[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20111220054441.7043.64449.stgit@notabene.brown>
Date: Tue, 20 Dec 2011 16:44:41 +1100
From: NeilBrown <neilb@...e.de>
To: Andrew Morton <akpm@...ux-foundation.org>
Cc: rpurdie@...ys.net, linux-kernel@...r.kernel.org
Subject: [PATCH 2/2] leds-tca6507 - fix off by one error.
When walking the list of possible time codes we can fall off the end.
Fix that, and also improved the commentary.
Reported-by: Dan Carpenter <dan.carpenter@...cle.com>
Signed-off-by: NeilBrown <neilb@...e.de>
---
drivers/leds/leds-tca6507.c | 8 ++++++--
1 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/drivers/leds/leds-tca6507.c b/drivers/leds/leds-tca6507.c
index 3565a61..7680f50 100644
--- a/drivers/leds/leds-tca6507.c
+++ b/drivers/leds/leds-tca6507.c
@@ -165,7 +165,8 @@ MODULE_DEVICE_TABLE(i2c, tca6507);
static int choose_times(int msec, int *c1p, int *c2p)
{
/* Chose two timecodes which add to 'msec' as near as possible.
- * The first returned should be the larger.
+ * The first returned should be the larger and is the 'on' of 'off' time.
+ * The second will be used as a 'fade-on' or 'fade-off' time.
* If cannot get within 1/8, fail.
* If two possibilities are equally good (e.g. 512+0, 256+256), choose
* the first pair so there is more change-time visible (i.e. it is softer).
@@ -175,7 +176,10 @@ static int choose_times(int msec, int *c1p, int *c2p)
int tmin = msec * 7 / 8;
int diff = 65536;
- for (c1 = 1; c1 <= TIMECODES; c1++) {
+ /* We start at '1' to ensure we never even think of choosing a
+ * total time of '0'.
+ */
+ for (c1 = 1; c1 < TIMECODES; c1++) {
int t = time_codes[c1];
if (t*2 < tmin)
continue;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists