From: "Alvin Šipraga" <ALSI@bang-olufsen.dk>
To: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"linus.walleij@linaro.org" <linus.walleij@linaro.org>,
"andrew@lunn.ch" <andrew@lunn.ch>,
"vivien.didelot@gmail.com" <vivien.didelot@gmail.com>,
"f.fainelli@gmail.com" <f.fainelli@gmail.com>,
"olteanv@gmail.com" <olteanv@gmail.com>,
"davem@davemloft.net" <davem@davemloft.net>,
"kuba@kernel.org" <kuba@kernel.org>,
"arinc.unal@arinc9.com" <arinc.unal@arinc9.com>
Subject: Re: [PATCH net-next v2 2/2] net: dsa: realtek: rtl8365mb: add support for rtl8_4t
Date: Fri, 18 Feb 2022 12:26:28 +0000 [thread overview]
Message-ID: <8735kgpdho.fsf@bang-olufsen.dk> (raw)
In-Reply-To: <20220218060959.6631-3-luizluca@gmail.com> (Luiz Angelo Daros de Luca's message of "Fri, 18 Feb 2022 03:09:59 -0300")
Luiz Angelo Daros de Luca <luizluca@gmail.com> writes:
> The trailing tag is also supported by this family. The default is still
> rtl8_4 but now the switch supports changing the tag to rtl8_4t.
>
> Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
> ---
> drivers/net/dsa/realtek/rtl8365mb.c | 78 ++++++++++++++++++++++++-----
> 1 file changed, 66 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/net/dsa/realtek/rtl8365mb.c b/drivers/net/dsa/realtek/rtl8365mb.c
> index 2ed592147c20..043cac34e906 100644
> --- a/drivers/net/dsa/realtek/rtl8365mb.c
> +++ b/drivers/net/dsa/realtek/rtl8365mb.c
> @@ -524,9 +524,7 @@ enum rtl8365mb_cpu_rxlen {
> * @mask: port mask of ports that parse should parse CPU tags
> * @trap_port: forward trapped frames to this port
> * @insert: CPU tag insertion mode in switch->CPU frames
> - * @position: position of CPU tag in frame
> * @rx_length: minimum CPU RX length
> - * @format: CPU tag format
> *
> * Represents the CPU tagging and CPU port configuration of the switch. These
> * settings are configurable at runtime.
> @@ -536,9 +534,7 @@ struct rtl8365mb_cpu {
> u32 mask;
> u32 trap_port;
> enum rtl8365mb_cpu_insert insert;
> - enum rtl8365mb_cpu_position position;
> enum rtl8365mb_cpu_rxlen rx_length;
> - enum rtl8365mb_cpu_format format;
This struct is meant to represent the whole CPU config register. Rather
than pulling it out and adding tag_protocol to struct rtl8365mb, can you
instead do something like:
- keep these members of _cpu
- put back the cpu member of struct rtl8365mb (I don't know why it was removed...)
- in get_tag_protocol: return mb->cpu.position == AFTER_SA ? RTL8_4 : RTL8_4T;
- in change_tag_protocol: just update mb->cpu.position and call
rtl8365mb_cpu_config again
- avoid the arcane call to rtl8365mb_change_tag_protocol in _setup
- avoid the need to do regmap_update_bits instead of a clean
regmap_write in one place
The reason I'm saying this is because, in the original version of the
driver, CPU configuration was in a single place. Now it is scattered. I
would kindly ask that you try to respect the existing design because I
can already see that things are starting to get a bit messy.
If we subsequently want to configure other CPU parameters on the fly, it
will be as easy as updating the cpu struct and calling cpu_config
again. This register is also non-volatile so the state we keep will
always conform with the switch configuration.
Sorry if you find the feedback too opinionated - I don't mean anything
personally. But the original design was not by accident, so I would
appreciate if we can keep it that way unless there is a good reason to
change it.
> };
>
> /**
> @@ -566,6 +562,7 @@ struct rtl8365mb_port {
> * @chip_ver: chip silicon revision
> * @port_mask: mask of all ports
> * @learn_limit_max: maximum number of L2 addresses the chip can learn
> + * @tag_protocol: current switch CPU tag protocol
> * @mib_lock: prevent concurrent reads of MIB counters
> * @ports: per-port data
> * @jam_table: chip-specific initialization jam table
> @@ -580,6 +577,7 @@ struct rtl8365mb {
> u32 chip_ver;
> u32 port_mask;
> u32 learn_limit_max;
> + enum dsa_tag_protocol tag_protocol;
> struct mutex mib_lock;
> struct rtl8365mb_port ports[RTL8365MB_MAX_NUM_PORTS];
> const struct rtl8365mb_jam_tbl_entry *jam_table;
> @@ -770,7 +768,54 @@ static enum dsa_tag_protocol
> rtl8365mb_get_tag_protocol(struct dsa_switch *ds, int port,
> enum dsa_tag_protocol mp)
> {
> - return DSA_TAG_PROTO_RTL8_4;
> + struct realtek_priv *priv = ds->priv;
> + struct rtl8365mb *chip_data;
Please stick to the convention and call this struct rtl8365mb pointer mb.
> +
> + chip_data = priv->chip_data;
> +
> + return chip_data->tag_protocol;
> +}
> +
> +static int rtl8365mb_change_tag_protocol(struct dsa_switch *ds, int cpu,
> + enum dsa_tag_protocol proto)
> +{
> + struct realtek_priv *priv = ds->priv;
> + struct rtl8365mb *chip_data;
s/chip_data/mb/ per convention
> + int tag_position;
> + int tag_format;
> + int ret;
> +
> + switch (proto) {
> + case DSA_TAG_PROTO_RTL8_4:
> + tag_format = RTL8365MB_CPU_FORMAT_8BYTES;
> + tag_position = RTL8365MB_CPU_POS_AFTER_SA;
> + break;
> + case DSA_TAG_PROTO_RTL8_4T:
> + tag_format = RTL8365MB_CPU_FORMAT_8BYTES;
> + tag_position = RTL8365MB_CPU_POS_BEFORE_CRC;
> + break;
> + /* The switch also supports a 4-byte format, similar to rtl4a but with
> + * the same 0x04 8-bit version and probably 8-bit port source/dest.
> + * There is no public doc about it. Not supported yet.
> + */
> + default:
> + return -EPROTONOSUPPORT;
> + }
> +
> + ret = regmap_update_bits(priv->map, RTL8365MB_CPU_CTRL_REG,
> + RTL8365MB_CPU_CTRL_TAG_POSITION_MASK |
> + RTL8365MB_CPU_CTRL_TAG_FORMAT_MASK,
> + FIELD_PREP(RTL8365MB_CPU_CTRL_TAG_POSITION_MASK,
> + tag_position) |
> + FIELD_PREP(RTL8365MB_CPU_CTRL_TAG_FORMAT_MASK,
> + tag_format));
> + if (ret)
> + return ret;
> +
> + chip_data = priv->chip_data;
nit: I would put this assignment up top like in the rest of the driver,
respecting reverse-christmass-tree order. It's nice to stick to the
existing style.
> + chip_data->tag_protocol = proto;
> +
> + return 0;
> }
>
> static int rtl8365mb_ext_config_rgmii(struct realtek_priv *priv, int port,
> @@ -1739,13 +1784,18 @@ static int rtl8365mb_cpu_config(struct realtek_priv *priv, const struct rtl8365m
>
> val = FIELD_PREP(RTL8365MB_CPU_CTRL_EN_MASK, cpu->enable ? 1 : 0) |
> FIELD_PREP(RTL8365MB_CPU_CTRL_INSERTMODE_MASK, cpu->insert) |
> - FIELD_PREP(RTL8365MB_CPU_CTRL_TAG_POSITION_MASK, cpu->position) |
> FIELD_PREP(RTL8365MB_CPU_CTRL_RXBYTECOUNT_MASK, cpu->rx_length) |
> - FIELD_PREP(RTL8365MB_CPU_CTRL_TAG_FORMAT_MASK, cpu->format) |
> FIELD_PREP(RTL8365MB_CPU_CTRL_TRAP_PORT_MASK, cpu->trap_port & 0x7) |
> FIELD_PREP(RTL8365MB_CPU_CTRL_TRAP_PORT_EXT_MASK,
> cpu->trap_port >> 3 & 0x1);
> - ret = regmap_write(priv->map, RTL8365MB_CPU_CTRL_REG, val);
> +
> + ret = regmap_update_bits(priv->map, RTL8365MB_CPU_CTRL_REG,
> + RTL8365MB_CPU_CTRL_EN_MASK |
> + RTL8365MB_CPU_CTRL_INSERTMODE_MASK |
> + RTL8365MB_CPU_CTRL_RXBYTECOUNT_MASK |
> + RTL8365MB_CPU_CTRL_TRAP_PORT_MASK |
> + RTL8365MB_CPU_CTRL_TRAP_PORT_EXT_MASK,
> + val);
> if (ret)
> return ret;
>
> @@ -1827,6 +1877,11 @@ static int rtl8365mb_setup(struct dsa_switch *ds)
> dev_info(priv->dev, "no interrupt support\n");
>
> /* Configure CPU tagging */
> + ret = rtl8365mb_change_tag_protocol(priv->ds, -1, DSA_TAG_PROTO_RTL8_4);
> + if (ret) {
> + dev_err(priv->dev, "failed to set default tag protocol: %d\n", ret);
> + return ret;
> + }
> cpu.trap_port = RTL8365MB_MAX_NUM_PORTS;
> dsa_switch_for_each_cpu_port(cpu_dp, priv->ds) {
> cpu.mask |= BIT(cpu_dp->index);
> @@ -1834,13 +1889,9 @@ static int rtl8365mb_setup(struct dsa_switch *ds)
> if (cpu.trap_port == RTL8365MB_MAX_NUM_PORTS)
> cpu.trap_port = cpu_dp->index;
> }
> -
> cpu.enable = cpu.mask > 0;
> cpu.insert = RTL8365MB_CPU_INSERT_TO_ALL;
> - cpu.position = RTL8365MB_CPU_POS_AFTER_SA;
> cpu.rx_length = RTL8365MB_CPU_RXLEN_64BYTES;
> - cpu.format = RTL8365MB_CPU_FORMAT_8BYTES;
Like I said above, I think it would be nice to put this cpu struct back
in the rtl8365mb private data.
> -
> ret = rtl8365mb_cpu_config(priv, &cpu);
> if (ret)
> goto out_teardown_irq;
> @@ -1982,6 +2033,7 @@ static int rtl8365mb_detect(struct realtek_priv *priv)
> mb->learn_limit_max = RTL8365MB_LEARN_LIMIT_MAX;
> mb->jam_table = rtl8365mb_init_jam_8365mb_vc;
> mb->jam_size = ARRAY_SIZE(rtl8365mb_init_jam_8365mb_vc);
> + mb->tag_protocol = DSA_TAG_PROTO_RTL8_4;
>
> break;
> default:
> @@ -1996,6 +2048,7 @@ static int rtl8365mb_detect(struct realtek_priv *priv)
>
> static const struct dsa_switch_ops rtl8365mb_switch_ops_smi = {
> .get_tag_protocol = rtl8365mb_get_tag_protocol,
> + .change_tag_protocol = rtl8365mb_change_tag_protocol,
> .setup = rtl8365mb_setup,
> .teardown = rtl8365mb_teardown,
> .phylink_get_caps = rtl8365mb_phylink_get_caps,
> @@ -2014,6 +2067,7 @@ static const struct dsa_switch_ops rtl8365mb_switch_ops_smi = {
>
> static const struct dsa_switch_ops rtl8365mb_switch_ops_mdio = {
> .get_tag_protocol = rtl8365mb_get_tag_protocol,
> + .change_tag_protocol = rtl8365mb_change_tag_protocol,
> .setup = rtl8365mb_setup,
> .teardown = rtl8365mb_teardown,
> .phylink_get_caps = rtl8365mb_phylink_get_caps,
next prev parent reply other threads:[~2022-02-18 12:26 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-18 6:09 [PATCH net-next v2 0/2] net: dsa: realtek: add rtl8_4t tag Luiz Angelo Daros de Luca
2022-02-18 6:09 ` [PATCH net-next v2 1/2] net: dsa: tag_rtl8_4: add rtl8_4t trailing variant Luiz Angelo Daros de Luca
2022-02-18 11:46 ` Alvin Šipraga
2022-02-22 0:37 ` Luiz Angelo Daros de Luca
2022-02-18 6:09 ` [PATCH net-next v2 2/2] net: dsa: realtek: rtl8365mb: add support for rtl8_4t Luiz Angelo Daros de Luca
2022-02-18 12:26 ` Alvin Šipraga [this message]
2022-02-22 2:42 ` Luiz Angelo Daros de Luca
2022-02-22 12:20 ` Alvin Šipraga
2022-02-22 19:54 ` Luiz Angelo Daros de Luca
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=8735kgpdho.fsf@bang-olufsen.dk \
--to=alsi@bang-olufsen.dk \
--cc=andrew@lunn.ch \
--cc=arinc.unal@arinc9.com \
--cc=davem@davemloft.net \
--cc=f.fainelli@gmail.com \
--cc=kuba@kernel.org \
--cc=linus.walleij@linaro.org \
--cc=luizluca@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=olteanv@gmail.com \
--cc=vivien.didelot@gmail.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.