Index | Thread | Search

From:
Ted Bullock <tbullock@comlore.com>
Subject:
Re: sparc64: X and getty sharing raw keyboard input
To:
tech@openbsd.org
Cc:
Matthieu Herrb <matthieu@openbsd.org>
Date:
Tue, 13 Oct 2026 12:46:58 -0600

Download raw body.

Thread
Hiya folks,

Selecting Protocol "wskbd" locally separates X keyboard input from getty and
addresses the problem on my sparc with a consumer Logitech USB keyboard. Would
it be appropriate to make that the sparc64 default?

I can see reasons to be cautious, mixed Sun and USB or KVM keyboards could be
interpreted incorrectly in mysterious ways, which I could see happening but
not all that often.

Matthieu discussed the mux limitation in 2010 and largely I agree with his
take here: https://mail-index.netbsd.org/tech-x11/2010/06/20/msg000790.html

Anyway, I've included the small local Xenocara proposal below. I don't know
whether an OpenBSD guard belongs here either. Does changing the default make
sense, or should this remain an explicit configuration choice?

Ted

--- driver/xf86-input-keyboard/src/kbd.c
+++ driver/xf86-input-keyboard/src/kbd.c
@@ -73,9 +73,11 @@
 #else
     "Protocol",		"standard",
 #endif
-#else /* NetBSD */
+#elif defined(__sparc64__)
+    "Protocol",		"wskbd",
+#else
     "Protocol",		"standard",
-#endif /* NetBSD */
+#endif
     "XkbRules",		XKB_DFLT_RULES,
     "XkbModel",		"pc105",
     "XkbLayout",	"us",