Slashdot Mirror
FreeBSD Looking for People with Lots of RAM
drdink writes "A few weeks ago, PAE (Physical Address Extension)
support was added to FreeBSD 5-CURRENT. This
allows memory above 4GB to be used normally by the kernel and userland
on the x86 platform. Jake Burkholder, the man
behind PAE, is now looking for users to help
him test this new feature. In his message to
the freebsd-current mailing list, Jake describes
the current caveats to PAE and also says
'We'd like this feature to be solid for
5.1-RELEASE, so I'm hoping there are people out
there with systems with more than 4G of ram that
are willing to test it.' This, along with other features
make FreeBSD 5-STABLE look very promising."
Also, if anyone is interested in a machine that can play at this level, check out auction depot. I'm not associated with them, other than having gotten my LXR dirt cheap (not as cheap as the link, needless to say) there. I can't say it's a sound investment, but if you want a toy you need a winch to get upstairs, you might find this fun.
I've had this sig for three days.
Freebsd does SMP quit well. They have/are redoing the whole SMP system. It was slated for 5.0 but i don't know if it did or did not make it in. When finished Freebsd will have and extremely good SMP, if not the best.
So what are you crunching with that thing?
Windows 2000 Datacenter supports PAE.
Beware, Nugget is watching... See?
I don't know if you could ever call it finished, but no they won't have "extremely good SMP". It will be OK though. The thing is they don't have the resources to do testing.
If you look at Linux 2.5 for example, just about the entire time it has been worked on, people with 32 processor POWER4 machines with 256GB ram, 32 way IA32, etc. have been running benchmarks and optimising and posting results.
This guy is having a hard time finding a > 6GB box to test with...
Now Linux 2.6 will have "good SMP". Not the best, good though. FreeBSD will probably scale OK to a few processors, don't kid yourself it will have "the best" SMP.
For those who are curious about what is new in -CURRENT compared to 4-STABLE, you can read the 5.0-RELEASE release notes for the bits that were new at the time of 5.0-RELEASE. More has come since.
Beware, Nugget is watching... See?
Beware, Nugget is watching... See?
As others have noted, Windows NT 5.0^H^H^H^H^H^H2000 also supports it.
SMPng is in 5.0, but there's still some work to be done. Work is in progress to free major parts of the kernel from needing the Giant kernel lock. You can see that much needs to be done on that link - the only driver listed as "SMPng locked" is the aac SCSI driver. Incidentally, that driver saw a speedup of up to 20% when it was freed from Giant. Once this work is finished, FBSD will really fly. Could be a long time waiting, though.
In your comment, you say:
Your assertions about Release and Stable are incorrect. -STABLE is merely a CVS branch which is considered stable. This includes RELENG_4 (4-STABLE), RELENG_3 (3-STABLE),Beware, Nugget is watching... See?
FreeBSD supports Hyperthreading in 5.0-CURRENT. There is a sysctl variable called "machdep.hlt_cpus". You can use this variable to control which logical CPUs should be taken out of the idle loop and used by the kernel. This, of course, requires a kernel built with the APIC_IO and SMP kernel options. Lacking a SMP system, I haven't tested this. This is just what I see on the mailing lists and in CVSWeb
Beware, Nugget is watching... See?
Why not run the real thing then? BSD is where linux was years ago so I fail to see how your not going to get "frustrated" with BSD.
1) The linux emulation layer on freebsd is quite thin: it's not unusual to see apps run nearly as fast (within a few percent when comparing total running times) under emulation.
2) The freebsd system is cleaner than any linux distribution. It's easier to maintain, and many would claim more secure, most would claim more stable.
I would rather run an app under linux emulation on freebsd than to run it under a real linux distribution.
We've had this system for a couple years:
:)
bash-2.03$ uname -a ; prtconf | more
SunOS largo 5.8 Generic_108528-14 sun4u sparc SUNW,Ultra-Enterprise
System Configuration: Sun Microsystems sun4u
Memory size: 10240 Megabytes
bash-2.03$ psrinfo
0 on-line since 03/10/03 13:25:03
1 on-line since 03/10/03 13:25:07
4 on-line since 03/10/03 13:25:07
5 on-line since 03/10/03 13:25:07
8 on-line since 03/10/03 13:25:07
9 on-line since 03/10/03 13:25:07
10 on-line since 03/10/03 13:25:07
11 on-line since 03/10/03 13:25:07
12 on-line since 03/10/03 13:25:07
13 on-line since 03/10/03 13:25:07
Look ma.. no PAE.
bash-3.00$ uname -a
SunOS panda 5.10 Generic sun4u sparc SUNW,Ultra-2
FreeBSD 5.0 is not -Current its -Release. There is a difference. Its not just a snapshot like the previous poster who has no clue mentioned. Its stable. Just not server 24x7 verified stable. Like I pointed out before only FreeBSD 4.0 is officially -Stable. All other 4.x releases are -Release versions. If you do not believe me you can go to the FreeBSD website and look for yourself 4.1x-4.8 are -Release versions. Yet many servers use them without problems.
The majority of FreeBSD servers out on the web use -Release and not -Stable. Only sites like Yahoo use -Stable or FreeBsd.com use the -stable release.
I admit I may not run a critical server on -Release but for non critical workstation and even server use its perfectly fine. Its just that there are many changes to the kernel in 5.0 and the development team wants to be absolutely sure that all the bugs are worked out before somone like yahoo installs it and shit hits the fan. Like the documentation said 5.0 -Release is not for everyone. But for the last month and a half it has finally left the -Current snapshot and into -Release if you have been reading slashdot.
Why would the FreeBSD team call something -Release and not have it ready to be released?
Sadly I looked at the FreeBSD handbook and found no mention of what -Release means. It is deffinetly no longer considered -Current anymore.
I would say its probably release-candidate equilivant in the Microsoft world when describing Windows products.
http://saveie6.com/
Okay, if you insist:
You're ignorant.
Does that make you feel better?
I thought Intel's HT was basically emulating dual CPUs, and it was transparent to the OS. Why would the OS need special support for it?
On a conventional SMP system, each CPU can handle exactly the same processing tasks. The second 'virtual' CPU on a HT CPU is created using the idle exectution units on the single die.
To (over)simplfy, if the main CPU is working on a series of integer calculations, the virtual CPU will have free floating-point units. In theory, you could treat the two CPUs as identical, but you would end up with a bottleneck as they would both be waiting for the same execution units.
An HT aware instruction scheduler would make sure that instructions to be executed on the same execution units were not sent to virtual CPUs on the same die, resulting in a much higher instruction throughput.
I am TheRaven on Soylent News
Years ago, there was the "LIM" (Lotus-Intel-Microsoft) for adding more than 640KB of RAM to a PC, by "windowing in" a section of RAM in a certain area.
It seems that, 20 years later, we're back to doing essentially the same thing.
No, this system doesn't work like that.
Intel processors since the Pentium have supported a system that allows you to use a larger page size than standard so that you can have more physical address space. You specify the start address of each page as 24 bits which are assumed to align to a 4K boundary which gives you 4M*4K = 16Gb of physical RAM. Each page is 2Mb in length. You can mix 4K and 2Mb pages in the same system, although not in the same quarter of the process adress space. So you get more actual physical memory, although each process is limited to 4G at once (whereas with LIM EMS the entire system was limited to 640K + 64K of 'banked' memory)
You can upgrade Apples pretty easily these days. This includes the "older" G3s.
Ever hear of the PCI bus? Or AGP?
The dangers of knowledge trigger emotional distress in human beings.
USB is wire compatible with PS/2. Take apart a USB->PS/2 converter sometime: it's just wires, no IC's. Newer motherboards may have the PS/2 jacks connected directly into the USB subsystem.
Most modern motherboards will let you swap PS/2 keyboards. I've unplugged mine and plugged it back in quite a few times without any problems. The last time I had such a problem was on my 486 SX25.
Beware, Nugget is watching... See?