[src/trunk]: src/sys/dev/usb Back out r1.82 ("Do not explicitly set the HID R...

[jmcneill <jmcneill%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/accbd625b4f2
branches:  trunk
changeset: 372262:accbd625b4f2
user:      jmcneill <jmcneill%NetBSD.org@localhost>
date:      Fri Nov 04 19:46:55 2022 +0000

description:
Back out r1.82 ("Do not explicitly set the HID Report Protocol upon attach")
due to the regression reported in PR# 57031

diffstat:

 sys/dev/usb/uhidev.c |  14 +++-----------
 1 files changed, 3 insertions(+), 11 deletions(-)

diffs (36 lines):

diff -r 32c924a91348 -r accbd625b4f2 sys/dev/usb/uhidev.c
--- a/sys/dev/usb/uhidev.c      Fri Nov 04 11:20:39 2022 +0000
+++ b/sys/dev/usb/uhidev.c      Fri Nov 04 19:46:55 2022 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: uhidev.c,v 1.93 2022/03/28 12:44:45 riastradh Exp $    */
+/*     $NetBSD: uhidev.c,v 1.94 2022/11/04 19:46:55 jmcneill Exp $     */
 
 /*
  * Copyright (c) 2001, 2012 The NetBSD Foundation, Inc.
@@ -35,7 +35,7 @@
  */
 
 #include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: uhidev.c,v 1.93 2022/03/28 12:44:45 riastradh Exp $");
+__KERNEL_RCSID(0, "$NetBSD: uhidev.c,v 1.94 2022/11/04 19:46:55 jmcneill Exp $");
 
 #ifdef _KERNEL_OPT
 #include "opt_usb.h"
@@ -223,16 +223,8 @@
        }
        (void)usbd_set_idle(iface, 0, 0);
 
-#if 0
-       /*
-        * HID 1.11 says we should do this, but the device firmware is
-        * supposed to come up in Report Protocol after reset anyway, and
-        * apparently explicitly requesting it confuses some devices.
-        */
-       if ((usbd_get_quirks(sc->sc_udev)->uq_flags & UQ_NO_SET_PROTO) == 0 &&
-           id->bInterfaceSubClass == UISUBCLASS_BOOT)
+       if ((usbd_get_quirks(sc->sc_udev)->uq_flags & UQ_NO_SET_PROTO) == 0)
                (void)usbd_set_protocol(iface, 1);
-#endif
 
        maxinpktsize = 0;
        sc->sc_iep_addr = sc->sc_oep_addr = -1;