This was a long journey to arrive and discover this problem.
To not waste too much char, there is a race problem with PHY and driver
probe. This was observed with Aquantia PHY firmware loading.
With some hacks the race problem was workarounded but an interesting
thing was notice. It took more than a minute for the firmware to load
via MDIO.
This was strange as the same operation was done by UBoot in at max 5
second and the same data was loaded.
A similar problem was observed on a mtk board that also had an
Aquantia PHY where the load was very slow. It was notice that the cause
was the MDIO bus running at a very low speed and the firmware
was missing a property (present in mtk sdk) that set the right frequency
to the MDIO bus.
It was fun to find that THE VERY SAME PROBLEM is present on IPQ in a
different form. The MDIO apply internally a division to the feed clock
resulting in the bus running at 390KHz instead of 6.25Mhz.
Searching around the web for some documentation and some include and
analyzing the uboot codeflow resulted in the divider being set wrongly
at /256 instead of /16 as the value was actually never set.
Applying the value restore the original load time for the Aquantia PHY.
This series mainly handle this by adding support for the "clock-frequency"
property.
Changes v3:
- Add Reviewed-by tag
- Fix english grammar error in comment
- Drop DTS patch
Changes v2:
- Use DIV_ROUND_UP
- Introduce logic to chose a default value for 802.3 spec 2.5MHz
Christian Marangi (2):
dt-bindings: net: ipq4019-mdio: document now supported clock-frequency
net: mdio: ipq4019: add support for clock-frequency property
.../bindings/net/qcom,ipq4019-mdio.yaml | 15 +++
drivers/net/mdio/mdio-ipq4019.c | 109 +++++++++++++++++-
2 files changed, 118 insertions(+), 6 deletions(-)
--
2.43.0
The IPQ4019 MDIO internally divide the clock feed by AHB based on the
MDIO_MODE reg. On reset or power up, the default value for the
divider is 0xff that reflect the divider set to /256.
This makes the MDC run at a very low rate, that is, considering AHB is
always fixed to 100Mhz, a value of 390KHz.
This hasn't have been a problem as MDIO wasn't used for time sensitive
operation, it is now that on IPQ807x is usually mounted with PHY that
requires MDIO to load their firmware (example Aquantia PHY).
To handle this problem and permit to set the correct designed MDC
frequency for the SoC add support for the standard "clock-frequency"
property for the MDIO node.
The divider supports value from /1 to /256 and the common value are to
set it to /16 to reflect 6.25Mhz or to /8 on newer platform to reflect
12.5Mhz.
To scan if the requested rate is supported by the divider, loop with
each supported divider and stop when the requested rate match the final
rate with the current divider. An error is returned if the rate doesn't
match any value.
On MDIO reset, the divider is restored to the requested value to prevent
any kind of downclocking caused by the divider reverting to a default
value.
To follow 802.3 spec of 2.5MHz of default value, if divider is set at
/256 and "clock-frequency" is not set in DT, assume nobody set the
divider and try to find the closest MDC rate to 2.5MHz. (in the case of
AHB set to 100MHz, it's 1.5625MHz)
While at is also document other bits of the MDIO_MODE reg to have a
clear idea of what is actually applied there.
Documentation of some BITs is skipped as they are marked as reserved and
their usage is not clear (RES 11:9 GENPHY 16:13 RES1 19:17)
Signed-off-by: Christian Marangi <[email protected]>
Reviewed-by: Andrew Lunn <[email protected]>
---
drivers/net/mdio/mdio-ipq4019.c | 109 ++++++++++++++++++++++++++++++--
1 file changed, 103 insertions(+), 6 deletions(-)
diff --git a/drivers/net/mdio/mdio-ipq4019.c b/drivers/net/mdio/mdio-ipq4019.c
index abd8b508ec16..9d8f43b28aac 100644
--- a/drivers/net/mdio/mdio-ipq4019.c
+++ b/drivers/net/mdio/mdio-ipq4019.c
@@ -14,6 +14,20 @@
#include <linux/clk.h>
#define MDIO_MODE_REG 0x40
+#define MDIO_MODE_MDC_MODE BIT(12)
+/* 0 = Clause 22, 1 = Clause 45 */
+#define MDIO_MODE_C45 BIT(8)
+#define MDIO_MODE_DIV_MASK GENMASK(7, 0)
+#define MDIO_MODE_DIV(x) FIELD_PREP(MDIO_MODE_DIV_MASK, (x) - 1)
+#define MDIO_MODE_DIV_1 0x0
+#define MDIO_MODE_DIV_2 0x1
+#define MDIO_MODE_DIV_4 0x3
+#define MDIO_MODE_DIV_8 0x7
+#define MDIO_MODE_DIV_16 0xf
+#define MDIO_MODE_DIV_32 0x1f
+#define MDIO_MODE_DIV_64 0x3f
+#define MDIO_MODE_DIV_128 0x7f
+#define MDIO_MODE_DIV_256 0xff
#define MDIO_ADDR_REG 0x44
#define MDIO_DATA_WRITE_REG 0x48
#define MDIO_DATA_READ_REG 0x4c
@@ -26,9 +40,6 @@
#define MDIO_CMD_ACCESS_CODE_C45_WRITE 1
#define MDIO_CMD_ACCESS_CODE_C45_READ 2
-/* 0 = Clause 22, 1 = Clause 45 */
-#define MDIO_MODE_C45 BIT(8)
-
#define IPQ4019_MDIO_TIMEOUT 10000
#define IPQ4019_MDIO_SLEEP 10
@@ -41,6 +52,7 @@ struct ipq4019_mdio_data {
void __iomem *membase;
void __iomem *eth_ldo_rdy;
struct clk *mdio_clk;
+ unsigned int mdc_rate;
};
static int ipq4019_mdio_wait_busy(struct mii_bus *bus)
@@ -203,6 +215,38 @@ static int ipq4019_mdio_write_c22(struct mii_bus *bus, int mii_id, int regnum,
return 0;
}
+static int ipq4019_mdio_set_div(struct ipq4019_mdio_data *priv)
+{
+ unsigned long ahb_rate;
+ int div;
+ u32 val;
+
+ /* If we don't have a clock for AHB use the fixed value */
+ ahb_rate = IPQ_MDIO_CLK_RATE;
+ if (priv->mdio_clk)
+ ahb_rate = clk_get_rate(priv->mdio_clk);
+
+ /* MDC rate is ahb_rate/(MDIO_MODE_DIV + 1)
+ * While supported, internal documentation doesn't
+ * assure correct functionality of the MDIO bus
+ * with divider of 1, 2 or 4.
+ */
+ for (div = 8; div <= 256; div *= 2) {
+ /* The requested rate is supported by the div */
+ if (priv->mdc_rate == DIV_ROUND_UP(ahb_rate, div)) {
+ val = readl(priv->membase + MDIO_MODE_REG);
+ val &= ~MDIO_MODE_DIV_MASK;
+ val |= MDIO_MODE_DIV(div);
+ writel(val, priv->membase + MDIO_MODE_REG);
+
+ return 0;
+ }
+ }
+
+ /* The requested rate is not supported */
+ return -EINVAL;
+}
+
static int ipq_mdio_reset(struct mii_bus *bus)
{
struct ipq4019_mdio_data *priv = bus->priv;
@@ -225,10 +269,58 @@ static int ipq_mdio_reset(struct mii_bus *bus)
return ret;
ret = clk_prepare_enable(priv->mdio_clk);
- if (ret == 0)
- mdelay(10);
+ if (ret)
+ return ret;
- return ret;
+ mdelay(10);
+
+ /* Restore MDC rate */
+ return ipq4019_mdio_set_div(priv);
+}
+
+static void ipq4019_mdio_select_mdc_rate(struct platform_device *pdev,
+ struct ipq4019_mdio_data *priv)
+{
+ unsigned long ahb_rate;
+ int div;
+ u32 val;
+
+ /* MDC rate defined in DT, we don't have to decide a default value */
+ if (!of_property_read_u32(pdev->dev.of_node, "clock-frequency",
+ &priv->mdc_rate))
+ return;
+
+ /* If we don't have a clock for AHB use the fixed value */
+ ahb_rate = IPQ_MDIO_CLK_RATE;
+ if (priv->mdio_clk)
+ ahb_rate = clk_get_rate(priv->mdio_clk);
+
+ /* Check what is the current div set */
+ val = readl(priv->membase + MDIO_MODE_REG);
+ div = FIELD_GET(MDIO_MODE_DIV_MASK, val);
+
+ /* div is not set to the default value of /256
+ * Probably someone changed that (bootloader, other drivers)
+ * Keep this and don't overwrite it.
+ */
+ if (div != MDIO_MODE_DIV_256) {
+ priv->mdc_rate = DIV_ROUND_UP(ahb_rate, div + 1);
+ return;
+ }
+
+ /* If div is /256 assume nobody have set this value and
+ * try to find one MDC rate that is close the 802.3 spec of
+ * 2.5MHz
+ */
+ for (div = 256; div >= 8; div /= 2) {
+ /* Stop as soon as we found a divider that
+ * reached the closest value to 2.5MHz
+ */
+ if (DIV_ROUND_UP(ahb_rate, div) > 2500000)
+ break;
+
+ priv->mdc_rate = DIV_ROUND_UP(ahb_rate, div);
+ }
}
static int ipq4019_mdio_probe(struct platform_device *pdev)
@@ -252,6 +344,11 @@ static int ipq4019_mdio_probe(struct platform_device *pdev)
if (IS_ERR(priv->mdio_clk))
return PTR_ERR(priv->mdio_clk);
+ ipq4019_mdio_select_mdc_rate(pdev, priv);
+ ret = ipq4019_mdio_set_div(priv);
+ if (ret)
+ return ret;
+
/* The platform resource is provided on the chipset IPQ5018 */
/* This resource is optional */
res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
--
2.43.0
Hello:
This series was applied to netdev/net-next.git (main)
by David S. Miller <[email protected]>:
On Wed, 31 Jan 2024 03:26:02 +0100 you wrote:
> This was a long journey to arrive and discover this problem.
>
> To not waste too much char, there is a race problem with PHY and driver
> probe. This was observed with Aquantia PHY firmware loading.
>
> With some hacks the race problem was workarounded but an interesting
> thing was notice. It took more than a minute for the firmware to load
> via MDIO.
>
> [...]
Here is the summary with links:
- [net-next,v3,1/2] dt-bindings: net: ipq4019-mdio: document now supported clock-frequency
https://git.kernel.org/netdev/net-next/c/9484b9555de0
- [net-next,v3,2/2] net: mdio: ipq4019: add support for clock-frequency property
https://git.kernel.org/netdev/net-next/c/bdce82e960d1
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html