diff mbox series

[v3] Input: atmel_mxt_ts - Fix lost interrupts

Message ID 20201201123026.1416743-1-linus.walleij@linaro.org (mailing list archive)
State Accepted
Commit 8c3b55a299c325830a987de21dab6a89ecb71164
Headers show
Series [v3] Input: atmel_mxt_ts - Fix lost interrupts | expand

Commit Message

Linus Walleij Dec. 1, 2020, 12:30 p.m. UTC
After commit 74d905d2d38a devices requiring the workaround
for edge triggered interrupts stopped working.

The hardware needs the quirk to be used before even
proceeding to check if the quirk is needed because
mxt_acquire_irq() is called before mxt_check_retrigen()
is called and at this point pending IRQs need to be
checked, and if the workaround is not active, all
interrupts will be lost from this point.

Solve this by switching the calls around.

Cc: Andre Müller <andre.muller@web.de>
Cc: Nick Dyer <nick.dyer@itdev.co.uk>
Cc: Jiada Wang <jiada_wang@mentor.com>
Cc: stable@vger.kernel.org
Reported-by: Andre Müller <andre.muller@web.de>
Tested-by: Andre Müller <andre.muller@web.de>
Suggested-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Fixes: 74d905d2d38a ("Input: atmel_mxt_ts - only read messages in mxt_acquire_irq() when necessary")
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
ChangeLog v2->v3:
- Make the simpler fix suggested by Dmitry.
- Collect Andre's Tested-by.
ChangeLog v1->v2:
- Explicitly disable the retrig workaround also if the
  IRQ descriptor says we have a level triggered interrupt.
- Drop the second explicit assigning of "true" to the
  use_retrigen_workaround bool, it is already enabled.
- Augment debug text to say that we leave it enabled
  rather than enable it.
---
 drivers/input/touchscreen/atmel_mxt_ts.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Comments

Dmitry Torokhov Dec. 1, 2020, 5:56 p.m. UTC | #1
On Tue, Dec 01, 2020 at 01:30:26PM +0100, Linus Walleij wrote:
> After commit 74d905d2d38a devices requiring the workaround
> for edge triggered interrupts stopped working.
> 
> The hardware needs the quirk to be used before even
> proceeding to check if the quirk is needed because
> mxt_acquire_irq() is called before mxt_check_retrigen()
> is called and at this point pending IRQs need to be
> checked, and if the workaround is not active, all
> interrupts will be lost from this point.
> 
> Solve this by switching the calls around.
> 
> Cc: Andre Müller <andre.muller@web.de>
> Cc: Nick Dyer <nick.dyer@itdev.co.uk>
> Cc: Jiada Wang <jiada_wang@mentor.com>
> Cc: stable@vger.kernel.org
> Reported-by: Andre Müller <andre.muller@web.de>
> Tested-by: Andre Müller <andre.muller@web.de>
> Suggested-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> Fixes: 74d905d2d38a ("Input: atmel_mxt_ts - only read messages in mxt_acquire_irq() when necessary")
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>

Applied, thank you.
diff mbox series

Patch

diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c
index e34984388791..109655547477 100644
--- a/drivers/input/touchscreen/atmel_mxt_ts.c
+++ b/drivers/input/touchscreen/atmel_mxt_ts.c
@@ -2185,11 +2185,11 @@  static int mxt_initialize(struct mxt_data *data)
 		msleep(MXT_FW_RESET_TIME);
 	}
 
-	error = mxt_acquire_irq(data);
+	error = mxt_check_retrigen(data);
 	if (error)
 		return error;
 
-	error = mxt_check_retrigen(data);
+	error = mxt_acquire_irq(data);
 	if (error)
 		return error;