Re: [PATCH] HID: logitech-hidpp: Handle timeout differently from busy

[Benjamin Tissoires <benjamin.tissoires@redhat.com>]

On Jun 03 2023, Mark Lord wrote:
> 
> On 2023-06-03 08:41 AM, Jiri Kosina wrote:
> > On Wed, 31 May 2023, Jiri Kosina wrote:
> > 
> >>> If an attempt at contacting a receiver or a device fails because the
> >>> receiver or device never responds, don't restart the communication, only
> >>> restart it if the receiver or device answers that it's busy, as originally
> >>> intended.
> >>>
> >>> This was the behaviour on communication timeout before commit 586e8fede795
> >>> ("HID: logitech-hidpp: Retry commands when device is busy").
> >>>
> >>> This fixes some overly long waits in a critical path on boot, when
> >>> checking whether the device is connected by getting its HID++ version.
> >>>
> >>> Signed-off-by: Bastien Nocera <hadess@hadess.net>
> >>> Suggested-by: Mark Lord <mlord@pobox.com>
> >>> Fixes: 586e8fede795 ("HID: logitech-hidpp: Retry commands when device is busy")
> >>> Link: https://bugzilla.kernel.org/show_bug.cgi?id=217412
> >>> ---
> >>>  drivers/hid/hid-logitech-hidpp.c | 1 +
> >>>  1 file changed, 1 insertion(+)
> >>>
> >>> diff --git a/drivers/hid/hid-logitech-hidpp.c b/drivers/hid/hid-logitech-hidpp.c
> >>> index 0fcfd85fea0f..2246044b1639 100644
> >>> --- a/drivers/hid/hid-logitech-hidpp.c
> >>> +++ b/drivers/hid/hid-logitech-hidpp.c
> >>> @@ -314,6 +314,7 @@ static int hidpp_send_message_sync(struct hidpp_device *hidpp,
> >>>  			dbg_hid("%s:timeout waiting for response\n", __func__);
> >>>  			memset(response, 0, sizeof(struct hidpp_report));
> >>>  			ret = -ETIMEDOUT;
> >>> +			goto exit;
> >>>  		}
> >>>  
> >>
> >> I have applied this even before getting confirmation from the reporters in 
> >> bugzilla, as it's the right thing to do anyway.
> > 
> > Unfortunately it doesn't seem to cure the reported issue (while reverting 
> > 586e8fede79 does): https://bugzilla.kernel.org/show_bug.cgi?id=217523#c2
> 
> I wonder if this code could be re-worked to not even do this (waiting)
> from the _probe() function?  It ought to be able to throw it on a workqueue
> or something, rather than stalling system boot for a minimum of 5-seconds
> (or much longer as as-is).

That's an option, but the fact that I can not replicate locally with the
exact same hardware seems to indicate that we would just be papering
over the issue.

Here, I admittely have the USB receiver running through USB-C ports, and
the communication never fails and I get immediate bring ups of the
devices. Which means I am not hitting that path.

The hidpp driver should have everything ready to delay the init in a
workqueue, but the impacted users would still get a delay when they plug
in the device (which is better than stalling the boot, I agree).

Cheers,
Benjamin