Re: [PATCH net-next v5 05/11] net: ti: prueth: Adds ethtool support for ICSSM PRUETH Driver

[Simon Horman <horms@kernel.org>]

On Mon, Apr 14, 2025 at 06:32:31PM +0530, Parvathi Pudi wrote:
> From: Roger Quadros <rogerq@ti.com>
> 
> Changes for enabling ethtool support for the newly added PRU Ethernet
> interfaces. Extends the support for statistics collection from PRU internal
> memory and displays it in the user space. Along with statistics,
> enable/disable of features, configuring link speed etc.are now supported.
> 
> The firmware running on PRU maintains statistics in internal data memory.
> When requested ethtool collects all the statistics for the specified
> interface and displays it in the user space.
> 
> Makefile is updated to include ethtool support into PRUETH driver.
> 
> Signed-off-by: Roger Quadros <rogerq@ti.com>
> Signed-off-by: Andrew F. Davis <afd@ti.com>
> Signed-off-by: Basharath Hussain Khaja <basharath@couthit.com>
> Signed-off-by: Parvathi Pudi <parvathi@couthit.com>

...

> diff --git a/drivers/net/ethernet/ti/icssm/icssm_prueth.h b/drivers/net/ethernet/ti/icssm/icssm_prueth.h

...

> @@ -108,6 +114,119 @@ struct prueth_packet_info {
>  	bool timestamp;
>  };
>  
> +/**
> + * struct port_statistics - Statistics structure for capturing statistics
> + *			    on PRUs
> + * @tx_bcast: Number of broadcast packets sent
> + * @tx_mcast:Number of multicast packets sent
> + * @tx_ucast:Number of unicast packets sent
> + *
> + * @tx_octets:Number of undersized frames rcvd
> + *
> + * @rx_bcast:Number of broadcast packets rcvd
> + * @rx_mcast:Number of multicast packets rcvd
> + * @rx_ucast:Number of unicast packets rcvd
> + *
> + * @rx_octets:Number of Rx packets
> + *
> + * @tx64byte:Number of 64 byte packets sent
> + * @tx65_127byte:Number of 65-127 byte packets sent
> + * @tx128_255byte:Number of 128-255 byte packets sent
> + * @tx256_511byte:Number of 256-511 byte packets sent
> + * @tx512_1023byte:Number of 512-1023 byte packets sent
> + * @tx1024byte:Number of 1024 and larger size packets sent
> + *
> + * @rx64byte:Number of 64 byte packets rcvd
> + * @rx65_127byte:Number of 65-127 byte packets rcvd
> + * @rx128_255byte:Number of 128-255 byte packets rcvd
> + * @rx256_511byte:Number of 256-511 byte packets rcvd
> + * @rx512_1023byte:Number of 512-1023 byte packets rcvd
> + * @rx1024byte:Number of 1024 and larger size packets rcvd
> + *
> + * @late_coll:Number of late collisions(Half Duplex)
> + * @single_coll:Number of single collisions (Half Duplex)
> + * @multi_coll:Number of multiple collisions (Half Duplex)
> + * @excess_coll:Number of excess collisions(Half Duplex)
> + *
> + * @rx_misalignment_frames:Number of non multiple of 8 byte frames rcvd
> + * @stormprev_counter:Number of packets dropped because of Storm Prevention

nit: It looks like the documentation of @stormprev_counter should
     be replaced by documentation of:
     @u32 stormprev_counter_bc;
     @u32 stormprev_counter_mc;
     @u32 stormprev_counter_uc;

> + * @mac_rxerror:Number of MAC receive errors
> + * @sfd_error:Number of invalid SFD
> + * @def_tx:Number of transmissions deferred
> + * @mac_txerror:Number of MAC transmit errors
> + * @rx_oversized_frames:Number of oversized frames rcvd
> + * @rx_undersized_frames:Number of undersized frames rcvd
> + * @rx_crc_frames:Number of CRC error frames rcvd
> + * @dropped_packets:Number of packets dropped due to link down on opposite port
> + *
> + * @tx_hwq_overflow:Hardware Tx Queue (on PRU) over flow count
> + * @tx_hwq_underflow:Hardware Tx Queue (on PRU) under flow count
> + *
> + * @u32 cs_error: Number of carrier sense errors

nit: @cs_error

     i.e. remove "u32 "

Documentation nits flagged by ./scripts/kernel-doc -none

> + * @sqe_test_error: Number of MAC receive errors
> + *
> + * Above fields are aligned so that it's consistent
> + * with the memory layout in PRU DRAM, this is to facilitate easy
> + * memcpy. Don't change the order of the fields.
> + *
> + * @vlan_dropped: Number of VLAN tagged packets dropped
> + * @multicast_dropped: Number of multicast packets dropped
> + */
> +struct port_statistics {
> +	u32 tx_bcast;
> +	u32 tx_mcast;
> +	u32 tx_ucast;
> +
> +	u32 tx_octets;
> +
> +	u32 rx_bcast;
> +	u32 rx_mcast;
> +	u32 rx_ucast;
> +
> +	u32 rx_octets;
> +
> +	u32 tx64byte;
> +	u32 tx65_127byte;
> +	u32 tx128_255byte;
> +	u32 tx256_511byte;
> +	u32 tx512_1023byte;
> +	u32 tx1024byte;
> +
> +	u32 rx64byte;
> +	u32 rx65_127byte;
> +	u32 rx128_255byte;
> +	u32 rx256_511byte;
> +	u32 rx512_1023byte;
> +	u32 rx1024byte;
> +
> +	u32 late_coll;
> +	u32 single_coll;
> +	u32 multi_coll;
> +	u32 excess_coll;
> +
> +	u32 rx_misalignment_frames;
> +	u32 stormprev_counter_bc;
> +	u32 stormprev_counter_mc;
> +	u32 stormprev_counter_uc;
> +	u32 mac_rxerror;
> +	u32 sfd_error;
> +	u32 def_tx;
> +	u32 mac_txerror;
> +	u32 rx_oversized_frames;
> +	u32 rx_undersized_frames;
> +	u32 rx_crc_frames;
> +	u32 dropped_packets;
> +
> +	u32 tx_hwq_overflow;
> +	u32 tx_hwq_underflow;
> +
> +	u32 cs_error;
> +	u32 sqe_test_error;
> +
> +	u32 vlan_dropped;
> +	u32 multicast_dropped;
> +} __packed;

...

The above notwithstanding, it seems based on comment's elsewhere in this
thread that there will be another revision of this patchset. Let patchwork
know about that.

pw-bot: changes-requested