Re: SMP detection
- From: Michal Mertl <mime@xxxxxxxxxxxx>
- Date: Thu, 31 Aug 2006 22:24:01 +0200
backyard píše v čt 31. 08. 2006 v 07:45 -0700:
--- Michal Mertl <mime@xxxxxxxxxxxx> wrote:
Skylar Thompson wrote:http://lists.freebsd.org/mailman/listinfo/freebsd-questions
Jordi Carrillo wrote:<backyard1454-bsd@xxxxxxxxx>:
2006/8/30, backyard
3GHz
--- Jordi Carrillo <jordilin@xxxxxxxxx> wrote:
I've read that SMP should be disabled for
performance issues (I did not know
that before installing freebsd). I have a P4
itwith hyperthreading
technology. I have the SMP-GENERIC kernel and
that ok?,only launches one cpu. So,
I've decided to disable SMP from BIOS. Is
withoutknowing that I have a
Smp enabled kernel? or should I install one
_______________________________________________smp? If so, is there a
way to install one already precompiled?
Thanks in advance
--
http://jordilin.wordpress.com
freebsd-questions@xxxxxxxxxxx mailing list
don'tTo unsubscribe, send any mail to
"freebsd-questions-unsubscribe@xxxxxxxxxxx"
if the system runs with one cpu now and you
then youenable smp with HT with the sysctl variable
recompilingshould be ok. If your not doing SMP then
thingsthe kernel for single processor mode will make
comerun a little quicker because the SMP code won't
securityinto play.
with HT disabling in FreeBSD is more for the
processissues about a potential exploit whereby one
information ofin one pipe can access the priveledged
sharea process in another pipe because the two cores
To myone processor cache and thus one cache table.
shouldknowledge this hasn't been exploited yet.
If you just install the generic kernel you it
a:be only the uniprocessor one. I would just do
KERNCONF=GENERIC
cd /usr/src && make buildworld && make
KERNCONF=GENERIC buildkernel && make
haven'tinstallkernel
as opposed to a binary version assuming you
Iupdated yet you won't have to install world but
tree tobelieve it must have the build in the source
differencebuild a kernel. On your P4 though the
troublebetween SMP and uniproc may not be worth the
made. onbecause I don't think much of a gain would be
hackersa P1 a much different story...
if you aren't concerned with bad users or
sysctlhitting the box I would just enable HT with the
at all,variable. This will not make things run slower
recall.just (in theory) less secure, which is why the
veriable was created in the first place as I
updateIf you are concerned I would wait until you
GENERIC/CUSTOMyour system and then just build a
kernel, I suppose therekernel without the SMP option set.
-brian
I will disable smp from bios. If I have a smp
smp kernel onlywill
be no problem after all. Would that be ok?
The problem with having SMP enabled is that the
as well as gkrellm (indetects one
cpu and the system monitor only features one cpu
system monitor shows theLinux it shows two cpus). When compiling the
other 50%?cpu at
a maximum of 50%, so what's going on with the
loader.conf does not solvewriting machdep.hlt_logical_cpus to 2 in
anything.
I believe FreeBSD uses the other logical CPU tohandle hardware
interrupts, which can still help perormance. Youcan check dmesg to see
how it's actually handling it.
No! Kernel threads (e.g. handling interrupts) aren't
that much different
to normal processes.
Logical CPUs on a single HTT capable CPU share most
of the CPU logic,
especially all the external stuff (handling
interrupts). Scheduling
handling of interrupts on the "secondary/logical"
core wouldn't
probably help performance at all (if that is at all
possible).
When FreeBSD sees logical CPUs it means HTT is
either enabled in BIOS or
that disabling HTT in BIOS does not hide the CPUs to
FreeBSD (bug in
BIOS/FreeBSD).
Until you enable scheduler to schedule tasks to HTT
cores (with
machdep.hyperthreading_allowed=1 in loader.conf)
(disabled by default
due to mentioned security/performance reasons)
machine won't utilize the
logical HTT CPUs. Therefore total CPU utilization
won't be more than
50%, because there are the (unused) logical CPUs
which don't get
scheduled tasks.
are you sure about this???
Almost sure but can't check at the moment. I believe there were problems
when all HTT CPUs weren't launched sometimes (when HTT wasn't disabled
with BIOS), so the logical CPU cores are started and fully visible but
only run the idle kernel thread (are 100% idle).
I would have figured the scheduler wouldn't see the
other core at all without this option set and so it
wouldn't be used in calculating load at all. 50% on a
compile is fairly normal from my experience. I don't
have too much experience with HT as I always opt for
true SMP so I can't speak with authority on the
matter.
but if
top
isn't showing CPU 1 or 0 next to a process then it
isn't computing the load on multiple cores... Also if
dmesg |grep cpu
doesn't show application cpu1 (and on through all your
cores)... launched then the system isn't looking at
the HT core at all.
-brian
_______________________________________________
freebsd-questions@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscribe@xxxxxxxxxxx"
- References:
- Re: SMP detection
- From: backyard
- Re: SMP detection
- Prev by Date: Re: The future of NetBSD
- Next by Date: Re: Mounting Linux Procfs at Boot
- Previous by thread: Re: SMP detection
- Next by thread: Re: SMP detection
- Index(es):
Relevant Pages
|
|