linux-kernel - Re: [PATCH 03/10] remoteproc: imx: use devm

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200727073757.r2vq6djh3a4dyfp6@pengutronix.de>
Date:   Mon, 27 Jul 2020 09:37:57 +0200
From:   Oleksij Rempel <o.rempel@...gutronix.de>
To:     Peng Fan <peng.fan@....com>
Cc:     "bjorn.andersson@...aro.org" <bjorn.andersson@...aro.org>,
        "mathieu.poirier@...aro.org" <mathieu.poirier@...aro.org>,
        "robh+dt@...nel.org" <robh+dt@...nel.org>,
        "linux-remoteproc@...r.kernel.org" <linux-remoteproc@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "shawnguo@...nel.org" <shawnguo@...nel.org>,
        "s.hauer@...gutronix.de" <s.hauer@...gutronix.de>,
        "kernel@...gutronix.de" <kernel@...gutronix.de>,
        "festevam@...il.com" <festevam@...il.com>,
        dl-linux-imx <linux-imx@....com>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>
Subject: Re: [PATCH 03/10] remoteproc: imx: use devm_ioremap

On Mon, Jul 27, 2020 at 06:51:00AM +0000, Peng Fan wrote:
> > Subject: Re: [PATCH 03/10] remoteproc: imx: use devm_ioremap
> > 
> > On Mon, Jul 27, 2020 at 06:28:20AM +0000, Peng Fan wrote:
> > > Hi Oleksij,
> > >
> > > > Subject: Re: [PATCH 03/10] remoteproc: imx: use devm_ioremap
> > > >
> > > > On Fri, Jul 24, 2020 at 04:08:06PM +0800, Peng Fan wrote:
> > > > > We might need to map an region multiple times, becaue the region
> > > > > might be shared between remote processors, such i.MX8QM with dual
> > M4 cores.
> > > > > So use devm_ioremap, not devm_ioremap_resource.
> > > >
> > > > Can you please give an example of this kind of shared resources and
> > > > how they should be handled by two separate devices?
> > >
> > > This is to share vdevbuffer space, there is a vdevbuffer in device
> > > tree, it will be shared between M4_0 and M4_1.
> > >
> > > For the buffer, it is Linux DMA API will handle the space.
> > 
> > Why remoteproc need to care about it? If I see it correctly, from the linux
> > perspective, it is one buffer and one driver is responsible for it. Or do I missing
> > some thing?
> 
> We not have the vdev buffer in resource table, so I added in device tree, see below:

Hm.. if vdev is not in resource table and should not be controlled by
remoteproc, why do we need remoteproc?

>         imx8qm_cm40: imx8qm_cm4@0 {
>                 compatible = "fsl,imx8qm-cm4";
>                 rsc-da = <0x90000000>;
>                 mbox-names = "tx", "rx", "rxdb";
>                 mboxes = <&lsio_mu5 0 1
>                           &lsio_mu5 1 1
>                           &lsio_mu5 3 1>;
>                 mub-partition = <3>;
>                 memory-region = <&vdev0vring0>, <&vdev0vring1>, <&vdevbuffer>,
>                                 <&vdev1vring0>, <&vdev1vring1>;
>                 core-index = <0>;
>                 core-id = <IMX_SC_R_M4_0_PID0>;
>                 status = "okay";
>                 power-domains = <&pd IMX_SC_R_M4_0_PID0>,
>                                 <&pd IMX_SC_R_M4_0_MU_1A>;
>         };
> 
>         imx8qm_cm41: imx8x_cm4@1 {
>                 compatible = "fsl,imx8qm-cm4";
>                 rsc-da = <0x90100000>;
>                 mbox-names = "tx", "rx", "rxdb";
>                 mboxes = <&lsio_mu6 0 1
>                           &lsio_mu6 1 1
>                           &lsio_mu6 3 1>;
>                 mub-partition = <4>;
>                 memory-region = <&vdev2vring0>, <&vdev2vring1>, <&vdevbuffer>,
>                                 <&vdev3vring0>, <&vdev3vring1>;
>                 core-index = <1>;
>                 core-id = <IMX_SC_R_M4_1_PID0>;
>                 status = "okay";
>                 power-domains = <&pd IMX_SC_R_M4_1_PID0>,
>                                 <&pd IMX_SC_R_M4_1_MU_1A>;
>         };
> 
>                 vdevbuffer: vdevbuffer {
>                         compatible = "shared-dma-pool";
>                         reg = <0 0x90400000 0 0x100000>;
>                         no-map;
>                 };
> 
> I have the upper vdevbuffer node shared between M40 and M41 node.
> The vdevbuffer will be used as virtio data buffer.
> 
> And I have the following in rproc_add_virtio_dev to share vdevbuffer:
>         /* Try to find dedicated vdev buffer carveout */
>         mem = rproc_find_carveout_by_name(rproc, "vdev%dbuffer", rvdev->index);
>         if (!mem)
>                 mem = rproc_find_carveout_by_name(rproc, "vdevbuffer");

With kernel v5.8-rc7 i get following call chain:
rproc_boot()
  rproc_fw_boot()
    rproc_handle_vdev
      rproc_vdev_do_start()
        rproc_add_virtio_dev()


So, at the end, we will call rproc_add_virtio_dev() only if we boot
firmware by linux, or if we get at least the resource table.

Since none of this seems to be the case, i still do not understand how
it should work.

> Hope this is clear.

:) i still need some time to understand it.

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)