On Fri, Jun 27, 2003, Brooks Davis wrote:
> On Fri, Jun 27, 2003 at 06:39:12PM -0500, Glenn Johnson wrote:
> > Thanks. I had read the smp manual page. I know _how_ to enable HTT; I
> > was wondering whether I _should_ enable it. It seems the answer is that
> > it is not beneficial in its current
Glenn Johnson wrote:
> Thanks. I had read the smp manual page. I know _how_ to enable HTT; I
> was wondering whether I _should_ enable it. It seems the answer is that
> it is not beneficial in its current state because the scheduler does not
> yet differentiate between physical and logical proce
In order to give some values to compare: I was recently
running an P4 3.06 GHz with Hyperthreading (actually it
was Linux, but I will repeat the tests with current
when I have time for such games :)
Running ubench 0.32 on this HTT enabled machine (i865PE)
showed some interesting details when acces
On Fri, Jun 27, 2003 at 06:39:12PM -0500, Glenn Johnson wrote:
> Thanks. I had read the smp manual page. I know _how_ to enable HTT; I
> was wondering whether I _should_ enable it. It seems the answer is that
> it is not beneficial in its current state because the scheduler does not
> yet differ
On Friday, June 27, 2003, at 05:46 PM, Doug White wrote:
On Fri, 27 Jun 2003, Glenn Johnson wrote:
I have a P4 processor on order that will support hyperthreading. I
was
wondering what the general opinion is on enabling HTT for FreeBSD-5
(current).
Thanks for any input.
He didn't ask how... h
On Fri, Jun 27, 2003 at 03:46:47PM -0700, Doug White wrote:
> On Fri, 27 Jun 2003, Glenn Johnson wrote:
>
> > I have a P4 processor on order that will support hyperthreading.
> > I was wondering what the general opinion is on enabling HTT for
> > FreeBSD-5 (current).
> >
> > Thanks for any input.
On Fri, 27 Jun 2003, Glenn Johnson wrote:
> I have a P4 processor on order that will support hyperthreading. I was
> wondering what the general opinion is on enabling HTT for FreeBSD-5
> (current).
>
> Thanks for any input.
man 4 smp
See the machdep.hlt_logical_cpus sysctl.
--
Doug White
On 26-Feb-2003 Trish Lynch wrote:
> On Wed, 26 Feb 2003, John Baldwin wrote:
>
>>
>> On 26-Feb-2003 Slawek Zak wrote:
>> > Hi,
>> >
>> > I'm in the process of installing 5.0-RELEASE-p3 on a Dell PowerEdge
>> > 2600 server. It has two HyperThreading Xeon 2.4Ghz processors, but
>> > HyperThreading
Juli Mallett wrote:
> > > > Any thoughts? I attach full dmesg and mptable output. Kernel
> > > > config too.
> > >
> > > 5.0 doesn't support HyperThreading. The upcoming 5.1 and 4.8 releases
> > > will support it.
> >
> > it doesn't? I have a 5.0-R box that surely does support HTT.
>
> Your BIOS
On Wed, Feb 26, 2003 at 02:05:51PM -0500, John Baldwin wrote:
> > I'm in the process of installing 5.0-RELEASE-p3 on a Dell PowerEdge
> > 2600 server. It has two HyperThreading Xeon 2.4Ghz processors, but
> > HyperThreading is not working in FreeBSD. I guess that four
> > processors should be visib
* De: Trish Lynch <[EMAIL PROTECTED]> [ Data: 2003-02-26 ]
[ Subjecte: RE: HyperThreading not working? ]
> On Wed, 26 Feb 2003, John Baldwin wrote:
>
> >
> > On 26-Feb-2003 Slawek Zak wrote:
> > > Hi,
> > >
> > > I'm in the
On Wed, 26 Feb 2003, John Baldwin wrote:
>
> On 26-Feb-2003 Slawek Zak wrote:
> > Hi,
> >
> > I'm in the process of installing 5.0-RELEASE-p3 on a Dell PowerEdge
> > 2600 server. It has two HyperThreading Xeon 2.4Ghz processors, but
> > HyperThreading is not working in FreeBSD. I guess that four
>
On 26-Feb-2003 Slawek Zak wrote:
> Hi,
>
> I'm in the process of installing 5.0-RELEASE-p3 on a Dell PowerEdge
> 2600 server. It has two HyperThreading Xeon 2.4Ghz processors, but
> HyperThreading is not working in FreeBSD. I guess that four
> processors should be visible in the system after boot
Matthew Dillon wrote:
> The HLT/clock interrupt issue is precisely what I describe in the
> idle_hlt comments in i386/i386/machdep.c (last July). I wish we had a
> better mechanism then the stupid IPI stuff, like a simple per-cpu
> latch/acknowledge level interrupt (softint), but w
:> The ideal situation would be to have as Matt (and the comment
:> actually) says a cpu mask of idle cpus and generate an IPI to wake up
:> CPUs sitting in HLT when something hits the runqueue, then you can
:> just hlt all of them and rely on the IPI to wake you up, or the next
:> timer
On Sat, Feb 01, 2003 at 01:28:45PM -0800, Terry Lambert wrote:
> Bosko Milekic wrote:
> > > > Or, as I explained in my previous post, only HLT the [virtual] CPU if
> > > > the other [virtual] CPU that is sharing the same execution & cache
> > > > units is not HLT'd itself. If the other one
Bosko Milekic wrote:
> > > Or, as I explained in my previous post, only HLT the [virtual] CPU if
> > > the other [virtual] CPU that is sharing the same execution & cache
> > > units is not HLT'd itself. If the other one is HLT'd, then not do the
> > > HLT.
> >
> > Actually, why is that? W
On Sat, Feb 01, 2003 at 12:47:59PM -0800, Terry Lambert wrote:
> Bosko Milekic wrote:
> > On Fri, Jan 31, 2003 at 11:52:53AM -0800, Matthew Dillon wrote:
> > > Another solution would be to have a global mask of 'idle' cpus and send
> > > an IPI to them when a new KSE is scheduled on a no
Bosko Milekic wrote:
> On Fri, Jan 31, 2003 at 11:52:53AM -0800, Matthew Dillon wrote:
> > Another solution would be to have a global mask of 'idle' cpus and send
> > an IPI to them when a new KSE is scheduled on a non-idle cpu that would
> > simply serve to wakeup the HLT. IPIs are
On Fri, Jan 31, 2003 at 11:52:53AM -0800, Matthew Dillon wrote:
> Another solution would be to have a global mask of 'idle' cpus and send
> an IPI to them when a new KSE is scheduled on a non-idle cpu that would
> simply serve to wakeup the HLT. IPIs are nasty, but there are large
On Fri, Jan 31, 2003 at 11:48:17AM -0800, Peter Wemm wrote:
> The cache and most of the execution hardware is shared. The execution
> units can run something like 4 instructions per clock. If the "idle"
> logical core is in a spinloop, then it is generating instructions for
> execution, so you a
On 31-Jan-2003 Julian Elischer wrote:
>
>
> On Fri, 31 Jan 2003, John Baldwin wrote:
>
>>
>> On 31-Jan-2003 Julian Elischer wrote:
>> >
>> >
>> > On Fri, 31 Jan 2003, John Baldwin wrote:
>> >
>> >> >> AT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,,ACC>
>> >> >>
>> >> >> It has
On Fri, 31 Jan 2003, John Baldwin wrote:
>
> On 31-Jan-2003 Julian Elischer wrote:
> >
> >
> > On Fri, 31 Jan 2003, John Baldwin wrote:
> >
> >> >> AT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,,ACC>
> >> >>
> >> >> It has HTT set but it's only a 1.2GHz box and I heard somewhere
In the last episode (Jan 31), Don Bowman said:
> Thus the worst thing you could have would be a nop-loop with
> no stalls, which would squeeze the other to death.
I don't think this is possible, from looking at Intel's white papers.
> In practise I've found intel's numbers to be true, that the SM
On 31-Jan-2003 Julian Elischer wrote:
>
>
> On Fri, 31 Jan 2003, John Baldwin wrote:
>
>>
>> On 31-Jan-2003 leafy wrote:
>> > On Fri, Jan 31, 2003 at 01:45:56PM -0500, Trish Lynch wrote:
>> >> I have the .dat's for you, unfortunately, the output is different, so
>> >> you'll have to modify the
On 31-Jan-2003 Julian Elischer wrote:
>
>
> On Fri, 31 Jan 2003, John Baldwin wrote:
>
>> >> AT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,,ACC>
>> >>
>> >> It has HTT set but it's only a 1.2GHz box and I heard somewhere that only
>> >> 2+ GHz P4's had hyperthreading. I notice
On Fri, 31 Jan 2003, John Baldwin wrote:
>
> On 31-Jan-2003 leafy wrote:
> > On Fri, Jan 31, 2003 at 01:45:56PM -0500, Trish Lynch wrote:
> >> I have the .dat's for you, unfortunately, the output is different, so
> >> you'll have to modify the .cfg for gnuplot :)
> >>
> >> -Trish
> > I have HT
On Fri, 31 Jan 2003, John Baldwin wrote:
> >> AT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,,ACC>
> >>
> >> It has HTT set but it's only a 1.2GHz box and I heard somewhere that only
> >> 2+ GHz P4's had hyperthreading. I noticed some MFCs to stable that
> >> suggested hype
> From: Matthew Dillon [mailto:[EMAIL PROTECTED]]
>
> :The cache and most of the execution hardware is shared. The
> execution
> :units can run something like 4 instructions per clock. If the "idle"
> :logical core is in a spinloop, then it is generating instructions for
> :execution, so you ar
On 31-Jan-2003 Matthew Dillon wrote:
>
>:AFAIK, full hyperthreading support, as it is, has been merged to
>:-stable. It consists of a patch to recognize the virtual CPUs, so they
>:will be dealt with like any SMP system, as long as HTT is enabled on the
>:BIOS.
>:
>:--
>:Daniel C. Sobral
On 31-Jan-2003 leafy wrote:
> On Fri, Jan 31, 2003 at 01:45:56PM -0500, Trish Lynch wrote:
>> I have the .dat's for you, unfortunately, the output is different, so
>> you'll have to modify the .cfg for gnuplot :)
>>
>> -Trish
> I have HTT for my CPU, is there any "hack" to the BIOS to enable Hype
On 31-Jan-2003 Daniel C. Sobral wrote:
> Matthew Dillon wrote:
>> :So, at the request of bmilekic, I ran netpipe on a hyperthreading box (non
>> :hyperthreading, I'll do when I can turn it off in BIOS next time I'm down
>> :there)
>> :...
>> :
>> :The results are here:
>> :
>> :http://bsdunix.net/
On Fri, 31 Jan 2003, Peter Wemm wrote:
>
> Personally, I doubt that HTT will buy much on FreeBSD, apart from being
> buzzword compliant. I'd actually like a compile option or boot tunable so
> that it to be turned on or off (and treated like a regular Xeon SMP
> system). Single-physical-cpu syst
:The cache and most of the execution hardware is shared. The execution
:units can run something like 4 instructions per clock. If the "idle"
:logical core is in a spinloop, then it is generating instructions for
:execution, so you are dividing the execution resources between one context
:that is
: Why do you think that hlt-ing the CPU(s) when idle would actually
: improve performance in this case? My only suspicion is that perhaps
: this reduces scheduling on the auxiliary 'logical' (fake) CPUs,
: thereby indirectly reducing cache ping-ponging and abuse. I would
: imagine that both
Bosko Milekic wrote:
>
>
> On Fri, Jan 31, 2003 at 11:08:38AM -0800, Matthew Dillon wrote:
> >
> > :AFAIK, full hyperthreading support, as it is, has been merged to
> > :-stable. It consists of a patch to recognize the virtual CPUs, so they
> > :will be dealt with like any SMP system, as long
On Fri, Jan 31, 2003 at 11:08:38AM -0800, Matthew Dillon wrote:
>
> :AFAIK, full hyperthreading support, as it is, has been merged to
> :-stable. It consists of a patch to recognize the virtual CPUs, so they
> :will be dealt with like any SMP system, as long as HTT is enabled on the
> :BIOS.
:AFAIK, full hyperthreading support, as it is, has been merged to
:-stable. It consists of a patch to recognize the virtual CPUs, so they
:will be dealt with like any SMP system, as long as HTT is enabled on the
:BIOS.
:
:--
:Daniel C. Sobral (8-DCS)
:Gerencia de Operacoes
> I have HTT for my CPU, is there any "hack" to the BIOS to enable
> HyperThreading?
>
You might try updating your BIOS.
Ken
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
On Fri, Jan 31, 2003 at 01:45:56PM -0500, Trish Lynch wrote:
> I have the .dat's for you, unfortunately, the output is different, so
> you'll have to modify the .cfg for gnuplot :)
>
> -Trish
I have HTT for my CPU, is there any "hack" to the BIOS to enable HyperThreading?
Jiawei Ye
--
"Without
On Fri, 31 Jan 2003, Bosko Milekic wrote:
>
> Trish,
>
> Thanks for the tests, it would be good to also get results with
> hyperthreading turned off. However, I need you to pass the -o
> option to NPtcp and get an actual dat file, so that you can generate
> the graphs using the gn
Trish,
Thanks for the tests, it would be good to also get results with
hyperthreading turned off. However, I need you to pass the -o
option to NPtcp and get an actual dat file, so that you can generate
the graphs using the gnuplot config file I asked you to download.
Having t
Matthew Dillon wrote:
:So, at the request of bmilekic, I ran netpipe on a hyperthreading box (non
:hyperthreading, I'll do when I can turn it off in BIOS next time I'm down
:there)
:...
:
:The results are here:
:
:http://bsdunix.net/performance
:
:all information on what command line options I use
:So, at the request of bmilekic, I ran netpipe on a hyperthreading box (non
:hyperthreading, I'll do when I can turn it off in BIOS next time I'm down
:there)
:...
:
:The results are here:
:
:http://bsdunix.net/performance
:
:all information on what command line options I used is in there.
:
:the d
44 matches
Mail list logo
