Visible to Intel only — GUID: GUID-6E83FD80-31F5-4E03-B6A4-A746EFF53730
Visible to Intel only — GUID: GUID-6E83FD80-31F5-4E03-B6A4-A746EFF53730
fast
Maximizes speed across the entire program.
Linux: |
-fast |
macOS: |
-fast |
Windows: |
/fast |
None
OFF |
The optimizations that maximize speed are not enabled. |
This option maximizes speed across the entire program.
Linux
It sets the following options:
-ipo, -O3, -no-prec-div,-static, -fp-model fast=2, and -xHost
macOS
It sets the following options:
-ipo, -mdynamic-no-pic,-O3, -no-prec-div,-fp-model fast=2, and -xHost
Windows
It sets the following options:
/O3, /Qipo, /Qprec-div-, /fp:fast=2, and /QxHost
When option fast is specified, you can override the [Q]xHost option setting by specifying a different processor-specific [Q]x option on the command line. However, the last option specified on the command line takes precedence.
For example:
Linux
If you specify option -fast -xSSE3, option -xSSE3 takes effect. However, if you specify -xSSE3 -fast, option -xHost takes effect.
Windows
If you specify option /fast /QxSSE3, option/QxSSE3 takes effect. However, if you specify /QxSSE3 /fast, option /QxHost takes effect.
For implications on non-Intel processors, refer to the [Q]xHost documentation.
Option fast sets some aggressive optimizations that may not be appropriate for all applications. The resulting executable may not run on processor types different from the one on which you compile. You should make sure that you understand the individual optimization options that are enabled by option fast.
This option only applies to host compilation. When offloading is enabled, it does not impact device-specific compilation. Offloading can only be enabled when using ifx.
Product and Performance Information |
---|
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex. Notice revision #20201201 |
None