* [PATCH net] net: dsa: microchip: Fix KSZ8863 reset problem
@ 2025-08-02 0:22 Tristram.Ha
2025-08-02 18:59 ` Oleksij Rempel
0 siblings, 1 reply; 3+ messages in thread
From: Tristram.Ha @ 2025-08-02 0:22 UTC (permalink / raw)
To: Oleksij Rempel, Woojung Huh, Andrew Lunn, Vladimir Oltean
Cc: Maxime Chevallier, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, UNGLinuxDriver, netdev, linux-kernel, Tristram Ha
From: Tristram Ha <tristram.ha@microchip.com>
ksz8873_valid_regs[] was added for register access for KSZ8863/KSZ8873
switches, but the reset register is not in the list so
ksz8_reset_switch() does not take any effect.
ksz_cfg() is updated to display an error so that there will be a future
check for adding new register access code.
A side effect of not resetting the switch is the static MAC table is not
cleared. Further additions to the table will show write error as there
are only 8 entries in the table.
Fixes: d0dec3333040 ("net: dsa: microchip: Add register access control for KSZ8873 chip")
Signed-off-by: Tristram Ha <tristram.ha@microchip.com>
---
drivers/net/dsa/microchip/ksz8.c | 7 ++++++-
drivers/net/dsa/microchip/ksz_common.c | 1 +
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/drivers/net/dsa/microchip/ksz8.c b/drivers/net/dsa/microchip/ksz8.c
index 76e490070e9c..6d282a8e3684 100644
--- a/drivers/net/dsa/microchip/ksz8.c
+++ b/drivers/net/dsa/microchip/ksz8.c
@@ -36,7 +36,12 @@
static void ksz_cfg(struct ksz_device *dev, u32 addr, u8 bits, bool set)
{
- regmap_update_bits(ksz_regmap_8(dev), addr, bits, set ? bits : 0);
+ int ret;
+
+ ret = regmap_update_bits(ksz_regmap_8(dev), addr, bits, set ? bits : 0);
+ if (ret)
+ dev_err(dev->dev, "can't update reg 0x%x: %pe\n", addr,
+ ERR_PTR(ret));
}
static void ksz_port_cfg(struct ksz_device *dev, int port, int offset, u8 bits,
diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c
index 7292bfe2f7ca..4cb14288ff0f 100644
--- a/drivers/net/dsa/microchip/ksz_common.c
+++ b/drivers/net/dsa/microchip/ksz_common.c
@@ -1447,6 +1447,7 @@ static const struct regmap_range ksz8873_valid_regs[] = {
regmap_reg_range(0x3f, 0x3f),
/* advanced control registers */
+ regmap_reg_range(0x43, 0x43),
regmap_reg_range(0x60, 0x6f),
regmap_reg_range(0x70, 0x75),
regmap_reg_range(0x76, 0x78),
--
2.34.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH net] net: dsa: microchip: Fix KSZ8863 reset problem
2025-08-02 0:22 [PATCH net] net: dsa: microchip: Fix KSZ8863 reset problem Tristram.Ha
@ 2025-08-02 18:59 ` Oleksij Rempel
2025-08-07 0:53 ` Tristram.Ha
0 siblings, 1 reply; 3+ messages in thread
From: Oleksij Rempel @ 2025-08-02 18:59 UTC (permalink / raw)
To: Tristram.Ha
Cc: Oleksij Rempel, Woojung Huh, Andrew Lunn, Vladimir Oltean,
Maxime Chevallier, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, UNGLinuxDriver, netdev, linux-kernel
Hi,
On Fri, Aug 01, 2025 at 05:22:53PM -0700, Tristram.Ha@microchip.com wrote:
> From: Tristram Ha <tristram.ha@microchip.com>
>
> ksz8873_valid_regs[] was added for register access for KSZ8863/KSZ8873
> switches, but the reset register is not in the list so
> ksz8_reset_switch() does not take any effect.
>
> ksz_cfg() is updated to display an error so that there will be a future
> check for adding new register access code.
>
> A side effect of not resetting the switch is the static MAC table is not
> cleared. Further additions to the table will show write error as there
> are only 8 entries in the table.
Thank you for fixing it!
> Fixes: d0dec3333040 ("net: dsa: microchip: Add register access control for KSZ8873 chip")
> Signed-off-by: Tristram Ha <tristram.ha@microchip.com>
> ---
> drivers/net/dsa/microchip/ksz8.c | 7 ++++++-
> drivers/net/dsa/microchip/ksz_common.c | 1 +
> 2 files changed, 7 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/dsa/microchip/ksz8.c b/drivers/net/dsa/microchip/ksz8.c
> index 76e490070e9c..6d282a8e3684 100644
> --- a/drivers/net/dsa/microchip/ksz8.c
> +++ b/drivers/net/dsa/microchip/ksz8.c
> @@ -36,7 +36,12 @@
>
> static void ksz_cfg(struct ksz_device *dev, u32 addr, u8 bits, bool set)
> {
> - regmap_update_bits(ksz_regmap_8(dev), addr, bits, set ? bits : 0);
> + int ret;
> +
> + ret = regmap_update_bits(ksz_regmap_8(dev), addr, bits, set ? bits : 0);
> + if (ret)
> + dev_err(dev->dev, "can't update reg 0x%x: %pe\n", addr,
> + ERR_PTR(ret));
Better using ksz_rmw8() instead. It is already providing error message.
In this file there is 4 direct accesses to regmap_update_bits() without
error handling. It would be great if you have chance to replace it with
ksz_rmw8() too.
> }
>
> static void ksz_port_cfg(struct ksz_device *dev, int port, int offset, u8 bits,
> diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c
> index 7292bfe2f7ca..4cb14288ff0f 100644
> --- a/drivers/net/dsa/microchip/ksz_common.c
> +++ b/drivers/net/dsa/microchip/ksz_common.c
> @@ -1447,6 +1447,7 @@ static const struct regmap_range ksz8873_valid_regs[] = {
> regmap_reg_range(0x3f, 0x3f),
>
> /* advanced control registers */
> + regmap_reg_range(0x43, 0x43),
This register is no documented in the public documentation. Out of
curiosity, are there some where more information about this two
"reserved" register ranges: 0x3A-0x3E and 0x40-0x5F?
Best regards,
Oleksij
--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
^ permalink raw reply [flat|nested] 3+ messages in thread
* RE: [PATCH net] net: dsa: microchip: Fix KSZ8863 reset problem
2025-08-02 18:59 ` Oleksij Rempel
@ 2025-08-07 0:53 ` Tristram.Ha
0 siblings, 0 replies; 3+ messages in thread
From: Tristram.Ha @ 2025-08-07 0:53 UTC (permalink / raw)
To: o.rempel
Cc: linux, Woojung.Huh, andrew, olteanv, maxime.chevallier, davem,
edumazet, kuba, pabeni, UNGLinuxDriver, netdev, linux-kernel
> On Fri, Aug 01, 2025 at 05:22:53PM -0700, Tristram.Ha@microchip.com wrote:
> > From: Tristram Ha <tristram.ha@microchip.com>
> >
> > ksz8873_valid_regs[] was added for register access for KSZ8863/KSZ8873
> > switches, but the reset register is not in the list so
> > ksz8_reset_switch() does not take any effect.
> >
> > ksz_cfg() is updated to display an error so that there will be a future
> > check for adding new register access code.
> >
> > A side effect of not resetting the switch is the static MAC table is not
> > cleared. Further additions to the table will show write error as there
> > are only 8 entries in the table.
>
> Thank you for fixing it!
>
> > Fixes: d0dec3333040 ("net: dsa: microchip: Add register access control for KSZ8873
> chip")
> > Signed-off-by: Tristram Ha <tristram.ha@microchip.com>
> > ---
> > drivers/net/dsa/microchip/ksz8.c | 7 ++++++-
> > drivers/net/dsa/microchip/ksz_common.c | 1 +
> > 2 files changed, 7 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/dsa/microchip/ksz8.c b/drivers/net/dsa/microchip/ksz8.c
> > index 76e490070e9c..6d282a8e3684 100644
> > --- a/drivers/net/dsa/microchip/ksz8.c
> > +++ b/drivers/net/dsa/microchip/ksz8.c
> > @@ -36,7 +36,12 @@
> >
> > static void ksz_cfg(struct ksz_device *dev, u32 addr, u8 bits, bool set)
> > {
> > - regmap_update_bits(ksz_regmap_8(dev), addr, bits, set ? bits : 0);
> > + int ret;
> > +
> > + ret = regmap_update_bits(ksz_regmap_8(dev), addr, bits, set ? bits : 0);
> > + if (ret)
> > + dev_err(dev->dev, "can't update reg 0x%x: %pe\n", addr,
> > + ERR_PTR(ret));
>
> Better using ksz_rmw8() instead. It is already providing error message.
>
> In this file there is 4 direct accesses to regmap_update_bits() without
> error handling. It would be great if you have chance to replace it with
> ksz_rmw8() too.
Will do.
> > }
> >
> > static void ksz_port_cfg(struct ksz_device *dev, int port, int offset, u8 bits,
> > diff --git a/drivers/net/dsa/microchip/ksz_common.c
> b/drivers/net/dsa/microchip/ksz_common.c
> > index 7292bfe2f7ca..4cb14288ff0f 100644
> > --- a/drivers/net/dsa/microchip/ksz_common.c
> > +++ b/drivers/net/dsa/microchip/ksz_common.c
> > @@ -1447,6 +1447,7 @@ static const struct regmap_range ksz8873_valid_regs[] =
> {
> > regmap_reg_range(0x3f, 0x3f),
> >
> > /* advanced control registers */
> > + regmap_reg_range(0x43, 0x43),
>
> This register is no documented in the public documentation. Out of
> curiosity, are there some where more information about this two
> "reserved" register ranges: 0x3A-0x3E and 0x40-0x5F?
The current KSZ8863 and KSZ8873 datasheets available from Microchip
website do define this 0x43 register as it is a basic operation to reset
the chip. One thing is the register bits are automatically cleared so it
does not require a second write.
Registers 0x1A-1E and 0x2A-2E are used for LinkMD feature of the PHY and
other PHY related status, so they do not apply for port 3 with 0x3A-3E.
Registers 0x40-0x5F are not used as they are not in the datasheet.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2025-08-07 0:53 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-08-02 0:22 [PATCH net] net: dsa: microchip: Fix KSZ8863 reset problem Tristram.Ha
2025-08-02 18:59 ` Oleksij Rempel
2025-08-07 0:53 ` Tristram.Ha
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).