Re: [PATCH v3] PCI: endpoint: pci-epf-vntb: Track link state from both sides

[Manivannan Sadhasivam]

On Wed, Oct 14, 2026 at 04:14:48PM +0900, Koichiro Den wrote:
> The control-region link status is currently updated only by
> COMMAND_LINK_UP and COMMAND_LINK_DOWN from the HOST. The virtual
> NTB link callbacks are empty. Consequently, the HOST can see the link as
> up before the VHOST has enabled it, and ntb_link_disable() on the VHOST
> leaves LINK_STATUS_UP set without notifying either NTB client.
> 
> ntb_netdev can hide this device-level state bug because ntb_transport uses
> a separate per-QP LINK_DOWN_FLAG message. Use ntb_tool to observe the
> device link state directly:
> 
> After bringing both sides up, start one waiter on each side:
> 
>   HOST# echo N > /sys/kernel/debug/ntb_tool/<H-device>/peer0/link_event
>   VHOST# echo N > /sys/kernel/debug/ntb_tool/<V-device>/peer0/link_event
> 
> While both waiters are blocked, disable the link on the VHOST:
> 
>   VHOST# echo N > /sys/kernel/debug/ntb_tool/<V-device>/link
> 
> Without this patch, the link-disable command succeeds, but peer0/link
> remains Y on both sides and neither waiter returns.
> 
> Track HOST and VHOST enablement separately and report the effective link
> as up only when both sides have enabled it, as pci-epf-ntb does for its
> two physical hosts. Both NTB clients observe this effective state, so
> notify both sides only when it changes, including the side whose request
> completes the transition. Serialize the two paths because HOST commands
> run from delayed work while the VHOST callbacks may run concurrently.
> 
> With this patch, peer0/link changes to N on both sides and both waiters
> return.
> 
> Fixes: e35f56bb0330 ("PCI: endpoint: Support NTB transfer between RC and EP")
> Signed-off-by: Koichiro Den <den@xxxxxxxxxxxxx>
> ---
> Changes in v3:
>   - Rebase onto the latest pci/endpoint.
>   - Drop redundant dma_wmb(). (Frank)
>   - Retest using ntb_tool.
>   - Link to v2: https://lore.kernel.org/r/20260904162049.3432464-1-den@xxxxxxxxxxxxx/
> 
>  drivers/pci/endpoint/functions/pci-epf-vntb.c | 90 +++++++++++++++----
>  1 file changed, 73 insertions(+), 17 deletions(-)
> 
> diff --git a/drivers/pci/endpoint/functions/pci-epf-vntb.c b/drivers/pci/endpoint/functions/pci-epf-vntb.c
> index 07ba338732d4..a11f9260ee86 100644
> --- a/drivers/pci/endpoint/functions/pci-epf-vntb.c
> +++ b/drivers/pci/endpoint/functions/pci-epf-vntb.c
> @@ -41,6 +41,7 @@
>  #include <linux/delay.h>
>  #include <linux/io.h>
>  #include <linux/module.h>
> +#include <linux/mutex.h>
>  #include <linux/slab.h>
>  
>  #include <linux/pci-ep-msi.h>
> @@ -147,7 +148,10 @@ struct epf_ntb {
>  	u16 vntb_pid;
>  	u16 vntb_vid;
>  
> -	bool linkup;
> +	/* Serialize HOST and VHOST link state changes. */
> +	struct mutex link_lock;
> +	bool host_linkup;
> +	bool vhost_linkup;
>  
>  	/*
>  	 * True when doorbells are interrupt-driven (MSI or embedded), false
> @@ -183,24 +187,54 @@ static struct pci_epf_header epf_ntb_header = {
>  	.interrupt_pin	= PCI_INTERRUPT_INTA,
>  };
>  
> +static void epf_ntb_update_link(struct epf_ntb *ntb)
> +{
> +	u16 link_status = READ_ONCE(ntb->reg->link_status);
> +
> +	/* The link is usable only after both sides have enabled it. */

I see a bit of ambiguity here and it is confusing. This comment says that both
host and vhost has to enable the link first. Yet, the variables are named as
'{host/vhost}_linkup'. Also, the functions setting these variables are named
as:

	epf_ntb_link_up()
	vntb_epf_set_link() 

Even though these two are doing the same thing but on different side, these two
functions are not symmetric. This needs to be fixed up.

- Mani

-- 
மணிவண்ணன் சதாசிவம்