Re: [PATCH] [v5]Input: evdev: fix bug of dropping full valid packet after syn_dropped
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: 2016-01-13 18:52:59
Also in:
lkml
On Wed, Jan 13, 2016 at 05:27:41PM +0530, Aniroop Mathur wrote:
quoted hunk ↗ jump to hunk
If last event in old queue that was dropped was EV_SYN/SYN_REPORT, then lets generate EV_SYN/SYN_REPORT immediately after queing EV_SYN/SYN_DROPPED so that clients would not ignore next valid full packet events. Signed-off-by: Aniroop Mathur <redacted> --- drivers/input/evdev.c | 45 +++++++++++++++++++++++++++++++++------------ 1 file changed, 33 insertions(+), 12 deletions(-)diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c index e9ae3d5..0bc7b98 100644 --- a/drivers/input/evdev.c +++ b/drivers/input/evdev.c@@ -156,7 +156,12 @@ static void __evdev_flush_queue(struct evdev_client *client, unsigned int type) static void __evdev_queue_syn_dropped(struct evdev_client *client) { struct input_event ev; + struct input_event *prev_ev; ktime_t time; + unsigned int mask = client->bufsize - 1; + + /* store previous event */ + prev_ev = &client->buffer[(client->head - 1) & mask];
How do you know that previous event is valid/exists? In fact, when we are dropping events due to the full queue, you will be referencing the newest event being processed, not the previous event. I also wonder if this code is safe with regard to __evdev_flush_queue() that is dropping bunch of events and possible empty SYN_REPORT groups. Thanks. -- Dmitry