Re: Sysrq via serial line?
Send a BREAK signal. Most terminals/terminal s/w can. -- Paul On Thu, Sep 30, 2004 at 06:01:39PM +0100, Martin Habets wrote: > Hi, > > I'm using minicom to connect to a headless sparc running 2.6.8. > Have MAGIC_SYSRQ built into the kernel, but cannot find a way to trigger > it from the keyboard. echo into /proc/sysrq-trigger works fine. > > Sending a break gets me into openboot prompt. > > Have tried to use setkeycodes but it does not work for me. Initially > I had CONFIG_VT turned off, and setkeycodes gave: > Cannot get file desciptor for console > or something similar to that. > With CONFIG_VT turned on, but no keyboard support, I get: > KDSETKEYCODE: No such device > failed to set scancode 1b73 to keycode 84 > Turning on keyboard may help, but that's not right is it? I'm using a > serial port, not a keyboard. > > I'm using the sunzilog driver, but it seems to trigger sysrq after a > break signal in uart_handle_break(). But sendig a break gets me into > openboot. > > Is there a solution? > > >From dmesg: > ttyS0 at MMIO 0x0 (irq = 44) is a SunZilog > > Thanks, > -- > Martin > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >
Re: Issues installing on a SPARCServer 1000E
On Thu, Jan 08, 2004 at 03:40:19PM +, Martin wrote: > > SPARCserver 1000E which Debian supposedly supports. Information on the > 3. Get a team of people together and get Linux support in later versions > for them. Last time this was mentioned quite a few people here said > they would be happy to help but no-one had the time to /start/ the > project. It would give Linux one up on NetBSD though... :-) Well I have two 1000Es here (South London / UK) and I would be delighted to give one to someone who had the time and inclination to make it work... -- Paul
Re: debian/sparc serial console
On Wed, Oct 01, 2003 at 05:16:55PM +0200, Jan-Benedict Glaw wrote: > On Wed, 2003-10-01 09:37:01 +0100, Martin <[EMAIL PROTECTED]> > wrote in message <[EMAIL PROTECTED]>: > > > I've transitioned my Ultra 10 from my primary Desktop machine to > > > network server (having acquired a powerbook for my primary needs - > > > still trying to get Evolution compiled on that.) I would like to > > > access this Ultra 10 only through a serial cable & ethernet, without > > > monitor or keyboard. I should be able to do this through some settings > > > in Open Firmware right? > > OpenBoot should do it automatically - hook up the serial cable and > > unplug the monitor and keyboard. You can set it to always use serial > > but I would have thought leaving it on auto switching was best. > > Um, with my SS10, ROM 2.25, this doesn't happen. I have to explicitely > set serial I/O if I want that. Simply unplugging keyboard (which works > nicely on another SS10 with an older ROM version) doesn't help here... Then you need to unset the OBP settings that have made this manadtory. The default default is to use serial if there not a FB and a keyboard. -- Paul
Re: How to clone a drive
On Fri, Oct 03, 2003 at 10:22:50PM -0700, David S. Miller wrote: > > bash# dd if=X of=Y > This will would though you might want to add "bs=64k" which seems to make it much more efficient and so faster (default is 512bytes) -- Paul
Re: serial ports on Sparc 2.
On Sun, Jun 22, 2003 at 07:07:35PM -0700, Peter Easthope wrote: You can normally get 38400 on non-ultra Sun serial ports. No more. -- Paul > > Can anyone tell me the top speed of the serial ports on > a Sparc 2? The manual specifies the functions of the pins > and how to set jumpers for RS-232 or RS-4?? operation > but does not mention speeds. None of the lists of specs. > on the Web that I have found mention the speed either. > > Thanks, Peter E. > > http://carnot.pathology.ubc.ca/peter.html > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >
Re: change eth0 to eth2 on a multi-homed ultra 1
On Tue, Jun 10, 2003 at 02:47:10AM -0700, Blars Blarson wrote: > In article <[EMAIL PROTECTED]> [EMAIL PROTECTED] writes: > >This comes under the heading of 'nice to have' so I wouldn't want anyone > >to break stride in answering this problem but maybe someone has trodden > >this path before! > > > >I have a multi-homed U1. 1 motherboard 10 base T nic (eth2; Lance) and > >two additional 100 base T nics (eth0 & eth1; both Happy Meal). > > > >I would like the 10 base T eth2 to be eth0. How do I achieve this small > >miracle, please? > > If the drivers are compiled as modules, just make sure to modprobe the > lance driver before the happymeal. Alternativly, compile the lance > into the kernel and have happymeal be a modules. Eather way, the > lance devices will have lower numbers than the happymeal. If you've got them compiled into the kernel then the OBP's "sbus_probe" variable may change the order they are discovered by the kernel as well as the OBP. -- Paul
Re: wrong sector size
On Sun, Mar 23, 2003 at 04:56:47PM +0100, Thomas H?hn wrote: > Hi > > I have got a SparcStation 5 since a few weeks. It's my first Sparc. > > The SparcStation has no floppy or CD drive. So I installed Debian 3.0 > via netboot. > > Since the 500 MB hard disk is not much, I've bought a 9 gig drive from > ebay (IBM DGHS-09Y). > > I plugged it into the backplane and tried to start Debian. > But it reported "sda: unsupported sector size 520" and then "scsi: > deleting disk entry". > > It stops booting with the message "kernel NULL pointer dereference". > > I think I have to reformat the drive with a sector size of 512. Is that > correct? I have found two utilities to do that: scsiformat and > setblocksize (http://micha.freeshell.org/scsi/). > > The problem is, that I have to unplug the drive in order to boot. After > that i can plug it in again. But those tools need to be given the drive > in terms of /dev/sdb or sonething similar. > If i use "scsiadd -a 1", it's again "deleting disk entry". > > Is there a way to use these tools? They seem to use the sg driver. How > can I get this driver to "bind to" the drive, instead of the scsi disk > driver? > > The other solution (formatting on a scsi hostadapter) is not possible at > the moment because I have no adapter (it's a SCA drive). > > Can you help me? I bet that disc was once used in an Iceburg array. Linux really does not like non 512-block formatting and generally just crashes... You can try 1: plugging it in after booting and using something like scsiadd -a 0 4 0 if it is scsi ID 4 on the first scsi bus. 2: format it in solaris (this definately works ;O) -- Paul
Ooops with 2.4.12-preX and acX
I have an Ultra 1E which I'm having trouble with.
I've tried various pre's with and without the ac patches. I have the
sunhme patch compiled in.
For example I'm currently running:
Linux larder 2.4.21-pre4-spc64s0131-a #1 SMP Fri Jan 31 23:42:45
GMT 2003 sparc64 sun4u TI UltraSparc I (SpitFire) GNU/Linux
Here is the oopsym output:
ksymoops 2.4.6 on sparc64 2.4.21-pre4-spc64s0131-a. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.4.21-pre4-spc64s0131-a/ (default)
-m /boot/System.map-2.4.21-pre4-spc64s0131-a (default)
Warning: You did not tell me where to find symbol information. I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc. ksymoops -h explains the options.
Unable to handle kernel NULL pointer dereference
tsk->{mm,active_mm}->context = 0401
tsk->{mm,active_mm}->pgd = f820a000
\|/ \|/
"@'/ .. \`@"
/_| \__/ |_\
\__U_/
swapper(0): Oops
CPU[0]: local_irq_count[1] irqs_running[1]
TSTATE: 004480f09607 TPC: 005935c4 TNPC: 005935c8 Y:
0700Tainted: P
Using defaults from ksymoops -t elf32-sparc -a sparc
g0: 00698400 g1: 0001 g2: 7b28 g3:
4800
g4: f800 g5: 0080 g6: 00414000 g7:
0040
o0: 01e8 o1: f800268e12a0 o2: 0040 o3:
0080
o4: 0080 o5: 00593360 sp: 004171d1 ret_pc:
0200e5a0
l0: f800268e1240 l1: l2: 004a l3:
f82100c0
l4: f8210100 l5: 004a l6: f800262d1348 l7:
0007
i0: i1: 004fb840 i2: 005d36b0 i3:
i4: i5: 7000 i6: 00417291 i7:
0200e678
Caller[0200e678]
Caller[0041f424]
Caller[004088f4]
Caller[0041a5fc]
Caller[00418068]
Caller[005f87d8]
Caller[00404638]
Caller[]
Instruction DUMP: c4da7fc0 c6da7fc8 87832000 8b30b020
caa23fc0 c6a23fcc 9b30f020 daa23fc8
>>PC; 005935c4 <__memcpy_entry+290/3a0> <=
>>g0; 00698400
>>g6; 00414000
>>o5; 00593360 <__memcpy_entry+2c/3a0>
>>sp; 004171d1
>>ret_pc; 0200e5a0 <[qlogicpti]qlogicpti_intr_handler+120/1e0>
>>i1; 004fb840
>>i2; 005d36b0
>>i6; 00417291
>>i7; 0200e678 <[qlogicpti]qpti_intr+18/80>
Trace; 0200e678 <[qlogicpti]qpti_intr+18/80>
Trace; 0041f424
Trace; 004088f4
Trace; 0041a5fc
Trace; 00418068
Trace; 005f87d8
Trace; 00404638
Trace; Before first symbol
Code; 005935b8 <__memcpy_entry+284/3a0>
<_PC>:
Code; 005935b8 <__memcpy_entry+284/3a0>
0: c4 da 7f c0 unknown
Code; 005935bc <__memcpy_entry+288/3a0>
4: c6 da 7f c8 unknown
Code; 005935c0 <__memcpy_entry+28c/3a0>
8: 87 83 20 00 mov %o4, %asr3
Code; 005935c4 <__memcpy_entry+290/3a0> <=
c: c4 a2 3f c4 unknown <=
Code; 005935c8 <__memcpy_entry+294/3a0>
10: 8b 30 b0 20 unknown
Code; 005935cc <__memcpy_entry+298/3a0>
14: ca a2 3f c0 unknown
Code; 005935d0 <__memcpy_entry+29c/3a0>
18: c6 a2 3f cc unknown
Code; 005935d4 <__memcpy_entry+2a0/3a0>
1c: 9b 30 f0 20 unknown
Code; 005935d8 <__memcpy_entry+2a4/3a0>
20: da a2 3f c8 unknown
Kernel panic: Aiee, killing interrupt handler!
1 warning issued. Results may not be reliable.
Any ideas?
Re: sparc64 and ipsec anyone?
On Tue, Jan 28, 2003 at 06:26:14PM -0500, Chad Miller wrote: > > Well, Mips can be either Big- or Little-Endian. I've heard rumor of a > jumper on some motherboards to flip it. Of course, that sounds just > too cool to be true. ARM can run either way too - and it doesnt even need a jumper :O) -- Paul
Re: sparc64 and ipsec anyone?
On Tue, Jan 28, 2003 at 02:31:02PM +0100, Martin Rusko wrote: > > > Paul Hedderly wrote: > >On Mon, Jan 27, 2003 at 12:44:06PM -0800, Patrick Morris wrote: > > > >>In my experience, it's been unusable on anything big-endian. > > > > > >Nah many people use freeswan on sparc/alpha/PPC ... > > I was unable to compile freeswan on Sun Ultra1 (sparc64) due use of two Ahh.. > So maybe it is working on sparc32, but in my experience not on sparc64. Yea I guess so... -- Paul
Re: sparc64 and ipsec anyone?
On Mon, Jan 27, 2003 at 12:44:06PM -0800, Patrick Morris wrote: > In my experience, it's been unusable on anything big-endian. Nah many people use freeswan on sparc/alpha/PPC ... -- Paul > > Daniel van Eeden wrote: > > >Is it an sparc64 only problem or is ipsec also unusable on sparc32? > > > >Daniel van Eeden <[EMAIL PROTECTED]> > > > >Noah L. Meyerhans wrote: > > > >>On Sun, Jan 26, 2003 at 11:06:51AM +0100, Ulrich Wurst wrote: > >> > >>>Has anyone of you successfully managed to use ipsec on a sparc64 > >>>machine? > >>> > >> > >> > >>I've still never managed to get it working. I've basically given up on > >>freeswan on SPARC64. Hopefully the ipsec implementation that's going in > >>to Linux 2.6 will give better results. > >> > >>I don't remember the details of the freeswan problem, but it has > >>something to do with the way ioctls are implemented on sparc64. There's > >>some kind of translation or something that goes on in kernel space. But > >>that's only what I can recall from my vague memories of hearing it > >>explained; I've never actually looked at the code, and don't know my way > >>around the kernel source well enough to know where to look. > >> > >>noah > >> > > > > > > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact > [EMAIL PROTECTED] >
Re: 2.4.20 doesn't boot (watchdog reset)
On Fri, Jan 24, 2003 at 09:19:22AM +0100, Varga Daniel (QI/SSC) * wrote: > > I have initrd only as module, so I think it's not active by default, right? > Still, I have the same error as described above with 2.4.20. With initrd as > module in 2.4.17 it works. :-O Erm what use is initrd as a module? Are you sure it can be built as a module... -- Paul
Re: Debian for SPARC, and the "Front" key...
On Wed, Jan 22, 2003 at 11:03:15AM -, Hobbs, Richard wrote: So long as it generates a keycode, it should not be too hard to map it in various window managed to a window select/jump forward function. Debian has probably 30 odd window managers - you are bound to find at least one you like :O) -- Paul > Hello, > I have just obtained Debian 3.0 r1 for SPARC... but before removing > Solaris, i would like to know one thing... > > Does the "Front" key, on the Sun keyboards, work with the window manager > by default on this port of debian? > > If not, how easy is it to get it working? > > Hope someone can help!! :-) > > Thanks, > Richard. > > -- > Richard Hobbs > [EMAIL PROTECTED] > http://mongeese.co.uk | http://fishsponge.co.uk > > "There's only one way of life, and that's your own" - The Levellers > > _ > Send all your jokes to [EMAIL PROTECTED] !! > To subscribe, email: [EMAIL PROTECTED] > > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >
Re: DHCP with happy meal and/or lance
On Wed, Sep 25, 2002 at 03:14:25PM -0400, Clint Adams wrote: > > My both cards have the same MAC address, i've read somewhere that it is > > normal (both card have the station MAC address) but none of my cards > > This is usually a bad idea. You can fix this by setting > local-mac-address?=false in the prom. trueX or am I going mad? hmm. -- Paul
Re: 2.4 kernel debs / Anyone running 2.4 on a SS1000 ?
On Tue, Jun 25, 2002 at 01:33:41PM -0400, Rob Radez wrote: > I'm in the process of uploading kernel debs for 2.4.19-rc1. Same disclaimer > from last time applies. > > New site this time though: > http://osinvestor.com/sparc/ Fab - I'll give em a try. Anyone here running on an SS1000? I would _love_ to get 2.4 running on my SS1k :O( -- Paul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: kernel 2.4.18 and 2.4.19 for sparc
On Thu, May 30, 2002 at 09:44:41AM +0200, Turbo Fredriksson wrote: > > "Wilmer" == Wilmer van der Gaast <[EMAIL PROTECTED]> writes: > > Wilmer> Phil [EMAIL PROTECTED]@Tue, 28 May 2002 18:39:26 > Wilmer> -0500: > >> Back when 2.4.18 came out, I heard there was a problem where a > >> patch vital to Sparc didn't make it it. > >> > Wilmer> AFAIK this patch wasn't really Sparc-specific. The patch > Wilmer> was sort-of i386 specific, I'd say. 2.4.18 doesn't work on > Wilmer> anything but i386, I'm told. > > Wilmer> Try 2.4.19-pre8-ac5.. Alan Cox kernels are very good. > > Got the 2.4.19-pre8-ac5 patch, and tried to apply it on a standard > 2.4.18 kernel tree. I got A LOT of rejects! > > Alan's patches don't seem to be up to the usual standard... You DID apply 2.4.19-pre8 FIRST didn't you... -- Paul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Optical mouse?
On Wed, Jan 09, 2002 at 06:54:40AM +, Paul Hedderly wrote: > If you don't have the mat, there is a postscript file around that will > print a "mat" on a laser printer for you. I've got a copy around > somewhere - you 'll find it on the internet too. Let me know if you need > it. Ok - google just found it for me: http://www.geocrawler.com/archives/3/237/2001/10/0/6940477/ http://www.obsolyte.com/sunFAQ/sun3mousepad.html -- Paul
Re: Optical mouse?
On Wed, Jan 09, 2002 at 01:24:56AM -0500, Ben Collins wrote: > On Tue, Jan 08, 2002 at 08:19:26PM -0800, Alex Cavnar wrote: > > Greetings, > > > > I have a SparcStation 5, and I recently got a keyboard/ optical mouse > > pair from the guy I bought the system from. He said he could never get > > the mouse to work on the machine, so it could just be that it's > > non-functioning, but are there any tricks or techniques one must use to > > make the Sun optical mouse work on my machine? Note that a regular > > analog mouse works on my machine, so I know it's not my mouse port. Any > > help would be appreciated! > > Do you have the special mouse pad for it? It's metal and has some > red/blue gridlines, from what I remember. If you don't have the mat, there is a postscript file around that will print a "mat" on a laser printer for you. I've got a copy around somewhere - you 'll find it on the internet too. Let me know if you need it. -- Paul
X401(pre2-RC3) on Sparc hardware with a UK keyboard. Help!
How do I confiure X so I don't get this: (II) Keyboard "Generic Keyboard" handled by legacy driver (**) Option "Protocol" "busmouse" (**) Generic Mouse: Protocol: "busmouse" (**) Generic Mouse: Core Pointer (**) Option "Device" "/dev/sunmouse" (==) Generic Mouse: Buttons: 3 (II) XINPUT: Adding extended input device "Generic Mouse" (type: MOUSE) Couldn't load XKB keymap, falling back to pre-XKB keymap (which results in a keymap that thinks all keys are one place to the right of where they really are. Not useful... :O) At the moment I have: Section "InputDevice" Identifier "Generic Keyboard" Driver "keyboard" Option "CoreKeyboard" Option "XkbRules" "sun" Option "XkbModel" "type5_uk" Option "XkbLayout" "gb" EndSection But I have tried all sorts - basically rooting around in /etc/X11/xkb/* files trying to find tags that might do something useful... Any help much appreciated! -- Cheers, Paul
Are you running SMP linux on an SS1000/2000 ?
If so - could you tell me what OpenBoot PROM version you have? I am on 2.31 - and my machine won't start an SMP kernel... (Whatever the flavour.) -- Cheers, Paul
Sun SS1000 SMP booting problem
I have an old SS1000 here and it runs fine with a uniprocessor kernel. I've tried several. But when I try any SMP kernel it fails. I've tried cross-compiled, Debian main packages, locally compiled etc... What happens? Well here: SPARCserver 1000E, No Keyboard ROM Rev. 2.31, 512 MB memory installed, Serial #8264719. Ethernet address 8:0:20:7e:1c:f, Host ID: 807e1c0f. Type help for more information <#0> ok boot SILO boot: /notwork Uncompressing image... | Watchdog Reset <#0> ok notwork is one of the kernels I'm trying. Any idea what this is - why - and how to get around it? As you can see it's OpenBoot 2.31 - There are currently 2x85MHz processors in there - but I've changed the system board, the processors and the memory. (I have 4 full system boards... which I'd like to use :O) Oh I'm running unstable - not that makes a jot of difference since the kernel doesn't even fully uncompress. Please help me!!! -- Regards, Paul
