These are five patches that fix various issues I found when testing the F54 support of my Lenovo X1 Carbon Gen 6 laptop. The first three are pretty straightforward. The third patch is a media documentation fix and, once Acked, it can go in either through the media subsystem or the input subsystem, whatever is easiest. The last two patches are more of an RFC quality: I noticed that irq_find_mapping() could return 0, which causes a kernel crash. I suspect that this patch just fixes the symptom and not necessarily the actual cause. I can do more testing to see if I can find the real cause. Hints of where to look would be welcome. The last patch is basically trial-and-error. When testing F54 I noticed that only the first 32 bytes of the capture image were valid, everything else was garbage. By deleting the line that increments rmiaddr it suddenly started working, but I only found an old RMI4 spec and I have no idea why this fix works. I can't imagine that it failed when this F54 driver was first added, so did something else break? Or is my Lenovo special in some way? I can help with testing, but the https://github.com/ndyer/heatmap/commits/heatmap-v4l is easy enough to use. Regards, Hans Hans Verkuil (5): input/mouse/synaptics: add LEN0091 support input/rmi4/rmi_f54: fix various V4L2 compliance problems pixfmt-tch-td16/tu16.rst: document that this is little endian input/rmi4/rmi_driver: check if irq_find_mapping returns 0 input/rmi4/rmi_smbus.c: don't increment rmiaddr in rmi_smb_read_block() .../media/uapi/v4l/pixfmt-tch-td16.rst | 34 +++++++++---------- .../media/uapi/v4l/pixfmt-tch-tu16.rst | 34 +++++++++---------- drivers/input/mouse/synaptics.c | 1 + drivers/input/rmi4/rmi_driver.c | 8 +++-- drivers/input/rmi4/rmi_f54.c | 15 +++++++- drivers/input/rmi4/rmi_smbus.c | 1 - 6 files changed, 55 insertions(+), 38 deletions(-) -- 2.23.0
