[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 2 Feb 2015 16:45:46 +0000
From: Will Deacon <will.deacon@....com>
To: Joerg Roedel <joro@...tes.org>
Cc: "iommu@...ts.linux-foundation.org" <iommu@...ts.linux-foundation.org>,
Kukjin Kim <kgene@...nel.org>,
David Woodhouse <dwmw2@...radead.org>,
Heiko Stuebner <heiko@...ech.de>,
Hiroshi Doyu <hdoyu@...dia.com>,
Thierry Reding <thierry.reding@...il.com>,
Alex Williamson <alex.williamson@...hat.com>,
Arnd Bergmann <arnd@...db.de>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Robin Murphy <Robin.Murphy@....com>,
Laurent Pinchart <laurent.pinchart@...asonboard.com>,
"jroedel@...e.de" <jroedel@...e.de>
Subject: Re: [PATCH 3/5] iommu: Limit iommu_attach/detach_device to devices
with their own group
On Fri, Jan 30, 2015 at 12:28:14PM +0000, Joerg Roedel wrote:
> On Wed, Jan 28, 2015 at 02:35:24PM +0000, Will Deacon wrote:
> > On Tue, Jan 27, 2015 at 12:08:57AM +0000, Joerg Roedel wrote:
> > > @@ -51,6 +51,7 @@ struct iommu_group {
> > > void (*iommu_data_release)(void *iommu_data);
> > > char *name;
> > > int id;
> > > + unsigned dev_cnt;
> >
> > Is this actually used on a fast path, or can we just inspect the list of
> > devices on the group instead?
>
> Not really a fast path, but we have to hold the group mutex while
> traversing the list, which could hurt performance somewhere else. Are
> these 4 bytes a problem?
No problem, it just seemed a bit redundant to have two ways of describing
the same thing and having the pain of keeping them in sync with each other.
Will
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists