2014-04-17 18:00:33

by Doug Anderson

Subject: [PATCH 4/7] mfd: cros_ec: spi: Increase cros_ec_spi deadline from 5ms to 100ms

We're adding i2c tunneling to the list of things that goes over
cros_ec. i2c tunneling can be slooooooow, so increase our deadline to
100ms to account for that.

Signed-off-by: Doug Anderson <[email protected]>
---
drivers/mfd/cros_ec_spi.c | 24 ++++++++++++++++--------
1 file changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/mfd/cros_ec_spi.c b/drivers/mfd/cros_ec_spi.c
index 4f863c3..0b8d328 100644
--- a/drivers/mfd/cros_ec_spi.c
+++ b/drivers/mfd/cros_ec_spi.c
@@ -39,14 +39,22 @@
#define EC_MSG_PREAMBLE_COUNT 32

/*
- * We must get a response from the EC in 5ms. This is a very long
- * time, but the flash write command can take 2-3ms. The EC command
- * processing is currently not very fast (about 500us). We could
- * look at speeding this up and making the flash write command a
- * 'slow' command, requiring a GET_STATUS wait loop, like flash
- * erase.
- */
-#define EC_MSG_DEADLINE_MS 5
+ * Allow for a long time for the EC to respond. We support i2c
+ * tunneling and support fairly long messages for the tunnel (249
+ * bytes long at the moment). If we're talking to a 100 kHz device
+ * on the other end and need to transfer ~256 bytes, then we need:
+ * 10 us/bit * ~10 bits/byte * ~256 bytes = ~25ms
+ *
+ * We'll wait 4 times that to handle clock stretching and other
+ * paranoia.
+ *
+ * It's pretty unlikely that we'll really see a 249 byte tunnel in
+ * anything other than testing. If this was more common we might
+ * consider having slow commands like this require a GET_STATUS
+ * wait loop. The 'flash write' command would be another candidate
+ * for this, clocking in at 2-3ms.
+ */
+#define EC_MSG_DEADLINE_MS 100

/*
* Time between raising the SPI chip select (for the end of a
--
1.9.1.423.g4596e3a

2014-04-18 17:31:01

by Simon Glass

Subject: Re: [PATCH 4/7] mfd: cros_ec: spi: Increase cros_ec_spi deadline from 5ms to 100ms

Hi Doug,

On 17 April 2014 11:59, Doug Anderson <[email protected]> wrote:
> We're adding i2c tunneling to the list of things that goes over
> cros_ec. i2c tunneling can be slooooooow, so increase our deadline to
> 100ms to account for that.
>
> Signed-off-by: Doug Anderson <[email protected]>

I believe the EC protocol should be changed on future platforms to
poll for completion of I2C, but in the meantime this patch prevents
failure.

Reviewed-by: Simon Glass <[email protected]>