Date: Thu, 14 Nov 1996 09:19:43 -0700 (MST) From: Nate Williams <nate@mt.sri.com> To: Ron Bolin <rlb@mindspring.com> Cc: freebsd-current@freebsd.org Subject: Re: New Syscons.c still gives PS/2 Mouse problems. Suggestions? Message-ID: <199611141619.JAA05621@rocky.mt.sri.com> In-Reply-To: <328A9413.794BDF32@mindspring.com> References: <328A9413.794BDF32@mindspring.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Ron Bolin writes:
> I did a cvsup and got the 1.86 version of syscons.c, and am still
> experiencing the wierd PS/2 mouse behavior.
>
> Any thoughts?
S'ren hasn't yet applied the fixes that Bruce suggested. In the
meantime, you can back out the following 'fix' which seems to be causing
the problem.
Nate
---------------
Index: syscons.c
===================================================================
RCS file: /home/CVS/src/sys/i386/isa/syscons.c,v
retrieving revision 1.182
retrieving revision 1.183
diff -c -r1.182 -r1.183
*** syscons.c 1996/10/26 20:16:58 1.182
--- syscons.c 1996/11/04 21:01:08 1.183
***************
*** 25,31 ****
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
! * $Id: syscons.c,v 1.182 1996/10/26 20:16:58 sos Exp $
*/
#include "sc.h"
--- 25,31 ----
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
! * $Id: syscons.c,v 1.183 1996/11/04 21:01:08 sos Exp $
*/
#include "sc.h"
***************
*** 572,606 ****
mark_all(cur_console);
}
! c = scgetc(SCGETC_NONBLOCK);
! cur_tty = VIRTUAL_TTY(get_scr_num());
! if (!(cur_tty->t_state & TS_ISOPEN))
! if (!((cur_tty = CONSOLE_TTY)->t_state & TS_ISOPEN))
! return;
! switch (c & 0xff00) {
! case 0x0000: /* normal key */
! (*linesw[cur_tty->t_line].l_rint)(c & 0xFF, cur_tty);
break;
! case NOKEY: /* nothing there */
! return;
! case FKEY: /* function key, return string */
! if (cp = get_fstr((u_int)c, (u_int *)&len)) {
! while (len-- > 0)
! (*linesw[cur_tty->t_line].l_rint)(*cp++ & 0xFF, cur_tty);
}
- break;
- case MKEY: /* meta is active, prepend ESC */
- (*linesw[cur_tty->t_line].l_rint)(0x1b, cur_tty);
- (*linesw[cur_tty->t_line].l_rint)(c & 0xFF, cur_tty);
- break;
- case BKEY: /* backtab fixed sequence (esc [ Z) */
- (*linesw[cur_tty->t_line].l_rint)(0x1b, cur_tty);
- (*linesw[cur_tty->t_line].l_rint)('[', cur_tty);
- (*linesw[cur_tty->t_line].l_rint)('Z', cur_tty);
- break;
}
if (cur_console->status & MOUSE_ENABLED) {
cur_console->status &= ~MOUSE_VISIBLE;
remove_mouse_image(cur_console);
--- 572,611 ----
mark_all(cur_console);
}
! /*
! * Loop while there is still input to get from the keyboard.
! * I don't think this is nessesary, and it doesn't fix
! * the Xaccel-2.1 keyboard hang, but it can't hurt. XXX
! */
! while ((c = scgetc(SCGETC_NONBLOCK)) != NOKEY) {
! cur_tty = VIRTUAL_TTY(get_scr_num());
! if (!(cur_tty->t_state & TS_ISOPEN))
! if (!((cur_tty = CONSOLE_TTY)->t_state & TS_ISOPEN))
! return;
! switch (c & 0xff00) {
! case 0x0000: /* normal key */
! (*linesw[cur_tty->t_line].l_rint)(c & 0xFF, cur_tty);
! break;
! case FKEY: /* function key, return string */
! if (cp = get_fstr((u_int)c, (u_int *)&len)) {
! while (len-- > 0)
! (*linesw[cur_tty->t_line].l_rint)(*cp++ & 0xFF, cur_tty);
! }
break;
! case MKEY: /* meta is active, prepend ESC */
! (*linesw[cur_tty->t_line].l_rint)(0x1b, cur_tty);
! (*linesw[cur_tty->t_line].l_rint)(c & 0xFF, cur_tty);
! break;
! case BKEY: /* backtab fixed sequence (esc [ Z) */
! (*linesw[cur_tty->t_line].l_rint)(0x1b, cur_tty);
! (*linesw[cur_tty->t_line].l_rint)('[', cur_tty);
! (*linesw[cur_tty->t_line].l_rint)('Z', cur_tty);
! break;
}
}
+
if (cur_console->status & MOUSE_ENABLED) {
cur_console->status &= ~MOUSE_VISIBLE;
remove_mouse_image(cur_console);
***************
*** 1442,1447 ****
--- 1447,1462 ----
scr_stat *scp = cur_console;
int s = spltty();
+ /*
+ * With release 2.1 of the Xaccel server, the keyboard is left
+ * hanging pretty often. Apparently the interrupt from the
+ * keyboard is lost, and I don't know why (yet).
+ * This Ugly hack calls scintr if input is ready and
+ * conveniently hides the problem. XXX
+ */
+ if (inb(KB_STAT) & KB_BUF_FULL)
+ scintr(0);
+
/* should we just return ? */
if ((scp->status&UNKNOWN_MODE) || blink_in_progress || switch_in_progress) {
timeout((timeout_func_t)scrn_timer, 0, hz/10);
***************
*** 2462,2471 ****
static u_int chr = 0;
next_code:
! kbd_wait();
! /* first see if there is something in the keyboard port */
! if (inb(KB_STAT) & KB_BUF_FULL)
scancode = inb(KB_DATA);
else if (flags & SCGETC_NONBLOCK)
return(NOKEY);
else
--- 2477,2487 ----
static u_int chr = 0;
next_code:
! /* check if there is anything in the keyboard buffer */
! if (inb(KB_STAT) & KB_BUF_FULL) {
! DELAY(25);
scancode = inb(KB_DATA);
+ }
else if (flags & SCGETC_NONBLOCK)
return(NOKEY);
else
***************
*** 2830,2836 ****
console[0]->smode.mode == VT_AUTO)
switch_scr(cur_console, 0);
Debugger("manual escape to debugger");
- return(NOKEY);
#else
printf("No debugger in kernel\n");
#endif
--- 2846,2851 ----
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611141619.JAA05621>