Intel® Parallel Amplifier

bus error for custom collection

Hi Intel experts,

 

I'm using the custom collection to get cache hits and misses for the L1, L2, and L3 caches on an image processing program. When I run the command on 50 frames, the collection works, but as soon as I increase it up to 100 frames it throws out "bus error (core dumped)" after amplxe generates 8% of the report. This is the command I'm using:

 

VTune stack overflow exception

Greetings,

I don't have mailing program installed on that machine, so I post VTune crash report here. Dump file is attached.

---

[Exception]
Address: 0x00007ffbcf730556
Code: 0x00000000c00000fd
CrashedPID: 7996
CrashedTID: 8932
Description: EXCEPTION_STACK_OVERFLOW
Module:
Product: Intel(R) VTune(TM) Amplifier XE 2015 Update 4

[Process]
Process Name: amplxe-gui.exe
Virtual Memory Size: 382956K
Working Set Size: 140448K

Summary view, any way to filter to a subset of cpus?

This question is about the SUMMARY TAB

I have an application that uses a subset of the available CPUs on the server.  Let's say I have 16 cores but I set OMP_NUM_THREADS to 8 and run an analysis.  

IF i want the summary to filter and show summary for just my 8 cpus, and not the entire 16 available, how do I do this?

And yes, I know I can use Filters on the other tabs but there seems to be no way to tell the Summary tab to summarize data on a subset of cpus or threads, is there?  If so, how?

Ron

Understanding to CPU Time and Instructions retired

Hi All,

The following is the snapshot from VTune on my Haswell processor. However, I don't understand that why the CPU time and the number of instructions retired for the highlighted code (vpbroadcastq) are so significantly greater than the others in the same basic block. I thought the number of the retired instructions should be not too different, though there might be cache misses or TLB misses. Can someone explain some possible reasons for it? Thanks.

Cannot find bandwidth value in VTune

Hello,

My CPU is Intel(R) Xeon(R) CPU E5-2620 0 @ 2.00GHz.

I choose the Microarchitecture Analysis/Bandwidth and start the testing.

In the Bandwidth viewpoint panel, where is the average Bandwidth?

I put my screen shot in the attachment. I cannot find the bandwidth.

 

I also use the command line and the result is as follows. The bandwidth is zero. How come?

 

Summary
-------
Elapsed Time:       70.813 
CPU Time:           644.534
Average CPU Usage:  8.906  
CPI Rate:           1.107  

Command line reports matching gui

Hi Intel experts,

I'm currently running a custom advanced hotspots analysis using the amplxe-cl -collect command. The analysis is created by copying Advanced Hotspots analysis and scrolling to the bottom to check the box "Analyze the energy taken by processor cores..."

I run the command: 

amplxe-cl -collect advanced-hotspots-0 -r report -- ./run.sh

Intel VTune for Embedded Windows OS

Hi all,

I am wondering if it is possible to use VTune 2013 or VTune2015 using Windows embedded and, if yes, what is the minimum supported windows edition. I read on the release notes that "embedded editions not supported" but I would like to know if there is a way for example to collect data on an embedded system and view results on standard windows system and/or to perform remote mode using Windows embedded target.

Thanks,

Giorgio

vtune 2013 and very long run hotspot analysis

A customer using vtune 2013 tried this command line for a 40+ hour long run, 16 MPI ranks - I know, it's asking a lot.  Customer reported:

"We've been trying to use it for a problem that runs for 40+ hours/

The elapsed time reported in vtune is far less than it should be, something like 1 hour in this case. Also, the finer-grain times reported when browsing the functions appear to be far smaller than expected.

Suscribirse a Intel® Parallel Amplifier