Hi Eugeniy

On 02/06/2018 03:12 PM, Eugeniy Paltsev wrote:

After commit ba1f966725223 "net: designware: add clock support" we got NET broken on axs101 and axs103 platforms.

Some clock don't support gating so their clock drivers don't implement .enable/.disable callbacks. In such case clk_enable returns -ENOSYS. Also some clock drivers implement .enable/.disable callbacks not for all clock IDs and return -ENOSYS (or -ENOTSUPP) for others.

If we have such clock in 'clocks' list of designware ethernet controller node we fail to probe designware ethernet.

Fix it.

Signed-off-by: Eugeniy Paltsev Eugeniy.Paltsev@synopsys.com

drivers/net/designware.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/designware.c b/drivers/net/designware.c index 6d53071..43670a7 100644 --- a/drivers/net/designware.c +++ b/drivers/net/designware.c @@ -684,7 +684,7 @@ int designware_eth_probe(struct udevice *dev) break;

	err = clk_enable(&priv->clocks[i]);

• 	if (err) {
  

• 	if (err && err != -ENOSYS && err != -ENOTSUPP) {
  	pr_err("failed to enable clock %d\n", i);
  	clk_free(&priv->clocks[i]);
  	goto clk_err;
  
  

Reviewed-by: Patrice Chotard patrice.chotard@st.com

Thanks

On 6 Feb 2018, at 15:12, Eugeniy Paltsev Eugeniy.Paltsev@synopsys.com wrote:

After commit ba1f966725223 "net: designware: add clock support" we got NET broken on axs101 and axs103 platforms.

Some clock don't support gating so their clock drivers don't implement .enable/.disable callbacks. In such case clk_enable returns -ENOSYS. Also some clock drivers implement .enable/.disable callbacks not for all clock IDs and return -ENOSYS (or -ENOTSUPP) for others.

If we have such clock in 'clocks' list of designware ethernet controller node we fail to probe designware ethernet.

Fix it.

Signed-off-by: Eugeniy Paltsev Eugeniy.Paltsev@synopsys.com

Reviewed-by: Philipp Tomsich philipp.tomsich@theobroma-systems.com

See below for recommended changes.

---

drivers/net/designware.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/designware.c b/drivers/net/designware.c index 6d53071..43670a7 100644 --- a/drivers/net/designware.c +++ b/drivers/net/designware.c @@ -684,7 +684,7 @@ int designware_eth_probe(struct udevice *dev) break;

	err = clk_enable(&priv->clocks[i]);

• 	if (err) {
  

• 	if (err && err != -ENOSYS && err != -ENOTSUPP) {
  

I agree on the ENOSYS, but not the ENOTSUPP: if an enable function exists, it should explicitly handle the requested clock (even if it is just a no-op due to the driver knowing that it is always enabled).

Just as a side-note: most clock drivers return ENOENT, if they don’t know about a specific clock… whereas ENOTSUPP should be returned if a clock is known, but the specific operation on the clock is not supported (e.g. clk_disable on an always-on clock … or one that is critical to the system such as a fundamental PLL).

Thanks, Philipp.

		pr_err("failed to enable clock %d\n", i);
		clk_free(&priv->clocks[i]);
		goto clk_err;

-- 2.9.3

---

U-Boot mailing list U-Boot@lists.denx.de https://lists.denx.de/listinfo/u-boot