This topic lists the options or option settings that provide new functionality in this release.
If no label appears, the option is available on all supported systems.
If "only" appears in the label, the option is only available on the identified system.
For more details on the options, refer to the individual option descriptions.
For information on conventions used in this table, see Notational Conventions.
New compiler options or option settings are listed in tables below:
The first table lists new options or option settings that are available on Windows* systems.
The second table lists new options or option settings that are available on Linux* and macOS* systems. If an option is only available on one of these operating systems, it is labeled.
Windows* Options |
Description |
---|---|
/assume |
New setting [no]old_inquire_recl has been added. It determines the value of the RECL= specifier on an INQUIRE statement for an unconnected unit or a unit connected for stream access. New setting [no]old_ldout_zero has been added. It determines the format of a floating-point zero produced by list-directed output. old_ldout_zero uses exponential format, noold_ldout_zero uses fractional format. Default: /assume:old_inquire_recl /assume:old_ldout_zero |
/check |
New setting [no]udio_iostat has been added. It determines whether conformance checking occurs when user defined derived type input/output routines are executed. Default: /nocheck |
-f[no-]openmp-device-lib:library |
Enables or disables certain device libraries for an OpenMP* target. Default: OFF |
/Qauto-arch |
Tells the compiler to generate multiple, feature-specific auto-dispatch code paths for x86 architecture processors if there is a performance benefit. Default: OFF |
/Qconditional-branch:keyword |
Lets you identify and fix code that may be vulnerable to speculative execution side-channel attacks, which can leak your secure data as a result of bad speculation of a conditional branch direction. Default: /Qconditional-branch:keep |
/Qopenmp-target-buffers:keyword |
Enables a way to overcome the problem where some OpenMP* offload SPIR-V* devices produce incorrect code when a target object is larger than 4GB. This feature is only available for ifx. Default: /Qopenmp-target-buffers:default |
/Qopt-multiple-gather-scatter-by-shuffles[-] |
Enables or disables the optimization for multiple adjacent gather/scatter type vector memory references. Default: varies; see the option description |
Linux* and macOS* Options |
Description |
---|---|
-assume |
New setting [no]old_inquire_recl has been added. It determines the value of the RECL= specifier on an INQUIRE statement for an unconnected unit or a unit connected for stream access. New setting [no]old_ldout_zero has been added. It determines the format of a floating-point zero produced by list-directed output. old_ldout_zero uses exponential format, noold_ldout_zero uses fractional format. Default: -assumeold_inquire_recl -assume old_ldout_zero |
-check |
New setting [no]udio_iostat has been added. It determines whether conformance checking occurs when user defined derived type input/output routines are executed. Default: -nocheck |
-f[no-]openmp-device-lib:library |
Enables or disables certain device libraries for an OpenMP* target. Default: OFF |
-fopenmp-target-buffers=keyword |
Enables a way to overcome the problem where some OpenMP* offload SPIR-V* devices produce incorrect code when a target object is larger than 4GB. This feature is only available for ifx. Default: -fopenmp-target-buffers=default |
-mauto-arch |
Tells the compiler to generate multiple, feature-specific auto-dispatch code paths for x86 architecture processors if there is a performance benefit. Default: OFF |
-mconditional-branch=keyword |
Lets you identify and fix code that may be vulnerable to speculative execution side-channel attacks, which can leak your secure data as a result of bad speculation of a conditional branch direction. Default: -mconditional-branch=keep |
-q[no-]opt-multiple-gather-scatter-by-shuffles |
Enables or disables the optimization for multiple adjacent gather/scatter type vector memory references. Default: varies; see the option description |
Product and Performance Information |
---|
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex. Notice revision #20201201 |