On 4/19/2025 1:30 AM, Bjorn Helgaas wrote:
On Sat, Apr 12, 2025 at 07:19:49AM +0530, Krishna Chaitanya Chundru wrote:
TC9563 is the PCIe switch which has one upstream and three downstream
ports. To one of the downstream ports ethernet MAC is connected as endpoint
device. Other two downstream ports are supposed to connect to external
device. One Host can connect to TC956x by upstream port.
I guess this topology is for one specific platform that includes the
TC9563? Since it's a PCIe switch, I assume it could also be used in
other platforms with other topologies?
This topology is for the switch not for per platform, the switch exposes
two DSP's for external device and one DSP has always integrated MAC.
TC9563 switch power is controlled by the GPIO's. After powering on
the switch will immediately participate in the link training. if the
host is also ready by that time PCIe link will established.
The TC9563 needs to configured certain parameters like de-emphasis,
disable unused port etc before link is established.
As the controller starts link training before the probe of pwrctl driver,
the PCIe link may come up as soon as we power on the switch. Due to this
configuring the switch itself through i2c will not have any effect as
this configuration needs to done before link training. To avoid this
introduce two functions in pci_ops to start_link() & stop_link() which
will disable the link training if the PCIe link is not up yet.
This series depends on the https://lore.kernel.org/all/20250124101038.3871768-3-krishna.chundru@xxxxxxxxxxxxxxxx/
How so?
https://lore.kernel.org/all/20250124101038.3871768-3-krishna.chundru@xxxxxxxxxxxxxxxx/
adds a schema "n-fts" property, but this series doesn't mention
"n-fts". This series *does* add this:
of_property_read_u8_array(node, "nfts", cfg->nfts, 2);
Is that supposed to be the same thing, or does "nfts" magically match
"n-fts"?
It is miss from my side, thanks for pointing out, I will
fix this in next patch series.
- Krishna Chaitanya.
Bjorn
