On 2019-08-15 3:28 p.m., Bart Van Assche wrote:
> Commit 52916982af48 ("PCI/P2PDMA: Support peer-to-peer memory"; v4.20)
> introduced the following text: "there's no way to determine whether the
> root complex supports forwarding between them." A later commit added a
> whitelist check in the function that comment applies to. Update the
> comment to reflect the addition of the whitelist check.
Thanks for the vigilant patch, but I've already got a series[1] that
cleans up most of these commits. It looks like this patch will conflict
with that series.
Logan
[1]
https://lore.kernel.org/linux-pci/20190812173048.9186-1-logang@xxxxxxxxxxxx/
> Cc: Logan Gunthorpe <logang@xxxxxxxxxxxx>
> Cc: Christoph Hellwig <hch@xxxxxx>
> Cc: Keith Busch <keith.busch@xxxxxxxxx>
> Signed-off-by: Bart Van Assche <bvanassche@xxxxxxx>
> ---
> drivers/pci/p2pdma.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
> index 234476226529..f719adc2b826 100644
> --- a/drivers/pci/p2pdma.c
> +++ b/drivers/pci/p2pdma.c
> @@ -300,8 +300,8 @@ static bool root_complex_whitelist(struct pci_dev *dev)
> * Any two devices that don't have a common upstream bridge will return -1.
> * In this way devices on separate PCIe root ports will be rejected, which
> * is what we want for peer-to-peer seeing each PCIe root port defines a
> - * separate hierarchy domain and there's no way to determine whether the root
> - * complex supports forwarding between them.
> + * separate hierarchy domain and there's no way other than using a whitelist
> + * to determine whether the root complex supports forwarding between them.
> *
> * In the case where two devices are connected to different PCIe switches,
> * this function will still return a positive distance as long as both
>
