Re: [PATCH v4 02/11] PCI/bwctrl: Add support to scale bandwidth before & after link re-training

[Krishna Chaitanya Chundru <krishna.chundru@oss.qualcomm.com>]

On 7/12/2025 3:06 AM, Bjorn Helgaas wrote:
> On Mon, Jun 09, 2025 at 04:21:23PM +0530, Krishna Chaitanya Chundru wrote:
>> If the driver wants to move to higher data rate/speed than the current data
>> rate then the controller driver may need to change certain votes so that
>> link may come up at requested data rate/speed like QCOM PCIe controllers
>> need to change their RPMh (Resource Power Manager-hardened) state. Once
>> link retraining is done controller drivers needs to adjust their votes
>> based on the final data rate.
>>
>> Some controllers also may need to update their bandwidth voting like
>> ICC BW votings etc.
>>
>> So, add pre_link_speed_change() & post_link_speed_change() op to call
>> before & after the link re-train. There is no explicit locking mechanisms
>> as these are called by a single client Endpoint driver.
>>
>> In case of PCIe switch, if there is a request to change target speed for a
>> downstream port then no need to call these function ops as these are
>> outside the scope of the controller drivers.
> 
>> +++ b/include/linux/pci.h
>> @@ -599,6 +599,24 @@ struct pci_host_bridge {
>>   	void (*release_fn)(struct pci_host_bridge *);
>>   	int (*enable_device)(struct pci_host_bridge *bridge, struct pci_dev *dev);
>>   	void (*disable_device)(struct pci_host_bridge *bridge, struct pci_dev *dev);
>> +	/*
>> +	 * Callback to the host bridge drivers to update ICC BW votes, clock
>> +	 * frequencies etc.. for the link re-train to come up in targeted speed.
>> +	 * These are intended to be called by devices directly attached to the
>> +	 * Root Port. These are called by a single client Endpoint driver, so
>> +	 * there is no need for explicit locking mechanisms.
>> +	 */
>> +	int (*pre_link_speed_change)(struct pci_host_bridge *bridge,
>> +				     struct pci_dev *dev, int speed);
>> +	/*
>> +	 * Callback to the host bridge drivers to adjust ICC BW votes, clock
>> +	 * frequencies etc.. to the updated speed after link re-train. These
>> +	 * are intended to be called by devices directly attached to the
>> +	 * Root Port. These are called by a single client Endpoint driver,
>> +	 * so there is no need for explicit locking mechanisms.
> 
> No need to repeat the entire comment.  s/.././
> 
> These pointers feel awfully specific for being in struct
> pci_host_bridge, since we only need them for a questionable QCOM
> controller.  I think this needs to be pushed down into qcom somehow as
> some kind of quirk.
> 
Currently these are needed by QCOM controllers, but it may also needed
by other controllers may also need these for updating ICC votes, any
system level votes, clock frequencies etc.
QCOM controllers is also doing one extra step in these functions to
disable and enable ASPM only as it cannot link speed change support
with ASPM enabled.

- Krishna Chaitanya.
>> +	 */
>> +	void (*post_link_speed_change)(struct pci_host_bridge *bridge,
>> +				       struct pci_dev *dev, int speed);