Re: [dpdk-dev] [PATCH v4] raw/ioat: add secondary process support

[Bruce Richardson <bruce.richardson@intel.com>]

On Mon, Jan 04, 2021 at 10:00:49PM +0530, Kumar Amber wrote:
> Add support for secondary processes in ioat devices. The update
> allocates a memzone for a primary process or returns it in a
> secondary process.
> 
> Signed-off-by: Kumar Amber <kumar.amber@intel.com>
> ---

Thanks for the patch. Some comments below.
Also, with each version can you include below the cutline [i.e. here] what
has changed since the previous version, to make reviewing easier.

Thanks,
/Bruce

>  drivers/raw/ioat/ioat_common.c | 16 +++++++++++++---
>  drivers/raw/ioat/ioat_rawdev.c | 17 +++++++++++++----
>  2 files changed, 26 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/raw/ioat/ioat_common.c b/drivers/raw/ioat/ioat_common.c
> index 142e171bc9..c174e4bb28 100644
> --- a/drivers/raw/ioat/ioat_common.c
> +++ b/drivers/raw/ioat/ioat_common.c
> @@ -215,11 +215,21 @@ idxd_rawdev_create(const char *name, struct rte_device *dev,
>  		goto cleanup;
>  	}
>  
> +	/* Allocate memory for the primary process or else return the memory
> +	 * of primary memzone for the secondary process.
> +	 */
>  	snprintf(mz_name, sizeof(mz_name), "rawdev%u_private", rawdev->dev_id);
> -	mz = rte_memzone_reserve(mz_name, sizeof(struct idxd_rawdev),
> -			dev->numa_node, RTE_MEMZONE_IOVA_CONTIG);
> +	if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
> +		mz = rte_memzone_lookup(mz_name);
> +		rawdev->dev_private = mz->addr;

I think we should have some error checking here in case the lookup fails.
The particular device in question could be unused by the primary.

> +		return 0;
> +	}
> +	mz = rte_memzone_reserve(mz_name,
> +			sizeof(struct rte_ioat_rawdev),
> +			dev->numa_node,
> +			RTE_MEMZONE_IOVA_CONTIG);
>  	if (mz == NULL) {
> -		IOAT_PMD_ERR("Unable to reserve memzone for private data\n");
> +		IOAT_PMD_ERR("Unable to reserve memzone\n");

Is there a need to change the error message here?

>  		ret = -ENOMEM;
>  		goto cleanup;
>  	}
> diff --git a/drivers/raw/ioat/ioat_rawdev.c b/drivers/raw/ioat/ioat_rawdev.c
> index 2c88b4369f..0d3e904c8d 100644
> --- a/drivers/raw/ioat/ioat_rawdev.c
> +++ b/drivers/raw/ioat/ioat_rawdev.c
> @@ -165,15 +165,24 @@ ioat_rawdev_create(const char *name, struct rte_pci_device *dev)
>  		goto cleanup;
>  	}
>  
> +	/* Allocate memory for the primary process or else return the memory
> +	 * of primary memzone for the secondary process.
> +	 */
>  	snprintf(mz_name, sizeof(mz_name), "rawdev%u_private", rawdev->dev_id);
> -	mz = rte_memzone_reserve(mz_name, sizeof(struct rte_ioat_rawdev),
> -			dev->device.numa_node, RTE_MEMZONE_IOVA_CONTIG);
> +	if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
> +		mz = rte_memzone_lookup(mz_name);
> +		rawdev->dev_private = mz->addr;
> +		return 0;
> +	}
> +	mz = rte_memzone_reserve(mz_name,
> +			sizeof(struct rte_ioat_rawdev),
> +			dev->device.numa_node,
> +			RTE_MEMZONE_IOVA_CONTIG);
>  	if (mz == NULL) {
> -		IOAT_PMD_ERR("Unable to reserve memzone for private data\n");
> +		IOAT_PMD_ERR("Unable to reserve memzone\n");
>  		ret = -ENOMEM;
>  		goto cleanup;
>  	}
> -

Same comments for this file as for previous. Also note the extra line
removal here, which probably should not be part of this patch too.

>  	rawdev->dev_private = mz->addr;
>  	rawdev->dev_ops = &ioat_rawdev_ops;
>  	rawdev->device = &dev->device;
> -- 
> 2.25.1
>