Hi! I tried to set up my imon remote, but run into the issue of buttons getting easily stuck. And while looking at the code, I noticed some more issues :) So here they all are (on 2.6.35-rc4): - There is no fallback timer if a release code is missed (i.e. remote pointed away from receiver or some other anomaly), causing a key stuck on autorepeat. The driver should inject a release code itself if there is no release/repeat code in 500ms after initial press or in 120ms after a repeat code. - No release code is sent for 0x02XXXXXX keys if pressing any other button before release, examples: example 1: hold '5', then press 'Play' once, then release '5' The 'Play' codes are relayed properly, but the release code for '5' (the 'all 0x02XXXXXX keys released' (i.e. empty HID input report) which the hardware does send properly) is wrongly interpreted as a release code for 'Play'. The driver should either release '5' when the empty report is received, or, as this is just a remote control, just inject a release code for '5' when 'Play' is pressed. example 2: hold '5', then hold '4', then release '5', then release '4' As the 0x02XXXXXX range is not completely released until after '4' is released, the zeroed bitfield is not sent until after '4', and the driver doesn't release '5' at this point anymore. The driver should've injected a release code for '5' when '4' was pressed. - imon_mce_timeout() timer function seems to access ictx->last_keycode without locking - imon_parse_press_type() tests for ictx->release_code which is in an undefined state if ktype isn't IMON_KEY_IMON - when the dpad is in keyboard mode and you hold it down to one direction, instead of autorepeat there is a constant stream of release/press events I may get around to fix these (if I find time and an MCE remote for testing), but that won't probably happen soon, thus I'm reporting these here if someone else wants to fix them. -- Anssi Hannula -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
