Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <YuzlAT7RW1U36+ap@8bytes.org> <CAHk-=whs5CR5FRZP_HUksrr9wX6=Rai-S_XTUFJN7jwB5wKN-Q@mail.gmail.com>
In-Reply-To: <CAHk-=whs5CR5FRZP_HUksrr9wX6=Rai-S_XTUFJN7jwB5wKN-Q@mail.gmail.com>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Sat, 6 Aug 2022 13:31:28 -0700
Message-ID: <CAHk-=wiSxEVW78nTumvz+ZYt2MZcmqe6S0nb_-efKYm636VuGA@mail.gmail.com>
Subject: Re: [git pull] IOMMU Updates for Linux v5.20
To:     Joerg Roedel <joro@8bytes.org>,
        Robin Murphy <robin.murphy@arm.com>,
        Christoph Hellwig <hch@lst.de>,
        John Garry <john.garry@huawei.com>
Cc:     Will Deacon <will@kernel.org>, linux-kernel@vger.kernel.org,
        iommu@lists.linux.dev
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sat, Aug 6, 2022 at 12:39 PM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> If it is, it will be reverted as hopelessly optimistic.

Indeed. I should just have verified with the kernel command line, but
I started bisecting when it wasn't entirely obvious what was going on,
so I just finished it.

Commit 4bf7fda4dce2 ("iommu/dma: Add config for PCI SAC address
trick") is reverted in my tree. You can try again in a decade or so,
but at least on x86-64, that config option is basically a "stop modern
machines from working", and I don't want to have that even as an
option.

If somebody wants some debug output to figure out which device it is
that cares, send me a patch, but we're not trying this thing as-is
again.

                 Linus