Re: [PATCH 01/11] HID: uHID: implement .raw_request
From: David Herrmann <hidden>
Date: 2014-02-03 15:26:23
Also in:
lkml
Hi On Sun, Feb 2, 2014 at 5:50 AM, Benjamin Tissoires [off-list ref] wrote:
quoted hunk ↗ jump to hunk
It was missing, so adding it. Signed-off-by: Benjamin Tissoires <redacted> --- drivers/hid/uhid.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)diff --git a/drivers/hid/uhid.c b/drivers/hid/uhid.c index f5a2b19..438c9f1 100644 --- a/drivers/hid/uhid.c +++ b/drivers/hid/uhid.c@@ -270,6 +270,22 @@ static int uhid_hid_output_report(struct hid_device *hid, __u8 *buf, return count; } +static int uhid_raw_request(struct hid_device *hid, unsigned char reportnum, + __u8 *buf, size_t len, unsigned char rtype, + int reqtype) +{ + switch (reqtype) { + case HID_REQ_GET_REPORT: + return uhid_hid_get_raw(hid, reportnum, buf, len, rtype); + case HID_REQ_SET_REPORT: + if (buf[0] != reportnum) + return -EINVAL; + return uhid_hid_output_raw(hid, buf, len, rtype);
But that one looks wrong. UHID does not have any SET_REPORT query in the protocol, yet. You turn a SET_REPORT into an OUTPUT_REPORT here. So if user-space gets the UHID_OUTPUT event, it doesn't know whether to send a SET_REPORT@ctrl to the device, or an async OUTPUT_REPORT@intr. This at least matters for low-energy BT in bluez, which uses uhid. So we'd have to add an UHID_SET_REPORT event. Note that currently UHID_FEATURE is the equivalent of UHID_GET_REPORT, but just horribly named.. Thanks David
quoted hunk ↗ jump to hunk
+ default: + return -EIO; + } +} + static struct hid_ll_driver uhid_hid_driver = { .start = uhid_hid_start, .stop = uhid_hid_stop,@@ -277,6 +293,7 @@ static struct hid_ll_driver uhid_hid_driver = { .close = uhid_hid_close, .parse = uhid_hid_parse, .output_report = uhid_hid_output_report, + .raw_request = uhid_raw_request, }; #ifdef CONFIG_COMPAT --1.8.3.1