Re: [PATCH v3] winebus.sys: Handle device reports for hidraw devices

[Sebastian Lackner]

On 18.10.2016 18:58, Sebastian Lackner wrote:
> Not sure if it is
> a problem, but due to the fact that the device has been opened a long time before starting to
> read from it, couldn't there be a large amount of reports in the kernel buffers? Do you think
> it makes sense to skip such outdated reports?

Actually, isn't there also a problem that reports are skipped in certain situations? Handling
of hidclass looks correct to me, but there is a risk that process_hid_report() will drop the
report if there is no pending IRP in the list. This happens for a short time inbetween of the
IOCTL_HID_READ_REPORT ioctls.

Regards,
Sebastian