Date: Sun, 07 Jan 2001 20:36:54 +0900 From: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp> To: Robert Watson <rwatson@freebsd.org> Cc: hackers@freebsd.org, John Polstra <jdp@polstra.com>, yokota@zodiac.mech.utsunomiya-u.ac.jp Subject: Re: KVM switch vs. FreeBSD psm driver (Solved!) Message-ID: <200101071136.UAA24356@zodiac.mech.utsunomiya-u.ac.jp> In-Reply-To: Your message of "Sat, 06 Jan 2001 22:10:40 EST." <Pine.NEB.3.96L.1010106220832.22348B-100000@fledge.watson.org> References: <Pine.NEB.3.96L.1010106220832.22348B-100000@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
>Do you know if this fixes the problem the following problem that I started >experiencing somewhere in the RELENG_4 line? For some machines, if the >KVM is not pointed at the box, the keyboard will not probe properly, and >does not respond for that session. As long as I boot with the KVM pointed >at the machine during the boot process, it probes fine. This doesn't seem >to impact the boot loaders, only after the kernel has loaded and probed. >It's really annoying as my crashbox has this problem, so I have to swap to >it every time I boot, and given the need to type continue in the serial >gdb, I often miss the window. I guess your KVM switch won't let the kernel talk to the keyboard, if the switch wasn't pointed to the FreeBSD box. Remove the flags 0x1 from atkbd in your kernel config file. This flag makes atkbd fail, if it doesn't detect a keyboard. Without the flag, atkbd will install regardless of the absence of the keyboard. In RELENG_2 and RELENG_3, this flag wasn't specified to atkbd by default. In RELENG_4, the flag was added to GENERIC so that syscons will use a USB keyboard, if any, when there is no AT keyboard. Kazu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200101071136.UAA24356>