[PATCH 21/57] power: Overflow in current calculation

[mathieu . poirier]

From: Paer-Olof Haakansson <par-olof.hakansson@xxxxxxxxxxxxxx>

When calculating the average current the nominator will
overflow when the charging current is high.

Signed-off-by: Henrik SÃlver <henrik.solver@xxxxxxxxxxxxxx>
Signed-off-by: Mathieu Poirier <mathieu.poirier@xxxxxxxxxx>
Reviewed-by: Par-Olof HAKANSSON <par-olof.hakansson@xxxxxxxxxxxxxx>
Reviewed-by: Jonas ABERG <jonas.aberg@xxxxxxxxxxxxxx>
---
 drivers/power/ab8500_fg.c |   15 ++++++++-------
 1 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/power/ab8500_fg.c b/drivers/power/ab8500_fg.c
index 46010ec..7c42150 100644
--- a/drivers/power/ab8500_fg.c
+++ b/drivers/power/ab8500_fg.c
@@ -805,15 +805,13 @@ static void ab8500_fg_acc_cur_work(struct work_struct *work)
 
 	/*
 	 * Convert to unit value in mA
-	 * Full scale input voltage is
-	 * 66.660mV => LSB = 66.660mV/(4096*res) = 1.627mA
-	 * Given a 250ms conversion cycle time the LSB corresponds
-	 * to 112.9 nAh. Convert to current by dividing by the conversion
+	 * by dividing by the conversion
 	 * time in hours (= samples / (3600 * 4)h)
-	 * 112.9nAh assumes 10mOhm, but fg_res is in 0.1mOhm
+	 * and multiply with 1000
 	 */
-	di->avg_curr = (val * QLSB_NANO_AMP_HOURS_X10 * 36) /
-		(1000 * di->bat->fg_res * (di->fg_samples / 4));
+
+	di->avg_curr = (di->accu_charge * 36) /
+			((di->fg_samples / 4) * 10);
 
 	di->flags.conv_done = true;
 
@@ -821,6 +819,9 @@ static void ab8500_fg_acc_cur_work(struct work_struct *work)
 
 	queue_work(di->fg_wq, &di->fg_work);
 
+	dev_dbg(di->dev,
+		"fg_res: %d, fg_samples: %d, gasg: %d, accu_charge: %d\n",
+		di->bat->fg_res, di->fg_samples, val, di->accu_charge);
 	return;
 exit:
 	dev_err(di->dev,
-- 
1.7.5.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/