[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20170316220519.GU28800@wotan.suse.de>
Date: Thu, 16 Mar 2017 23:05:19 +0100
From: "Luis R. Rodriguez" <mcgrof@...nel.org>
To: bjorn@...gaas.com
Cc: "Luis R. Rodriguez" <mcgrof@...not-panic.com>, mingo@...nel.org,
bp@...e.de, arnd@...db.de, dan.j.williams@...el.com,
Christoph Hellwig <hch@....de>, luto@...capital.net,
hpa@...or.com, tglx@...utronix.de, geert@...ux-m68k.org,
ralf@...ux-mips.org, hmh@....eng.br, ross.zwisler@...ux.intel.com,
akpm@...ux-foundation.org, jgross@...e.com,
Benjamin Herrenschmidt <benh@...nel.crashing.org>,
mpe@...erman.id.au, tj@...nel.org, x86 <x86@...nel.org>,
tomi.valkeinen@...com, mst@...hat.com, toshi.kani@...com,
stefan.bader@...onical.com, linux-mm@...ck.org,
linux-fbdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1] x86/mm, asm-generic: Add IOMMU ioremap_uc() variant
default
On Thu, Mar 16, 2017 at 03:46:51PM -0500, Bjorn Helgaas wrote:
> On Thu, Jul 9, 2015 at 7:28 PM, Luis R. Rodriguez
> <mcgrof@...not-panic.com> wrote:
>
> > +/**
> > + * DOC: ioremap() and ioremap_*() variants
> > + *
> > + * If you have an IOMMU your architecture is expected to have both ioremap()
> > + * and iounmap() implemented otherwise the asm-generic helpers will provide a
> > + * direct mapping.
> > + *
> > + * There are ioremap_*() call variants, if you have no IOMMU we naturally will
> > + * default to direct mapping for all of them, you can override these defaults.
> > + * If you have an IOMMU you are highly encouraged to provide your own
> > + * ioremap variant implementation as there currently is no safe architecture
> > + * agnostic default. To avoid possible improper behaviour default asm-generic
> > + * ioremap_*() variants all return NULL when an IOMMU is available. If you've
> > + * defined your own ioremap_*() variant you must then declare your own
> > + * ioremap_*() variant as defined to itself to avoid the default NULL return.
>
> Are the references above to "IOMMU" typos? Should they say "MMU"
> instead, so they match the #ifdef below?
Yes. Patch welcomed.
Luis
Powered by blists - more mailing lists