Intel® Advisor Help
View the highest vector ISA (Instruction Set Architecture) used in the loop among ISAs used for individual instructions.
ISA higher than the ISA of your current hardware appears when you add corresponding codepaths with x, Qx / ax, Qax compiler options. To see ISA of non-executed codepaths, enable the Intel® Advisor Analyze non-executed codepaths option in Intel® Advisor Project Properties (find link to Help topic in See Also).
Metric |
Description |
Interpretation |
---|---|---|
AVX512 |
This metric is reported in case when a loop includes instructions provided with the Intel® Advanced Vector Extensions 512 (Intel® AVX-512). |
... |
AVX2 |
This metric is reported in case when a loop includes instructions provided with the Intel® Advanced Vector Extensions 2 (Intel® AVX2). |
... |
AVX |
This metric is reported in case when a loop includes instructions provided with the Intel® Advanced Vector Extensions (Intel® AVX). |
... |
SSE5 |
This metric is reported in case when a loop includes instructions provided with the Intel® Streaming SIMD Extensions 5 (Intel® SSE5). |
... |
SSE4A |
This metric is reported in case when a loop includes instructions provided with the Intel® Streaming SIMD Extensions 4A (Intel® SSE4A). |
... |
SSE42 |
This metric is reported in case when a loop includes instructions provided with the Intel® Streaming SIMD Extensions 4.2 (Intel® SSE4.2). |
... |
SSE4 |
This metric is reported in case when a loop includes instructions provided with the Intel® Streaming SIMD Extensions 4 (Intel® SSE4). |
... |
SSSE3 |
This metric is reported in case when a loop includes instructions provided with the Intel® Supplemental Streaming SIMD Extensions 3 (Intel® SSSE3). |
... |
SSE3 |
This metric is reported in case when a loop includes instructions provided with the Intel® Streaming SIMD Extensions 3 (Intel® SSE3). |
... |
SSE2 |
This metric is reported in case when a loop includes instructions provided with the Intel® Streaming SIMD Extensions 2 (Intel® SSE2). |
... |
SSE |
This metric is reported in case when a loop includes instructions provided with the Intel® Streaming SIMD Extensions (Intel® SSE). |
... |
MMX |
This metric is reported in case when a loop includes instructions provided with the Intel® MMX™. |
... |