Re: [PATCH 2/2] remoteproc: imx_dsp_rproc: Add support of coredump

[Mathieu Poirier <mathieu.poirier@linaro.org>]

On Wed, Jun 18, 2025 at 02:26:44PM +0800, Shengjiu Wang wrote:
> Add call rproc_coredump_set_elf_info() to initialize the elf info for
> coredump, otherwise coredump will report an error "ELF class is not set".
> 
> Remove the DSP IRAM and DRAM segment in coredump list, because after
> stop, DSP power is disabled, the IRAM and DRAM can't be accessed.
> 
> Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
> ---
>  drivers/remoteproc/imx_dsp_rproc.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/remoteproc/imx_dsp_rproc.c b/drivers/remoteproc/imx_dsp_rproc.c
> index 9b9cddb224b0..9e7efb77b6e5 100644
> --- a/drivers/remoteproc/imx_dsp_rproc.c
> +++ b/drivers/remoteproc/imx_dsp_rproc.c
> @@ -738,9 +738,7 @@ static int imx_dsp_rproc_add_carveout(struct imx_dsp_rproc *priv)
>  		mem = rproc_mem_entry_init(dev, (void __force *)cpu_addr, (dma_addr_t)att->sa,
>  					   att->size, da, NULL, NULL, "dsp_mem");
>  
> -		if (mem)
> -			rproc_coredump_add_segment(rproc, da, att->size);
> -		else
> +		if (!mem)

Flag rproc->recovery_disabled is never set to true, meaning that since this
driver was introduced, some kind of recovery was available.

I worry that your work will introduce regression for other users.  Daniel and
Iuliana, once again have to ask you to look at this patchset.

Thanks,
Mathieu

>  			return -ENOMEM;
>  
>  		rproc_add_carveout(rproc, mem);
> @@ -1203,6 +1201,8 @@ static int imx_dsp_rproc_probe(struct platform_device *pdev)
>  		goto err_detach_domains;
>  	}
>  
> +	rproc_coredump_set_elf_info(rproc, ELFCLASS32, EM_XTENSA);
> +
>  	pm_runtime_enable(dev);
>  
>  	return 0;
> -- 
> 2.34.1
>