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;
Date:   Sat, 6 May 2023 15:07:41 -0700
From:   Jacob Pan <jacob.jun.pan@linux.intel.com>
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     LKML <linux-kernel@vger.kernel.org>, iommu@lists.linux.dev,
        Jason Gunthorpe <jgg@nvidia.com>,
        Lu Baolu <baolu.lu@linux.intel.com>,
        Joerg Roedel <joro@8bytes.org>,
        Jean-Philippe Brucker <jean-philippe@linaro.com>,
        Robin Murphy <robin.murphy@arm.com>,
        Will Deacon <will@kernel.org>, Raj Ashok <ashok.raj@intel.com>,
        "Tian, Kevin" <kevin.tian@intel.com>, Yi Liu <yi.l.liu@intel.com>,
        "Yu, Fenghua" <fenghua.yu@intel.com>, jacob.jun.pan@linux.intel.com
Subject: Re: [PATCH] iommu: Add Kconfig help text for IOMMU_SVA
Message-ID: <20230506150741.2e3d2dcc@jacob-builder>
In-Reply-To: <CAHk-=wgUiAtiszwseM1p2fCJ+sC4XWQ+YN4TanFhUgvUqjr9Xw@mail.gmail.com>
References: <20230506133134.1492395-1-jacob.jun.pan@linux.intel.com>
        <CAHk-=wgUiAtiszwseM1p2fCJ+sC4XWQ+YN4TanFhUgvUqjr9Xw@mail.gmail.com>
Organization: OTC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi Linus,

On Sat, 6 May 2023 08:43:08 -0700, Linus Torvalds
<torvalds@linux-foundation.org> wrote:

> On Sat, May 6, 2023 at 6:27=E2=80=AFAM Jacob Pan <jacob.jun.pan@linux.int=
el.com>
> wrote:
> >
> > Shared Virtual Addressing (SVA) is not immediately intuitive to people
> > who work outside IOMMU subsystem, add some explanation to make it less
> > opaque. =20
>=20
> So the patch has at least two problems.
>=20
> The first is that you corrupted the SPDX line.
>=20
Sorry, my mistake. Will not try sending patches before fully waking up.

> But the second is that the change from
>=20
> -       bool
> +       bool "Shared Virtual Addressing"
>=20
> means that now Kconfig *asks* about this thing, which was never the
> intent. The other Kconfig options that need it all do a
>=20
>         select IOMMU_SVA
>=20
> to get it picked.
>=20
> I suspect that's why it then causes errors - because now the test
> robot can enable the option even in situations where it makes no sense
> and doesn't work.
>=20
> So no, it's not that the option needs a help entry that can be queried
> at Kconfig time. It's that the option needs a name that makes sense
> and isn't some random jumble of letters that is only understandable to
> people who already know exactly what it is.
Right, how about IOMMU_SHARING_CPU_PGTABLE?

Since this is more of a one-sided relationship where IOMMU walks CPU page
tables.


Thanks,

Jacob