Status of Ross hyperSPARC on linux-2.4.x and linux-2.6.x?

Status of Ross hyperSPARC on linux-2.4.x and linux-2.6.x?

[christian]

Just a little ping for me (and anyone else).. what's the status of the
linux kernels 2.4 and 2.6 for running Ross' hyperSPARC processors?

This is what I have (running up to date Aurora SPARC Linux 1.0)

cpu		: ROSS HyperSparc RT625 or RT626
fpu		: ROSS HyperSparc combined IU/FPU
promlib		: Version 3 Revision 2
prom		: 2.25
type		: sun4m
ncpus probed	: 4
ncpus active	: 4
Cpu0Bogo	: 124.92
Cpu1Bogo	: 125.33
Cpu2Bogo	: 125.33
Cpu3Bogo	: 125.33
MMU type	: ROSS HyperSparc
contexts	: 4096
nocache total	: 5242880
nocache used	: 159232
CPU0		: online
CPU1		: online
CPU2		: online
CPU3		: online

and dmesg reports 

PROMLIB: Sun Boot Prom Version 3 Revision 2
Linux version 2.4.23-0.1sparcsmp (root@fw.j-son.org) (gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)) #1 SMP Mon Jan 12 16:20:26 CET 2004
ARCH: SUN4M
TYPE: Sun4m SparcStation10/20
Ethernet address: 8:0:20:74:b0:90
SRMMU: Using VAC size of 1048576 bytes, line size 64 bytes.
Boot time fixup v1.6. 4/Mar/98 Jakub Jelinek (jj@ultra.linux.cz). Patching kernel for srmmu[ROSS HyperSparc]/iommu
65413MB HIGHMEM available.
On node 0 totalpages: 113727
zone(0): 49152 pages.
zone(1): 0 pages.
zone(2): 65413 pages.
Found CPU 0 <nodeÿd74150,mid=8>
Found CPU 1 <nodeÿd743a0,mid=9>
Found CPU 2 <nodeÿd745f0,mid\x10>
Found CPU 3 <nodeÿd74840,mid\x11>
Found 4 CPU prom device tree node(s).
Power off control detected.
Kernel command line: root=/dev/sda2 ro
Console: colour dummy device 80x25
Calibrating delay loop... 124.92 BogoMIPS
Memory: 444300k available (1732k kernel code, 352k data, 344k init, 261652k highmem) [f0000000,1bf85000]
Dentry cache hash table entries: 32768 (order: 6, 262144 bytes)
Inode cache hash table entries: 16384 (order: 5, 131072 bytes)
Mount cache hash table entries: 512 (order: 0, 4096 bytes)
Buffer cache hash table entries: 8192 (order: 3, 32768 bytes)
Page-cache hash table entries: 65536 (order: 6, 262144 bytes)
POSIX conformance testing by UNIFIX
Entering SMP Mode...
Starting CPU 1 at f021890c
Calibrating delay loop... 125.33 BogoMIPS
Starting CPU 2 at f0218918
Calibrating delay loop... 125.33 BogoMIPS
Starting CPU 3 at f0218924
Calibrating delay loop... 125.33 BogoMIPS
Total of 4 Processors activated (500.94 BogoMIPS).
Waiting on wait_init_idle (map = 0xe)
All processors have done init_idle
IOMMU: impl 1 vers 3 page table at f0dc0000 of size 262144 bytes
sbus0: Clock 25.0 MHz
dma0: Revision 2 
dma1: Revision 2 
Sparc Zilog8530 serial driver version 1.68.2.2
Sun Mouse-Systems mouse driver version 1.00
tty00 at 0xffede004 (irq = 44) is a Zilog8530
tty01 at 0xffede000 (irq = 44) is a Zilog8530
tty02 at 0xffedb004 (irq = 44) is a Zilog8530
tty03 at 0xffedb000 (irq = 44) is a Zilog8530
Sun TYPE 5 keyboard detected without keyclick
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
Starting kswapd
allocated 32 pages and 32 bhs reserved for the highmem bounces
VFS: Disk quotas vdquot_6.5.1
ioremap: done with statics, switching to malloc
Console: switching to colour frame buffer device 128x54
fb0: cgsix at e.20000000 TEC Rev 4 CPU sparc Rev b [TGX+]
pty: 1024 Unix98 ptys configured
Floppy drive(s): fd0 is 1.44M
FDC 0 is a National Semiconductor PC87306
RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
SCSI subsystem driver Revision: 1.00
kmod: failed to exec /sbin/modprobe -s -k scsi_hostadapter, errno = 2
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP, IGMP
IP: routing cache hash table of 128 buckets, 6Kbytes
TCP: Hash tables configured (established 2048 bind 10922)
Linux IP multicast router 0.06 plus PIM-SM
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
RAMDISK: Compressed image found at block 0
Freeing initrd memory: 553k freed
VFS: Mounted root (ext2 filesystem).
esp0: IRQ 36 SCSI ID 7 Clk 40MHz CCYC%000 CCF=8 TOut 167 NCR53C9XF(espfast)
ESP: Total of 1 ESP hosts found, 1 actually in use.
scsi0 : Sparc ESP100A-FAST
  Vendor: IBM       Model: DDRS-34560D       Rev: DC1B
  Type:   Direct-Access                      ANSI SCSI revision: 02
  Vendor: TOSHIBA   Model: XM-4101TASUNSLCD  Rev: 3424
  Type:   CD-ROM                             ANSI SCSI revision: 02
Attached scsi disk sda at scsi0, channel 0, id 3, lun 0
esp0: target 3 [period 100ns offset 15 10.00MHz FAST SCSI-II]
SCSI device sda: 8925000 512-byte hdwr sectors (4570 MB)
Partition check:
 sda: sda1 sda2 sda3 sda4
Journalled Block Device driver loaded
kjournald starting.  Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.
Freeing unused kernel memory: 344k freed
Adding Swap: 917224k swap-space (priority -1)
EXT3 FS 2.4-0.9.19, 19 August 2002 on sd(8,2), internal journal
md: md driver 0.90.0 MAX_MD_DEVS%6, MD_SB_DISKS'
kjournald starting.  Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on sd(8,1), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
sunlance.c:v2.01 08/Nov/01 Miguel de Icaza (miguel@nuclecu.unam.mx)
eth0: LANCE 08:00:20:74:b0:90 
eth0: using auto-carrier-detection.
eth0: Carrier Lost, trying TPE
LANCE unopened after 100 ticks, csr0\033.
dcsr 461010
sunmouse: Successfully adjusted to 1200 baud.

TIA,

/ChJ

Re: Status of Ross hyperSPARC on linux-2.4.x and linux-2.6.x?

[Tom 'spot' Callaway]

On Tue, 2004-01-13 at 13:42, christian@j-son.org wrote:
> Just a little ping for me (and anyone else).. what's the status of the
> linux kernels 2.4 and 2.6 for running Ross' hyperSPARC processors?

The code to deal with HyperSPARCs hasn't changed between 2.4 and 2.6.
There are definitely lurking bugs in the HyperSPARC implementation
(anaconda triggers a hard lock in Aurora). The kernel will boot, but its
probably not overly stable.

I honestly don't think that 2.2 (or older) has these problems resolved,
but rather that they're not as easily exposed.

Peter Jones will undoubtedly correct me if I said anything wrong here,
as he knows more about this than me.

~spot
---
Tom "spot" Callaway <tcallawa(a)redhat*com> LCA, RHCE 
Red Hat Sales Engineer || Aurora SPARC Linux Project Leader

"The author's mathematical treatment of the conception of purpose is
novel and highly ingenious, but heretical and, so far as the present
social order is concerned, dangerous and potentially subversive. Not to
be published." -- Aldous Huxley's "Brave New World"

Re: Status of Ross hyperSPARC on linux-2.4.x and linux-2.6.x?

[christian]

On Tue, Jan 13, 2004 at 08:42:06PM +0100, christian@j-son.org wrote:
> Just a little ping for me (and anyone else).. what's the status of the
> linux kernels 2.4 and 2.6 for running Ross' hyperSPARC processors?
> 
> This is what I have (running up to date Aurora SPARC Linux 1.0)
> 
> cpu		: ROSS HyperSparc RT625 or RT626
> fpu		: ROSS HyperSparc combined IU/FPU
> promlib		: Version 3 Revision 2
> prom		: 2.25
> type		: sun4m
> ncpus probed	: 4
> ncpus active	: 4
> Cpu0Bogo	: 124.92
> Cpu1Bogo	: 125.33
> Cpu2Bogo	: 125.33
> Cpu3Bogo	: 125.33
> MMU type	: ROSS HyperSparc
> contexts	: 4096
> nocache total	: 5242880
> nocache used	: 159232
> CPU0		: online
> CPU1		: online
> CPU2		: online
> CPU3		: online

I get a lot of system cpu here and little user....

(null)  9:04pm  up 31 min,  2 users,  load average: 6.18, 5.81, 3.66
57 processes: 53 sleeping, 4 running, 0 zombie, 0 stopped
CPU0 states: 23.0% user, 76.1% system,  0.0% nice,  0.1% idle
CPU1 states: 34.1% user, 65.0% system,  0.0% nice,  0.1% idle
CPU2 states: 30.0% user, 69.1% system,  0.0% nice,  0.1% idle
CPU3 states: 13.1% user, 86.0% system,  0.0% nice,  0.1% idle
Mem:   444644K av,   98268K used,  346376K free,       0K shrd,    7860K buff
Swap:  917224K av,       0K used,  917224K free                   61848K cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
 5338 chj       17   0  1436 1436   832 R    73.7  0.3   0:02 as
 5330 chj       15   0  1156 1152   928 R    37.8  0.2   0:02 top
 1542 chj       11   0  1228 1228  1136 S    30.0  0.2   1:49 sh
 5351 chj       12   0   152  152   128 R    16.8  0.0   0:00 cat
 5350 chj        7   0  1028 1028   960 S    14.8  0.2   0:00 sh
 5306 chj        9   0  1028 1028   960 S     9.8  0.2   0:01 sh
 4869 chj        9   0  6636 6636  3220 S     5.7  1.4   0:37 emacs
   15 root       9   0     0    0     0 SW    3.2  0.0   0:17 kjournald
 1543 chj        7   0  1072 1072  1060 R     0.8  0.2   0:34 sh
    1 root       9   0   564  564   496 S     0.0  0.1   0:02 init
    2 root       9   0     0    0     0 SW    0.0  0.0   0:02 keventd
    3 root      19  19     0    0     0 SWN   0.0  0.0   0:00 ksoftirqd_CPU0
    4 root      19  19     0    0     0 SWN   0.0  0.0   0:00 ksoftirqd_CPU1
    5 root      19  19     0    0     0 SWN   0.0  0.0   0:00 ksoftirqd_CPU2

what's up?

/ChJ

Re: Status of Ross hyperSPARC on linux-2.4.x and linux-2.6.x?

[Wiktor Wodecki]

[-- Attachment #1: Type: text/plain, Size: 474 bytes --]

On Tue, Jan 13, 2004 at 08:42:06PM +0100, christian@j-son.org wrote:
> Just a little ping for me (and anyone else).. what's the status of the
> linux kernels 2.4 and 2.6 for running Ross' hyperSPARC processors?

I have a quad hyperSparc ss10 running 2.4.25-pre4* and it is running
pretty well - From time to time it hangs, tho. The system watchdog
interrupts and the system goes back to OBP.

* with Ben's recent smp_mb additions

-- 
Regards,

Wiktor Wodecki

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Status of Ross hyperSPARC on linux-2.4.x and linux-2.6.x?

[christian]

On Tue, Jan 13, 2004 at 09:06:41PM +0100, christian@j-son.org wrote:
> On Tue, Jan 13, 2004 at 08:42:06PM +0100, christian@j-son.org wrote:
> > Just a little ping for me (and anyone else).. what's the status of the
> > linux kernels 2.4 and 2.6 for running Ross' hyperSPARC processors?
> > 
> > This is what I have (running up to date Aurora SPARC Linux 1.0)
> > 
> > cpu		: ROSS HyperSparc RT625 or RT626
> > fpu		: ROSS HyperSparc combined IU/FPU
> > promlib		: Version 3 Revision 2
> > prom		: 2.25
> > type		: sun4m
> > ncpus probed	: 4
> > ncpus active	: 4
> > Cpu0Bogo	: 124.92
> > Cpu1Bogo	: 125.33
> > Cpu2Bogo	: 125.33
> > Cpu3Bogo	: 125.33
> > MMU type	: ROSS HyperSparc
> > contexts	: 4096
> > nocache total	: 5242880
> > nocache used	: 159232
> > CPU0		: online
> > CPU1		: online
> > CPU2		: online
> > CPU3		: online

and I got spin locks...

spin_lock (f0269000) CPU#3 stuck at f0020a34, owner PC(f007124c): CPU(0)
spin_lock (f01f83d0) CPU#2 stuck at f0024b7c, owner PC(f0024b7c): CPU(1)

if anyone care for additional info, pls let me know. this was on
Aurora SPARC Linux running spot's kernel-2.4.23-0.1sparc.src.rpm built
on my own SS20 system.

Cheers,

/ChJ

Re: Status of Ross hyperSPARC on linux-2.4.x and linux-2.6.x?

[Keith M Wesolowski]

On Tue, Jan 13, 2004 at 08:42:06PM +0100, christian@j-son.org wrote:

> Just a little ping for me (and anyone else).. what's the status of the
> linux kernels 2.4 and 2.6 for running Ross' hyperSPARC processors?

2.4 is more or less hosed.  I don't plan to fix it, no time.  2.6
boots UP and seems to work as well as SuperSPARC given the limited
testing I've done so far.  It may be possible later to backport
whatever changes are made in 2.6, but frankly I don't recommend to use
2.4.  If you want to help development, use 2.6.  If you need first
rate stability, use 2.2.

-- 
Keith M Wesolowski <wesolows@foobazco.org> http://foobazco.org/~wesolows
------(( Project Foobazco Coordinator and Network Administrator ))------
	"May Buddha bless all stubborn people!"
				-- Uliassutai Karakorum Blake