Re: [PATCH 4/4] iommu/x86: make unity range checking more strict

From: Jan Beulich <jbeulich@suse.com>
To: Roger Pau Monne <roger.pau@citrix.com>
Cc: Paul Durrant <paul@xen.org>, xen-devel@lists.xenproject.org
Subject: Re: [PATCH 4/4] iommu/x86: make unity range checking more strict
Date: Tue, 6 Feb 2024 12:49:08 +0100	[thread overview]
Message-ID: <5db22ee1-40b3-4df8-88b8-446a2e224d22@suse.com> (raw)
In-Reply-To: <20240201170159.66330-5-roger.pau@citrix.com>

On 01.02.2024 18:01, Roger Pau Monne wrote:
> Currently when a unity range overlaps with memory being used as RAM by the
> hypervisor the result would be that the IOMMU gets disabled.  However that's
> not enough, as even with the IOMMU disabled the device will still access the
> affected RAM areas.

Hmm, no, I think this is going too far. Not the least because it is
s/will/may/. But also because if we really wanted such behavior, we
ought to also parse the respective ACPI tables when the "iommu=off".

> Note that IVMD or RMRR ranges being placed over RAM is a firmware bug.

As written this is wrong: They're typically in RAM, just that the E820
type for that range should not be RAM_TYPE_CONVENTIONAL.

> Doing so also allows to simplify the code and use a switch over the reported
> memory type(s).

I'm afraid this isn't right either: page_get_ram_type() can set
multiple bits in its output.

Jan