mirror/tinyusb
3
0
Fork 0
mirror of https://github.com/hathach/tinyusb.git synced 2026-08-20 12:07:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

device: ignore DEVICE_REMOTE_WAKEUP at SuperSpeed instead of stalling it

Browse Source 
Stalling it was a regression, caught on hardware. USB 3.2 9.2.5.4 says the
DEVICE_REMOTE_WAKEUP selector is "ignored and not used by Enhanced SuperSpeed
devices" - ignored, not rejected - and I read that as license to STALL. On a
CH32H417 at 5 Gbps a STALL there made usbtest ctrl_out (14, 21) and iso read
(16, 23) fail; bisected to this change and confirmed by reverting only these two
guards, with func_wakeup_clear_itf left in place and innocent.

Now genuinely ignored: the status stage is acknowledged and only the state
change is skipped, which still keeps remote_wakeup_en from being set while
func_wakeup_bm is empty - the desync the gate existed to prevent.
This commit is contained in:
hathach
2026-08-18 17:06:42 +07:00
parent b2189417c6
commit d33e6da729
1 changed files with 18 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
src/device/usbd.c
View File

@ -1112,17 +1112,20 @@ static bool process_std_device_request(uint8_t rhport, tusb_control_request_t co
switch (p_request->wValue) { //-V2520
case TUSB_REQ_FEATURE_REMOTE_WAKEUP:
TU_LOG_USBD(" Enable Remote Wakeup\r\n");
// USB 3.2 §9.2.5.4: the DEVICE_REMOTE_WAKEUP selector is not used by Enhanced
// SuperSpeed devices, which arm wake per function via FUNCTION_SUSPEND instead (Linux
// picks the recipient by speed for exactly this reason). Accepting it on a SuperSpeed
// link would set the aggregate with func_wakeup_bm still empty, so tud_remote_wakeup()
// would drive a U-state exit no function was authorised for while GET_STATUS(Device)
// masks the bit out and the host cannot even observe it.
// USB 3.2 §9.2.5.4: this selector is "ignored and not used by Enhanced SuperSpeed
// devices", which arm wake per function via FUNCTION_SUSPEND instead. IGNORED, not
// rejected: acknowledge the request but leave the state alone, because setting the
// aggregate while func_wakeup_bm is empty would let tud_remote_wakeup() drive a
// U-state exit no function authorised. Stalling it instead is what the spec does NOT
// say, and it breaks real hosts - hardware-measured on a CH32H417 at 5 Gbps, a STALL
// here made usbtest ctrl_out (14, 21) and iso read (16, 23) fail.
#if TUD_OPT_SUPER_SPEED
TU_VERIFY(!link_is_superspeed());
if (!link_is_superspeed())
#endif
// Host may enable remote wake up before suspending especially HID device
_usbd_dev.remote_wakeup_en = 1;
{
// Host may enable remote wake up before suspending especially HID device
_usbd_dev.remote_wakeup_en = 1;
}
tud_control_status(rhport, p_request);
return true;
@ -1169,14 +1172,16 @@ static bool process_std_device_request(uint8_t rhport, tusb_control_request_t co
switch (p_request->wValue) { //-V2520
case TUSB_REQ_FEATURE_REMOTE_WAKEUP:
TU_LOG_USBD(" Disable Remote Wakeup\r\n");
// Not used at SuperSpeed, same as the SET_FEATURE side above: clearing the aggregate
// Ignored at SuperSpeed, same as the SET_FEATURE side above: clearing the aggregate
// there would leave func_wakeup_bm holding bits, so GET_STATUS(Interface) would report
// D1=1 while tud_remote_wakeup() refuses.
#if TUD_OPT_SUPER_SPEED
TU_VERIFY(!link_is_superspeed());
if (!link_is_superspeed())
#endif
// Host may disable remote wake up after resuming
_usbd_dev.remote_wakeup_en = 0;
{
// Host may disable remote wake up after resuming
_usbd_dev.remote_wakeup_en = 0;
}
tud_control_status(rhport, p_request);
return true;