All of lore.kernel.org
 help / color / mirror / Atom feed
From: Simon Horman <horms@kernel.org>
To: Roger Quadros <rogerq@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	Siddharth Vadapalli <s-vadapalli@ti.com>,
	Julien Panis <jpanis@baylibre.com>, Andrew Lunn <andrew@lunn.ch>,
	srk@ti.com, vigneshr@ti.com, danishanwar@ti.com,
	pekka Varis <p-varis@ti.com>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-omap@vger.kernel.org
Subject: Re: [PATCH net-next v2 6/7] net: ethernet: ti: cpsw_ale: add helper to setup classifier defaults
Date: Mon, 1 Jul 2024 08:35:05 +0100	[thread overview]
Message-ID: <20240701073505.GI17134@kernel.org> (raw)
In-Reply-To: <20240628-am65-cpsw-multi-rx-v2-6-c399cb77db56@kernel.org>

On Fri, Jun 28, 2024 at 03:01:55PM +0300, Roger Quadros wrote:
> Default behaviour is to have 8 classifiers to map 8 DSCP/PCP
> priorities to N receive threads (flows). N depends on number of
> RX channels enabled for the port.
> 
> Signed-off-by: Roger Quadros <rogerq@kernel.org>
> ---
>  drivers/net/ethernet/ti/cpsw_ale.c | 57 ++++++++++++++++++++++++++++++++++++++
>  drivers/net/ethernet/ti/cpsw_ale.h |  1 +
>  2 files changed, 58 insertions(+)
> 
> diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c
> index 75a17184d34c..51da527388df 100644
> --- a/drivers/net/ethernet/ti/cpsw_ale.c
> +++ b/drivers/net/ethernet/ti/cpsw_ale.c
> @@ -1650,3 +1650,60 @@ static void cpsw_ale_policer_thread_idx_enable(struct cpsw_ale *ale, u32 idx,
>  	regmap_field_write(ale->fields[ALE_THREAD_VALUE], thread_id);
>  	regmap_field_write(ale->fields[ALE_THREAD_ENABLE], enable ? 1 : 0);
>  }
> +
> +/* Disable all policer entries and thread mappings */
> +static void cpsw_ale_policer_reset(struct cpsw_ale *ale)
> +{
> +	int i;
> +
> +	for (i = 0; i < ale->params.num_policers ; i++) {
> +		cpsw_ale_policer_read_idx(ale, i);
> +		regmap_field_write(ale->fields[POL_PORT_MEN], 0);
> +		regmap_field_write(ale->fields[POL_PRI_MEN], 0);
> +		regmap_field_write(ale->fields[POL_OUI_MEN], 0);
> +		regmap_field_write(ale->fields[POL_DST_MEN], 0);
> +		regmap_field_write(ale->fields[POL_SRC_MEN], 0);
> +		regmap_field_write(ale->fields[POL_OVLAN_MEN], 0);
> +		regmap_field_write(ale->fields[POL_IVLAN_MEN], 0);
> +		regmap_field_write(ale->fields[POL_ETHERTYPE_MEN], 0);
> +		regmap_field_write(ale->fields[POL_IPSRC_MEN], 0);
> +		regmap_field_write(ale->fields[POL_IPDST_MEN], 0);
> +		regmap_field_write(ale->fields[POL_EN], 0);
> +		regmap_field_write(ale->fields[POL_RED_DROP_EN], 0);
> +		regmap_field_write(ale->fields[POL_YELLOW_DROP_EN], 0);
> +		regmap_field_write(ale->fields[POL_PRIORITY_THREAD_EN], 0);
> +
> +		cpsw_ale_policer_thread_idx_enable(ale, i, 0, 0);
> +	}
> +}
> +
> +/* Default classifer is to map 8 user priorities to N receive channels */
> +void cpsw_ale_classifier_setup_default(struct cpsw_ale *ale, int num_rx_ch)
> +{
> +	int pri, idx;
> +	int pri_thread_map[8][9] = {	{ 0, 0, 0, 0, 0, 0, 0, 0, },
> +					{ 0, 0, 0, 0, 1, 1, 1, 1, },
> +					{ 0, 0, 0, 0, 1, 1, 2, 2, },
> +					{ 1, 0, 0, 1, 2, 2, 3, 3, },
> +					{ 1, 0, 0, 1, 2, 3, 4, 4, },
> +					{ 1, 0, 0, 2, 3, 4, 5, 5, },
> +					{ 1, 0, 0, 2, 3, 4, 5, 6, },
> +					{ 2, 0, 1, 3, 4, 5, 6, 7, } };

Hi Roger,

Perhaps it is obvious, but I'm wondering if it is appropriate
to add a comment explaining the layout of the table, especially
the latter rows where the mapping of priority to receive channel
seems somewhat non-trivial.

> +
> +	cpsw_ale_policer_reset(ale);
> +
> +	/* use first 8 classifiers to map 8 (DSCP/PCP) priorities to channels */
> +	for (pri = 0; pri < 8; pri++) {
> +		idx = pri;
> +
> +		/* Classifier 'idx' match on priority 'pri' */
> +		cpsw_ale_policer_read_idx(ale, idx);
> +		regmap_field_write(ale->fields[POL_PRI_VAL], pri);
> +		regmap_field_write(ale->fields[POL_PRI_MEN], 1);
> +		cpsw_ale_policer_write_idx(ale, idx);
> +
> +		/* Map Classifier 'idx' to thread provided by the map */
> +		cpsw_ale_policer_thread_idx_enable(ale, idx,
> +						   pri_thread_map[num_rx_ch - 1][pri], 1);
> +	}
> +}
> diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h
> index 2cb76acc6d16..1e4e9a3dd234 100644
> --- a/drivers/net/ethernet/ti/cpsw_ale.h
> +++ b/drivers/net/ethernet/ti/cpsw_ale.h
> @@ -193,5 +193,6 @@ int cpsw_ale_vlan_add_modify(struct cpsw_ale *ale, u16 vid, int port_mask,
>  int cpsw_ale_vlan_del_modify(struct cpsw_ale *ale, u16 vid, int port_mask);
>  void cpsw_ale_set_unreg_mcast(struct cpsw_ale *ale, int unreg_mcast_mask,
>  			      bool add);
> +void cpsw_ale_classifier_setup_default(struct cpsw_ale *ale, int num_rx_ch);
>  
>  #endif
> 
> -- 
> 2.34.1
> 
> 

  reply	other threads:[~2024-07-01  7:35 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-28 12:01 [PATCH net-next v2 0/7] net: ethernet: ti: am65-cpsw: Add multi queue RX support Roger Quadros
2024-06-28 12:01 ` [PATCH net-next v2 1/7] net: ethernet: ti: am65-cpsw: Introduce multi queue Rx Roger Quadros
2024-07-01  7:35   ` Simon Horman
2024-07-01  9:44     ` Roger Quadros
2024-06-28 12:01 ` [PATCH net-next v2 2/7] net: ethernet: ti: cpsw_ale: use regfields for ALE registers Roger Quadros
2024-07-01  7:36   ` Simon Horman
2024-06-28 12:01 ` [PATCH net-next v2 3/7] net: ethernet: ti: cpsw_ale: use regfields for number of Entries and Policers Roger Quadros
2024-07-01  7:36   ` Simon Horman
2024-06-28 12:01 ` [PATCH net-next v2 4/7] net: ethernet: ti: cpsw_ale: add Policer and Thread control register fields Roger Quadros
2024-07-01  7:36   ` Simon Horman
2024-06-28 12:01 ` [PATCH net-next v2 5/7] net: ethernet: ti: cpsw_ale: add policer/classifier helpers Roger Quadros
2024-07-01  7:35   ` Simon Horman
2024-07-01  9:46     ` Roger Quadros
2024-06-28 12:01 ` [PATCH net-next v2 6/7] net: ethernet: ti: cpsw_ale: add helper to setup classifier defaults Roger Quadros
2024-07-01  7:35   ` Simon Horman [this message]
2024-07-01 10:32     ` Roger Quadros
2024-07-02  9:24       ` Simon Horman
2024-06-28 12:01 ` [PATCH net-next v2 7/7] net: ethernet: ti: am65-cpsw: setup priority to flow mapping Roger Quadros
2024-07-01  7:36   ` Simon Horman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240701073505.GI17134@kernel.org \
    --to=horms@kernel.org \
    --cc=andrew@lunn.ch \
    --cc=danishanwar@ti.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=jpanis@baylibre.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-omap@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=p-varis@ti.com \
    --cc=pabeni@redhat.com \
    --cc=rogerq@kernel.org \
    --cc=s-vadapalli@ti.com \
    --cc=srk@ti.com \
    --cc=vigneshr@ti.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.