dwc_otg: pay attention to qh->interval when rescheduling periodic queues

A regression introduced in https://github.com/raspberrypi/linux/pull/3887
meant that if the newly scheduled transfer immediately returned data, and
the driver resubmitted a single URB after every transfer, then the effective
polling interval would end up being approx 1ms.

Use the larger of SCHEDULE_SLOP or the configured endpoint interval.

Signed-off-by: Jonathan Bell <jonathan@raspberrypi.com>
This commit is contained in:
Jonathan Bell
2021-10-06 15:27:53 +01:00
committed by Phil Elwell
parent 6237d09759
commit abff5012ab

View File

@@ -691,7 +691,7 @@ int dwc_otg_hcd_qh_add(dwc_otg_hcd_t * hcd, dwc_otg_qh_t * qh)
} else {
/* If the QH wasn't in a schedule, then sched_frame is stale. */
qh->sched_frame = dwc_frame_num_inc(dwc_otg_hcd_get_frame_number(hcd),
SCHEDULE_SLOP);
max_t(uint32_t, qh->interval, SCHEDULE_SLOP));
status = schedule_periodic(hcd, qh);
qh->start_split_frame = qh->sched_frame;
if ( !hcd->periodic_qh_count ) {