headers
Donald Lee | 18 May 2013 00:18

... and a panic... again

I got another one of these yesterday.  It seems to happen under network
load while there is some heavy duty computation going on.

I'll file a PR for starters - done - kern/47830.

What can I do to help get this resolved?

May 16 07:27:45 mercy syslogd[151]: restart
May 16 07:27:45 mercy /netbsd: panic: pmap_pte_spill: victim p-pte (0x1ffcb68) has no pvo entry!
May 16 07:27:45 mercy /netbsd: cpu0: Begin traceback...
May 16 07:27:45 mercy /netbsd: 0xe1d3fe10: at panic+0x4c
May 16 07:27:45 mercy /netbsd: 0xe1d3fe50: at pmap_pte_spill+0x4c0  
May 16 07:27:45 mercy /netbsd: 0xe1d3fe90: at trap+0xa60
May 16 07:27:45 mercy /netbsd: 0xe1d3ff20: user ISI trap by 0x183d4e4: srr1=0x4200d032
May 16 07:27:45 mercy /netbsd:            r1=0xffffb2b0 cr=0x24000044 xer=0 ctr=0xfdd70cb8
May 16 07:27:45 mercy /netbsd: cpu0: End traceback... 
May 16 07:27:45 mercy /netbsd: dumpsys: TBD
May 16 07:27:45 mercy /netbsd: wd2: flush cache command didn't complete
May 16 07:27:45 mercy /netbsd: wd1: flush cache command didn't complete
May 16 07:27:45 mercy /netbsd: rebooting 

Thanks,

-dgl-
Permalink | Reply | 0 comments |
headers
SOS Permis | 12 May 2013 16:48
Picon

Sauver votre permis

Bonjour

Si vous etes detenteur d'un permis de conduire, lisez attentivement cette page.

Une contravention c'est plus qu'un montant a payer : ca nuit a votre permis de conduire et a vos assurances.
Bien entendu, vous pouvez demander les services dispendieux d'un avocat pour vous aider a contester
votre contravention ou vous pouvez des maintenant apprendre a contester vous-meme une contravention
sans l'aide d'un avocat avec les techniques 100% legal de SOS PERMIS.

"Certains avocats ne veulent pas que je devoile mes techniques de contestation par crainte de perdre des clients"

C'est possible de contester une contravention et C'EST VOTRE DROIT. Vous trouverez toutes les
informations pour y arriver dans ce nouveau guide pratique.

http://contravention.p.ht/

Cordialement

Vous recevez cet email car vous etes inscrit a notre programme relationnel.
Vos coordonnees n'ont pas ete transmises a l'annonceur de ce message.
Vous pouvez vous desinscrire a tout moment.:http://www.lqs-fr.com/diffusion/abonnement.php?M=1118117&C=038cd5026792a56d862fdaab65f17945&L=131&N=13199&cint=0
Conformement a la loi Informatique et Liberte du 06 Janvier 1978, vous disposez d'un droit d'acces, de
rectification et d'opposition aux donnees vous concernant.
Permalink | Reply | 0 comments |
headers
Sam Jerry | 25 Apr 2013 00:34
Picon

Order Enq!


Hi Have a nice day,

Can you make this design that our big customers here are interested in buying and selling in our country.

We have patented the mold designs, and will like to know if you can produce to specification for our market only.

For the safety of our patented  sample design, please view the attached and download the mold design and give
us your monthly production rate, quote, and CIF or FOB Sydney.

Will you be at the Canton Fair? Please provide your booth and full contact details.

Kindest Regards,

Sam Jerry.
WIKA Australia Pty Ltd
Sydney Head Office
Unit K, 10-16 South Street
Rydalmere NSW 2116
ABN 49 004 465 936
http://www.wika.com.au/
Permalink | Reply | 0 comments |
headers
Valery Ushakov | 22 Apr 2013 02:55
Picon

Problem with gdb breakpoints

I was debugging a problem in mail/mutt on macppc and encountered a
problem with gdb.  I have zero ppc clue so I could use some help from
the audience before I file a PR.

The bug in mutt looked like classic big-endian vs. %d vs. 64-bit
value: "Reading ..." progress updates when opening a mailbox would
print percentages like 2064384% 

To catch the offender I did

  (gdb) break mutt_curses_message
  Breakpoint 2 at 0x181c30c: file curs_lib.c, line 272.

but the breakpoint was never hit.

I walked into that function with stepi's:

Dump of assembler code for function mutt_curses_message:
   0x0181c2d8 <+0>:     stwu    r1,-160(r1)
   0x0181c2dc <+4>:     mflr    r0
   0x0181c2e0 <+8>:     stw     r0,164(r1)
   0x0181c2e4 <+12>:    stw     r31,156(r1)
   0x0181c2e8 <+16>:    mr      r31,r1
   0x0181c2ec <+20>:    stw     r4,44(r31)
   0x0181c2f0 <+24>:    stw     r5,48(r31)
   0x0181c2f4 <+28>:    stw     r6,52(r31)
   0x0181c2f8 <+32>:    stw     r7,56(r31)
   0x0181c2fc <+36>:    stw     r8,60(r31)
   0x0181c300 <+40>:    stw     r9,64(r31)
   0x0181c304 <+44>:    stw     r10,68(r31)
(Continue reading)

Permalink | Reply | 0 comments |
headers
Phileas Fogg | 20 Apr 2013 12:02
Picon

[PATCH] build.sh release is broken on macppc


Hi,

i tries to build a NetBSD-current release for my G5 and
stumbled upon the following bug. The build.sh stops with the message
that it doesn't know how to build pdisk in
distrib/macppc/floppies/ramdisk. I figured out that the problem lies in
the missing path to pdisk in list.

My patch to fix the problem is attached.

Regards
Attachment (macppc-ramdisk-pdisk-fix.patch): text/x-patch, 693 bytes
Permalink | Reply | 0 comments |
headers
Phileas Fogg | 13 Apr 2013 22:11
Picon

[PATCH] PMAC_G5: Add identity mapping for msgbuf


Hi,

i found another problem on PowerMac G5.

This line of code

__syncicache((void *)msgbuf_paddr, round_page(MSGBUFSIZE));

in file sys/arch/macppc/macppc/machdep.c
causes a DSI trap in kernel on PowerMac G5 when i try to reboot it.
The reason is that the msgbuf is allocated at a high address
which is > SEGMENT_LENGTH and therefore the msgbuf is not
identity mapped.

On my PMAC G5 the physical address of the msgbuf is 0x7f5ed000
which is outside of the segment0. Thefore, we have to map it too.
The patch is attached.

Regards
Attachment (ofwoea_machdep.c.patch): text/x-patch, 592 bytes
Permalink | Reply | 0 comments |
headers
Phileas Fogg | 13 Apr 2013 14:13
Picon

SMU support for PowerMac G5 11,2


Hi,

at length i implemented the initial support of SMU and sensors for 
PowerMac G5.
Now it's possible to control fan RPM and see current temperature of the 
CPUs too.
At boot all fans are at the maximum RPM which makes it impossible to
work on G5 due to loud noise made by fans. With SMU we can control it and
set to values which make fans noiseless.
I implemented the necessary drivers with syctl and envsys support.

We also need to send "quisce" command to OFW at boot else
it's not possible to communicate with SMU. It just ignores my SMU 
commands in this case. Sending this command solves the issue.
I spend several days figuring it out, thanks to FreeBSD i could fix it 
;) Patches for quisce are attached.

Quastion:
----------
I'm not very familiar with envsys but is it possible to
make it control fan speed automatically depending on the current 
temperature ? Or is it better to control fan speed in the kernel space, 
like
Linux or FreeBSD kernels do it. I would appreciate any pointers here.
Currently i'm doing it with a shell script which runs periodically.
One odd thing i noticed is that if i set e.g. all fan speeds to 1500
but then do not read fan rpm back for a while, SMU automatically 
increases all fan speeds. If i read fan rpms periodically then it remains
at the speed which i configured last. Very odd.
(Continue reading)

Permalink | Reply | 7 comments |
headers
Donald Lee | 12 Apr 2013 15:00

More panics - NetBSD 6.0.1 stock

I have been seeing this panic on a regular basis.  My NetBDS 4.0
machine has been up now for over 1200 days. (Beige G3)

Has anyone else seen this?

-dgl-

Mar 29 06:45:36 mercy syslogd[151]: restart
Mar 29 06:45:36 mercy /netbsd: panic: pmap_pte_spill: victim p-pte (0x1ffd040) has no pvo entry!
Mar 29 06:45:36 mercy /netbsd: cpu0: Begin traceback...
Mar 29 06:45:36 mercy /netbsd: 0xe1ff4e10: at panic+0x4c
Mar 29 06:45:36 mercy /netbsd: 0xe1ff4e50: at pmap_pte_spill+0x4c0
Mar 29 06:45:36 mercy /netbsd: 0xe1ff4e90: at trap+0x560
Mar 29 06:45:36 mercy /netbsd: 0xe1ff4f20: user DSI read trap  <at>  0xfdd0e0e8 by 0xfdfe6ee4: srr1=0x200f
032  
Mar 29 06:45:36 mercy /netbsd:            r1=0xffffaf70 cr=0x24028024 xer=0x20000000 ctr=0 dsisr=0x40
000000 
Mar 29 06:45:36 mercy /netbsd: cpu0: End traceback????e=^[^Msys: TBD
Mar 29 06:45:36 mercy /netbsd: rebooting   

---

Apr  7 18:32:49 mercy syslogd[151]: restart
Apr  7 18:32:49 mercy /netbsd: panic: pmap_pte_spill: victim p-pte (0x1fff7d0) has no pvo entry!
Apr  7 18:32:49 mercy /netbsd: cpu0: Begin traceback...
Apr  7 18:32:49 mercy /netbsd: 0xe2183e10: at panic+0x4c
Apr  7 18:32:49 mercy /netbsd: 0xe2183e50: at pmap_pte_spill+0x4c0
Apr  7 18:32:49 mercy /netbsd: 0xe2183e90: at trap+0x560
Apr  7 18:32:49 mercy /netbsd: 0xe2183f20: user DSI read trap  <at>  0xfc0302c4 by 0xfdfe6f5c: srr1=0x200f
032
(Continue reading)

Permalink | Reply | 0 comments |
headers
Apna-Loan | 9 Apr 2013 21:40
Picon
Favicon

Apply for loan <at> 2%...

Please Contact Us With This Email: apnapaisaloan.com12 <at> hotmail.com

Apna Paisa Loan Company , ? Are you in any financial mess or do you 

need a loan to start up your own business? at 2% rate? ; Email 

:apnapaisaloan.com12 <at> hotmail.com

(1) Full Names:
(2) State/Country:
(3)Amount needed as loan):
(4)Loan duration:
(5)Cell-Phone number:

Mr. Harsh Roongta
Permalink | Reply | 0 comments |
headers
Mark Fischer | 8 Apr 2013 23:19

(unknown)


We offer all types of LOAN  <at>  3% annual rate. To apply,  COMPOSE A NEW MESSAGE: E mail: firstoverseafinancecenter <at> gmail.com<mailto:firstoverseafinancecenter <at> gmail.com>
Permalink | Reply | 0 comments |
headers
Phileas Fogg | 6 Apr 2013 23:38
Picon

NetBSD-current booted from internal HDD on PowerMac G5 11,2

Hi,

i managed to boot NetBSD-current from the internal HDD on my
PowerMac G5 11,2.

The loader ofwboot hat to be patched a little and
it has problems with gzipped kernels but boots non-gzipped kernels
just fine ;)

The ServerWorks K2 SATA driver had to be patched too.
Thanks to Linux, FreeBSD and OpenBSD i could fix minor problems
in the driver and it works perfectly now.

bootpath: /ht/pci <at> 9/k2-sata-root/k2-sata <at> 0/disk <at> 0:0/netbsd

g5-netbsd# mount
/dev/wd0a on / type ffs (local)
kernfs on /kern type kernfs (local)

g5-netbsd# df -h
Filesystem         Size       Used      Avail %Cap Mounted on
/dev/wd0a           34G       554M        32G   1% /
kernfs             1.0K       1.0K         0B 100% /kern

g5-netbsd# pdisk /dev/wd0c
Edit /dev/wd0c -
Command (? for help): p

Partition map (with 512 byte blocks) on '/dev/wd0c'
  #:                type name       length   base      ( size )
(Continue reading)

Permalink | Reply | 0 comments |

Gmane