Re: Multiple MSI, take 3

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
[view in full thread]
From: Eric W. Biederman <ebiederm@...>
To: Matthew Wilcox <matthew@...>
Cc: <linux-pci@...>, <linux-kernel@...>, <grundler@...>, <mingo@...>, <tglx@...>, <jgarzik@...>, <linux-ide@...>, <suresh.b.siddha@...>, <benh@...>, <jbarnes@...>, <rdunlap@...>, <mtk.manpages@...>
Subject: Re: Multiple MSI, take 3
Date: Friday, July 11, 2008 - 7:34 am

There is one idea that seems to model this cleanly
without breaking all kinds of expectations.

That is an irq with a very small data payload.

In that case we wire all of the vectors up to a single
irq handler that computes the payload as:
payload = vector - base-vector.

And then we figure out how to pass that to the handler in irqaction.

To most of the system it is a single irq so it avoids breaking
expectations about what an irq is.

To everything else it is a little odd, and has it's own unique
set of rules (which is good as well).

Eric
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
Messages in current thread:
Multiple MSI, take 3, Matthew Wilcox, (Thu Jul 10, 8:57 pm)
Re: Multiple MSI, take 3, Jike Song, (Fri Sep 26, 1:30 am)
Re: Multiple MSI, take 3, Matthew Wilcox, (Sat Sep 27, 3:04 pm)
Re: Multiple MSI, take 3, Eric W. Biederman, (Fri Jul 11, 6:06 am)
Re: Multiple MSI, take 3, Suresh Siddha, (Fri Jul 11, 5:59 pm)
Re: Multiple MSI, take 3, Eric W. Biederman, (Fri Jul 11, 6:59 pm)
Re: Multiple MSI, take 3, Benjamin Herrenschmidt, (Fri Jul 11, 11:52 pm)
Re: Multiple MSI, take 3, Eric W. Biederman, (Sat Jul 12, 12:41 am)
Re: Multiple MSI, take 3, Benjamin Herrenschmidt, (Sat Jul 12, 3:36 am)
Re: Multiple MSI, take 3, Eric W. Biederman, (Sun Jul 13, 6:30 pm)
Re: Multiple MSI, take 3, Benjamin Herrenschmidt, (Sun Jul 13, 6:44 pm)
Re: Multiple MSI, take 3, Eric W. Biederman, (Sun Jul 13, 7:29 pm)
Re: Multiple MSI, take 3, Benjamin Herrenschmidt, (Sun Jul 13, 8:17 pm)
Re: Multiple MSI, take 3, David Miller, (Sun Jul 13, 8:44 pm)
Re: Multiple MSI, take 3, Eric W. Biederman, (Sun Jul 13, 10:03 pm)
Re: Multiple MSI, take 3, David Miller, (Sun Jul 13, 11:19 pm)
Re: Multiple MSI, take 3, Suresh Siddha, (Fri Jul 11, 7:15 pm)
Re: Multiple MSI, take 3, Eric W. Biederman, (Fri Jul 11, 7:59 pm)
Re: Multiple MSI, take 3, Matthew Wilcox, (Fri Jul 11, 6:23 am)
Re: Multiple MSI, take 3, Eric W. Biederman, (Fri Jul 11, 7:34 am)
Re: Multiple MSI, take 3, Matthew Wilcox, (Fri Jul 11, 8:17 am)
Re: Multiple MSI, take 3, Matthew Wilcox, (Fri Jul 11, 11:10 am)
Re: Multiple MSI, take 3, Eric W. Biederman, (Fri Jul 11, 7:05 am)
Re: Multiple MSI, take 3, David Miller, (Fri Jul 11, 6:32 am)
Re: Multiple MSI, take 3, Matthew Wilcox, (Fri Jul 11, 6:41 am)
[PATCH] Rewrite MSI-HOWTO, Matthew Wilcox, (Thu Jul 10, 9:00 pm)
Re: [PATCH] Rewrite MSI-HOWTO, Grant Grundler, (Fri Sep 26, 2:42 am)
[PATCH] PCI: Add support for multiple MSI, Matthew Wilcox, (Thu Jul 10, 8:59 pm)
Re: [PATCH] PCI: Add support for multiple MSI, Hidetoshi Seto, (Fri Jul 11, 4:28 am)
Re: [PATCH] PCI: Add support for multiple MSI, Benjamin Herrenschmidt, (Fri Jul 11, 11:45 pm)
Re: [PATCH] PCI: Add support for multiple MSI, Matthew Wilcox, (Fri Jul 11, 5:45 am)
[PATCH] AHCI: Request multiple MSIs, Matthew Wilcox, (Thu Jul 10, 9:00 pm)
[PATCH] PCI MSI: Replace 'type' with 'is_msix', Matthew Wilcox, (Thu Jul 10, 8:59 pm)
[PATCH] x86-64: Support for multiple MSIs, Matthew Wilcox, (Thu Jul 10, 9:00 pm)
Re: [PATCH] x86-64: Support for multiple MSIs, Kenji Kaneshige, (Fri Jul 11, 12:50 am)
Re: [PATCH] x86-64: Support for multiple MSIs, Matthew Wilcox, (Fri Jul 11, 4:50 am)
Re: [PATCH] x86-64: Support for multiple MSIs, Kenji Kaneshige, (Sun Jul 13, 9:08 pm)