2019-01-07 03:21:26

by Joey Zhang

[permalink] [raw]
Subject: [PATCH] NTB: ntb_transport: Free MWs in ntb_transport_link_cleanup()

If NTB peer host crashes or reboots, the NTB transport link will be
down and the MWs of NTB transport will be invalid. But the
ntb_transport_link_cleanup() does not free these invalid MWs. When
the NTB peer host is recovered later, NTB transport link will be
up and the ntb_set_mw() will not reset up MWs. Because the MWs of
NTB transport are invalid, the NTB transport will not work.

We can fix it by freeing MWs when NTB transport link is down, then
the ntb_set_mw() will reset up MWs when NTB transport link is up.

Signed-off-by: Joey Zhang <[email protected]>
---
drivers/ntb/ntb_transport.c | 3 +++
1 file changed, 3 insertions(+)

diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c
index 3bfdb45..6e8902d 100644
--- a/drivers/ntb/ntb_transport.c
+++ b/drivers/ntb/ntb_transport.c
@@ -862,6 +862,9 @@ static void ntb_transport_link_cleanup(struct ntb_transport_ctx *nt)
if (!nt->link_is_up)
cancel_delayed_work_sync(&nt->link_work);

+ for (i = 0; i < nt->mw_count; i++)
+ ntb_free_mw(nt, i);
+
/* The scratchpad registers keep the values if the remote side
* goes down, blast them now to give them a sane value the next
* time they are accessed
--
2.7.4



2019-01-08 01:17:47

by Logan Gunthorpe

[permalink] [raw]
Subject: Re: [PATCH] NTB: ntb_transport: Free MWs in ntb_transport_link_cleanup()



On 06/01/19 08:12 PM, Joey Zhang wrote:
> If NTB peer host crashes or reboots, the NTB transport link will be
> down and the MWs of NTB transport will be invalid. But the
> ntb_transport_link_cleanup() does not free these invalid MWs. When
> the NTB peer host is recovered later, NTB transport link will be
> up and the ntb_set_mw() will not reset up MWs. Because the MWs of
> NTB transport are invalid, the NTB transport will not work.
>
> We can fix it by freeing MWs when NTB transport link is down, then
> the ntb_set_mw() will reset up MWs when NTB transport link is up.
>
> Signed-off-by: Joey Zhang <[email protected]>

Looks ok to me.

Reviewed By: Logan Gunthorpe <[email protected]>

> ---
> drivers/ntb/ntb_transport.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c
> index 3bfdb45..6e8902d 100644
> --- a/drivers/ntb/ntb_transport.c
> +++ b/drivers/ntb/ntb_transport.c
> @@ -862,6 +862,9 @@ static void ntb_transport_link_cleanup(struct ntb_transport_ctx *nt)
> if (!nt->link_is_up)
> cancel_delayed_work_sync(&nt->link_work);
>
> + for (i = 0; i < nt->mw_count; i++)
> + ntb_free_mw(nt, i);
> +
> /* The scratchpad registers keep the values if the remote side
> * goes down, blast them now to give them a sane value the next
> * time they are accessed
>

2019-02-11 15:56:42

by Jon Mason

[permalink] [raw]
Subject: Re: [PATCH] NTB: ntb_transport: Free MWs in ntb_transport_link_cleanup()

On Mon, Jan 07, 2019 at 06:16:02PM -0700, Logan Gunthorpe wrote:
>
>
> On 06/01/19 08:12 PM, Joey Zhang wrote:
> > If NTB peer host crashes or reboots, the NTB transport link will be
> > down and the MWs of NTB transport will be invalid. But the
> > ntb_transport_link_cleanup() does not free these invalid MWs. When
> > the NTB peer host is recovered later, NTB transport link will be
> > up and the ntb_set_mw() will not reset up MWs. Because the MWs of
> > NTB transport are invalid, the NTB transport will not work.
> >
> > We can fix it by freeing MWs when NTB transport link is down, then
> > the ntb_set_mw() will reset up MWs when NTB transport link is up.
> >
> > Signed-off-by: Joey Zhang <[email protected]>
>
> Looks ok to me.

Added to the ntb branch, thanks!

> Reviewed By: Logan Gunthorpe <[email protected]>

Looks like your missed the '-' ;-)
I took the liberity of adding it

Thanks,
Jon

>
> > ---
> > drivers/ntb/ntb_transport.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c
> > index 3bfdb45..6e8902d 100644
> > --- a/drivers/ntb/ntb_transport.c
> > +++ b/drivers/ntb/ntb_transport.c
> > @@ -862,6 +862,9 @@ static void ntb_transport_link_cleanup(struct ntb_transport_ctx *nt)
> > if (!nt->link_is_up)
> > cancel_delayed_work_sync(&nt->link_work);
> >
> > + for (i = 0; i < nt->mw_count; i++)
> > + ntb_free_mw(nt, i);
> > +
> > /* The scratchpad registers keep the values if the remote side
> > * goes down, blast them now to give them a sane value the next
> > * time they are accessed
> >