[9fans] mouse vs kvm switch

[9fans] mouse vs kvm switch

[erik quanstrom]

i'm having a bit of trouble with the mouse v the kvm switch on a dual
processor machine (even running with *nomp=1).  when i startup with
the kbd and mouse attached to the kvm, the the mouse will move 2 pixels
after a kvm switch or accept one mouse click and then it freezes.  after
i switch the kvm input away from the plan9 box and back again, i get another
two pixels or one mouse click.  any ideas?

(fwiw it's an iogear 4-port kvm switch with builtin cabling.)

- erik

Re: [9fans] mouse vs kvm switch

[jmk]

i use kvm switches all the time, cybex autoview, cybex autocommander
and raritan and have never seen a problem like that. sometimes old
motherboards with 30-foot cables don't work except after a power-off/on
on the motherboard, but that's all.

Re: [9fans] mouse vs kvm switch

[quanstro]

[-- Attachment #1: Type: text/plain, Size: 1348 bytes --]

it's an infamous valinux box.  i had to use an extension cord to connect
the kvm mouse cord with the machine because the alignment pin on the
kvm switch's mouse plug is too fat to fit in the socket on the motherboard.  the
total length is 16'.  it's really odd that it works for one click or two pixels.
it's odder that switching to another machine and back will reset things for
another two pixels.  this happens on cold boot.  it does seem like a timing issue.

i do have a 128MB nvidia mx4000 (nv18) "working" in the box; but when i start
with monitor=vesa and then used my modified aux/vga (that maps did=0x0185
to nvarch=10) with monitor=multisync75, the display is correct. however when
i start the machine with monitor=multisync75, the display is shifted to the left
by 1 inch.  CrystalFreq is off by 5Mhz. (does this matter?)

also, not that i've finally been able to kill the vesa montior bit, i've noticed that
when i delete the "*nomp=1" line from plan9.ini, i still start with cpu0 only.

- erik

On Fri Apr 21 20:46:20 CDT 2006, jmk@plan9.bell-labs.com wrote:
> i use kvm switches all the time, cybex autoview, cybex autocommander
> and raritan and have never seen a problem like that. sometimes old
> motherboards with 30-foot cables don't work except after a power-off/on
> on the motherboard, but that's all.

[-- Attachment #2: vga-p --]
[-- Type: text/plain, Size: 1504 bytes --]

vga->attr: vid=0x10DE
vga->attr: did=*
vga misc             23
vga feature          00
vga sequencer        03 01 0F 00 0A
vga crt              D1 9F 9F 95 AE 01 31 5A - 00 60 00 00 00 00 00 00
                     04 2A FF 80 00 FF 32 A3 - FF
vga graphics         00 00 00 00 00 50 05 0F - FF
vga attribute        00 01 02 03 04 05 06 07 - 08 09 0A 0B 0C 0D 0E 0F
                     01 00 0F 00 00
vga virtual         0 0
vga panning         off
vga vmz             134217728
vga apz                     0
vga linear                  1
nvidia dclk m n p    135000000 14 - 140 1
nvidia CrystalFreq   27000000 Hz
nvidia arch          10
nvidia did           0185
nvidia repaint0      40
nvidia repaint1      28
nvidia screen        15
nvidia pixel         3
nvidia horiz         0
nvidia cursor0       ff
nvidia cursor1       41
nvidia cursor2       3
nvidia interlace     ff
nvidia extra         0
nvidia crtcowner     4
nvidia timingH       0
nvidia timingV       0
nvidia vpll          18c0e
nvidia vpllB         0
nvidia vpll2         4a40b
nvidia vpll2B        0
nvidia pllsel        10000700
nvidia general       100
nvidia scale         11100011
nvidia config        9c11001
nvidia head          1010
nvidia head2         30311
nvidia cursorconfig  2000101
nvidia dither        f0000
nvidia crtcsync      0
nvidia islcd         0
nvidia twoheads      1
nvidia twostagepll   0
nvidia crtcnumber    0
nvidia fpwidth       0
nvidia fpheight      0

[-- Attachment #3: vesavga-p --]
[-- Type: text/plain, Size: 1731 bytes --]

aux/vga: controller not in /lib/vgadb, not vesa
0xC0000 55 AA 7C EB 4B 37 34 30 30 E9 4C 19 77 CC 56 49  U.|.K7400.L.w.VI
0xC0010 44 45 4F 20 0D 00 00 00 3C 01 75 0F 00 00 49 42  DEO ....<.u...IB
0xC0020 4D 20 56 47 41 20 43 6F 6D 70 61 74 69 62 6C 65  M VGA Compatible
0xC0030 01 00 00 00 C0 08 B8 AF 30 35 2F 32 32 2F 30 35  ........05/22/05
0xC0040 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0xC0050 E9 7B EB 00 00 00 00 00 3F 88 6F FE 80 10 D0 00  .{......?.o.....
0xC0060 FE FF FF 7F 00 00 00 80 22 00 A5 AA E9 D9 CF E9  ........".......
0xC0070 E0 CF 50 4D 49 44 6C 00 6F 00 00 00 00 A0 00 B3  ..PMIDl.o.......
0xC0080 00 B8 00 C0 00 33 0A F6 00 00 00 00 00 00 00 00  .....3..........
0xC0090 FF 7F 4E 56 00 05 28 B1 7E 7C 42 20 18 04 00 00  ..NV..(.~|B ....
0xC00A0 00 00 00 00 00 00 F7 04 00 00 00 00 00 00 54 01  ..............T.
0xC00B0 50 00 CF 4C A0 19 5E 3C B8 3C 22 44 D7 43 EB 43  P..L..^<.<"D.C.C
0xC00C0 14 02 37 02 E1 6E 00 00 00 00 00 00 00 00 61 0E  ..7..n........a.
0xC00D0 AC 20 AC 60 AE 0A 00 30 57 05 00 78 6F 06 70 0E  . .`...0W..xo.p.
0xC00E0 70 2E 70 52 70 52 70 06 70 26 E2 1E E3 00 00 00  p.pRpRp.p&......
0xC00F0 00 00 00 00 00 16 E4 17 E4 E0 AC B0 AC 00 00 00  ................
vga misc             67
vga feature          00
vga sequencer        03 00 03 00 02
vga crt              5F 4F 50 82 54 80 BF 1F - 00 4F 0E 0F 00 00 07 80
                     9C 8E 8F 28 1F 96 B9 A3 - FF
vga graphics         00 00 00 00 00 10 0E 00 - FF
vga attribute        00 01 02 03 04 05 14 07 - 38 39 3A 3B 3C 3D 3E 3F
                     0C 00 0F 08 00
vga virtual         0 0
vga panning         off
vga apz                     0
vga linear                  0

Re: [9fans] mouse vs kvm switch

[quanstro]

this was a phoenixbios problem.  there is an option to forget saved processor
information.  both processors are now started.

- erik

On Sat Apr 22 08:04:46 CDT 2006, quanstro@quanstro.net wrote:

> also, not that i've finally been able to kill the vesa montior bit, i've noticed that
> when i delete the "*nomp=1" line from plan9.ini, i still start with cpu0 only.
>