Subject: [PATCH 1/3] Input: applespi: Don't wait for responses to commands indefinitely.

The response to a command may never arrive or it may be corrupted (and
hence dropped) for some reason. While exceedingly rare, when it did
happen it blocked all further commands. One way to fix this was to
do a suspend/resume. However, recovering automatically seems like a
nicer option. Hence this puts a time limit (1 sec) on how long we're
willing to wait for a response, after which we assume it got lost.

Signed-off-by: Ronald Tschalär <[email protected]>
---
drivers/input/keyboard/applespi.c | 21 +++++++++++++++------
1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/input/keyboard/applespi.c b/drivers/input/keyboard/applespi.c
index d22223154177f..8494bf610fd70 100644
--- a/drivers/input/keyboard/applespi.c
+++ b/drivers/input/keyboard/applespi.c
@@ -48,6 +48,7 @@
#include <linux/efi.h>
#include <linux/input.h>
#include <linux/input/mt.h>
+#include <linux/ktime.h>
#include <linux/leds.h>
#include <linux/module.h>
#include <linux/spinlock.h>
@@ -409,7 +410,7 @@ struct applespi_data {
unsigned int cmd_msg_cntr;
/* lock to protect the above parameters and flags below */
spinlock_t cmd_msg_lock;
- bool cmd_msg_queued;
+ ktime_t cmd_msg_queued;
enum applespi_evt_type cmd_evt_type;

struct led_classdev backlight_info;
@@ -729,7 +730,7 @@ static void applespi_msg_complete(struct applespi_data *applespi,
wake_up_all(&applespi->drain_complete);

if (is_write_msg) {
- applespi->cmd_msg_queued = false;
+ applespi->cmd_msg_queued = 0;
applespi_send_cmd_msg(applespi);
}

@@ -771,8 +772,16 @@ static int applespi_send_cmd_msg(struct applespi_data *applespi)
return 0;

/* check whether send is in progress */
- if (applespi->cmd_msg_queued)
- return 0;
+ if (applespi->cmd_msg_queued) {
+ if (ktime_ms_delta(ktime_get(), applespi->cmd_msg_queued) < 1000)
+ return 0;
+
+ dev_warn(&applespi->spi->dev, "Command %d timed out\n",
+ applespi->cmd_evt_type);
+
+ applespi->cmd_msg_queued = 0;
+ applespi->write_active = false;
+ }

/* set up packet */
memset(packet, 0, APPLESPI_PACKET_SIZE);
@@ -869,7 +878,7 @@ static int applespi_send_cmd_msg(struct applespi_data *applespi)
return sts;
}

- applespi->cmd_msg_queued = true;
+ applespi->cmd_msg_queued = ktime_get();
applespi->write_active = true;

return 0;
@@ -1921,7 +1930,7 @@ static int __maybe_unused applespi_resume(struct device *dev)
applespi->drain = false;
applespi->have_cl_led_on = false;
applespi->have_bl_level = 0;
- applespi->cmd_msg_queued = false;
+ applespi->cmd_msg_queued = 0;
applespi->read_active = false;
applespi->write_active = false;

--
2.26.2


Subject: [PATCH 2/3] Input: applespi: Fix occasional crc errors under load.

For some reason, when the system is under heavy CPU load, the read
following the write sometimes occurs unusually quickly, resulting in
the read data not being quite ready and hence a bad packet getting read.
Adding another delay after reading the status message appears to fix
this.

Signed-off-by: Ronald Tschalär <[email protected]>
---
drivers/input/keyboard/applespi.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/drivers/input/keyboard/applespi.c b/drivers/input/keyboard/applespi.c
index 8494bf610fd70..f0a0067c48ff6 100644
--- a/drivers/input/keyboard/applespi.c
+++ b/drivers/input/keyboard/applespi.c
@@ -749,6 +749,8 @@ static void applespi_async_write_complete(void *context)
applespi->tx_status,
APPLESPI_STATUS_SIZE);

+ udelay(SPI_RW_CHG_DELAY_US);
+
if (!applespi_check_write_status(applespi, applespi->wr_m.status)) {
/*
* If we got an error, we presumably won't get the expected
--
2.26.2

2021-02-19 19:17:50

by Dmitry Torokhov

[permalink] [raw]
Subject: Re: [PATCH 2/3] Input: applespi: Fix occasional crc errors under load.

On Wed, Feb 17, 2021 at 11:07:17AM -0800, Ronald Tschal?r wrote:
> For some reason, when the system is under heavy CPU load, the read
> following the write sometimes occurs unusually quickly, resulting in
> the read data not being quite ready and hence a bad packet getting read.
> Adding another delay after reading the status message appears to fix
> this.
>
> Signed-off-by: Ronald Tschal?r <[email protected]>

Applied, thank you.

--
Dmitry