Re: Dar architecture questions

On 10/16/2012 04:13 PM, Denis Corbin wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Le 16/10/2012 22:09, Kevin Wormington a =E9crit :
>> On 10/16/2012 02:42 PM, Denis Corbin wrote:
>> On 16/10/2012 17:07, Kevin Wormington wrote:
>>>>>> Le 16/10/2012 00:30, Kevin Wormington a =E9crit :
>>>>>>> I strace'd the small backup (14k files) both with and without
>>>>>>> --disable-special-alloc and it appears the difference is in the huge
>>>>>>> number of rt_sigprocmask calls made by special-alloc.
>>>>>>
>>>>>> interesting... I have tests currently running on my side...
>>
>> The so far ran tests also reported huge number of rt_sigprocmask calls
>> but they did not have a big impact (see below). I do not know yet
>> whether this is caused by the --empty argument or by the 64 bits flavor
>> of dar used here. I will do additional tests on Saturday. The used
>> system is a Debian stable one with a slightly tuned kernel (kernel with
>> less modules and many unused features disabled, but I do not see how
>> this may have any concern here).
>>
>>> My systems are stock ubuntu server 10.04 LTS and debian 6 all on 64bit
>>> versions of the OS.
>
> Are they multi-core or multiprocessors host? Mine in an not very recent
> 64bits mono processor host. Linux kernel used had SMP disabled at
> compilation time. This might have an impact.
>

They are a mix of dual processor single-core (hyperthreading enabled) =

older Intel Xeons and newer AMD Opteron quad-core dual processor.  So if =

you have SMP disabled in the kernel that could definitely be a =

difference.  Perhaps it's some sort of race condition with the =

rt_sigprocmask or the mutex lock/unlock that surrounds it.

Did you see my comment in the previous post about the patch?  I was =

thinking that with -v it would display all 1.7 million files instead of =

just the status messages which is why I had checked the quiet flag =

instead or adding a new command line flag to just get verbose program =

status and not file listing.



>>>>   My configure arguments are the same as yours.  I
>>> did a "make install" between rounds of testing vs running it from the
>>> source directory.  This is probably a dumb question -- but I don't know
>>> your dev environment -- are you doing something to cause the rebuilt
>>> library in the source directory to be used vs the one installed in the
>>> system ld path (assuming there is one)?
>
> doing
> 	make clean distclean
> to clean the tree from object and binary files and then
> 	./configure [options] && make
> rebuilds the libdar library and dar command-line tools from scratch. The
> source code relies on libtool, which replaces the src/dar_suite/dar
> binary by a script to handle proper linking with the library located in
> src/libdar. The real binary can be found in src/dar_suite/.libs but if
> you call it directly the dynamic loader of the system will try linking
> it with the system's libdar, which is not what you want here.
>
> You can also use --disable-shared with configure, which makes a faster
> compilation (as no shared object has to be compiled). In this case no
> dynamic link problem occurs (of course :) ), and libtool is not in use
> at all so you have the real (statically linked) binary in place. I use
> that way of doing for step by step debugging mainly (using the
> - --enable-debug option for configure, making it ready for use with gdb).
> This let me keep a stable dar binary installed on the system (yes I do
> use dar for my own backups!) while I can work and test development
> versions of dar/libdar.
>
> So in brief, no need to install the binary to test it.
>
>
>>
>>
>>> [snip ]
>>
>
> [...]
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>
> iQIVAwUBUH3N6ggxsL0D2LGCAQLTvxAAiMg1wY11/ltKxLRLU7Wsonp1N2Vs0DUt
> r8XEhvqmauHK4GlMQM42+l8BnvmWOTYfqyr7vCRePlo/nkzLGbGYBuxycpw1FA8U
> ywqSzXizZsbsA5tMQkl+cixv1Chz/ZGBgT+II9QS/nQidyA5nPkBKzp0F2vmD+tm
> BNwG/XQoclGqnoEyVq8elPpls9LiGFzQ+DbmwNzecusiNiwY68puQoyqfr7o6pLP
> S+HVuwkAlAD1o3KdCWmDCt0AvH2El08rSUi31svOEW3T4/BXJeLAisx20A92b81Z
> A/Que+43PUaYcinT+WdBj7se9Pb77K5sWGE8v7lGdUNod1eaK3gyXHz2GcWjovr+
> fU7T+MGCZP9l43Yy3TZhpQyrOfO8lf7YdWiIepf9yg+YDO2Um/roldXnpZDD3YJ5
> CW3wIzWNjAd+vCMt2uU21UtpslHI4jIWu9XElb6iTKgP0qXvwAnLCMx4jL8yUXNx
> JL3Zws1R+FMCCzTREHA5CvQwBHpP+aC+86Ds67yG2dHPxFm8z8f87pPNZiasrs9W
> Q1qstvkhUYeDOxFsi6uI7moFr0aVttseRD2ZUVI+wbQJTwkVMhA8oRBnAJaUvJFG
> EiFpRK2ee/Ufo/ZKW6MSnhASYwBEnwHvSCPmSbvrWyUFtIqp8qBi9vzQvtQYrXxV
> PCBLdGxbVuw=3D
> =3Du+hz
> -----END PGP SIGNATURE-----
>
> -------------------------------------------------------------------------=
-----
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_sfd2d_oct
> _______________________________________________
> Dar-discussions mailing list
> Dar-discussions-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/dar-discussions
>

---------------------------------------------------------------------------=
---
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_sfd2d_oct