netdev - Re: [PATCH v5 net-next 2/2] net: airoha: Introduce ethernet support for EN7581 SoC

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]

Date: Sat, 6 Jul 2024 14:58:04 +0200
From: Lorenzo Bianconi <lorenzo@...nel.org>
To: Simon Horman <horms@...nel.org>
Cc: netdev@...r.kernel.org, nbd@....name, lorenzo.bianconi83@...il.com,
	davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org,
	pabeni@...hat.com, conor@...nel.org,
	linux-arm-kernel@...ts.infradead.org, robh+dt@...nel.org,
	krzysztof.kozlowski+dt@...aro.org, conor+dt@...nel.org,
	devicetree@...r.kernel.org, catalin.marinas@....com,
	will@...nel.org, upstream@...oha.com,
	angelogioacchino.delregno@...labora.com,
	benjamin.larsson@...exis.eu, rkannoth@...vell.com,
	sgoutham@...vell.com, andrew@...n.ch, arnd@...db.de
Subject: Re: [PATCH v5 net-next 2/2] net: airoha: Introduce ethernet support
 for EN7581 SoC

> On Sat, Jul 06, 2024 at 10:27:00AM +0200, Lorenzo Bianconi wrote:
> > > On Thu, Jul 04, 2024 at 10:08:11AM +0200, Lorenzo Bianconi wrote:
> > > > Add airoha_eth driver in order to introduce ethernet support for
> > > > Airoha EN7581 SoC available on EN7581 development board (en7581-evb).
> > > > en7581-evb networking architecture is composed by airoha_eth as mac
> > > > controller (cpu port) and a mt7530 dsa based switch.
> > > > EN7581 mac controller is mainly composed by Frame Engine (FE) and
> > > > QoS-DMA (QDMA) modules. FE is used for traffic offloading (just basic
> > > > functionalities are supported now) while QDMA is used for DMA operation
> > > > and QOS functionalities between mac layer and the dsa switch (hw QoS is
> > > > not available yet and it will be added in the future).
> > > > Currently only hw lan features are available, hw wan will be added with
> > > > subsequent patches.
> > > > 
> > > > Tested-by: Benjamin Larsson <benjamin.larsson@...exis.eu>
> > > > Signed-off-by: Lorenzo Bianconi <lorenzo@...nel.org>
> > > 
> > > ...
> > > 
> > > > +static const char * const airoha_ethtool_stats_name[] = {
> > > > +	"tx_eth_pkt_cnt",
> > > > +	"tx_eth_byte_cnt",
> > > > +	"tx_ok_pkt_cnt",
> > > > +	"tx_ok_byte_cnt",
> > > > +	"tx_eth_drop_cnt",
> > > > +	"tx_eth_bc_cnt",
> > > > +	"tx_eth_mc_cnt",
> > > > +	"tx_eth_lt64_cnt",
> > > > +	"tx_eth_eq64_cnt",
> > > > +	"tx_eth_65_127_cnt",
> > > > +	"tx_eth_128_255_cnt",
> > > > +	"tx_eth_256_511_cnt",
> > > > +	"tx_eth_512_1023_cnt",
> > > > +	"tx_eth_1024_1518_cnt",
> > > > +	"tx_eth_gt1518_cnt",
> > > > +	"rx_eth_pkt_cnt",
> > > > +	"rx_eth_byte_cnt",
> > > > +	"rx_ok_pkt_cnt",
> > > > +	"rx_ok_byte_cnt",
> > > > +	"rx_eth_drop_cnt",
> > > > +	"rx_eth_bc_cnt",
> > > > +	"rx_eth_mc_cnt",
> > > > +	"rx_eth_crc_drop_cnt",
> > > > +	"rx_eth_frag_cnt",
> > > > +	"rx_eth_jabber_cnt",
> > > > +	"rx_eth_lt64_cnt",
> > > > +	"rx_eth_eq64_cnt",
> > > > +	"rx_eth_65_127_cnt",
> > > > +	"rx_eth_128_255_cnt",
> > > > +	"rx_eth_256_511_cnt",
> > > > +	"rx_eth_512_1023_cnt",
> > > > +	"rx_eth_1024_1518_cnt",
> > > > +	"rx_eth_gt1518_cnt",
> > > > +};
> > > 
> > > Hi Lorenzo,
> > > 
> > > Sorry for not noticing this earlier.
> > 
> > Hi Simon,
> > 
> > no worries :)
> > 
> > > It seems to me that some of the stats above could
> > > use standard stats, which is preferred.
> > 
> > Please correct me if I am wrong but it seems quite a common approach to have
> > same stats in both .ndo_get_stats64() and .get_ethtool_stats():
> > - https://github.com/torvalds/linux/blob/master/drivers/net/ethernet/mediatek/mtk_eth_soc.c#L212
> > - https://github.com/torvalds/linux/blob/master/drivers/net/ethernet/marvell/mvneta.c#L435
> > - https://github.com/torvalds/linux/blob/master/drivers/net/ethernet/intel/i40e/i40e_ethtool.c#L243
> > - https://github.com/torvalds/linux/blob/master/net/mac80211/ethtool.c#L52
> > - ...
> > 
> > Do you mean I should just report common stats (e.g. tx_packets or tx_bytes) in
> > .ndo_get_stats64()? Or it is fine to just add .ndo_get_stats64() callback (not
> > supported at the moment)?
> 
> The first option: It is my understanding that it preferred to only
> report common stats via ndo_get_stats64.
> 
>   "Please limit stats you report in ethtool -S to just the stats for which
>    proper interfaces don't exist. Don't duplicate what's already reported
>    via rtase_get_stats64(), also take a look at what can be reported via
>    various *_stats members of struct ethtool_ops."
> 
>    - Re: [PATCH net-next v18 10/13] rtase: Implement ethtool function
>      by Jakub Kicinski
>      https://lore.kernel.org/netdev/20240509204047.149e226e@kernel.org/

ack, thx for the clarification. I will move common stats in ndo_get_stats64()
in v6.

Regards,
Lorenzo

> 
> > > Basically, my understanding is that one should:
> > > 1. Implement .ndo_get_stats64
> > >    (that seems relevant here)
> > > 2. As appropriate implement ethtool_stats non-extended stats operations
> > >    (perhaps not relevant here)
> > 
> > Can you please provide me a pointer for it?
> 
> Sorry, that was not at all clear.
> I meant, as per the quote above, *_stats members of struct ethtool_ops,
> other than those that implement extended ops.
> e.g. get_rmon_stats.
> 
> > 
> > Regards,
> > Lorenzo
> > 
> > > 3. Then implement get_ethtool_stats for what is left over
> > > 
> > > ...
> 
> 

Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)