| Author |
Message |
Oliver S.
Guest
|
 Posted:
Fri Oct 14, 2005 12:15 am Post subject:
Re: Pentium M to become THE CPU |
|
|
| Quote: | you memory-pages that physically map to the memory attached to the CPU.
But I'm not aware of any OS that supports this processor-affine allocations.
It's pretty much standard on any NUMA aware OS (Linux, Solaris, Windows Server 2k3)
|
It isn't that easy as someone might think:
- At least for Windows Server 2003, there's no API to allocate processor local
memory. And for other systems you'd have to use such an API instead of stupid
malloc()ing.
- A thread will usually get migrated from one CPU to another sooner or later (if
it hasn't fixed CPU-affinity) and the OS might chose to re-schedule a thread on
the CPU the thead ran on the last time to try to help to (to^3 *g*) recycle the
working-sets of this thread in the cache-hierarchy of that CPU. So it won't help
much to allocate memory of a thread on that CPU the read was running on when it
was requesting the memory.
- Memory allocated by one thread might be used more by another thread running on
another CPU. |
|
| Back to top |
|
 |
Bill Todd
Guest
|
| Posted:
Fri Oct 14, 2005 5:36 am Post subject:
Re: Pentium M to become THE CPU |
|
|
Bill Davidsen wrote:
....
The P-M is really low power
| Quote: | compared to Opteron
|
Perhaps you haven't been paying attention recently: AMD just introduced
a 2.4 GHz mobile chip with a 35W power envelope (and given the slope of
their performance/W curve lately there's little reason to expect that to
be the best they can do this year, let alone next).
- bill |
|
| Back to top |
|
|
keith
Guest
|
| Posted:
Fri Oct 14, 2005 7:48 am Post subject:
Re: Pentium M to become THE CPU |
|
|
On Thu, 13 Oct 2005 23:35:25 +0000, Bill Davidsen wrote:
| Quote: | Ketil Malde wrote:
"Nathan Bates" <nathanbates99@yahoo.com> writes:
Pentium M has all the right ingredients for total world domination:
low power consumption, short pipeline stages, hi-performance.
Mediocre FP performance, few available motherboards, high price, no
SMP support?
For the price of a high end Pentium M, I can get a dual core AMD where
each core has equivalent integer performance and much better FP. Sure
Pentium M is attractive for some purposes, but total world domination
is still a way off, IMO.
And you can heat your house in the winter. The P-M is really low power
compared to Opteron, and of course P4 is in a class by itself for heat.
|
Don't I wish I could heat my house with the 150W my 18-month-old Opteron
144 (complete system, sans monitor) draws. Please!
| Quote: | The SMP and FP issues are supposedly being addressed soon, as will
EMT64, current advantage is power. AMD realized this and recently
offered a mobil chip to be more competitive, so I guess AMD saw the
need.
|
Sure, and met it. Intel hasn't met AMD64. Next!
| Quote: | I don't think the P-M is going to make everything else go away, but at
the moment the play is in low power, no matter what the O.P. thinks.
|
That's the current marketeering, sure. A couple of years ago it was GHz.
What benchmarket will they lose next?
| Quote: | I'd like to get a dual dual-core system, but my net will be a single
chip Intel-DC to be compatible with other things I support.
|
The Intel marketeering department thanks you.
--
Keith |
|
| Back to top |
|
|
keith
Guest
|
| Posted:
Fri Oct 14, 2005 7:52 am Post subject:
Re: Pentium M to become THE CPU |
|
|
On Thu, 13 Oct 2005 18:34:08 +0000, Scott Alfter wrote:
| Quote: | -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In article <pqask1dio0nuip9atb9p4h4j8k9027pl0h@4ax.com>,
Trent <none@dev.nul.pissoff> wrote:
On 12 Oct 2005 17:46:25 GMT salfter@salfter.diespammersdie.dyndns.org
(Scott Alfter) wrote in Message id:
434d4bf1$0$28781$9a6e19ea@news.newshosting.com>:
(One of these days, I'll build a controller for my beer fridges so I can
free up the Apple IIs that are currently running them (a IIGS on one and a
IIe on the other). To simplify the software-porting effort, it'll most
likely be built around a 6502, or something compatible with it. It's not
like monitoring the temperature and switching the compressor on and off
requires dual Opterons or something insane like that.)
Even a 6502 is overkill. Why not use a simple 8 pin device like this?
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2735
The only additional thing you'd need aside from a power source is a
transistor, a diode, and a relay. You can program the device with a
parallel port.
1) It doesn't appear to have an ability to enforce a minimum off-time. If
you repeatedly turn the compressor back on too soon after it has shut off,
that'll shorten its life. Choosing setpoints that are far-enough apart
might minimize this, but that would result in the temperature not being
as tightly-regulated as it could be.
|
Get a PIC.
| Quote: | 2) It doesn't appear to have a way to slowly ramp the temperature up/down.
If the fridge is at 50 degrees and you want it to go up to 70 for a
diacetyl rest and then down to 35 for lagering, you want those
temperature changes to be made slowly (at a rate of maybe 1 degree per
hour).
|
Get a PIC.
| Quote: | These are things for which some sort of microprocessor control is needed. A
6502 might be overkill, but it's what I know, so there's less time getting
up to speed with an unfamiliar instruction set. I'm currently using DS18B20
sensors controlled by Apple IIs through a little bit of custom hardware, so
I already have software for the 6502 that talks to 1-Wire devices. Porting
that to another 6502-based machine would take minimal effort.
|
Get a PIC, though I understand. I'd use an 8051. ;-)
| Quote: | (If anyone's interested, the 1-Wire software is the first link at
http://alfter.us/a2soft.shtml.)
I also have the IIs graphing the temperature for the past ~4 hours; this
functionality would most likely go away, as it's (more or less) a curiosity
that was relatively easy to implement.
|
Trivial, though I'd use X10 for the power control.
--
Keith |
|
| Back to top |
|
|
Zak
Guest
|
| Posted:
Fri Oct 14, 2005 8:15 am Post subject:
Re: Pentium M to become THE CPU |
|
|
Bill Davidsen wrote:
| Quote: | And you can heat your house in the winter. The P-M is really low power
compared to Opteron,
|
The strange thing is that a Turion is 'just' a tuned Opteron with the
process shifted for low power. Turion still uses more power, but from
what I've seen the run time for otherwise identical laptops differs by
10% with P-M.
The difference in CPU power is thus greater - OTOH it is a system that
I'm using so 10% it is.
Thomas |
|
| Back to top |
|
|
Anton Ertl
Guest
|
| Posted:
Fri Oct 14, 2005 12:32 pm Post subject:
Re: Pentium M to become THE CPU |
|
|
Andi Kleen <freitag@alancoxonachip.com> writes:
| Quote: | Don't underestimate the predictors. They are good at finding patterns
even sometimes suprising for a human.
|
Sure, certainly for branch predictors. But if they can predict a
pseudo-random sequence, the pseudo-random number generator was
obviously pretty bad. Unless, of course they predict the pattern from
the last time they saw it, but unless the benchmarks repeats the
patterns it uses for the subblocks, the pattern should be much too
large for any history buffer.
Now, for bplat <http://www.complang.tuwien.ac.at/anton/bplat/>, it
uses the low bits of a linear congruential RNG, which is not so great
(the bottom bits are much less random than the top bits). Each random
number is used at most 32 times (which should still be little enough
to overwhelm any history buffer).
I have also looked at lmbench-3.0-a5. You need to run "lat_mem_rd -t"
(undocumented) to get the non-strided behaviour. I do not understand
much of the source code, but there seem to be only #pages calls to
rand(); for large memory sizes, that's probably also enough to
overwhelm history buffers in stream buffers. What that benchmark also
does is thrash the TLB in addition to the stride predictor.
TLB misses could be the reason why you are seeing worse numbers on the
Pentium D than on the Athlon 64. E.g., on my Socket 754 Athlon 64
system, lat_mem_rd -t gives ~80ns (with TLB misses), a benchmark with
few TLB misses gives ~50ns, and lat_mem_rd (few TLB misses, and
predictable stride) gives 28ns.
- anton
--
M. Anton Ertl Some things have to be seen to be believed
anton@mips.complang.tuwien.ac.at Most things have to be believed to be seen
http://www.complang.tuwien.ac.at/anton/home.html |
|
| Back to top |
|
|
Andi Kleen
Guest
|
| Posted:
Fri Oct 14, 2005 2:48 pm Post subject:
Re: Pentium M to become THE CPU |
|
|
"Oliver S." <Follow.Me@gmx.net> writes:
| Quote: | - At least for Windows Server 2003, there's no API to allocate processor local
memory.
|
I don't know the details of Windows' NUMA support, but I would assume
it is just default for all allocations.
| Quote: | And for other systems you'd have to use such an API instead of stupid
malloc()ing.
|
At least on Linux malloc with large enough allocations will normally get local
memory.
-Andi |
|
| Back to top |
|
|
Ketil Malde
Guest
|
| Posted:
Fri Oct 14, 2005 4:15 pm Post subject:
Re: Pentium M to become THE CPU |
|
|
Bill Todd <billtodd@metrocast.net> writes:
| Quote: | Perhaps you haven't been paying attention recently: AMD just
introduced a 2.4 GHz mobile chip with a 35W power envelope
|
Correct me if I'm wrong, but it is intended to fit into a standard
(socket 939) motherboard¹? So if there is a demand for quiet PCs (in
addition to laptops), the market should be ready.
-k
¹ Of course, it'll be one of those with a big heatsink and fan on the
chipset, and a separate PSU for the two graphics cards :-)
--
If I haven't seen further, it is by standing in the footprints of giants |
|
| Back to top |
|
|
Oliver S.
Guest
|
| Posted:
Fri Oct 14, 2005 9:33 pm Post subject:
Re: Pentium M to become THE CPU |
|
|
| Quote: | And for other systems you'd have to use such an API instead of stupid
malloc()ing.
At least on Linux malloc with large enough allocations will normally
get local memory.
|
And what if a thread is migrated to another CPU? How can the OS estimate
what's more important: the cache working-set or the memory-distance? |
|
| Back to top |
|
|
Bill Sommerfeld
Guest
|
| Posted:
Fri Oct 14, 2005 10:33 pm Post subject:
Re: VT on portable systems (Was: Pentium M to become THE CPU |
|
|
Oliver S. wrote:
| Quote: | Even the upcoming dual-core incarnation of the Pentium-M won't
be capable of x86-64 and it seems that this version even miss VT
technology; but both are not strong arguments for Notebook-systems.
|
On the latter point, I know of a bunch of people using virtualization
technologies on their notebooks (vmware, xen, user-mode linux, etc.,)
because it means they only have to carry one computer around and it
matters less which one it is.
and developers are increasingly using portable systems.
maybe I just have lots of weird friends.
- Bill |
|
| Back to top |
|
|
Bill Todd
Guest
|
| Posted:
Sat Oct 15, 2005 12:15 am Post subject:
Re: Pentium M to become THE CPU |
|
|
Oliver S. wrote:
| Quote: | And for other systems you'd have to use such an API instead of stupid
malloc()ing.
At least on Linux malloc with large enough allocations will normally
get local memory.
And what if a thread is migrated to another CPU? How can the OS estimate
what's more important: the cache working-set or the memory-distance?
|
That's one of the things that the ability to set hard per-thread
processor affinity (to supplement the default soft affinity) is for -
features supported in Win2K if not earlier.
- bill |
|
| Back to top |
|
|
Scott Alfter
Guest
|
| Posted:
Sat Oct 15, 2005 12:15 am Post subject:
Re: Pentium M to become THE CPU |
|
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In article <pan.2005.10.14.02.52.27.519738@att.bizzzz>,
keith <krw@att.bizzzz> wrote:
| Quote: | On Thu, 13 Oct 2005 18:34:08 +0000, Scott Alfter wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In article <pqask1dio0nuip9atb9p4h4j8k9027pl0h@4ax.com>,
Trent <none@dev.nul.pissoff> wrote:
On 12 Oct 2005 17:46:25 GMT salfter@salfter.diespammersdie.dyndns.org
(Scott Alfter) wrote in Message id:
434d4bf1$0$28781$9a6e19ea@news.newshosting.com>:
(One of these days, I'll build a controller for my beer fridges so I can
free up the Apple IIs that are currently running them (a IIGS on one and a
IIe on the other). To simplify the software-porting effort, it'll most
likely be built around a 6502, or something compatible with it. It's not
like monitoring the temperature and switching the compressor on and off
requires dual Opterons or something insane like that.)
Even a 6502 is overkill. Why not use a simple 8 pin device like this?
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2735
The only additional thing you'd need aside from a power source is a
transistor, a diode, and a relay. You can program the device with a
parallel port.
1) It doesn't appear to have an ability to enforce a minimum off-time. If
you repeatedly turn the compressor back on too soon after it has shut off,
that'll shorten its life. Choosing setpoints that are far-enough apart
might minimize this, but that would result in the temperature not being
as tightly-regulated as it could be.
Get a PIC.
2) It doesn't appear to have a way to slowly ramp the temperature up/down.
If the fridge is at 50 degrees and you want it to go up to 70 for a
diacetyl rest and then down to 35 for lagering, you want those
temperature changes to be made slowly (at a rate of maybe 1 degree per
hour).
Get a PIC.
|
I already have the tools and the know-how (acquired over the past 20 years)
to code for the 6502. I know squat about PICs, and don't have any
assemblers and/or compilers for them.
_/_
/ v \ Scott Alfter (remove the obvious to send mail)
(IIGS( http://alfter.us/ Top-posting!
\_^_/ rm -rf /bin/laden >What's the most annoying thing on Usenet?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
iD8DBQFDUEhxVgTKos01OwkRApkGAJ0USW7RS0KH3IdoBtSXPyITVp3F9QCgvVVx
aw/VGUgYqpjW0adJT0onUUg=
=DJ3k
-----END PGP SIGNATURE----- |
|
| Back to top |
|
|
keith
Guest
|
| Posted:
Sat Oct 15, 2005 5:57 am Post subject:
Re: Pentium M to become THE CPU |
|
|
On Fri, 14 Oct 2005 16:04:42 -0400, Bill Todd wrote:
| Quote: | Oliver S. wrote:
And for other systems you'd have to use such an API instead of stupid
malloc()ing.
At least on Linux malloc with large enough allocations will normally
get local memory.
And what if a thread is migrated to another CPU? How can the OS estimate
what's more important: the cache working-set or the memory-distance?
That's one of the things that the ability to set hard per-thread
processor affinity (to supplement the default soft affinity) is for -
features supported in Win2K if not earlier.
|
Or page it to the other processor (perhaps something inbetween).
Processor affinity issues have been dealt with for many years. There
really shouldn't be too much new here, other than perhaps the penalty for
not doing it right.
--
Keith |
|
| Back to top |
|
|
keith
Guest
|
| Posted:
Sat Oct 15, 2005 5:59 am Post subject:
Re: Pentium M to become THE CPU |
|
|
On Fri, 14 Oct 2005 13:48:56 +0200, Ketil Malde wrote:
| Quote: | Bill Todd <billtodd@metrocast.net> writes:
Perhaps you haven't been paying attention recently: AMD just
introduced a 2.4 GHz mobile chip with a 35W power envelope
Correct me if I'm wrong, but it is intended to fit into a standard
(socket 939) motherboard¹? So if there is a demand for quiet PCs (in
addition to laptops), the market should be ready.
|
I think there is a demand, but I'm not sure the market (infrastructure) is
ready.
| Quote: |
-k
¹ Of course, it'll be one of those with a big heatsink and fan on the
chipset, and a separate PSU for the two graphics cards :-)
|
;-)
--
Keith |
|
| Back to top |
|
|
keith
Guest
|
| Posted:
Sat Oct 15, 2005 6:01 am Post subject:
Re: Pentium M to become THE CPU |
|
|
On Sat, 15 Oct 2005 00:05:32 +0000, Scott Alfter wrote:
| Quote: | -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In article <pan.2005.10.14.02.52.27.519738@att.bizzzz>,
keith <krw@att.bizzzz> wrote:
On Thu, 13 Oct 2005 18:34:08 +0000, Scott Alfter wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In article <pqask1dio0nuip9atb9p4h4j8k9027pl0h@4ax.com>,
Trent <none@dev.nul.pissoff> wrote:
On 12 Oct 2005 17:46:25 GMT salfter@salfter.diespammersdie.dyndns.org
(Scott Alfter) wrote in Message id:
434d4bf1$0$28781$9a6e19ea@news.newshosting.com>:
(One of these days, I'll build a controller for my beer fridges so I can
free up the Apple IIs that are currently running them (a IIGS on one and a
IIe on the other). To simplify the software-porting effort, it'll most
likely be built around a 6502, or something compatible with it. It's not
like monitoring the temperature and switching the compressor on and off
requires dual Opterons or something insane like that.)
Even a 6502 is overkill. Why not use a simple 8 pin device like this?
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2735
The only additional thing you'd need aside from a power source is a
transistor, a diode, and a relay. You can program the device with a
parallel port.
1) It doesn't appear to have an ability to enforce a minimum off-time. If
you repeatedly turn the compressor back on too soon after it has shut off,
that'll shorten its life. Choosing setpoints that are far-enough apart
might minimize this, but that would result in the temperature not being
as tightly-regulated as it could be.
Get a PIC.
2) It doesn't appear to have a way to slowly ramp the temperature up/down.
If the fridge is at 50 degrees and you want it to go up to 70 for a
diacetyl rest and then down to 35 for lagering, you want those
temperature changes to be made slowly (at a rate of maybe 1 degree per
hour).
Get a PIC.
I already have the tools and the know-how (acquired over the past 20 years)
to code for the 6502. I know squat about PICs, and don't have any
assemblers and/or compilers for them.
|
Look for 'em. They're free. Disclaimer: I've never done PICs either
but have done several 8051 designs. If I had to do some of them again, it
would be a PIC. 6502? You *must* be kidding!
--
Keith |
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in