Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 17 Feb 2014 18:41:13 +0200
From:      Aleksandr Rybalko <ray@ddteam.net>
To:        Andriy Gapon <avg@FreeBSD.org>, FreeBSD Current <freebsd-current@FreeBSD.org>, Aleksandr Rybalko <ray@FreeBSD.org>
Cc:        freebsd-x11@FreeBSD.org
Subject:   Re: vt(9): couple of small issues
Message-ID:  <78196899-7922-4fd0-937c-7b4a69714dbc@email.android.com>
In-Reply-To: <5301DF89.5030707@FreeBSD.org>
References:  <5301DF89.5030707@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On 17 лютого 2014 р. 12:08:09 GMT+02:00, Andriy Gapon <avg@FreeBSD.org> wrote:
>
>I used to have mousechar_start="3" in my r.conf.  To be honest I can
>not even
>recall why.  But this worked fine with syscons without any glitches.
>With vt I get the following during boot up:
>vidcontrol: setting mouse character: Inappropriate ioctl for device
>And mouse cursor is never drawn, apparently vidcontrol -m on -M <X>
>just fails.
>
>Looks like with syscons MOUSE_MOUSECHAR ioctl is handled by a terminal
>device
>and consolectl device while with vt it is handled only by sysmouse
>device.
>
>
>Another issue is that I see the following messages appearing during
>boot and
>from time to time afterwards in the system log:
>kernel: sysmouse: unknown ioctl: t:40007413
>kernel: sysmouse: unknown ioctl: t:80007410
>
>The ioctls seem to be TIOCFLUSH and TIOCGETA, so I am not sure why
>sysmouse gets
>involved.  Maybe the same issue exists with syscons's sysmouse as well,
>but it
>is just silent about it.
>
>Seems like ioctls are called by X server.  For example:
>sysmouse_ioctl:entry
>              kernel`devfs_ioctl_f+0x11c
>              kernel`kern_ioctl+0x1db
>              kernel`sys_ioctl+0x142
>              kernel`amd64_syscall+0x3c9
>              kernel`0xffffffff8080e6db
>
>              libc.so.7`ioctl+0xa
>              Xorg`xf86OpenSerial+0x221
>              mouse_drv.so`MouseProc+0x2f3
>              Xorg`EnableDevice+0x214
>              Xorg`xf86Wakeup+0x57b
>              Xorg`WakeupHandler+0x42
>              Xorg`WaitForSomething+0x317
>              Xorg`Dispatch+0x92
>              Xorg`main+0x475
>              Xorg`_start+0x14f
>
>sysmouse_ioctl:entry
>              kernel`devfs_ioctl_f+0x11c
>              kernel`kern_ioctl+0x1db
>              kernel`sys_ioctl+0x142
>              kernel`amd64_syscall+0x3c9
>              kernel`0xffffffff8080e6db
>
>              libc.so.7`ioctl+0xa
>              Xorg`xf86FlushInput+0x21
>              mouse_drv.so`MouseProc+0x376
>              Xorg`EnableDevice+0x214
>              Xorg`xf86Wakeup+0x57b
>              Xorg`WakeupHandler+0x42
>              Xorg`WaitForSomething+0x317
>              Xorg`Dispatch+0x92
>              Xorg`main+0x475
>              Xorg`_start+0x14f
>              ld-elf.so.1`0x8007d5000

Hello Andriy!

I remember we were discussed that problem before, can you please check if moused enabled (in rc.conf or by devd for USB mouse) and do you use Kern.vt.textmode (IIRC its name)?

Mousecharstart is a first char of 4 to replace its bitmap with parts of mouse pointer. You can see its job by defining it to some of frequently used chars, then mouse pointer will be moved over many chars on the screen :-)

About second question - there is no problem to just remove that warning, but think better to left it to make an idea for developers about what term ioctls doing with sysmuse devfs node.

Thanks for testing!
WBW
------
Aleksandr Rybalko <ray@ddteam.net>





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?78196899-7922-4fd0-937c-7b4a69714dbc>