- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I try to use hardware sampling event to calculate the L1,L2 miss.
I use the command" amplxe-cl -collect-with runsa -knob event-config=MEM_LOAD_UPOS_RETIRED.LLC_HIT_PS, MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT_PS, MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM_PS,CPU_CLK_UNHALTED.THREAD /home/test"
but it fails saying that "osi_DLL_ERROR:libivytown.so:cannot open shared object file:no such file or directory
unable to open shared library :libivytown.so".
My machine is Ivybridge E5-2692v2. Using "amplxe-cl -collect hotspots" command is successful.
I have no idea about what the error means. Is there any thing wrong with my library file?
Any help would be appreciated.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This indicates that you are using a newly released processor for which support has not been made publicly available, yet. The next update of VTune(TM) Amplifier XE 2013 should include this support. Watch for the update information in the Registration Center and Intel(R) Software Update Manager.
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This indicates that you are using a newly released processor for which support has not been made publicly available, yet. The next update of VTune(TM) Amplifier XE 2013 should include this support. Watch for the update information in the Registration Center and Intel(R) Software Update Manager.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks a lot. I'll watch for the update information.
However, while I turn to use a sandybridge microarchitecture, I encounter 2 problems when I try to use the formula given by @ Kirill Rogozhin (Intel) to calculate the cache miss.
1.When I calculate the L3 miss ratio, I get 90%. But my test application code is just one line using function printf. Therefore , it can't be that
big. And when I calculate the L2 miss ration, the result is even bigger than 1 which is obvious not correct.
2.When I use hardware event :MEM_LOAD_RETIRED.LLC_HIT_PS , it shows that it's a invalid event. But on the platform of Sandybridge,
I think this event should be valid. So, I've no idea what's happening.
Any help would be appreciated.
Sun.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi sun:
I'm sorry. I can't help you with that. I just use the tool. We already have everything defined in the General Exploration analysis type. All micrprocessor issues are defined and identified via this analysis type, so I never calculate my own. :\
If you review our tuning guides, you will see that the General Exploration analysis type correlates directly to the tuning guides. I highly recommend you follow these guides and the advise we provide via them.

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page