Slide 9 of https://software.intel.com/sites/default/files/Using_Intel_VTune_Amplifi...

shows a Sandy Bridge/Ivy Bridge Analysis category of pre-configured profiles.  However, with the XE2015 Update 2, there are categories for Sandy Bridge and Haswell, but none (besides General) which work for Ivy Bridge (see attached).   Which version of VTune, if any, is recommended for Ivy Bridge Memory Access analysis on

Intel(R) Xeon(R) CPU E7-8893 v2 @ 3.40GHz CPU?

Hi, 

I need some help to calculate the data misses in L1,L2 and L3 caches in intel xeon E5-2640. I use vtune to collect the events data (See the attachment). When i read the intel guide i didn't fine some events used in the formula. 

Regards,

Naif

Running "./insmod-sep3" as described in the instructions to build the driver fails:

Checking for PMU arbitration service (PAX) ... detected.
PAX service is accessible to all users.
Executing: insmod ./sep3_15-x32_64-3.11.10-25-desktopsmp.ko
Creating /dev/sep3_15 base devices with major number 246 ... done.
Creating /dev/sep3_15 percpu devices with major number 245 ... done.
Setting group ownership of devices to group "vtune" ... done.
Setting file permissions on devices to "666" ... done.
The sep3_15 driver has been successfully loaded.
Checking for vtsspp driver ... not detected.
Executing: insmod ./vtsspp/vtsspp-x32_64-3.11.10-25-desktopsmp.ko gid=1000 mode=0666
insmod: error inserting './vtsspp/vtsspp-x32_64-3.11.10-25-desktopsmp.ko': -1 Unknown symbol in module

Error:  vtsspp driver failed to load!

You may need to build vtsspp driver for your kernel.
Please see the vtsspp driver README for instructions.

dmesg gives the following output:

[2585520.616825] sep3_15: PMU collection driver v3.15.5 (EMON) has been loaded.
[2585520.616827] sep3_15: Chipset support is enabled.
[2585520.616836] sep3_15: IDT vector 0x21 will be used for handling PMU interrupts.
[2585522.660763] vtsspp: Unknown symbol __pcpu_unique_vtss_recovery_tskd (err 0)
[2585522.660811] vtsspp: Unknown symbol __pcpu_unique_vtss_pcb (err 0)
[2585522.660817] vtsspp: Unknown symbol __pcpu_scope_vtss_pcb (err 0)
[2585522.660833] vtsspp: Unknown symbol __pcpu_unique_vtss_dsa_per_cpu (err 0)
[2585522.660838] vtsspp: Unknown symbol __pcpu_unique_vtss_dsa_cpu_msr (err 0)
[2585522.660844] vtsspp: Unknown symbol __pcpu_unique_vtss_bts_per_cpu (err 0)
[2585522.660850] vtsspp: Unknown symbol __pcpu_unique_vtss_pebs_per_cpu (err 0)
[2585644.662367] vtsspp: Unknown symbol __pcpu_unique_vtss_recovery_tskd (err 0)
[2585644.662423] vtsspp: Unknown symbol __pcpu_unique_vtss_pcb (err 0)
[2585644.662430] vtsspp: Unknown symbol __pcpu_scope_vtss_pcb (err 0)
[2585644.662448] vtsspp: Unknown symbol __pcpu_unique_vtss_dsa_per_cpu (err 0)
[2585644.662455] vtsspp: Unknown symbol __pcpu_unique_vtss_dsa_cpu_msr (err 0)
[2585644.662462] vtsspp: Unknown symbol __pcpu_unique_vtss_bts_per_cpu (err 0)
[2585644.662469] vtsspp: Unknown symbol __pcpu_unique_vtss_pebs_per_cpu (err 0)

Manually running insmod-vtsspp in the sepdk/src/vtsspp directory gives the same error:

Executing: insmod ./vtsspp-x32_64-3.11.10-25-desktopsmp.ko gid=1000 mode=0666
insmod: error inserting './vtsspp-x32_64-3.11.10-25-desktopsmp.ko': -1 Unknown symbol in module

Error:  vtsspp driver failed to load!

You may need to build vtsspp driver for your kernel.
Please see the vtsspp driver README for instructions.

I am running OpenSuse 13.1 with kernel version 3.11.10-25-desktop, installed V-Tune from vtune_amplifier_xe_2015_update1.tar.gz. Any idea on how to solve this?

(I tried googling for "__pcpu_unique_vtss_recovery_tskd", but got no results.

"

It extracts the files, then dies. If I keep the files around and run setup.exe manually it starts, then dies on "loading product.xml".

The bottom of the log file reads:

[t1410 2015.02.24 16.04.08 00000637] [wi_presenter]: INFO: Get clients for following component set: {3149A017-83E0-4629-AFDF-FACF72327B47}
[t1410 2015.02.24 16.04.08 00000638] [msistd]: INFO: Internal call: error code: 259 (No more data is available.
), call: MsiEnumClients( component.c_str(), i, guid ), function:msistd::CInstaller::ComponentClients, file:sources\stdInstallLib.cpp, line:981

[t1410 2015.02.24 16.04.08 00000639] [context]: INFO: MSIComponentCodeUpgradeEntry "Intel(R) VTune(TM) Amplifier XE 2013 for Windows*": Check object with id {30AAA322-96C3-48CD-AEE6-F27E0D7A3577}
[t1410 2015.02.24 16.04.08 0000063a] [context]: INFO: MSIComponentCodeUpgradeEntry "Intel(R) VTune(TM) Amplifier XE 2013 for Windows*": VerifyVersionRange
[t1410 2015.02.24 16.04.08 0000063b] [context]: INFO: MSIComponentCodeUpgradeEntry "Intel(R) VTune(TM) Amplifier XE 2013 for Windows*": version range is undefined, therefore any version is acceptable
[t1410 2015.02.24 16.04.08 0000063c] [wi_product]: INFO: Loading msi begin
[t1410 2015.02.24 16.04.08 0000063d] [wi_product]: INFO: Looking for INSTALLPROPERTY_PRODUCTNAME
[t1410 2015.02.24 16.04.08 0000063e] [wi_product]: INFO:
[t1410 2015.02.24 16.04.08 0000063f] [wi_product]: INFO: analize_package completed
[t1410 2015.02.24 16.04.08 00000640] [msistd]: INFO: Internal call: error code: 1605 (This action is only valid for products that are currently installed.
), call: MsiOpenProduct( product.c_str(), &session ), function:msistd::CInstaller::OpenProduct, file:sources\stdInstallLib.cpp, line:1045

[t1410 2015.02.24 16.04.08 00000641] [wi_product]: INFO: analize_product_code completed
[t1410 2015.02.24 16.04.08 00000642] [wi_product]: INFO: Loading msi end
[t1410 2015.02.24 16.04.08 00000643] [wi_product]: WARNING: Failed to create session object:
[t1410 2015.02.24 16.04.08 00000644] [context]: WARNING: InfoWindowsInstaller: could not execute query
[t1410 2015.02.24 16.04.08 00000645] [context]: INFO: ComponentInfo: could not find any actual info object
[t1410 2015.02.24 16.04.08 00000646] [context]: ERROR: Attempt to create component with undefined Id - input info isn't defined or doesn't have Id or Id() is empty
[t1410 2015.02.24 16.04.08 00000647] [bind_script]: ERROR: Scripting error:
[t1410 2015.02.24 16.04.08 00000648] [bind_script]: ERROR: Uncaught TypeError: Cannot call method 'Id' of undefined
[t1410 2015.02.24 16.04.08 00000649] [bind_script]: ERROR: C:\trash\vtune-install\config\base\upgrade3.js
[t1410 2015.02.24 16.04.08 0000064a] [bind_script]: ERROR: Line: 452: this.Log(" found prod id = \"" + cmp.Id() + "\" name = \""
[t1410 2015.02.24 16.04.08 0000064b] [bind_script]: ERROR:
[t1410 2015.02.24 16.04.08 0000064c] [context]: ERROR: Exception handled calling mediaconfig.Go
[t1410 2015.02.24 16.04.08 0000064d] [context]: INFO: MICL script execution complete. Thank you.

Hi,

            I  am trying to use VTune Amplifier with demo project tachyon_vtune_amp_xe.zip in C:\Program Files (x86)\Intel\VTune Amplifier XE 2015\samples\en\C++, but not sure why it reports below issue, so look for help here

             It says:  Collection failed.The data cannot be displayed.  Highly accurate CPU time collection is disabled for this analysis.To enable this feature, run the product with the administrative privileges.  [Instrumentation Engine]: SYSCALL_INSPECTOR:Too long trace in the NTDLL!NtSetContextThread function.

             I am using it on windows 7, I installed below tools:

                              vistual studio 2013,

                              intel parallel studio XE 2015

                             MKL

                             IPP

Thank you

John

Greetings,

TL;DR I have two generations of processors where identical systems are drastically different in performance and I am trying to use VTune to figure out why.

Full story.
A user recently sent me a complaint that the new cluster was slower than the old cluster. This didn't surprise me too much as we got a really good deal on the procs for the new cluster and went with quantity for the parallel applications over a small number of "faster" processors. The old cluster was a hodge-podge collection of nodes comprised of whatever the fastest proc we could afford at that time. That caused a lot of problems for our parallel users so we wanted to stay close to a uniform cluster this time. I did a lot of application testing with a good sample of our apps and the difference was trivial between the fastest nodes on the old vs the new.

The "fast" nodes on the old cluster are the Xeon Westmere X5687
http://ark.intel.com/products/52578/Intel-Xeon-Processor-X5687-12M-Cache...

The nodes on the new cluster are the Xeon Sandy Bridge E5-2670
http://ark.intel.com/products/64595/Intel-Xeon-Processor-E5-2670-20M-Cac...

When he complained that it was four times slower, that bothered me. I ended up running a series of tests across multiple systems using the same binary and verified that:
* The user just had some crazy bad luck when he initially did his benchmarking. However, he was right in finding that there is a problem.
* One X5687 processor will run the code in 3 min 30 sec on average after many different runs including a fresh reboot. It is really consistent in running between 3:20 and 3:45.
* A second X5687 processor that *should* be identical consistently runs just over 9 minutes.
* One E5-2670 processor consistently averages 7 minutes
* A second E5-2670 processor consistently averages 15 minutes (hence the 4x slower response from the user)
* I have a wide variety of ranges from other E5-2670 processors with the average sitting closer to 9 minutes.
* I have a Xeon X5675 that is my "If I break it no one cares" test system which I can beat up with VTune and testing. It consistently runs just under 4 minutes.

My theories for the discrepancies between the processors are:
* Possibly some sort of cache/memory alignment problem?
* There is only one random number generated at the very beginning. The code should be pretty uniform after that point and the tests were all with the same binary. Maybe I need to compile with ifort to target specific architectures? Compiling a binary for each processor family did not seem to make much of a difference, but maybe there are other flags I should try.

However, those points would explain the difference between processors, not the differences between the same processor type.

Maybe it is something as simple as a CPU feature (like virtualization flag) enabled on one host and not the other. However, I can't seem to find that difference.

So I have turned to VTune in an effort to figure this out. VTune has pointed out several issues with the code (which we are working on) and there are improvements to be made, but so far I don't see anything that would tell me why it runs slow on one and faster on an "identical" system.

If it was just that one processor type was faster than the other then this wouldn't be of any issue. But I have been running tests, pouring over VTune output, and hitting up forums for the past few days and I feel like I am not getting anywhere in explaining this mystery.

I would greatly appreciate advice/suggestions on how I might be able to better figure out why there is such a large difference between "identical" systems. What should i be looking for in VTune? Is there a specific test I should run?

Thanks!

Hi,

I've just upgraded to Update 2 (from update 1) running inside VS2010.  On some C++ solutions I'm finding the New Analysis and Project properties greyed out and I can run any analysis on those projects.  The solutions still have the the results files from the previous version and I can open those fine.  On other solutions I've checked the New Analysis button isn't greyed out and I can run those without problem.

This solution has a single executable and 5 static libraries linked into it.   They've all got the same compiler settings (debug info on in C++ and debug output on in the linker).   I can run an analysis on the .exe file by loading that exe into a new solution in VS2010 so it appears the app has everything required in it.  Have any requirement changed in the projects, or is there any other reason I wouldn't be able to run an analysis on this solution any more?

Thanks,

Craig.

Hello,

I'm trying to use Vtune on a host with Xeon Phi cards. I have Intel Xeon Phi coprocessor (native) as the target system and 0(mic0) as the card number in the amplxe-gui. When I try to do an analysis, vtune complains:

ssh: Could not resolve hostname 0(mic0): Name or service not known

(see attached screen shot for vtune version info and the error message)

I can do 'ssh mic0' without any problem though, and mic0 is correctly set up in /etc/hosts. I'm member of the vtune group and the sep and vtssp drivers are loaded on the mic and on the host (here also pax).

Any suggestions about what could be wrong?

Hi,

I've just downloaded the evaluation version of VTune_Amplifier_XE_2015_update2 and tried using it for performance evaluation.

The performance evaluation has succeeded, but the problem is that I see all the text reversed (from right to left), hence I cannot understand the names of functions etc. Actually all the view panel is reversed.

Can anyone give me an advice about how to solve this problem?

Thanks.

Hi everyboby,

I'm new here in the forum so i would like to apologize if my post isn't in accordance with the rules.

I'm working on a scientífic paper on the university and need scan a code regardind the amount and tipe of branchs and export a log file that is readble nd contains the base adress and the target adress of each branch, and it was taken or not.

I started using the Vtune a short time and I couldn't find a way to do what i need.

So i want to know if it is possible to do this?If Yes, how can i do?And if no, you could apoint me some tool that i can do this?

Thank you.

Hi guys,

I just started exploring VTune trying to identify the hotspot in my program. After I did the "source", I typed "amplxe-gui", but it gave me:

"Error: Unable to initialize gtk, is DISPLAY set properly?"

I searched this problem for a long time but could not find anything related. Can any one help me?

Thanks a lot!

Terrence

I seem to have problems installing VTune Amplifier on Linux:

1. Downloaded "vtune_amplifier_xe_2015_update2.tar.gz"
2. Uncompressed and cd to "vtune_amplifier_xe_2015_update2"
3. Ran "./install.sh" as root
4. I got the following message:

Nothing to install
--------------------------------------------------------------------------------
Based on availability of RPM packages and coexistence checks, there is nothing
to install from this package. Installation will quit.
--------------------------------------------------------------------------------
q. Quit
--------------------------------------------------------------------------------
Please type a selection or press "Enter" to accept default choice [q]: ^C
Interrupt install process?(Yes/No)[No]:

I'm using Ubuntu 14.04:

Linux slab2 3.13.0-44-generic #73-Ubuntu SMP Tue Dec 16 00:22:43 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

As the message is not very explicit I will appreciate any help.

Thanks

Hello,

we would like to try out Intel VTune and have downloaded the LINUX version and tried the installation without root access.

The installation guide says the followings:

> Execute the ./install.sh script file (available at the top level in the extracted contents) as a root user. Activation is required.

However it says later "The install can be run as a non-root user, but in this case not all collectors will be available to the user", when I try to install it as an evaluation version the following comes up:

> To install an evaluation version of this software, it is recommended that you have root privileges to complete a one time initialization activity securing the product licensing. You can also install an evaluation version of this software without root by obtaining a license file via email.

>

> 1. Become root to install evaluation [default] 2. Become root to

> install evaluation using sudo privileges and password 3. Install

> evaluation software using a license file 4. Install evaluation

> software with help of system administrator

Is it possible to get a evaluation license file? A file ending with .lic is needed. Or what else can we do to use the evalution without root access?

Thanks a lot for your support and best regards,

Andrea Froelich

Without root privileges, it looks like we need a license file.

Salve a tutti,

sono interessato a valutare i cache miss e il numero di operazioni floating point con riferimento ad un programma MPI.

Da riga di comando ho provato:

mpirun -n 4 -l amplxe-cl -result-dir XXX -collect hotspots eseguibile

Il tipo di analisi "HOTSPOTS"è sufficiente per ottenere quanto desiderato?

Se sì. una volta aperti i risultati usando l'interfaccia grafica (amplxe-gui), dove sono reperibili le informazioni di cui sopra?

Se no, che comando è necessario utilizzare?

Grazie a tutti in anticipo!

I am trying to use vtune to analyze an mpi program which is a mixture of python, c and fortran. I am able to do hotspots collection for this program just fine, but when I try to do advanced hotspots or snb-general-exploration it hangs.

I run with

mpirun -np 1 -l amplxe-cl -collect snb-general-exploration --result-dir appge -- python <...args...> : -np 7 python <...args...>

I'm running on a single node with 16 cores for now but would like to do this analysis for larger cases on multiple nodes. Compiler version is 14.0.2.

I am able to run a "hello world" mpi application written in c with the snb-general-exploration or advanced-hotspots collection using a command similar to what I'm trying to use with the python/c/fortran app, so it is not an issue with sampling drivers.

Any ideas how to get this to work? Thanks in advance.

I'm trying this command:

mpirun -n 1 amplxe-cl -collect advanced-hotspots -result-dir DIR -- HELLO_WORLD

but the following error appears:

"Error: Failed to create a tb6 file because 0 samples were collected.".

What's wrong?

Thanks!

Hi,

I'm sorry, but I can't find simple instructions on how to run.  Maybe you could make a 'sticky' in the forum with this information?

What do I do?  Which compiler flags does vtune need.  Do I just compile with -g then type amplxe-gui?  Where are amplxe-cli command instructions located?  I direct an input file into the code, i.e. %code < read_input.  How do I put that in the gui?

If there is a 'vtune for dummies', please send me the link.

Thanks!

The following command

mpirun -n 4 /opt/intel/vtune_amplifier_xe/bin64/amplxe-cl -collect-with runsa -knob event-config=FP_COMP_OPS_EXE.X87 -result-dir /home/fabrizio/Desktop/testt/ -- /home/fabrizio/Desktop/test

returns error in subject.

I tried a solution explained in an another post (./insmod -sep3 -r), but it seems not to be my case, because with ONE process I don't have error about PMU resource(s), but an error about "Failed to create a tb6 file".

How could I solve?

Thanks!

Hi everyone,

I need to install VTune Amplifier but I don't have root privileges, so I have to ask my system administrator to install drivers.

My question is: since I installed a FREE 30-day version on another system, Will I be able to collect data after this license expires? 

Regards

Hello,

I'm trying to analyze simple serial aplication (matrix multipication) dedicated for Intel Xeon Phi coprocessor using VTune. I created native aplication, using -mmic flag and next copy it to mic0 using scp command. Next I ran analysis as shown in the link below :

https://software.intel.com/en-us/articles/how-to-analyze-xeon-phi-coproc...

Upon completion I got the following result (pic.: vtune1, vtune2). In analisys result i did not get information as shown in the link above. I obtained the same result for General Exploration. Nothing information about implemented function etc.

Could You give me some suggestion what could be wrong?