Re: [PATCH -next] PCI: altera: Replace TLP_REQ_ID() with macro PCI_DEVID()

[Bjorn Helgaas <helgaas@kernel.org>]

On Wed, Aug 28, 2024 at 06:42:02PM +0800, Jinjie Ruan wrote:
> The TLP_REQ_ID's function is same as current PCI_DEVID()
> macro, replace it.
> 
> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
> ---
>  drivers/pci/controller/pcie-altera.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/drivers/pci/controller/pcie-altera.c b/drivers/pci/controller/pcie-altera.c
> index ef73baefaeb9..650b2dd81c48 100644
> --- a/drivers/pci/controller/pcie-altera.c
> +++ b/drivers/pci/controller/pcie-altera.c
> @@ -55,12 +55,11 @@
>  #define TLP_READ_TAG			0x1d
>  #define TLP_WRITE_TAG			0x10
>  #define RP_DEVFN			0
> -#define TLP_REQ_ID(bus, devfn)		(((bus) << 8) | (devfn))
>  #define TLP_CFG_DW0(pcie, cfg)		\
>  		(((cfg) << 24) |	\
>  		  TLP_PAYLOAD_SIZE)
>  #define TLP_CFG_DW1(pcie, tag, be)	\
> -	(((TLP_REQ_ID(pcie->root_bus_nr,  RP_DEVFN)) << 16) | (tag << 8) | (be))
> +	(((PCI_DEVID(pcie->root_bus_nr,  RP_DEVFN)) << 16) | (tag << 8) | (be))

This doesn't cause a problem per se, but PCI_DEVID() is defined
basically as a PCI core convenience without any particular connection
to the PCIe Requester ID format, so I don't think this is an
improvement.

I think it *would* be an improvement to move TLP_REQ_ID() to
drivers/pci/pci.h, where other drivers could share it.  

It's obvious how the Requester ID must be laid out, but I took a quick
look through the PCIe spec and was surprised that I couldn't find a
clear definition of it, unlike the *Completer ID*, which is specified
in PCIe r6.0, sec 2.2.9.1.

Bjorn