Contributed by jason on from the how-many-sparcs-does-it-take-to-screw-in-a-lightbulb dept.
Mark Kettenis (kettenis@) writes in to announce support for the UltraSPARC T1:
Yesterday I committed the last bit of code to support machines with Sun's UltraSPARC T1 CPUs. Below is a dmesg for the SPARC Enterprise T1000, and although other machines have not been tested yet, machines like the SPARC Enterprise T2000 and Sun Blade T6300 are expected to work too. As you can see, we support SMP right from the start.
All this is included in a standard OpenBSD/sparc64 kernel. Snapshots with UltraSPARC T1 support are now available at ftp.openbsd.org and its mirrors. It would be great if people could try these snapshots on a few more UltraSPARC T1 machines.
LDOM support is not yet complete; I'm still working on drivers for virtual network interfaces and virtual disks. But domains with access to real hardware should work fine.
UltraSPARC T2 machines are not yet supported, but hopefully that'll change soon.
Cheers,
Mark
console is /virtual-devices@100/console@1 Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. Copyright (c) 1995-2008 OpenBSD. All rights reserved. http://www.OpenBSD.org OpenBSD 4.3-current (GENERIC.MP) #3: Thu Apr 3 21:13:13 CEST 2008 kettenis@gershwin.sibelius.xs4all.nl:/usr/src/sys/arch/sparc64/compile/GENERIC.MP real mem = 17045651456 (16256MB) avail mem = 16679108608 (15906MB) mainbus0 at root: SPARC Enterprise T1000 cpu0 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu1 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu2 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu3 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu4 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu5 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu6 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu7 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu8 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu9 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu10 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu11 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu12 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu13 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu14 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu15 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu16 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu17 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu18 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu19 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu20 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu21 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu22 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu23 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu24 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu25 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu26 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu27 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu28 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu29 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu30 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz cpu31 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz vbus0 at mainbus0 "channel-devices" at vbus0 not configured "flashprom" at vbus0 not configured vcons0 at vbus0 vrtc0 at vbus0 "fma" at vbus0 not configured "sunvts" at vbus0 not configured "sunmc" at vbus0 not configured "explorer" at vbus0 not configured "led" at vbus0 not configured "flashupdate" at vbus0 not configured "ncp" at vbus0 not configured vpci0 at mainbus0: bus 2 to 2, dvma map 80000000-ffffffff pci0 at vpci0 ebus0 at mainbus0 com0 at ebus0 addr c2c000-c2c007 ivec 0xa: st16650, 32 byte fifo vpci1 at mainbus0: bus 2 to 4, dvma map 80000000-ffffffff pci1 at vpci1 ppb0 at pci1 dev 0 function 0 "ServerWorks PCIE-PCIX" rev 0xb3 pci2 at ppb0 bus 3 bge0 at pci2 dev 4 function 0 "Broadcom BCM5714" rev 0xa2, BCM5715 A1 (0x9001): ivec 0x7d4, address 00:14:4f:83:47:7a brgphy0 at bge0 phy 1: BCM5714 10/100/1000baseT PHY, rev. 0 bge1 at pci2 dev 4 function 1 "Broadcom BCM5714" rev 0xa2, BCM5715 A1 (0x9001): ivec 0x7d5, address 00:14:4f:83:47:7b brgphy1 at bge1 phy 1: BCM5714 10/100/1000baseT PHY, rev. 0 ppb1 at pci2 dev 8 function 0 "ServerWorks HT-1000 PCIX" rev 0xb3 pci3 at ppb1 bus 4 bge2 at pci3 dev 1 function 0 "Broadcom BCM5704C" rev 0x10, BCM5704 B0 (0x2100): ivec 0x7c2, address 00:14:4f:83:47:7c brgphy2 at bge2 phy 1: BCM5704 10/100/1000baseT PHY, rev. 0 bge3 at pci3 dev 1 function 1 "Broadcom BCM5704C" rev 0x10, BCM5704 B0 (0x2100): ivec 0x7c1, address 00:14:4f:83:47:7d brgphy3 at bge3 phy 1: BCM5704 10/100/1000baseT PHY, rev. 0 mpi0 at pci3 dev 2 function 0 "Symbios Logic SAS1064" rev 0x02: ivec 0x7c0 scsibus0 at mpi0: 63 targets sd0 at scsibus0 targ 0 lun 0: <FUJITSU, MAY2073RCSUN72G, 0501> SCSI2 0/direct fixed sd0: 70007MB, 14100 cyl, 24 head, 423 sec, 512 bytes/sec, 143374738 sec total sd1 at scsibus0 targ 1 lun 0: <FUJITSU, MAY2073RCSUN72G, 0501> SCSI2 0/direct fixed sd1: 70007MB, 14100 cyl, 24 head, 423 sec, 512 bytes/sec, 143374738 sec total softraid0 at root bootpath: /pci@7c0,0/pci@0,0/pci@8,0/scsi@2,0/disk@1,0 root on sd1a swap on sd1b dump on sd1b
(Comments are closed)
By Anonymous Coward (63.227.10.89) on
By Anonymous Coward (213.118.57.79) on
What's performance like? Are these beauties really as fast as they're claimed to be?
Comments
By cameronsto (24.30.46.212) on http://cameronstokes.com/
>
>
> What's performance like? Are these beauties really as fast as they're claimed to be?
I can't speak to performance with OpenBSD but we use a set of T2000s running Solaris for our B2B applications at work and they are smoking fast. We're looking at using them for all of our http and application servers in the future.
Its great to see OpenBSD support for them.
-cameron
Comments
By Anonymous Coward (124.171.133.60) on
> >
> >
> > What's performance like? Are these beauties really as fast as they're claimed to be?
>
> I can't speak to performance with OpenBSD but we use a set of T2000s running Solaris for our B2B applications at work and they are smoking fast. We're looking at using them for all of our http and application servers in the future.
>
> Its great to see OpenBSD support for them.
>
> -cameron
We now use T1000s as load-balanced web servers (Solaris) as a nice upgrade from a single V480. They absolutely cruised through our busiest peak of the year, load hardly showed up as a blip. These things scream through web traffic. Great machines in 1RU.
I unfortunately can't speak for OpenBSD performance either, but it's fantastic to see this support and I'll be doing my best to boot an OpenBSD CD next time one is offline and sling over a dmesg.
Congrats to all involved, love your work.
By Anthony (68.145.117.155) on
>
>
> What's performance like? Are these beauties really as fast as they're claimed to be?
They're great but I'm not sure how well they'd do for most OpenBSD machines, which more often run as firewalls/routers. This is mostly kernel work, and mostly single-threaded on OpenBSD. T1's thrive on threads.
Comments
By Terrell Prude' Jr. (151.188.18.42) tprude@cmosnetworks.com (this is a spamtrap address) on http://www.cmosnetworks.com/
>
That means that any multithreaded app, such as a busy Web server, would love a box like this. Likewise for anything running, say, Nagios or NMIS, like we do at work. Both of these apps are very, very multithreaded.
We have Nagios running on a 4-core Opteron, and NMIS running on a dual P4-Hyperthreading box. The mere *thought* of either app running on an T1-based OpenBSD box...the most secure OS available to mere mortals...mmmm....
--TP
Comments
By Otto Moerbeek (otto) on http://www.drijf.net
> >
>
> That means that any multithreaded app, such as a busy Web server, would love a box like this. Likewise for anything running, say, Nagios or NMIS, like we do at work. Both of these apps are very, very multithreaded.
>
> We have Nagios running on a 4-core Opteron, and NMIS running on a dual P4-Hyperthreading box. The mere *thought* of either app running on an T1-based OpenBSD box...the most secure OS available to mere mortals...mmmm....
>
> --TP
Hate to spoil your party, but the current implementation of userland threads in OpenBSD won't benefit at all.
Applications running many processes may benefit, though it remains to be seen how large the overhead of SMP coordination (locking etc) will be. So don't hold your breath.
Comments
By Terrell Prude' Jr. (151.188.18.43) tprude@cmosnetworks.com (this is a spamtrap address) on http://www.cmosnetworks.com/
>
> Applications running many processes may benefit, though it remains to be seen how large the overhead of SMP coordination (locking etc) will be. So don't hold your breath.
>
Oh, I know about the giant lock problem, if that's what you mean. Linux used to have that issue, too, but that kernel's dev team fixed that pretty well. That's one of the big reasons we run NMIS and Nagios on GNU/Linux today.
I hear the FreeBSD folks have also done lots of good work in this area in their latest Release 7.0. From my readings of their mailing lists on the subject, they've been working for a long time to get this right, and that they're now seeing it come to fruition.
And so, ultimately, will the OpenBSD team do something similar. *That* team of master hackers? Heh...it's just a matter of time.
--TP
By Niall O'Higgins (69.12.154.245) niallo@openbsd.org on http://www.niallohiggins.com
This should be an excellent stress test of SMP. Forgetting about performance entirely, if this thing can run solidly on bsd.mp, that is very good news. If it prompts sparc64 bsd.mp bug fixes, quite possibly this will carry over to i386 and amd64 bsd.mp's and improve those also.
Of course it is unclear how "practical" a 32 CPU SMP machine is with OpenBSD's current giant lock implementation. For IO bound work loads (which the T1 is ostensibly targeted at) it could well be sub-optimal. But this should certainly push toward improvements - more fine-grained locking or whatever.
In conclusion, great news, and fantastic work by Kettenis.
Comments
By Anonymous Coward (213.118.57.79) on
By Pete (195.1.147.126) on
SPARC Enterprise T2000, No Keyboard
Copyright 2007 Sun Microsystems, Inc. All rights reserved.
OpenBoot 4.25.8, 16376 MB memory available, Serial #78505224.
Ethernet address 0:14:4f:ad:e5:8, Host ID: 84ade508.
Rebooting with command: boot
Boot device: disk File and args:
OpenBSD IEEE 1275 Bootblock 1.1
..>> OpenBSD BOOT 1.2
Trying bsd...
Booting /pci@780/pci@0/pci@9/scsi@0/disk@0,0:a/bsd
5188528@0x1000000+169704@0x1800000+4024600@0x18296e8
symbols @ 0xfeeba280 68+315384+190821 start=0x1000000
[ using 506920 bytes of bsd ELF symbol table ]
console is /virtual-devices@100/console@1
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.
Copyright (c) 1995-2008 OpenBSD. All rights reserved. http://www.OpenBSD.org
OpenBSD 4.3-current (GENERIC) #1574: Tue Apr 8 17:21:53 MDT 2008
deraadt@sparc64.openbsd.org:/usr/src/sys/arch/sparc64/compile/GENERIC
real mem = 17171480576 (16376MB)
avail mem = 16807075840 (16028MB)
mainbus0 at root: SPARC Enterprise T2000
cpu0 at mainbus0: SUNW,UltraSPARC-T1 (rev 0.0) @ 1000 MHz
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
"SUNW,UltraSPARC-T1" at mainbus0 not configured
vbus0 at mainbus0
"nvram" at vbus0 not configured
vcons0 at vbus0
"ncp" at vbus0 not configured
vrtc0 at vbus0
"loop" at vbus0 not configured
"loop" at vbus0 not configured
"echo" at vbus0 not configured
"fma" at vbus0 not configured
"sunvts" at vbus0 not configured
"sunmc" at vbus0 not configured
"explorer" at vbus0 not configured
"led" at vbus0 not configured
"flashupdate" at vbus0 not configured
"flashprom" at vbus0 not configured
"system-management" at vbus0 not configured
vpci0 at mainbus0: bus 2 to 7, dvma map 80000000-ffffffff
pci0 at vpci0
ppb0 at pci0 dev 0 function 0 "PLX 8532 PCIE" rev 0xbc
pci1 at ppb0 bus 3
ppb1 at pci1 dev 1 function 0 "PLX 8532 PCIE" rev 0xbc
pci2 at ppb1 bus 4
em0 at pci2 dev 0 function 0 "Intel PRO/1000 PT (82571EB)" rev 0x06: cannot find i/o space
em0: Allocation of PCI resources failed
em1 at pci2 dev 0 function 1 "Intel PRO/1000 PT (82571EB)" rev 0x06: ivec 0x796, address 00:14:4f:ad:e5:09
ppb2 at pci1 dev 2 function 0 "PLX 8532 PCIE" rev 0xbc
pci3 at ppb2 bus 5
ppb3 at pci1 dev 8 function 0 "PLX 8532 PCIE" rev 0xbc: ivec 0x794
pci4 at ppb3 bus 6
ppb4 at pci1 dev 9 function 0 "PLX 8532 PCIE" rev 0xbc
pci5 at ppb4 bus 7
mpi0 at pci5 dev 0 function 0 "Symbios Logic SAS1064E" rev 0x02: ivec 0x795
scsibus0 at mpi0: 63 targets
sd0 at scsibus0 targ 0 lun 0: <FUJITSU, MBB2147RCSUN146G, 0505> SCSI2 0/direct fixed
sd0: 140009MB, 14089 cyl, 24 head, 848 sec, 512 bytes/sec, 286739329 sec total
sd1 at scsibus0 targ 1 lun 0: <FUJITSU, MBB2147RCSUN146G, 0505> SCSI2 0/direct fixed
sd1: 140009MB, 14089 cyl, 24 head, 848 sec, 512 bytes/sec, 286739329 sec total
vpci1 at mainbus0: bus 2 to 9, dvma map 80000000-ffffffff
pci6 at vpci1
ppb5 at pci6 dev 0 function 0 "PLX 8532 PCIE" rev 0xbc
pci7 at ppb5 bus 3
ppb6 at pci7 dev 1 function 0 "PLX 8532 PCIE" rev 0xbc
pci8 at ppb6 bus 4
ppb7 at pci8 dev 0 function 0 "Intel 41210 PCIE-PCIX" rev 0x09
pci9 at ppb7 bus 5
ebus0 at pci9 dev 2 function 0 "Acer Labs M1533 ISA" rev 0x00
com0 at ebus0 addr 3f8-3ff ivec 0x2: ns16550a, 16 byte fifo
ohci0 at pci9 dev 5 function 0 "Acer Labs M5237 USB" rev 0x03: ivec 0x7c1, version 1.0, legacy support
ohci1 at pci9 dev 6 function 0 "Acer Labs M5237 USB" rev 0x03: ivec 0x7c3, version 1.0, legacy support
pciide0 at pci9 dev 8 function 0 "Acer Labs M5229 UDMA IDE" rev 0xc4: DMA, channel 0 configured to native-PCI, channel 1 configured to native-PCI
pciide0: using ivec 0x7c4 for native-PCI interrupt
atapiscsi0 at pciide0 channel 0 drive 0
scsibus1 at atapiscsi0: 2 targets
cd0 at scsibus1 targ 0 lun 0: <MATSHITA, CD-RW CW-8124, DZ15> SCSI0 5/cdrom removable
cd0(pciide0:0:0): using PIO mode 4, Ultra-DMA mode 2
pciide0: channel 1 disabled (no drives)
usb0 at ohci0: USB revision 1.0
uhub0 at usb0 "Acer Labs OHCI root hub" rev 1.00/1.00 addr 1
usb1 at ohci1: USB revision 1.0
uhub1 at usb1 "Acer Labs OHCI root hub" rev 1.00/1.00 addr 1
ppb8 at pci8 dev 0 function 2 "Intel 41210 PCIE-PCIX" rev 0x09
pci10 at ppb8 bus 6
ppb9 at pci7 dev 2 function 0 "PLX 8532 PCIE" rev 0xbc
pci11 at ppb9 bus 7
em2 at pci11 dev 0 function 0 "Intel PRO/1000 PT (82571EB)" rev 0x06: ivec 0x7d6, address 00:14:4f:ad:e5:0a
em3 at pci11 dev 0 function 1 "Intel PRO/1000 PT (82571EB)" rev 0x06: ivec 0x7d7, address 00:14:4f:ad:e5:0b
ppb10 at pci7 dev 8 function 0 "PLX 8532 PCIE" rev 0xbc: ivec 0x7d4
pci12 at ppb10 bus 8
ppb11 at pci7 dev 9 function 0 "PLX 8532 PCIE" rev 0xbc: ivec 0x7d6
pci13 at ppb11 bus 9
uhub2 at uhub1 port 1 "Atmel UHB124 hub" rev 1.10/3.00 addr 2
softraid0 at root
bootpath: /pci@780,0/pci@0,0/pci@9,0/scsi@0,0/disk@0,0
root on sd0a swap on sd0b dump on sd0b
Automatic boot in progress: starting file system checks.
/dev/rsd0a: file system is clean; not checking
setting tty flags
starting network
DHCPDISCOVER on em1 to 255.255.255.255 port 67 interval 2
DHCPDISCOVER on em1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on em1 to 255.255.255.255 port 67 interval 4
DHCPOFFER from 000.254.127.1
DHCPREQUEST on em1 to 255.255.255.255 port 67
DHCPACK from 000.254.127.1
bound to 000.254.127.213 -- renewal in 1800 seconds.
starting system logger
starting initial daemons: ntpd.
savecore: no core dump
checking quotas: done.
building ps databases: kvm dev.
clearing /tmp
starting pre-securelevel daemons:.
setting kernel security level: kern.securelevel: 0 -> 1
creating runtime link editor directory cache.
preserving editor files
ssh-keygen: generating new DSA host key... done.
ssh-keygen: generating new RSA host key... done.
ssh-keygen: generating new RSA1 host key... done.
openssl: generating new isakmpd RSA key... done.
starting network daemons: sendmail inetd sshd.
starting local daemons:.
standard daemons: cron.
Wed Apr 9 13:29:13 CEST 2008
OpenBSD/sparc64 (t2000-spare.000) (console)
Comments
By cameron (165.2.186.10) on http://cameronstokes.com/
Awesome. Sun just announced dual-socket versions of those too.
Sun SPARC Enterprise T5140 Server
Sun SPARC Enterprise T5240 Server
-cameron
By Anonymous Coward (204.80.187.9) on
>
try GENERIC.MP
Comments
By Brian Poole (65.117.234.99) on
GENERIC.MP works. As does the onboard mpt0. And the em0 (which was broken) also works with Mark's latest commits. So far it's been stable for me, although pretty slow (high system CPU use, relatively little remaining time for user) when loaded up with processes.