On 5/15/23 16:53, Filip Zaludek wrote:
Two following performance patches applied together occasionally harm usb keyboard on RPi3.
'dwc2: use the nonblock argument in submit_int_msg' commit 9dcab2c4d2cb50ab1864c818b82a72393c160236
'console: usb: kbd: Limit poll frequency to improve performance' commit 96991e652f541323a03c5b7e075d54a117091618
This empirically increased by sub-millisecond wait for dwc2 controller reset makes keyboard reliable.
Signed-off-by: Filip Zaludek filip.zaludek@oracle.com
drivers/usb/host/dwc2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/dwc2.c b/drivers/usb/host/dwc2.c index 23060fc369..71b66a52ed 100644 --- a/drivers/usb/host/dwc2.c +++ b/drivers/usb/host/dwc2.c @@ -764,7 +764,7 @@ static int dwc_otg_submit_rh_msg_out(struct dwc2_priv *priv, DWC2_HPRT0_PRTENCHNG | DWC2_HPRT0_PRTOVRCURRCHNG, DWC2_HPRT0_PRTRST);
mdelay(50);
udelay(50125);
Why not just use 'mdelay(51);' ?
If you can tell me how to reproduce this , I could try and use USB bus analyzer (Beagle 5000) to look at his. I have RPi3 somewhere I think. Then we would know what's up.
But that might take a while, since I am a bit busy these days.