[PATCH net v2] octeontx2-af: Send Link events one by one

[PATCH net v2] octeontx2-af: Send Link events one by one

[Subbaraya Sundeep]

Send link events one after another otherwise new message
is overwriting the message which is being processed by PF.

Fixes: a88e0f936ba9 ("octeontx2: Detect the mbox up or down message via register")
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
---
v2:
 No changes. Added subject prefix net.

 drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
index 992fa0b..ebb56eb 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
@@ -272,6 +272,8 @@ static void cgx_notify_pfs(struct cgx_link_event *event, struct rvu *rvu)
 
 		otx2_mbox_msg_send_up(&rvu->afpf_wq_info.mbox_up, pfid);
 
+		otx2_mbox_wait_for_rsp(&rvu->afpf_wq_info.mbox_up, pfid);
+
 		mutex_unlock(&rvu->mbox_lock);
 	} while (pfmap);
 }
-- 
2.7.4

Re: [PATCH net v2] octeontx2-af: Send Link events one by one

[Michal Swiatkowski]

On Wed, May 14, 2025 at 11:58:28AM +0530, Subbaraya Sundeep wrote:
> Send link events one after another otherwise new message
> is overwriting the message which is being processed by PF.
> 
> Fixes: a88e0f936ba9 ("octeontx2: Detect the mbox up or down message via register")
> Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
> ---
> v2:
>  No changes. Added subject prefix net.
> 
>  drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
> index 992fa0b..ebb56eb 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
> @@ -272,6 +272,8 @@ static void cgx_notify_pfs(struct cgx_link_event *event, struct rvu *rvu)
>  
>  		otx2_mbox_msg_send_up(&rvu->afpf_wq_info.mbox_up, pfid);
>  
> +		otx2_mbox_wait_for_rsp(&rvu->afpf_wq_info.mbox_up, pfid);
> +
>  		mutex_unlock(&rvu->mbox_lock);

Fix looks fine.
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>

In rvu_rep_up_notify() the same send function is called (and
rvu_rep_up_notify() is called in do, while loop in
rvu_rep_wq_handler()). Doesn't it also need waiting for response?
Are there a message that don't need waiting? Maybe it will be best to
always wait for response if another call can overwrite.

Thanks

>  	} while (pfmap);
>  }
> -- 
> 2.7.4

Re: [PATCH net v2] octeontx2-af: Send Link events one by one

[Subbaraya Sundeep]

Hi Michal,

On 2025-05-14 at 08:11:32, Michal Swiatkowski (michal.swiatkowski@linux.intel.com) wrote:
> On Wed, May 14, 2025 at 11:58:28AM +0530, Subbaraya Sundeep wrote:
> > Send link events one after another otherwise new message
> > is overwriting the message which is being processed by PF.
> > 
> > Fixes: a88e0f936ba9 ("octeontx2: Detect the mbox up or down message via register")
> > Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
> > ---
> > v2:
> >  No changes. Added subject prefix net.
> > 
> >  drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c | 2 ++
> >  1 file changed, 2 insertions(+)
> > 
> > diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
> > index 992fa0b..ebb56eb 100644
> > --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
> > +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c
> > @@ -272,6 +272,8 @@ static void cgx_notify_pfs(struct cgx_link_event *event, struct rvu *rvu)
> >  
> >  		otx2_mbox_msg_send_up(&rvu->afpf_wq_info.mbox_up, pfid);
> >  
> > +		otx2_mbox_wait_for_rsp(&rvu->afpf_wq_info.mbox_up, pfid);
> > +
> >  		mutex_unlock(&rvu->mbox_lock);
> 
> Fix looks fine.
> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
> 
> In rvu_rep_up_notify() the same send function is called (and
> rvu_rep_up_notify() is called in do, while loop in
> rvu_rep_wq_handler()). Doesn't it also need waiting for response?
> Are there a message that don't need waiting? Maybe it will be best to
> always wait for response if another call can overwrite.
> 
> Thanks
> 
Okay I will modify rvu_rep_up_notify, test and send next version along
with this.

Thanks,
Sundeep

> >  	} while (pfmap);
> >  }
> > -- 
> > 2.7.4

Re: [PATCH net v2] octeontx2-af: Send Link events one by one

[Jakub Kicinski]

On Wed, 14 May 2025 11:58:28 +0530 Subbaraya Sundeep wrote:
> Send link events one after another otherwise new message
> is overwriting the message which is being processed by PF.

Please respond to reviewers in a timely fashion.

Re: [PATCH net v2] octeontx2-af: Send Link events one by one

[Subbaraya Sundeep]

Hi Jakub,

On 2025-05-15 at 14:12:39, Jakub Kicinski (kuba@kernel.org) wrote:
> On Wed, 14 May 2025 11:58:28 +0530 Subbaraya Sundeep wrote:
> > Send link events one after another otherwise new message
> > is overwriting the message which is being processed by PF.
> 
> Please respond to reviewers in a timely fashion.

Just want to know is it you or bot? What is the time limit here?
Have to respond within 24hr?

Thanks,
Sundeep

Re: [PATCH net v2] octeontx2-af: Send Link events one by one

[Jakub Kicinski]

On Thu, 15 May 2025 16:41:18 +0000 Subbaraya Sundeep wrote:
> On 2025-05-15 at 14:12:39, Jakub Kicinski (kuba@kernel.org) wrote:
> > On Wed, 14 May 2025 11:58:28 +0530 Subbaraya Sundeep wrote:  
> > > Send link events one after another otherwise new message
> > > is overwriting the message which is being processed by PF.  
> > 
> > Please respond to reviewers in a timely fashion.  
> 
> Just want to know is it you or bot?

Have you seen me reply with the same text to someone else?
Oh, that's right, you don't read the list, how would you know..

> What is the time limit here?
> Have to respond within 24hr?

The review cadence on netdev is 24h.

Re: [PATCH net v2] octeontx2-af: Send Link events one by one

[Subbaraya Sundeep]

On 2025-05-15 at 17:19:37, Jakub Kicinski (kuba@kernel.org) wrote:
> On Thu, 15 May 2025 16:41:18 +0000 Subbaraya Sundeep wrote:
> > On 2025-05-15 at 14:12:39, Jakub Kicinski (kuba@kernel.org) wrote:
> > > On Wed, 14 May 2025 11:58:28 +0530 Subbaraya Sundeep wrote:  
> > > > Send link events one after another otherwise new message
> > > > is overwriting the message which is being processed by PF.  
> > > 
> > > Please respond to reviewers in a timely fashion.  
> > 
> > Just want to know is it you or bot?
> 
> Have you seen me reply with the same text to someone else?
> Oh, that's right, you don't read the list, how would you know..
> 
:)
> > What is the time limit here?
> > Have to respond within 24hr?
> 
> The review cadence on netdev is 24h.
Got it this patch missed your recent pull request because of my
delay in response.

Thanks,
Sundeep