3:54 p.m.
Doesnt "currently asm" indicate asm spinlocks.. Which can be unreliable. These
can be even more unreliable with unpatched bdb?
So perhaps you should try posix style semaphores.. Forcing posix/pthreads both at compile
time and at run time..
-----Original Message-----
From: Howard Chu <hyc(a)symas.com>
Sent: Tuesday, 17 February 2009 1:26 AM
To: Dieter Kluenter <dieter(a)dkluenter.de>
Cc: openldap-devel(a)openldap.org
Subject: Re: 2.4.14 prerelease call for testing #2
Dieter Kluenter wrote:
Aaron Richton<richton(a)nbcs.rutgers.edu> writes:
[..snip..]
#0 0x00002b77bc3da12d in __lock_detect ()
from /usr/local/BerkeleyDB.4.7/lib/libdb-4.7.so
Current language: auto; currently asm
Which is pretty squarely inside BerkeleyDB. But nobody else reported any such
problem yet. Seems you'll have to check your BDB build.
8:15 a.m.
New subject: 2.4.14 prerelease call for testing #2
Brett Maxfield <brett.maxfield(a)gmail.com> writes:
Doesnt "currently asm" indicate asm spinlocks.. Which can
be unreliable. These can be even more unreliable with unpatched bdb?
So perhaps you should try posix style semaphores.. Forcing posix/pthreads both at compile
time and at run time..
db-4.7.25 has been build with --enable-posixmutexes
--enable-pthread_api
[...]
-Dieter
--
Dieter Klünter | Systemberatung
http://www.dpunkt.de/buecher/2104.html
sip: +49.180.1555.7770535
GPG Key ID:8EF7B6C6
53°08'09,95"N
10°08'02,42"E
2:38 a.m.
New subject: 2.4.14 prerelease call for testing #2
On Thu, Feb 19, 2009 at 2:15 AM, Dieter Kluenter <dieter(a)dkluenter.de> wrote:
Brett Maxfield <brett.maxfield(a)gmail.com> writes:
> Doesnt "currently asm" indicate asm spinlocks.. Which can be unreliable.
These can be even more unreliable with unpatched bdb?
>
> So perhaps you should try posix style semaphores.. Forcing posix/pthreads both at
compile time and at run time..
db-4.7.25 has been build with --enable-posixmutexes
--enable-pthread_api
I use ../dist/configure --disable-java --enable-posixmutexes
--with-mutex=POSIX/pthreads --enable-largefile --enable-cxx
--with-mutex=POSIX/pthreads is important as it selects posix mutex at
runtime, not compile time.
If i remember above was also the workaround for before 4.7 patch #3
was released, which would cause asm to be selected by the selection
algorithm and it was asm spinlocks which had the bug.
On solaris at least, i think POSIX/pthreads is better as the thread
library is pretty good (IMHO) but i suspect newer Linuxes will have
good POSIX/pthreads also.. the BDB library makes debatable choice of
asm at runtime even if the platform has a good pthreads
implementation.
I think on multi-threaded platform, spinlocks are non-optimal as there
are more possible race conditions and consumtion of large numbers of
locking / locker objects with openldap using asm spinlocks.. it is
probably fine / optimal for single threaded use though.
Cheers
Brett
7:59 a.m.
New subject: 2.4.14 prerelease call for testing #2
"Brett @Google" <brett.maxfield(a)gmail.com> writes:
On Thu, Feb 19, 2009 at 2:15 AM, Dieter Kluenter
<dieter(a)dkluenter.de> wrote:
>
> Brett Maxfield <brett.maxfield(a)gmail.com> writes:
>
> > Doesnt "currently asm" indicate asm spinlocks.. Which can be
unreliable. These can be even more unreliable with unpatched bdb?
> >
> > So perhaps you should try posix style semaphores.. Forcing posix/pthreads both
at compile time and at run time..
>
> db-4.7.25 has been build with --enable-posixmutexes
> --enable-pthread_api
I use ../dist/configure --disable-java --enable-posixmutexes
--with-mutex=POSIX/pthreads --enable-largefile --enable-cxx
--with-mutex=POSIX/pthreads is important as it selects posix mutex at
runtime, not compile time.
I don't think that this configure option is necessary on linux, as
libpthread is the default AFAIK.
from configure --help:
--with-mutex=MUTEX Select non-default mutex implementation.
-Dieter
--
Dieter Klünter | Systemberatung
http://www.dpunkt.de/buecher/2104.html
sip: +49.180.1555.7770535
GPG Key ID:8EF7B6C6
53°08'09,95"N
10°08'02,42"E
1:20 p.m.
New subject: BDB configuration
Dieter Kluenter wrote:
"Brett @Google"<brett.maxfield@gmail.com> writes:
> On Thu, Feb 19, 2009 at 2:15 AM, Dieter Kluenter<dieter(a)dkluenter.de> wrote:
>> Brett Maxfield<brett.maxfield(a)gmail.com> writes:
>>
>>> Doesnt "currently asm" indicate asm spinlocks.. Which can be
unreliable. These can be even more unreliable with unpatched bdb?
>>>
>>> So perhaps you should try posix style semaphores.. Forcing posix/pthreads
both at compile time and at run time..
>> db-4.7.25 has been build with --enable-posixmutexes
>> --enable-pthread_api
> I use ../dist/configure --disable-java --enable-posixmutexes
> --with-mutex=POSIX/pthreads --enable-largefile --enable-cxx
>
> --with-mutex=POSIX/pthreads is important as it selects posix mutex at
> runtime, not compile time.
I don't think that this configure option is necessary on linux, as
libpthread is the default AFAIK.
from configure --help:
--with-mutex=MUTEX Select non-default mutex implementation.
The default on Linux is "hybrid" which means spinlocks and pthread locks. The
spinlocks are extremely CPU intensive and have shown no benefit in our
testing. You should always use --with-mutex=POSIX/pthreads. This has already
been discussed on this list before. Please read the archives instead of
posting redundant posts.
2.4.14 was released days ago. Please change the subject from "2.4.14
prerelease" if you're going to continue replying to these threads.
--
-- Howard Chu
CTO, Symas Corp. http://www.symas.com
Director, Highland Sun http://highlandsun.com/hyc/
Chief Architect, OpenLDAP http://www.openldap.org/project/
5399
days inactive
5402
days old
4 comments
4 participants
participants (4)
-
Brett ＠Google -
Brett Maxfield
-
Dieter Kluenter -
Howard Chu