The raw Sony IR decoder decodes 15bit messages slightly incorrectly. To decode the function number, it shifts the bits right by 7 so that the function is in bits 7:1, masks with 0xFD (0b11111101), and does an 8 bit reverse so it ends up in bits 6:0. The mask should be 0xFE to correspond with bits 7:1 (0b11111110).
The old mask had the effect of dropping the MSB of the function number from bit 6, and leaving the LSB of the device number in bit 7. Signed-off-by: James Hogan <james.ho...@imgtec.com> --- (note, i don't have a 15bit sony remote to test this with, but i'm pretty confident of it's correctness based on this: http://picprojects.org.uk/projects/sirc/sonysirc.pdf ) drivers/media/rc/ir-sony-decoder.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/media/rc/ir-sony-decoder.c b/drivers/media/rc/ir-sony-decoder.c index d5e2b50..dab98b3 100644 --- a/drivers/media/rc/ir-sony-decoder.c +++ b/drivers/media/rc/ir-sony-decoder.c @@ -130,7 +130,7 @@ static int ir_sony_decode(struct rc_dev *dev, struct ir_raw_event ev) case 15: device = bitrev8((data->bits >> 0) & 0xFF); subdevice = 0; - function = bitrev8((data->bits >> 7) & 0xFD); + function = bitrev8((data->bits >> 7) & 0xFE); break; case 20: device = bitrev8((data->bits >> 5) & 0xF8); -- 1.7.2.3 -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
