MIME-Version: 1.0
References: <20221222072603.1175248-1-korantwork@gmail.com>
 <3d1834d9-7905-1225-741a-f298dd5b8a8e@linux.dev> <Y6TSgGdCTvkwPiVg@kbusch-mbp.dhcp.thefacebook.com>
 <CAEm4hYUWf+Fx3FV7vNTc8+O9NSb0iQp75MTC6gra6XapXK=cxw@mail.gmail.com>
 <d14ac29d-027a-08a7-c5c8-848a6920d4a2@linux.dev> <CAEm4hYXncuvL-Gk1aEZExrvkbx=N1aiOQNeNjFdB4443EbKNBA@mail.gmail.com>
 <f05ee82a-4532-b12b-490f-904b946ff7b0@linux.dev> <CAEm4hYXk1RuKEw41VukH2iGTo_9GmZjUfrESWK5vFtpFA_O_4A@mail.gmail.com>
 <CAEm4hYWeZFrYxSvAcBJ8iw=t507vZMqfBwiQXFSJd2Hcyfw7fA@mail.gmail.com>
 <9e6cfda1-4309-d1bb-a0cf-404a7fe7d3aa@linux.intel.com> <Y+FG2EUpmsmCZqwh@kbusch-mbp.dhcp.thefacebook.com>
In-Reply-To: <Y+FG2EUpmsmCZqwh@kbusch-mbp.dhcp.thefacebook.com>
From:   Xinghui Li <korantwork@gmail.com>
Date:   Tue, 7 Feb 2023 11:18:14 +0800
Message-ID: <CAEm4hYWjaVCDFtCCyHrq5vOie4wV5LFn-UVZ5qQbpASxVGy1TA@mail.gmail.com>
Subject: Re: [PATCH] PCI: vmd: Do not disable MSI-X remapping in VMD 28C0 controller
To:     Keith Busch <kbusch@kernel.org>
Cc:     "Patel, Nirmal" <nirmal.patel@linux.intel.com>,
        Jonathan Derrick <jonathan.derrick@linux.dev>,
        lpieralisi@kernel.org, linux-pci@vger.kernel.org,
        linux-kernel@vger.kernel.org, Xinghui Li <korantli@tencent.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Keith Busch <kbusch@kernel.org> =E4=BA=8E2023=E5=B9=B42=E6=9C=887=E6=97=A5=
=E5=91=A8=E4=BA=8C 02:28=E5=86=99=E9=81=93=EF=BC=9A
>
> On Mon, Feb 06, 2023 at 11:11:36AM -0700, Patel, Nirmal wrote:
> > I like the idea of module parameter to allow switching between the mode=
s
> > but keep MSI remapping enabled (non-bypass) by default.
>
> Isn't there a more programatic way to go about selecting the best option =
at
> runtime?
Do you mean that the operating mode is automatically selected by
detecting the number of devices and CPUs instead of being set
manually?
>I suspect bypass is the better choice if "num_active_cpus() > pci_msix_vec=
_count(vmd->dev)".
For this situation, My speculation is that the PCIE nodes are
over-mounted and not just because of the CPU to Drive ratio.
We considered designing online nodes, because we were concerned that
the IO of different chunk sizes would adapt to different MSI-X modes.
I privately think that it may be logically complicated if programmatic
judgments are made.