Limit the amount of raw data (in MB) to be collected.
Specify Microsoft* runtime environment profiling mode.
Sort data in descending order by the specified column name.
If the product crashes, you can use the amplxe-feedback command tool to package relevant information and send a report to the Intel Customer Support Team.
Basic Crash Report Process
Create a bug report package using the create-bug-report action and desired options.
Identify causes of energy waste on target systems.
The same functions are compared as different instances in the grid panes.
You are using the Function Stackgrouping for the recompiled binary. The Function Stack grouping uses function start addresses and is based on function instances.
This metric shows average load latency in cycles
This metric represents Core cycles fraction CPU dispatched uops on execution port 5 (SNB+: Branches and ALU; HSW+: ALU)
This metric shows how often the machine was stalled on L1, L2, and L3 caches. While cache hits are serviced much more quickly than hits in DRAM, they can still incur a significant performance penalty. This metric also includes coherence penalties for shared data.
Not all arithmetic operations take the same amount of time. Divides and square roots, both performed by the DIV unit, take considerably longer than integer or floating point addition, subtraction, or multiplication. This metric represents cycles fraction where the Divider unit was active.