Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 16 Nov 2018 16:06:54 -0800
From:      John Baldwin <jhb@FreeBSD.org>
To:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org
Subject:   Re: svn commit: r340488 - in head/sys/x86: include x86 xen
Message-ID:  <cf0c08e9-e2d2-1029-a308-d3d8545a9733@FreeBSD.org>
In-Reply-To: <201811162339.wAGNdeVE014093@repo.freebsd.org>
References:  <201811162339.wAGNdeVE014093@repo.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On 11/16/18 3:39 PM, John Baldwin wrote:
> Author: jhb
> Date: Fri Nov 16 23:39:39 2018
> New Revision: 340488
> URL: https://svnweb.freebsd.org/changeset/base/340488
> 
> Log:
>   Axe MINIMUM_MSI_INT.
>   
>   Just allow MSI interrupts to always start at the end of the I/O APIC
>   pins.  Since existing machines already have more than 255 I/O APIC
>   pins, IRQ 255 is no longer reliably invalid, so just remove the
>   minimum starting value for MSI.
>   
>   Reviewed by:	kib, markj
>   Differential Revision:	https://reviews.freebsd.org/D17991

Note that this will renumber IRQ values.  If you are used to seeing IRQ
values > 256 as meaning MSI, that will no longer be true.  If you must
know which interrupts are MSI you can check the machdep.first_msi_irq
sysctl.  I've also considered renaming them to 'msi' instead of 'irq' in
vmstat -i output.  We still have to have a number for cpuset -x to work,
otherwise we could name MSI interrupts after the device perhaps.

-- 
John Baldwin

                                                                            



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?cf0c08e9-e2d2-1029-a308-d3d8545a9733>