diff mbox series

i2c: xiic: Don't try to handle more interrupt events after error

Message ID 20230606182558.1301413-1-robert.hancock@calian.com (mailing list archive)
State New, archived
Headers show
Series i2c: xiic: Don't try to handle more interrupt events after error | expand

Commit Message

Robert Hancock June 6, 2023, 6:25 p.m. UTC
In xiic_process, it is possible that error events such as arbitration
lost or TX error can be raised in conjunction with other interrupt flags
such as TX FIFO empty or bus not busy. Error events result in the
controller being reset and the error returned to the calling request,
but the function could potentially try to keep handling the other
events, such as by writing more messages into the TX FIFO. Since the
transaction has already failed, this is not helpful and will just cause
issues.

This problem has been present ever since:

commit 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")

which allowed non-error events to be handled after errors, but became
more obvious after:

commit 743e227a8959 ("i2c: xiic: Defer xiic_wakeup() and
__xiic_start_xfer() in xiic_process()")

which reworked the code to add a WARN_ON which triggers if both the
xfer_more and wakeup_req flags were set, since this combination is
not supposed to happen, but was occurring in this scenario.

Skip further interrupt handling after error flags are detected to avoid
this problem.

Fixes: 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
Signed-off-by: Robert Hancock <robert.hancock@calian.com>
---
 drivers/i2c/busses/i2c-xiic.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Andi Shyti June 6, 2023, 7:24 p.m. UTC | #1
Hi Robert,

On Tue, Jun 06, 2023 at 12:25:58PM -0600, Robert Hancock wrote:
> In xiic_process, it is possible that error events such as arbitration
> lost or TX error can be raised in conjunction with other interrupt flags
> such as TX FIFO empty or bus not busy. Error events result in the
> controller being reset and the error returned to the calling request,
> but the function could potentially try to keep handling the other
> events, such as by writing more messages into the TX FIFO. Since the
> transaction has already failed, this is not helpful and will just cause
> issues.

what kind of issues?

> This problem has been present ever since:
> 
> commit 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
> 
> which allowed non-error events to be handled after errors, but became
> more obvious after:
> 
> commit 743e227a8959 ("i2c: xiic: Defer xiic_wakeup() and
> __xiic_start_xfer() in xiic_process()")
> 
> which reworked the code to add a WARN_ON which triggers if both the
> xfer_more and wakeup_req flags were set, since this combination is
> not supposed to happen, but was occurring in this scenario.
> 
> Skip further interrupt handling after error flags are detected to avoid
> this problem.
> 
> Fixes: 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
> Signed-off-by: Robert Hancock <robert.hancock@calian.com>

please add:

Cc: <stable@vger.kernel.org> # v4.3+

> ---
>  drivers/i2c/busses/i2c-xiic.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/i2c/busses/i2c-xiic.c b/drivers/i2c/busses/i2c-xiic.c
> index 8a3d9817cb41..ee6edc963dea 100644
> --- a/drivers/i2c/busses/i2c-xiic.c
> +++ b/drivers/i2c/busses/i2c-xiic.c
> @@ -721,6 +721,8 @@ static irqreturn_t xiic_process(int irq, void *dev_id)
>  			wakeup_req = 1;
>  			wakeup_code = STATE_ERROR;
>  		}
> +		/* don't try to handle other events */
> +		goto out;

why don't we have goto's after every irq evaluation but only
here? Do the issues you mentioned happen olny in this particular
error case?

Thanks,
Andi

>  	}
>  	if (pend & XIIC_INTR_RX_FULL_MASK) {
>  		/* Receive register/FIFO is full */
> -- 
> 2.40.1
>
Robert Hancock June 6, 2023, 7:40 p.m. UTC | #2
On Tue, 2023-06-06 at 21:24 +0200, Andi Shyti wrote:
> Hi Robert,
> 
> On Tue, Jun 06, 2023 at 12:25:58PM -0600, Robert Hancock wrote:
> > In xiic_process, it is possible that error events such as
> > arbitration
> > lost or TX error can be raised in conjunction with other interrupt
> > flags
> > such as TX FIFO empty or bus not busy. Error events result in the
> > controller being reset and the error returned to the calling
> > request,
> > but the function could potentially try to keep handling the other
> > events, such as by writing more messages into the TX FIFO. Since
> > the
> > transaction has already failed, this is not helpful and will just
> > cause
> > issues.
> 
> what kind of issues?
> 

The one I ran into is what I alluded to further down, where that
WARN_ON was triggering repeatedly, which ended up flooding the kernel
log and causing the device's watchdog timer to timeout. I'm not sure
what other forms of havoc might ensue, other than "nothing good"..

> > This problem has been present ever since:
> > 
> > commit 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
> > 
> > which allowed non-error events to be handled after errors, but
> > became
> > more obvious after:
> > 
> > commit 743e227a8959 ("i2c: xiic: Defer xiic_wakeup() and
> > __xiic_start_xfer() in xiic_process()")
> > 
> > which reworked the code to add a WARN_ON which triggers if both the
> > xfer_more and wakeup_req flags were set, since this combination is
> > not supposed to happen, but was occurring in this scenario.
> > 
> > Skip further interrupt handling after error flags are detected to
> > avoid
> > this problem.
> > 
> > Fixes: 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
> > Signed-off-by: Robert Hancock <robert.hancock@calian.com>
> 
> please add:
> 
> Cc: <stable@vger.kernel.org> # v4.3+
> 

Can add for a v2 - although with the Fixes tag it would most likely be
picked up for stable anyway..

> > ---
> >  drivers/i2c/busses/i2c-xiic.c | 2 ++
> >  1 file changed, 2 insertions(+)
> > 
> > diff --git a/drivers/i2c/busses/i2c-xiic.c
> > b/drivers/i2c/busses/i2c-xiic.c
> > index 8a3d9817cb41..ee6edc963dea 100644
> > --- a/drivers/i2c/busses/i2c-xiic.c
> > +++ b/drivers/i2c/busses/i2c-xiic.c
> > @@ -721,6 +721,8 @@ static irqreturn_t xiic_process(int irq, void
> > *dev_id)
> >                       wakeup_req = 1;
> >                       wakeup_code = STATE_ERROR;
> >               }
> > +             /* don't try to handle other events */
> > +             goto out;
> 
> why don't we have goto's after every irq evaluation but only
> here? Do the issues you mentioned happen olny in this particular
> error case?
> 

As far as I can tell, yes. For example you could legitimately have
XIIC_INTR_TX_EMPTY_MASK and XIIC_INTR_BNB_MASK both being handled. The
error case is special as we end up resetting the controller, so it
doesn't make sense to try to continue with the rest of the transaction
while also completing it with an error.

> Thanks,
> Andi
> 
> >       }
> >       if (pend & XIIC_INTR_RX_FULL_MASK) {
> >               /* Receive register/FIFO is full */
> > --
> > 2.40.1
> >
Andi Shyti June 6, 2023, 9:20 p.m. UTC | #3
Hi Robert,

On Tue, Jun 06, 2023 at 07:40:15PM +0000, Robert Hancock wrote:
> On Tue, 2023-06-06 at 21:24 +0200, Andi Shyti wrote:
> > Hi Robert,
> > 
> > On Tue, Jun 06, 2023 at 12:25:58PM -0600, Robert Hancock wrote:
> > > In xiic_process, it is possible that error events such as
> > > arbitration
> > > lost or TX error can be raised in conjunction with other interrupt
> > > flags
> > > such as TX FIFO empty or bus not busy. Error events result in the
> > > controller being reset and the error returned to the calling
> > > request,
> > > but the function could potentially try to keep handling the other
> > > events, such as by writing more messages into the TX FIFO. Since
> > > the
> > > transaction has already failed, this is not helpful and will just
> > > cause
> > > issues.
> > 
> > what kind of issues?
> > 
> 
> The one I ran into is what I alluded to further down, where that
> WARN_ON was triggering repeatedly, which ended up flooding the kernel
> log and causing the device's watchdog timer to timeout. I'm not sure
> what other forms of havoc might ensue, other than "nothing good"..
> 
> > > This problem has been present ever since:
> > > 
> > > commit 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
> > > 
> > > which allowed non-error events to be handled after errors, but
> > > became
> > > more obvious after:
> > > 
> > > commit 743e227a8959 ("i2c: xiic: Defer xiic_wakeup() and
> > > __xiic_start_xfer() in xiic_process()")
> > > 
> > > which reworked the code to add a WARN_ON which triggers if both the
> > > xfer_more and wakeup_req flags were set, since this combination is
> > > not supposed to happen, but was occurring in this scenario.
> > > 
> > > Skip further interrupt handling after error flags are detected to
> > > avoid
> > > this problem.
> > > 
> > > Fixes: 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
> > > Signed-off-by: Robert Hancock <robert.hancock@calian.com>
> > 
> > please add:
> > 
> > Cc: <stable@vger.kernel.org> # v4.3+
> > 
> 
> Can add for a v2 - although with the Fixes tag it would most likely be
> picked up for stable anyway..

It's just a courtesy to stable maintainers :)

> > > ---
> > >  drivers/i2c/busses/i2c-xiic.c | 2 ++
> > >  1 file changed, 2 insertions(+)
> > > 
> > > diff --git a/drivers/i2c/busses/i2c-xiic.c
> > > b/drivers/i2c/busses/i2c-xiic.c
> > > index 8a3d9817cb41..ee6edc963dea 100644
> > > --- a/drivers/i2c/busses/i2c-xiic.c
> > > +++ b/drivers/i2c/busses/i2c-xiic.c
> > > @@ -721,6 +721,8 @@ static irqreturn_t xiic_process(int irq, void
> > > *dev_id)
> > >                       wakeup_req = 1;
> > >                       wakeup_code = STATE_ERROR;
> > >               }
> > > +             /* don't try to handle other events */
> > > +             goto out;
> > 
> > why don't we have goto's after every irq evaluation but only
> > here? Do the issues you mentioned happen olny in this particular
> > error case?
> > 
> 
> As far as I can tell, yes. For example you could legitimately have
> XIIC_INTR_TX_EMPTY_MASK and XIIC_INTR_BNB_MASK both being handled. The
> error case is special as we end up resetting the controller, so it
> doesn't make sense to try to continue with the rest of the transaction
> while also completing it with an error.

I think the patch is correct and I will ack it:

Acked-by: Andi Shyti <andi.shyti@kernel.org> 

I think, though, that this needs a proper fix and testing, in
order to cover all the possible combinations. The scenario you
highlighted is indeed one, but not only, potential situation that
could arise.

Can I just ask you to write a bit more in the comment to 
highlight the possible failure?

Thanks a lot,
Andi

> > Thanks,
> > Andi
> > 
> > >       }
> > >       if (pend & XIIC_INTR_RX_FULL_MASK) {
> > >               /* Receive register/FIFO is full */
> > > --
> > > 2.40.1
> > > 
> 
> -- 
> Robert Hancock <robert.hancock@calian.com>
Wolfram Sang June 9, 2023, 3:32 p.m. UTC | #4
> I think the patch is correct and I will ack it:
> 
> Acked-by: Andi Shyti <andi.shyti@kernel.org> 
> 
> I think, though, that this needs a proper fix and testing, in
> order to cover all the possible combinations. The scenario you
> highlighted is indeed one, but not only, potential situation that
> could arise.
> 
> Can I just ask you to write a bit more in the comment to 
> highlight the possible failure?

I tend to apply it to for-current because it improves the situation.
Further improvements could be made incrementally? D'accord everyone?
Andi Shyti June 9, 2023, 9:25 p.m. UTC | #5
Hi Wolfram,

On Fri, Jun 09, 2023 at 05:32:42PM +0200, wsa@kernel.org wrote:
> 
> > I think the patch is correct and I will ack it:
> > 
> > Acked-by: Andi Shyti <andi.shyti@kernel.org> 
> > 
> > I think, though, that this needs a proper fix and testing, in
> > order to cover all the possible combinations. The scenario you
> > highlighted is indeed one, but not only, potential situation that
> > could arise.
> > 
> > Can I just ask you to write a bit more in the comment to 
> > highlight the possible failure?
> 
> I tend to apply it to for-current because it improves the situation.
> Further improvements could be made incrementally? D'accord everyone?
> 

OK with that!

Thanks,
Andi
Robert Hancock June 27, 2023, 4:13 p.m. UTC | #6
On Fri, 2023-06-09 at 23:25 +0200, Andi Shyti wrote:
> Hi Wolfram,
> 
> On Fri, Jun 09, 2023 at 05:32:42PM +0200, wsa@kernel.org wrote:
> > 
> > > I think the patch is correct and I will ack it:
> > > 
> > > Acked-by: Andi Shyti <andi.shyti@kernel.org>
> > > 
> > > I think, though, that this needs a proper fix and testing, in
> > > order to cover all the possible combinations. The scenario you
> > > highlighted is indeed one, but not only, potential situation that
> > > could arise.
> > > 
> > > Can I just ask you to write a bit more in the comment to
> > > highlight the possible failure?
> > 
> > I tend to apply it to for-current because it improves the
> > situation.
> > Further improvements could be made incrementally? D'accord
> > everyone?
> > 
> 
> OK with that!
> 
> Thanks,
> Andi

Just checking on this patch, was it merged? It shows accepted in
Patchwork, but I'm not seeing it in the Git tree.
Wolfram Sang June 27, 2023, 9:25 p.m. UTC | #7
> Just checking on this patch, was it merged? It shows accepted in
> Patchwork, but I'm not seeing it in the Git tree.

Sorry, it seems I forgot to apply it? I will make sure it will go in
this merge window now.
Wolfram Sang July 6, 2023, 7:33 p.m. UTC | #8
On Tue, Jun 06, 2023 at 12:25:58PM -0600, Robert Hancock wrote:
> In xiic_process, it is possible that error events such as arbitration
> lost or TX error can be raised in conjunction with other interrupt flags
> such as TX FIFO empty or bus not busy. Error events result in the
> controller being reset and the error returned to the calling request,
> but the function could potentially try to keep handling the other
> events, such as by writing more messages into the TX FIFO. Since the
> transaction has already failed, this is not helpful and will just cause
> issues.
> 
> This problem has been present ever since:
> 
> commit 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
> 
> which allowed non-error events to be handled after errors, but became
> more obvious after:
> 
> commit 743e227a8959 ("i2c: xiic: Defer xiic_wakeup() and
> __xiic_start_xfer() in xiic_process()")
> 
> which reworked the code to add a WARN_ON which triggers if both the
> xfer_more and wakeup_req flags were set, since this combination is
> not supposed to happen, but was occurring in this scenario.
> 
> Skip further interrupt handling after error flags are detected to avoid
> this problem.
> 
> Fixes: 7f9906bd7f72 ("i2c: xiic: Service all interrupts in isr")
> Signed-off-by: Robert Hancock <robert.hancock@calian.com>

Applied to for-current, thanks!
diff mbox series

Patch

diff --git a/drivers/i2c/busses/i2c-xiic.c b/drivers/i2c/busses/i2c-xiic.c
index 8a3d9817cb41..ee6edc963dea 100644
--- a/drivers/i2c/busses/i2c-xiic.c
+++ b/drivers/i2c/busses/i2c-xiic.c
@@ -721,6 +721,8 @@  static irqreturn_t xiic_process(int irq, void *dev_id)
 			wakeup_req = 1;
 			wakeup_code = STATE_ERROR;
 		}
+		/* don't try to handle other events */
+		goto out;
 	}
 	if (pend & XIIC_INTR_RX_FULL_MASK) {
 		/* Receive register/FIFO is full */