Re: [PATCH v4 0/5] remoteproc: imx_rproc: Support i.MX95

[Peng Fan <peng.fan@oss.nxp.com>]

On Fri, Jul 18, 2025 at 11:06:04AM -0600, Mathieu Poirier wrote:
>On Fri, Jul 18, 2025 at 04:20:34PM +0800, Peng Fan wrote:
>> On Thu, Jul 17, 2025 at 08:48:43AM -0600, Mathieu Poirier wrote:
>> >On Mon, Jul 14, 2025 at 11:52:05AM +0000, Peng Fan wrote:
>> >> > Subject: Re: [PATCH v4 0/5] remoteproc: imx_rproc: Support i.MX95
>> >> > 
>> >> [...] 
>> >> > New warnings running 'make CHECK_DTBS=y for
>> >> > arch/arm64/boot/dts/freescale/' for 20250710-imx95-rproc-1-v4-0-
>> >> > a7123e857dfb@nxp.com:
>> >> > 
>> >> > arch/arm64/boot/dts/freescale/imx95-tqma9596sa-mb-smarc-2.dtb:
>> >> > scmi (arm,scmi): Unevaluated properties are not allowed
>> >> > ('protocol@80', 'protocol@81', 'protocol@82', 'protocol@84' were
>> >> > unexpected)
>> >> 
>> >> Same as replied in v3.
>> >> This is because [1] is still not picked, not because of my patchset.
>> >
>> >I won't move on this patchset until this is resolved.
>> >
>> 
>> Not understand why hold on this patchset. I suppose you may not
>> understand what the error means. The warning is totally irrelevant
>> to this patchset, there is no dependency.
>> 
>> Others added a property to arch/arm64/boot/dts/freescale/imx95-tqma9596sa.dtsi
>> &scmi_bbm {
>>         linux,code = <KEY_POWER>;
>> };
>> But this "linux,code" property not landed(missed to be picked up) to DT binding.
>> 
>> This patchset does not touch scmi_bbm. I could help address the warning
>> in the other patch, but I do not see why "linux,code" under scmi_bbm node
>> could block this patchset.
>> 
>> Please help clarify if you still think to hold on this patchset.
>> 
>> BTW: with [1] "remoteproc: imx_rproc: skip clock enable when M-core is managed by the SCU"
>> merged in Ulf's tree, there is a minor conflict with patch 2. Please suggest
>> what I should do with this patchset.
>>
>
>I was afraid of that.  The best way forward with this work is to wait for the
>"linux,code" property to be picked up by Sudeep.  I suggest you make sure that
>he, or anyone else, picks it up for the next merge window.  If that happens


I respect you as maintainer, but there is no reason to block this patch
because of "linux,code" property. It is totally irrelevant.

Even if I help to resubmit that "linux,code" patch, there is no chance to
land into 6.17-rc1, both Sudeep and Shawn sent their PR to arm-soc earlier
before your comments. You could raise in V3.. which there was time left.

>everything should be set for you to resend this patchset when 6.17-rc1 comes
>out.

Because of the code conflict in Ulf's tree, I will hold on until 6.17-rc1.
Patch 4 and 5 will be removed from this patchset in V5. The two patches
should go through Shawn's tree, I will resend them in a separate thread
with "linux,code" patch included. There will be no more CHECK_DTBS
warning in V5 for i.MX95 remoteproc support.

If you have concern on patch 1-3 or else, please raise earlier. As of now,
patch 1-3 in V5 will be almost same as V4 with only a minor code
conflict resolved, with below change
@@ -1030,7 +1030,8 @@ static int imx_rproc_clk_enable(struct imx_rproc *priv)
        int ret;

        /* Remote core is not under control of Linux or it is managed by SCU API */
-       if (dcfg->method == IMX_RPROC_NONE || dcfg->method == IMX_RPROC_SCU_API)
+       if (dcfg->method == IMX_RPROC_NONE || dcfg->method == IMX_RPROC_SCU_API ||
+           dcfg->method == IMX_RPROC_SCU_SM)
                return 0;

Regards,
Peng
> 
>> [1]https://lore.kernel.org/linux-remoteproc/20250629172512.14857-3-hiagofranco@gmail.com/T/#u
>> 
>> Thanks,
>> Peng