Hi Wouter,
Please keep in mind the storage controler (Perc H755) is _not_ yet
supported by OpenBSD.
Cheers,
Laurent
Le 14/12/2023 à 15:43, Wouter Prins a écrit :
Thank you Laurent and Claudio,
We have an identical setup (hardware specs), i am sure we need this in
the near future. :)
/Wo
Thank you Laurent and Claudio,
We have an identical setup (hardware specs), i am sure we need this in the
near future. :)
/Wouter
On Thu, Dec 14, 2023 at 3:08 PM Claudio Jeker
wrote:
> On Tue, Nov 28, 2023 at 05:55:03PM +0100, Laurent CARON wrote:
> >
> > Le 28/11/2023 à 17:46, Claudio Jeker a
On Tue, Nov 28, 2023 at 05:55:03PM +0100, Laurent CARON wrote:
>
> Le 28/11/2023 à 17:46, Claudio Jeker a écrit :
> > The problem is that the symbol nkmempages moved into .bss and is therefor
> > no longer modifiable by config(8). I think you can still use ukc via
> > boot -c to alter it (but that
Le 28/11/2023 à 17:46, Claudio Jeker a écrit :
The problem is that the symbol nkmempages moved into .bss and is therefor
no longer modifiable by config(8). I think you can still use ukc via
boot -c to alter it (but that is not sticky).
The alternative is to set "option NKMEMPAGES=131072" in yo
On Tue, Nov 28, 2023 at 04:50:05PM +0100, Laurent CARON wrote:
> Le 28/11/2023 à 12:12, Claudio Jeker a écrit :
> > So the problem is that the malloc space is filled by
> > a) 26540K of devbuf -- because of the multiqueue support in ixl
> > b) 63493K of ACPI -- what the heck ACPI?!?
> > and then th
Le 28/11/2023 à 12:12, Claudio Jeker a écrit :
So the problem is that the malloc space is filled by
a) 26540K of devbuf -- because of the multiqueue support in ixl
b) 63493K of ACPI -- what the heck ACPI?!?
and then there is not enough space for rtable. A full table requires
in your example 50816
On Mon, Nov 27, 2023 at 05:51:25PM +0100, Laurent CARON wrote:
> Please find attached the relevant info:
>
> vmstat-m_SP_with_bgpd -> vmstat -m SP with bgpd
>
> vmstat-m_SMP_without_bgpd -> vmstat -m SMP without bgpd
>
> vmstat-m_SMP_with_bgpd_0{01..11} -> vmsta
Hi Claudio,
Should you need remote access to the server, this is of course possible.
Le 27/11/2023 à 17:51, Laurent CARON a écrit :
Please find attached the relevant info:
vmstat-m_SP_with_bgpd -> vmstat -m SP with bgpd
vmstat-m_SMP_without_bgpd -> vmstat -m SMP without bgpd
On Mon, Nov 27, 2023 at 04:57:35PM +0100, Laurent CARON wrote:
> Hi,
>
> I'm currently migrating a BGPd server.
>
> Specs of "old" machine:
>
> - Dell R720 with Intel(R) Xeon(R) CPU E5-2637 v2and 16GB RAM
>
> - SMP Kernel (default)
>
> - BGPd
Hi,
I'm currently migrating a BGPd server.
Specs of "old" machine:
- Dell R720 with Intel(R) Xeon(R) CPU E5-2637 v2and 16GB RAM
- SMP Kernel (default)
- BGPd runs fine with 5 full views
- X710 NIC (ixl) 4 port interface
Specs of "new" machine:
- Dell R750xs with I
; > Hi
> >
> > I'm think about get a Ubiquiti Edgerouter box and drop openbsd there.
> > I read the instalation file and was not clear to me if the current
> > MIPS/Octeon kernel implementation supports SMP or not , does it support ?
> >
> > Regards
> &g
On 09/13/17 22:28, Dante F. B. Colò wrote:
> Hi
>
> I'm think about get a Ubiquiti Edgerouter box and drop openbsd there.
> I read the instalation file and was not clear to me if the current
> MIPS/Octeon kernel implementation supports SMP or not , does it support ?
>
Hi
I'm think about get a Ubiquiti Edgerouter box and drop openbsd there.
I read the instalation file and was not clear to me if the current
MIPS/Octeon kernel implementation supports SMP or not , does it support ?
Regards
Dante F. B. Colò
On Sun, 26 Feb 2017 03:56:33 +, Tinker wrote:
> Did I get it right, that ARM64 has SMP (as of the patches this week),
> but ARM32 does not have SMP and will not get it too?
As Peter says, someone has to step up andf do the work for ARM32
SMP. That said, it probably doesn't ma
On 2017 Feb 26 (Sun) at 03:56:33 + (+), Tinker wrote:
:Hi misc,
:
:I just wanted to understand what's going on with SMP on ARM -
:
:Did I get it right, that ARM64 has SMP (as of the patches this week), but
:ARM32 does not have SMP and will not get it too?
:
:What was the reason fo
Hi misc,
I just wanted to understand what's going on with SMP on ARM -
Did I get it right, that ARM64 has SMP (as of the patches this week),
but ARM32 does not have SMP and will not get it too?
What was the reason for not implementing SMP on ARM32? (I only need SMP
on ARM64 so just w
On 05/05/16(Thu) 19:03, Pavan Maddamsetti wrote:
> I have been reading about ongoing improvements to SMP in OpenBSD. My
> understanding is that context switching from userspace to the kernel can be
> hazardous if shared resources are not protected by locking.
The context switching i
I have been reading about ongoing improvements to SMP in OpenBSD. My
understanding is that context switching from userspace to the kernel can be
hazardous if shared resources are not protected by locking. OpenBSD
currently has a "giant lock" for safe concurrent access to kernel data
stru
Hi misc,
I've been reading that some improvements are being made to the network
stack regarding SMP.
I've also read that this is just starting, and that the KERNEL_LOCK will
still be present in 5.9 (seen it here:
http://undeadly.org/cgi?action=article&sid=20160302155046), but do yo
On 20/03/15(Fri) 20:57, Martin Pieuchot wrote:
> If you've been following my contributions to OpenBSD's kernel, you
> already know that in the past years I've been working on the Network
> Stack [1] to make it more SMP friendly [2].
>
> All the network hackers presen
Hi Martin
I can help you about switch. Where we can find a switch? Ebay?
On 20 Mar 2015 21:58, "Martin Pieuchot" wrote:
> If you've been following my contributions to OpenBSD's kernel, you
> already know that in the past years I've been working on the Network
How is this going ?
/T
On Fri, Mar 20, 2015 at 8:57 PM, Martin Pieuchot wrote:
> If you've been following my contributions to OpenBSD's kernel, you
> already know that in the past years I've been working on the Network
> Stack [1] to make it more SMP friendly [2].
>
It does the same thing on 5.3 through -current. I haven't put that box in the
rack yet so I can try a few older kernels too and see if any work.
Will report back.
> On Mar 22, 2015, at 8:45 AM, Claudio Jeker wrote:
>
>> On Sun, Mar 22, 2015 at 09:53:24AM +, Stuart Henderson wrote:
>>> On
On Sun, Mar 22, 2015 at 09:53:24AM +, Stuart Henderson wrote:
> On 2015-03-21, John E.P. Hynes wrote:
> > If anyone has any ideas, or would like more info, or if a dev suspects
> > it could be the driver, contact me off-list and I can arrange to send
> > hardware if it helps.
>
> It might b
On 2015-03-21, John E.P. Hynes wrote:
> If anyone has any ideas, or would like more info, or if a dev suspects
> it could be the driver, contact me off-list and I can arrange to send
> hardware if it helps.
It might be worth talking to Supermicro.
Well, I discovered the issue - the few machines that work properly had a
different quad-port nic in them.
With certain BIOS settings, you can catch part of the kernel panic
before the screen goes crazy.
Codes that were visible depending on BIOS settings:
"kernel: type 1994916275 trap, code=0
If you've been following my contributions to OpenBSD's kernel, you
already know that in the past years I've been working on the Network
Stack [1] to make it more SMP friendly [2].
All the network hackers present at s2k15 agreed to volunteer me to work
on the next step: properl
I've got three identical boxes that all display the same behavior:
Install of 5.6 or the March 18th 5.7 snapshot works, but is painfully
slow (no disk access on the CD or disks while "stalled") and on the
first reboot, it gets about as far as loading wskbd before rebooting
spontaneously. I'm t
Hi,
The openbsd journal had a series of articles on rthreads back in 2012,
which I came across last week. I've also been following DragonFlyBSD
lately. Their approach to SMP is different but seems to pay off
(according to their website), with contention in their kernel almost
eliminated in
the hardware, or at
> > least provide the abstractions so that it seems they run concurrently
> > on non-smp hardware.
>
>
> So is there not a implementation of them?
why do you conclude that?
What I'm rying to say that is, given the limitatioins of the harware,
the mp
ncy than you might
> expect, but it provides services to userland to either run process
> threads concurrently given the limitations of the hardware, or at
> least provide the abstractions so that it seems they run concurrently
> on non-smp hardware.
So is there not a implementation of
On Wed, May 15, 2013 at 10:03:40PM +0200, Alfonso S. Siciliano wrote:
> Hi,
> I'm trying to study the concurrency and the parallelism of OpenBSD.
> Fortunately SMP is supported on my platform, amd64.
> Where can I find documentation about what components are been
> paralle
On Wed, May 15, 2013 at 10:03 PM, Alfonso S. Siciliano wrote:
> Hi,
> I'm trying to study the concurrency and the parallelism of OpenBSD.
> Fortunately SMP is supported on my platform, amd64.
> Where can I find documentation about what components are been
> parallelized
Alfonso S. Siciliano [alfi...@gmail.com] wrote:
> Hi,
> I'm trying to study the concurrency and the parallelism of OpenBSD.
> Fortunately SMP is supported on my platform, amd64.
> Where can I find documentation about what components are been
> parallelized? (queue, stac
Hi,
I'm trying to study the concurrency and the parallelism of OpenBSD.
Fortunately SMP is supported on my platform, amd64.
Where can I find documentation about what components are been
parallelized? (queue, stack, etc.)
Regards
Alfonso
Alfonso S
[At the risk of starting yet another flame war...]
On Mon, Jun 18, 2012 at 2:31 AM, Ryan McBride wrote:
> It's not critical because they can change the state table implementation
> later - OpenBSD has reorganised this several times with more planned -
> but we've put quite a bit of effort into re
* Holger Glaess [2012-06-17 11:41]:
> From the very beginning of the project it was clear, that code is going
> to diverge significantly from original OpenBSD code. OpenBSD has always
> developed pf without taking into account that code can ever get
> multithreaded, thus quite a lot needed to be
t; kernel wherever possible, partly due to their attackability. My
> personal preference would be to go with a lockless or lock-on-write RB
> tree for the state table, but without an SMP-safe network stack
> there's little motivation to work on such things (though it remains
> somewhere on
less or lock-on-write RB
tree for the state table, but without an SMP-safe network stack
there's little motivation to work on such things (though it remains
somewhere on my todo list).
On Mon, Jun 18, 2012 at 10:06:01AM +0200, Franco Fichtner wrote:
> On Jun 17, 2012, at 7:53 PM, Ted Unang
On Jun 17, 2012, at 7:53 PM, Ted Unangst wrote:
> On Sun, Jun 17, 2012 at 11:43, Holger Glaess wrote:
>
>> i dident wont start about smp on openbsd but
>>
>> what about this porject ?
>
> Did you read the part below? I think it's pretty clear this project
&
On Sun, Jun 17, 2012 at 11:43, Holger Glaess wrote:
> i dident wont start about smp on openbsd but
>
> what about this porject ?
Did you read the part below? I think it's pretty clear this project
isn't going to have much relevance for OpenBSD.
> From the very beginning
hi
i dident wont start about smp on openbsd but
what about this porject ?
[quote]
Hello, networkers!
[net@ in Cc, but further discussion should go on pf@]
As you already probably know, or some may be don't yet know, the pf(4)
subsystem in FreeBSD is currently working under a single
On Sat, Jul 9, 2011 at 9:11 PM, Maurice Janssen wrote:
> On Sat, Jul 09, 2011 at 07:52:58AM -0400, Nick Holland wrote:
>>On 07/09/11 03:57, Maurice Janssen wrote:
>>> Hi,
>>>
>>> Is it possible to somehow force a program to run on a single CPU in an
>>
On Sat, Jul 09, 2011 at 07:52:58AM -0400, Nick Holland wrote:
>On 07/09/11 03:57, Maurice Janssen wrote:
>> Hi,
>>
>> Is it possible to somehow force a program to run on a single CPU in an
>> SMP system?
>> The reason I ask that on some SMP-capable architectu
On 07/09/11 03:57, Maurice Janssen wrote:
> Hi,
>
> Is it possible to somehow force a program to run on a single CPU in an
> SMP system?
> The reason I ask that on some SMP-capable architectures, I'm having some
> problems with ntpd. On hppa and sgi, the clock won'
Hi,
Is it possible to somehow force a program to run on a single CPU in an
SMP system?
The reason I ask that on some SMP-capable architectures, I'm having some
problems with ntpd. On hppa and sgi, the clock won't sync because ntpd
sees replies with negative delay:
Jul 9 08:58:19
right here: http://www.openbsd.org/hppa.html
thanks jsing kettenis and others that made SMP work on hppa!
On Sat, Feb 26, 2011 at 01:48:35PM -0600, L. V. Lammert wrote:
> Can't seem to find the SMP HCL results posted anywhere - does anyone have
> a recommendation?
>
> Lee
Can't seem to find the SMP HCL results posted anywhere - does anyone have
a recommendation?
Lee
Hi,
Emailed dev but think the mail was stripped because of attachments.
We have some racks of appro AMD blade servers that have been
decommissioned and are set to be disposed of. I got ok to donate some
or all.These were used in energy HPC environment for seismic data
processing. Email offl
On Mon, Jul 5, 2010 at 12:57 PM, Robert Yuri wrote:
> what is the state of the smp support in openbsd ?
"Better than yesterday, not as good as tomorrow"
> we'll degust improvements after the hackathon ?
If you want to know what happened during the hackathon, you should
what is the state of the smp support in openbsd ?
we'll degust improvements after the hackathon ?
On Sun, 13 Jun 2010 08:56:14 +0200
Tomas Bodzar wrote:
> He said that you need to send it to him and to misc@ ;-) Attachments
> are not allowed on m...@.
Yeah, I realized I was being stupid the second I hit send... I
apologize.
--TimH
He said that you need to send it to him and to misc@ ;-) Attachments
are not allowed on m...@.
On Sun, Jun 13, 2010 at 7:55 AM, TimH wrote:
> On Sat, 12 Jun 2010 18:45:51 -0500
> Marco Peereboom wrote:
>
>> something in the gpe handler screwing up. B please tar up the output of
>> acpidump -o hp
On Sat, 12 Jun 2010 18:45:51 -0500
Marco Peereboom wrote:
> something in the gpe handler screwing up. please tar up the output of
> acpidump -o hplaptop and send that to me
Attached... Please let me know if I can do anything else.
--TimH
[demime 1.01d removed an attachment of type ap
something in the gpe handler screwing up. please tar up the output of
acpidump -o hplaptop and send that to me
On Sat, Jun 12, 2010 at 11:06:27AM -0700, TimH wrote:
> I just got a semi-new HP laptop and it fails to boot unless I disable
> acpi. This isn't a big deal in itself, but it seems that
TimH wrote:
> I just got a semi-new HP laptop and it fails to boot unless I disable
> acpi. This isn't a big deal in itself, but it seems that it doesn't
> use both cores of the CPU when this is done. Is this normal?
Yes, normal, newer systems lack the legacy Intel MP mappings.. so if you
disabl
I just got a semi-new HP laptop and it fails to boot unless I disable
acpi. This isn't a big deal in itself, but it seems that it doesn't
use both cores of the CPU when this is done. Is this normal?
It's an HP ProBook 4510s if that's at all interesting...
Included are the dmesg from when it fai
Janne Johansson wrote:
2010/5/27 Leonardo Carneiro - Veltrac
Forgive me for the noob question (i'm a newbie at openbsd), but if i want
to build, for example, a large squid cache using openbsd, in a server with
BIIIG ram (12gb+), i will no be able to use the full memory space? is thi
2010/5/27 Leonardo Carneiro - Veltrac
>
> Forgive me for the noob question (i'm a newbie at openbsd), but if i want
>>> to build, for example, a large squid cache using openbsd, in a server with
>>> BIIIG ram (12gb+), i will no be able to use the full memory space? is this
>>> what you guys are
On Thu, May 27, 2010 at 10:05 PM, Owain Ainsworth wrote:
>
> Being worked on, needs some changes that hopefully will be done at
> c2k10, we definitely have access to a machine with the intel VT stuff,
> the amd64 new shiny iommu i'm not sure if we have hardware yet.
> Similarly turning bigmem back
>>>> send oga some beer for c2k10 instead he might feel more pressure that
>>>> way :-)
>>>>
>>>> On Thu, May 27, 2010 at 11:01:58AM +0100, Peter Kay (Syllopsium) wrote:
>>>>
>>>>>>
>>>>>> From: "Siju G
y 27, 2010 at 11:01:58AM +0100, Peter Kay (Syllopsium) wrote:
From: "Siju George"
On Thu, May 27, 2010 at 3:22 PM, Peter Kay (Syllopsium)
wrote:
From: "Siju George"
but OpenBSD 4.7/amd64 SMP detects only 3 GB.
Is there anything more I should d
On Thu, May 27, 2010 at 11:01:58AM +0100, Peter Kay (Syllopsium) wrote:
> >From: "Siju George"
> >On Thu, May 27, 2010 at 3:22 PM, Peter Kay (Syllopsium)
> > wrote:
> >>
> >>>From: "Siju George"
> >>>but OpenBSD 4.7/amd64
;>
>> On Thu, May 27, 2010 at 11:01:58AM +0100, Peter Kay (Syllopsium) wrote:
>>
>>>> From: "Siju George"
>>>> On Thu, May 27, 2010 at 3:22 PM, Peter Kay (Syllopsium)
>>>> wrote:
>>>>
>>>>>> From: &qu
27, 2010 at 3:22 PM, Peter Kay (Syllopsium)
wrote:
From: "Siju George"
but OpenBSD 4.7/amd64 SMP detects only 3 GB.
Is there anything more I should do to get the other 1 GB of RAM
recognized by the System?
This is normal. Large memory support is not yet included i
t 3:22 PM, Peter Kay (Syllopsium)
>> wrote:
>>>
>>>> From: "Siju George"
>>>> but OpenBSD 4.7/amd64 SMP detects only 3 GB.
>>>>
>>>> Is there anything more I should do to get the other 1 GB of RAM
>>>> recognized
On Thu, May 27, 2010 at 2:55 PM, Jan Stary wrote:
> On May 27 14:19:20, Siju George wrote:
>>
>>
>> $ pkg_info
>> Package database already locked... awaiting release... ^C
>
> So you were running some other pkg_command simultaneously?
>
not at all! :-( I was bewildered! it went off the next time
From: "Siju George"
On Thu, May 27, 2010 at 3:22 PM, Peter Kay (Syllopsium)
wrote:
From: "Siju George"
but OpenBSD 4.7/amd64 SMP detects only 3 GB.
Is there anything more I should do to get the other 1 GB of RAM
recognized by the System?
This is normal. Large memor
On Thu, May 27, 2010 at 3:22 PM, Peter Kay (Syllopsium)
wrote:
>
>> From: "Siju George"
>> but OpenBSD 4.7/amd64 SMP detects only 3 GB.
>>
>> Is there anything more I should do to get the other 1 GB of RAM
>> recognized by the System?
>>
> This
From: "Siju George"
but OpenBSD 4.7/amd64 SMP detects only 3 GB.
Is there anything more I should do to get the other 1 GB of RAM
recognized by the System?
This is normal. Large memory support is not yet included in
OpenBSD by default for amd64.
On May 27 14:19:20, Siju George wrote:
> Hi,
>
> The poser went off while I was installing packages ( awesome ) and
> after the system came up it is behaving strangely printing junk
> characters on the xterm for pkg_* related commands
>
> $ pkg_info
> Package database already locked... awaiting rel
On Thu, May 27, 2010 at 12:06:33PM +0300, Gregory Edigarov wrote:
> On Thu, 27 May 2010 14:19:20 +0530
> Siju George wrote:
>
> > $ rm -rf *
> > $sudo pkg_delete cairo gettext glib2 glitz jpeg libconfuse libgamin
> > libiconv libungif partial-bzip2 pcre png
> > <
> > Unknown element
On Thu, 27 May 2010 14:19:20 +0530
Siju George wrote:
> $ rm -rf *
> $sudo pkg_delete cairo gettext glib2 glitz jpeg libconfuse libgamin
> libiconv libungif partial-bzip2 pcre png
> <
> Unknown element: @A oFG" }j6ZWKFD)B7
> Ge_F1Qq.Hq4gZ4>,VlX2s,.? mWST/[@
)
==
There are no setting to limit RAM in BIOS and so on also.
but OpenBSD 4.7/amd64 SMP detects only 3 GB.
Is there anything more I should do to get the other 1 GB of RAM
recognized by the System?
Thanks
--Siju
dmesg below--
OpenBSD 4.7 (GENERIC.MP) #130: Wed Mar
Hi,
The poser went off while I was installing packages ( awesome ) and
after the system came up it is behaving strangely printing junk
characters on the xterm for pkg_* related commands
$ pkg_info
Package database already locked... awaiting release... ^C
$ pkg_info
cairo-1.8.8p0 vector grap
; Am 20 Dec 2009 um 10:18 schrieb Tomas Bodzar:
>
>> Hi all,
>>
>> someone have running OpenBSD release/stable/current on new line of
>> VirtualBox (3.1.x) on non-SMP machine? Older version 3.0.x was ok.
Now
>> it sets VT-x/AMD-V as default and you can't change it.
Hi Tomas,
2009/12/20 Tomas Bodzar :
> Hi all,
>
> someone have running OpenBSD release/stable/current on new line of
> VirtualBox (3.1.x) on non-SMP machine? Older version 3.0.x was ok. Now
> it sets VT-x/AMD-V as default and you can't change it. Even when I
> disable it
From: "Bayard Bell"
"Requires VT-x or AMD-V hardware virtualization support."
It would appear they've therefore made VT-x and friends non-
configurable. You can file a bug report and see where that goes.
Would it be too cynical to suggest using a product which doesn't suck?
It really is a pi
s Bodzar:
> Hi all,
>
> someone have running OpenBSD release/stable/current on new line of
> VirtualBox (3.1.x) on non-SMP machine? Older version 3.0.x was ok. Now
> it sets VT-x/AMD-V as default and you can't change it. Even when I
> disable it directly in .xml config file
Hi all,
someone have running OpenBSD release/stable/current on new line of
VirtualBox (3.1.x) on non-SMP machine? Older version 3.0.x was ok. Now
it sets VT-x/AMD-V as default and you can't change it. Even when I
disable it directly in .xml config file for guest it still try this
feature.
On Fri, 11 Dec 2009 14:56:57 -0500, Daniel Ouellet
wrote:
> Then using PostgreSQL should really work well for you then and you
> wouldn't really need or benefit much from multicore kernel with the
> giant lock removed as PostgreSQL is not and do not use threads anyway by
> design oppose to MyS
On 12/11/09 12:51 PM, Donald Allen wrote:
Thanks to everyone who took the time to weigh in on this. Perhaps most
useful to me are the comments of those who have used OpenBSD for heavy
database work (I intend to use Postgresql) and have gotten
satisfactory results.
Then using PostgreSQL should r
Thanks to everyone who took the time to weigh in on this. Perhaps most
useful to me are the comments of those who have used OpenBSD for heavy
database work (I intend to use Postgresql) and have gotten
satisfactory results.
To Daniel -- I don't think we'll be working for or with each other in
the f
On Wed, Dec 9, 2009 at 12:03 PM, Daniel Gracia Garallar
wrote:
> It is true, and AFAIK, todays it's a topper nice task... almost 20.
>
> Regards,
>
> Dani
>
> Donald Allen escribis:
IMHO I hope OpenBSD doesn't use locks at all in the future taking
FreeBSD's lesson, but does
something what Dragon
; trade-offs may be worth it anyway (...)
> >
> > Or put it another way.
> >
> > I couldn't help but smile when someone told me their 16-way SMP box
> > had been holed by a bug in their ld.so.
>
> Linux?
There was a recent FreeBSD vulnerability along those lines, see
:46:25 -0700
Theo de Raadt wrote:
Soo... Your performance requirements may met by OpenBSD despite it's
current poor SMP support - other OSes will scale on SMP. Trade-offs,
trade-offs... It's a psychological issue. We have all this multicore
hardware that doesn't get taken advantag
It is true, and AFAIK, todays it's a topper nice task... almost 20.
Regards,
Dani
Donald Allen escribis:
My understanding is that OpenBSD still employs the Giant Lock approach
to SMP, serializing access to kernel services. Is this still true? If
it is, do Theo and the other kernel devel
On Wed, Dec 9, 2009 at 6:46 PM, Theo de Raadt wrote:
>> Soo... Your performance requirements may met by OpenBSD despite it's
>> current poor SMP support - other OSes will scale on SMP. Trade-offs,
>> trade-offs... It's a psychological issue. We have all this multicore
On Wed, 09 Dec 2009 16:46:25 -0700
Theo de Raadt wrote:
> > Soo... Your performance requirements may met by OpenBSD despite it's
> > current poor SMP support - other OSes will scale on SMP. Trade-offs,
> > trade-offs... It's a psychological issue. We have all this
> Soo... Your performance requirements may met by OpenBSD despite it's
> current poor SMP support - other OSes will scale on SMP. Trade-offs,
> trade-offs... It's a psychological issue. We have all this multicore
> hardware that doesn't get taken advantage of by this O
nBSD fail
> (with regard to performance) in real-world settings. The OP should
> test it out... he'll be pleasantly surprised.
>
> Brad
Soo... Your performance requirements may met by OpenBSD despite it's
current poor SMP support - other OSes will scale on SMP. Trade-offs,
trade-o
On Wed, 9 Dec 2009 23:07:02 +0100
Michiel van Baak wrote:
> On 22:56, Wed 09 Dec 09, Robert wrote:
> > Just last month i have seen a database server being upgraded from
> > 32GB to 256GB of RAM because that was easier (to justify) for them
> > than to fix their horrible db layout.
>
> That must
On Wed, Dec 9, 2009 at 4:56 PM, Daniel Ouellet wrote:
> So, what's heavy for you may be just simple routine for others and no, I do
> not miss the fine lock either yet anyway. Would be nice, but really, I
> haven't run into it's need for me anyway yet.
That's true for me as well. We use OpenBSD
On 22:56, Wed 09 Dec 09, Robert wrote:
> On Wed, 9 Dec 2009 15:13:15 -0500
> Donald Allen wrote:
>
> > On Wed, Dec 9, 2009 at 2:50 PM, Brad Tilley
> > wrote:
> > > On Wed, Dec 9, 2009 at 1:03 PM, Donald Allen
> > > wrote:
> > >
> > >> Certainly I agree with you that a blazingly fast but unstabl
On Wed, 9 Dec 2009 15:13:15 -0500
Donald Allen wrote:
> On Wed, Dec 9, 2009 at 2:50 PM, Brad Tilley
> wrote:
> > On Wed, Dec 9, 2009 at 1:03 PM, Donald Allen
> > wrote:
> >
> >> Certainly I agree with you that a blazingly fast but unstable
> >> and/or insecure system isn't worth much in most, i
I don't, and many times we don't have the luxury of having such
examples or data. I'm in a different kind of real-world situation: I'm
setting up a database server on a 4-core machine that is going to
carry a heavy load -- it's performance will be critical to the success
of the project -- and I ne
On Wed, Dec 9, 2009 at 2:50 PM, Brad Tilley wrote:
> On Wed, Dec 9, 2009 at 1:03 PM, Donald Allen wrote:
>
>> Certainly I agree with you that a blazingly fast but unstable and/or
>> insecure system isn't worth much in most, if any, settings. On the
>> other hand, a rock-solid, secure system that
still employs the Giant Lock approach
> >> to SMP, serializing access to kernel services. Is this still true? If
> >
> > yes
> >
> >> it is, do Theo and the other kernel developers consider it a priority
> >> to improve this?
> >
> > sure
>
On Wed, Dec 9, 2009 at 1:03 PM, Donald Allen wrote:
> Certainly I agree with you that a blazingly fast but unstable and/or
> insecure system isn't worth much in most, if any, settings. On the
> other hand, a rock-solid, secure system that simply doesn't deliver
> the computations at the needed ra
On Wed, Dec 9, 2009 at 12:03 PM, wrote:
> Quoting "Donald Allen" :
>
>> My understanding is that OpenBSD still employs the Giant Lock approach
>> to SMP, serializing access to kernel services. Is this still true? If
>> it is, do Theo and the other kernel devel
1 - 100 of 280 matches
Mail list logo
