jesk wrote:
The thread mentions that update query first then select query is
faster on Linux. What did you need to change to fix that (or did one of
the updates on the way to STABLE fix it)?
Paul
i forgot to disable HTT ;)
so can you post the final numbers?
so can you post the final numbers?
the select procedure didnt changed much in its performance, but the
update-select test performed to around 4000 queries per second.
___
[EMAIL PROTECTED] mailing list
so can you post the final numbers?
the select procedure didnt changed much in its performance, but the
update-select test performed to around 4000 queries per second.
___
[EMAIL PROTECTED] mailing list
On Tuesday 09 November 2004 23:15, Thordur Ivar B. wrote:
On Tue, 9 Nov 2004 21:53:08 +0100
jesk [EMAIL PROTECTED] wrote:
fbsd outperformed linux in the end of my tests. ;)
dont forget to disable HTT, this will increase the performance of
threading.
In my tests also. This is good. IMHO
On Tuesday 09 November 2004 08:59 am, [EMAIL PROTECTED] wrote:
[EMAIL PROTECTED] wrote:
You're right, it does. However, myself and collegue recently presented
our work in raw Atheros raw frame injection at ToorCon based on the
Atheros HAL and Sam's driver/HAL wrapper. If you're familliar
I have very little assembler/x86 knowledge.
Could anyone please help me understand what it means to assign a
non-zero value to td_retval in a system call when return value of the
call is zero/success?
I see in syscall() in src/sys/i386/i386/trap.c (btw is this the right
place?) that in such
I have very little assembler/x86 knowledge.
Could anyone please help me understand what it means to assign a
non-zero value to td_retval in a system call when return value of the
call is zero/success?
I see in syscall() in src/sys/i386/i386/trap.c (btw is this the right
place?) that in such
Hello,
Just recently I've run into this:
when compiling kernel in 4.10-RELEASE and in 4-STABLE, options
IPDIVERT does not produce enabled divert in firewall code. Previously
(meaning other machines and previous 4.* variants) the configuration
compiled/worked okay.
I've used the attached config
if i open /dev/ad1s1 for example, i try to read() or write()
not in multiples of 512, it will return an error.
how would i do this? currently, if i need to read 100 bytes,
i read in the full sector, then memcpy the bytes i need.
is there a better way?
-- thx, Nehal
On 10.11.2004, at 16:36, Nehal wrote:
if i open /dev/ad1s1 for example, i try to read() or write()
not in multiples of 512, it will return an error.
how would i do this? currently, if i need to read 100 bytes,
i read in the full sector, then memcpy the bytes i need.
is there a better way?
maybe
jesk wrote:
so can you post the final numbers?
the select procedure didnt changed much in its performance, but the
update-select test performed to around 4000 queries per second.
___
[EMAIL PROTECTED] mailing list
On Wed, 10 Nov 2004, Nehal wrote:
if i open /dev/ad1s1 for example, i try to read() or write() not in
multiples of 512, it will return an error.
how would i do this? currently, if i need to read 100 bytes, i read in
the full sector, then memcpy the bytes i need.
Disks operate only on
Dear folks,
I'm recently investigating large scale deployment and upgrading FreeBSD
RELEASE. It's our tradition to bump RELEASE-pN after a security patch
is applied, however, it seems that there is less method to determine
whether the userland is patched, which is somewhat important for large
I've tracked down several bugs in contigmalloc*() in DragonFly based on
kernel cores provided by David Rhodus. These bugs have just been addressed
in DFly but also need to be addressed in FreeBSD-4, and at least some
work must also be done in FreeBSD-5/6. So someone needs to take
Hi, Julian,
On Wed, Nov 10, 2004 at 09:45:00AM -0800, Julian Elischer wrote:
X-Sieve: CMU Sieve 2.2
Date: Wed, 10 Nov 2004 09:45:00 -0800
From: Julian Elischer [EMAIL PROTECTED]
User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.8a3) Gecko/20041017
X-Accept-Language: en, hu
To: Xin
now I'm completely confused..
can you give a summary again:-)
ok here it is:
-
around 16000 queries per second
(pthreads without process scope threads, sched_4bsd and preemption)
around 7000 queries per second
(pthreads with process scope threads, sched_4bsd and preemption)
On Thu, Nov 11, 2004 at 02:30:46AM +0800, Xin LI wrote:
I have an idea that is somewhat too complex to be included in FreeBSD - we
maintain a ``master'' patchlevel, and two patchlevels indicating the least
``master'' patchlevel that touches kernel or userland. It might be something
like this:
Minor correction. The 'start = 1' part of the patch is not required,
I missed the fact that the second loop was starting at start + 1.
-Matt
___
[EMAIL PROTECTED] mailing list
Omar Punzalan [EMAIL PROTECTED] writes:
Dear Hackers,
I have successfully installed my new (and shiny) freebsd5.3 on one of
our hp rx4640 itanium2 system. With the SMP kernel, it has detected
the 4 CPUs correctly. But then reviewing the /var/log/messages, its
sees only 1G of ram:
Nov
On Nov 10, 2004, at 12:17 PM, Arne Schwabe wrote:
Omar Punzalan [EMAIL PROTECTED] writes:
I have successfully installed my new (and shiny) freebsd5.3 on one of
our hp rx4640 itanium2 system. With the SMP kernel, it has detected
the 4 CPUs correctly. But then reviewing the /var/log/messages, its
Hi,All
Yestoday, I'd upgraded my box to FreeBSD 5.3 RELEASE, but I found
it seems the Embedded MD_ROOT doesn't work fine.
My kernel config:
===
machine i386
cpu I486_CPU
cpu I586_CPU
cpu I686_CPU
ident MINI
device pf
device
On Wed, 10 Nov 2004, Matthew Dillon wrote:
I've tracked down several bugs in contigmalloc*() in DragonFly
based on kernel cores provided by David Rhodus. These bugs have
just been addressed in DFly but also need to be addressed in
FreeBSD-4, and at least some work must also be done in
:Here is the DragonFly commit.
:
:
http://www.dragonflybsd.org/cvsweb/src/sys/vm/vm_contig.c.diff?r1=1.10r2=1.11f=u
:
:FreeBSD-4:
:
: FreeBSD-4 is in the same situation that DFly was in and requires
: the same fixes as the above patch, though note that in FreeBSD-4
: the
23 matches
Mail list logo