SPEC CPU Flag Description for Intel oneAPI Compiler

Intel-ic2025-official-linux64-cpu2026-0.902 SPEC CPU Flag Description for Intel oneAPI Compiler

submit= MYMASK=`printf '0x%x' $((1<<$SPECCOPYNUM))`; /usr/bin/taskset $MYMASK $command

When running multiple copies of benchmarks, the SPEC config file feature submit is used to cause individual jobs to be bound to specific processors. This specific submit command, using taskset, is used for Linux64 systems without numactl.
Here is a brief guide to understanding the specific command which will be found in the config file:

/usr/bin/taskset [options] [mask] [pid | command [arg] ... ]:
taskset is used to set or retreive the CPU affinity of a running process given its PID or to launch a new COMMAND with a given CPU affinity. The CPU affinity is represented as a bitmask, with the lowest order bit corresponding to the first logical CPU and highest order bit corresponding to the last logical CPU. When the taskset returns, it is guaranteed that the given program has been scheduled to a specific, legal CPU, as defined by the mask setting.
[mask]: The bitmask (in hexadecimal) corresponding to a specific SPECCOPYNUM. The specific example above, computes this mask value in the variable $MYMASK. The value of this mask for the first copy of a rate run will be 0x00000001, for the second copy of the rate will be 0x00000002 etc. Thus, the first copy of the rate run will have a CPU affinity of CPU0, the second copy will have the affinity CPU1 etc.
$command: Program to be started, in this case, the benchmark instance to be started.

submit= numactl --localalloc --physcpubind=$SPECCOPYNUM $command

When running multiple copies of benchmarks, the SPEC config file feature submit is used to cause individual jobs to be bound to specific processors. This specific submit command is used for Linux64 systems with support for numactl.
Here is a brief guide to understanding the specific command which will be found in the config file:

syntax: numactl [--interleave=nodes] [--preferred=node] [--physcpubind=cpus] [--cpunodebind=nodes] [--membind=nodes] [--localalloc] command args ...
numactl runs processes with a specific NUMA scheduling or memory placement policy. The policy is set for a command and inherited by all of its children.
"--localalloc" instructs numactl to keep a process memory on the local node while "-m" specifies which node(s) to place a process memory.
"--physcpubind" specifies which core(s) to bind the process. In this case, copy 0 is bound to processor 0 etc.
For full details on using numactl, please refer to your Linux documentation, 'man numactl'

]]>

numactl --interleave=all "runspec command"

Launching a process with numactl --interleave=all sets the memory interleave policy so that memory will be allocated using round robin on nodes. When memory cannot be allocated on the current interleave target fall back to other nodes.

KMP_STACKSIZE

Specify stack size to be allocated for each thread.

KMP_AFFINITY

Syntax: KMP_AFFINITY=[<modifier>,...]<type>[,<permute>][,<offset>]
The value for the environment variable KMP_AFFINITY affects how the threads from an auto-parallelized program are scheduled across processors.
It applies to binaries built with -qopenmp and -parallel (Linux and Mac OS X) or /Qopenmp and /Qparallel (Windows).
modifier:
    granularity=fine Causes each OpenMP thread to be bound to a single thread context.
type:
    compact Specifying compact assigns the OpenMP thread <n>+1 to a free thread context as close as possible to the thread context where the <n> OpenMP thread was placed.
    scatter Specifying scatter distributes the threads as evenly as possible across the entire system.
permute: The permute specifier is an integer value controls which levels are most significant when sorting the machine topology map. A value for permute forces the mappings to make the specified number of most significant levels of the sort the least significant, and it inverts the order of significance.
offset: The offset specifier indicates the starting position for thread assignment.

Please see the Thread Affinity Interface article in the Intel Composer XE Documentation for more details.

Example: KMP_AFFINITY=granularity=fine,scatter
Specifying granularity=fine selects the finest granularity level and causes each OpenMP or auto-par thread to be bound to a single thread context.
This ensures that there is only one thread per core on cores supporting HyperThreading Technology
Specifying scatter distributes the threads as evenly as possible across the entire system.
Hence a combination of these two options, will spread the threads evenly across sockets, with one thread per physical core.

Example: KMP_AFFINITY=compact,1,0
Specifying compact will assign the n+1 thread to a free thread context as close as possible to thread n.
A default granularity=core is implied if no granularity is explicitly specified.
Specifying 1,0 sets permute and offset values of the thread assignment.
With a permute value of 1, thread n+1 is assigned to a consecutive core. With an offset of 0, the process's first thread 0 will be assigned to thread 0.
The same behavior is exhibited in a multisocket system.

OMP_NUM_THREADS

Sets the maximum number of threads to use for OpenMP* parallel regions if no other value is specified in the application. This environment variable applies to both -qopenmp and -parallel (Linux and Mac OS X) or /Qopenmp and /Qparallel (Windows). Example syntax on a Linux system with 8 cores: export OMP_NUM_THREADS=8

OMP_STACKSIZE

The OMP_STACKSIZE environment variable controls the size of the stack for threads created by the OpenMP implementation

Set stack size to unlimited

The command "ulimit -s unlimited" is used to set the stack size limit to unlimited.

Free the file system page cache

The command "echo 3> /proc/sys/vm/drop_caches" is used to free up the filesystem page cache as well as reclaimable slab objects like dentries and inodes.

MALLOC_CONF

Used for Jemalloc tuning at runtime. MALLOC_CONF=retain:true will retain unused virtual memory for later resue rather than discarding it.

Red Hat Specific features

Transparent Huge Pages: On RedHat EL 6 and later, Transparent Hugepages increase the memory page size from 4 kilobytes to 2 megabytes. Transparent Hugepages provide significant performance advantages on systems with highly contended resources and large memory workloads. If memory utilization is too high or memory is badly fragmented which prevents hugepages being allocated, the kernel will assign smaller 4k pages instead.
Hugepages are used by default unless the /sys/kernel/mm/redhat_transparent_hugepage/enabled field is changed from its RedHat EL6 default of 'always'.

]]>

/path/to/{icc|icpc|ifort|icx|icpx|ifx} Invoke the Intel oneAPI DPC++ C compiler.

]]> Invoke the Intel oneAPI DPC++ C++ compiler.

]]> Invoke the Intel C compiler.

]]> Invoke the Intel C++ compiler.

]]> Invoke the Intel Fortran compiler.

]]> Invoke the Intel oneAPI Fortran compiler.

]]> Supress compiler warnings.

]]> -m32 Compiles for a 32-bit (LP32) data model.

]]> -m64 Compiles for a 64-bit (LP64) data model.

]]> Sets the language standard to the specified version, for example c18, c++17, f2008.

]]> Sets the language dialect to conform to the indicated Fortran standard.

]]> This flag prevents floating-point operations from being reordered, maintaining the specified order of operations.

]]> Enable support for Posix Threads. Note that C++ programs using std::thread may require this flag. Enable the compiler to generate multi-threaded code based on the OpenMP* directives. Similar behavior was granted by -openmp in previous versions. Enable the compiler to generate multi-threaded code based on the OpenMP* directives. Similar behavior was granted by -openmp in previous versions. Enable the compiler to generate multi-threaded code based on the OpenMP* directives. Determines whether the compiler uses the current Fortran Standard rules or the old Fortran 2003 rules when interpreting assignment statements. Option standard-realloc-lhs (the default), tells the compiler that when the left-hand side of an assignment is an allocatable object, it should be reallocated to the shape of the right-hand side of the assignment before the assignment occurs. This is the current Fortran Standard definition. This feature may cause extra overhead at run time. This option has the same effect as option assume realloc_lhs.

If you specify nostandard-realloc-lhs, the compiler uses the old Fortran 2003 rules when interpreting assignment statements. The left-hand side is assumed to be allocated with the correct shape to hold the right-hand side. If it is not, incorrect behavior will occur. This option has the same effect as option assume norealloc_lhs.

]]> -align array32byte specifies that analyzed arrays should be aligned to 32byte boundaries The align toggle changes how data elements are aligned. Variables and arrays are analyzed and memory layout can be altered. Specifying array32byte will look for opportunities to transform and reailgn arrays to 32byte boundaries.

]]> Allow optimizations for floating point arithmetic that assume arguments and results are not NaNs or Infinities

]]> Link 64-bit applications to use large pages Linker options to link a 64-bit application to use large pages for the .bss, .data and .text sections.

]]> Link 32-bit applications to use large pages Linker options to link a 32-bit application to use large pages for the .bss, .data and .text sections.

]]> Enable optimizations for speed and disables some optimizations that increase code size and affect speed.
To limit code size, this option:

Enable global optimization; this includes data-flow analysis, code motion, strength reduction and test replacement, split-lifetime analysis, and instruction scheduling.
Disables intrinsic recognition and intrinsics inlining.

The O1 option may improve performance for applications with very large code size, many branches, and execution time not dominated by code within loops.

-O1 sets the following options:
-funroll-loops0, -fno-builtin, -mno-ieee-fp, -fomit-framepointer, -ffunction-sections, -ftz ]]> Enable optimizations for speed. This is the generally recommended optimization level. This option also enables:
- Inlining of intrinsics
- Intra-file interprocedural optimizations, which include:
- inlining
- constant propagation
- forward substitution
- routine attribute propagation
- variable address-taken analysis
- dead static function elimination
- removal of unreferenced variables
- The following capabilities for performance gain:
- constant propagation
- copy propagation
- dead-code elimination
- global register allocation
- global instruction scheduling and control speculation
- loop unrolling
- optimized code selection
- partial redundancy elimination
- strength reduction/induction variable simplification
- variable renaming
- exception handling optimizations
- tail recursions
- peephole optimizations
- structure assignment lowering and optimizations
- dead store elimination

]]> Enable O2 optimizations plus more aggressive optimizations, such as prefetching, scalar replacement, and loop and memory access transformations. Enable optimizations for maximum speed, such as:

Loop unrolling, including instruction scheduling
Code replication to eliminate branches
Padding the size of certain power-of-two arrays to allow more efficient cache use.

On IA-32 and Intel EM64T processors, when O3 is used with options -ax or -x (Linux) or with options /Qax or /Qx (Windows), the compiler performs more aggressive data dependency analysis than for O2, which may result in longer compilation times. The O3 optimizations may not cause higher performance unless loop and memory access transformations take place. The optimizations may slow down code in some cases compared to O2 optimizations. The O3 option is recommended for applications that have loops that heavily use floating-point calculations and process large data sets.

]]> Enable O3 optimizations plus more aggressive optimizations, such as -ffinite-math-only –no-prec-div

]]> Invoke Intel C++ compiler next generation code generator. Performs link time optimizations, which is also known as Interprocedural Optimizations. Instructs the compiler to use the GNU gold linker (gold) instead of the system linker when linking object files.

]]> Generate floating-point arithmetic for selected unit unit. Here use scalar floating-point instructions present in the SSE instruction set

]]> Use specified C language standard.

]]> Tells the compiler the maximum number of times to unroll loops. For example -funroll-loops0 would disable unrolling of loops. -fno-builtin disables inline expansion for all intrinsic functions. This option trades off floating-point precision for speed by removing the restriction to conform to the IEEE standard. EBP is used as a general-purpose register in optimizations. Places each function in its own COMDAT section. Flushes denormal results to zero. -qopt-mem-layout-trans=3 3 Controls the level of memory layout transformations performed by the compiler. This option can improve cache reuse and cache locality.

0: Disables memory layout transformations. This is the same as specifying -qno-opt-mem-layout-trans
1: Enable basic memory layout transformations like structure splitting, structure peeling, field inlining, field reordering, array field transpose, increase field alignment etc.
2: Enable more memory layout transformations like advanced structure splitting. This is the same as specifying -qopt-mem-layout-trans
3: Enable more memory layout transformations like copy-in/copy-out of structures for a region of code. You should only use this setting if your system has more than 4GB of physical memory per core.
4: Compiler is more aggressive in using memory layout transformations. You should only use this setting if your system has more than 4GB of physical memory per core.

]]> -unroll<n> This option sets the maximum number of times a loop can be unrolled, to $1. n. For example, -unroll1 will unroll loops just once. To disable loop unrolling, use -unroll0. The -par-schedule option lets you specify a scheduling algorithm or a tuning method for loop iterations.
It specifies how iterations are to be divided among the threads of the team. This option affects performance
tuning and can provide better performance during auto-parallelization.

]]> -par-schedule-static=n The chunks are assigned to threads in the team in a round-robin fashion in the order of the thread number.
Note that the last chunk to be assigned may have a smaller number of iterations. If n is not specified,
the iteration space is divided into chunks that are approximately equal in size, and each thread is assigned at most one chunk.
]]> n. For example, -par-schedule-static=32768 will split iterations into chunks of size 32768. This option enables additional interprocedural optimizations for single file compilation. These optimizations are a subset of full intra-file interprocedural optimizations. One of these optimizations enables the compiler to perform inline function expansion for calls to functions defined within the current source file. Multi-file ip optimizations that includes:
- inline function expansion
- interprocedural constant propogation
- dead code elimination
- propagation of function characteristics
- passing arguments in registers
- loop-invariant code motion

]]> This option instructs the compiler to analyze and transform the program so that 64-bit pointers are shrunk to 32-bit pointers, and 64-bit longs (on Linux) are shrunk into 32-bit longs wherever it is legal and safe to do so. In order for this option to be effective the compiler must be able to optimize using the -ipo/-Qipo option and must be able to analyze all library/external calls the program makes.

This option requires that the size of the program executable never exceeds 2^32 bytes and all data values can be represented within 32 bits. If the program can run correctly in a 32-bit system, these requirements are implicitly satisfied. If the program violates these size restrictions, unpredictable behavior might occur.

]]> This option instructs the compiler to analyze and transform the program so that 64-bit pointers are shrunk to 32-bit pointers wherever it is legal and safe to do so. In order for this option to be effective, the compiler must optimize using the -ipo option and must be able to analyze all library/external calls the program makes. This option has no effect unless you specify setting SSE3 or higher for option -x.

This option requires that the application cannot exceed a 32-bit address space, otherwise unpredictable results can occur.

]]> This option specifies that the main program is not written in Fortran. It is a link-time option that prevents the compiler from linking for_main.o into applications.

For example, if the main program is written in C and calls a Fortran subprogram, specify -nofor-main when compiling the program with the ifort command. If you omit this option, the main program must be a Fortran program.

]]> Disable scalar replacement -scalar-rep enables scalar replacement performed during loop transformation. To use this option, you must also specify O3. -scalar-rep- disables this optimization.

]]> This options tells the compiler to assume no aliasing in the program.

]]> The -fast option enhances execution speed across the entire program by including the following options that can improve run-time performance:

-O3 (maximum speed and high-level optimizations)

-ipo (enables interprocedural optimizations across files)

-xT (generate code specialized for Intel(R) Core(TM)2 Duo processors, Intel(R) Core(TM)2 Quad processors and Intel(R) Xeon(R) processors with SSSE3)

-static Statically link in libraries at link time

-no-prec-div (disable -prec-div) where -prec-div improves precision of FP divides (some speed impact)

To override one of the options set by /fast, specify that option after the -fast option on the command line. The exception is the xT or QxT option which can't be overridden. The options set by /fast may change from release to release.

]]> Enable fast math mode. This option may yield faster code for programs that do not require the guarantees of exact implementation of IEEE or ISO rules/specifications for math functions. Tells the compiler to use more aggressive optimizations when implementing floating-point calculations. These optimizations increase speed, but they may affect the accuracy or reproducibility of floating-point computations.

Specifying fast is the same as specifying fast=1. Setting fast=1 (the default) recognizes and supports NaN and infinite values.

For Intel compiler options ffp-model=fast and fp-model=fast are interchangeable.

]]> Compiler option to statically link in libraries at link time Code is optimized for Intel(R) Core(TM)2 Duo processors, Intel(R) Core(TM)2 Quad processors and Intel(R) Xeon(R) processors with SSSE3. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

Do not use this option if you are executing a program on a processor that is not an Intel processor. If you use this option on a non-compatible processor to compile the main program (in Fortran) or the function main() in C/C++, the program will display a fatal run-time error if they are executed on unsupported processors.

]]> Code is optimized for Intel(R) Atom processors with support for MOVBE instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel(R) processors with support for AVX2 instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> May generate instructions for processors that support the specified Intel processor or microarchitecture code name. Optimizes for the specified Intel processor or microarchitecture code name.

]]> On x86 systems, allows use of instructions that require the listed architecture.

]]> Tells the compiler to remove the assumption that source code follows c99 signed overflow rules.

]]> -Wno-implicit-int is needed to allow the compiler to accept invalid C code where the type specifier is missing. With this diagnostic disabled, the missing type will be interpreted as `int`, as in C89 (the last version of C in which implicit type specifiers were allowed).

]]> This option tells the compiler to generate instructions for the highest instruction set available on the compilation host processor. The instructions generated by Host differ depending on the compilation host processor.

]]> Code is optimized for Intel(R) Atom(TM) processors that support SSE4.2 and MOVBE instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel(R) Atom(TM) processors that support Intel(R) SSE3 and MOVBE instructions. and MOVBE instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel(R) processors with support for CORE-AVX512 instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel(R) processors with support for COMMON-AVX512 instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel(R) processors with support for MIC-AVX512 instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel(R) processors with support for AVX instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel(R) processors with support for SSE 4.2 instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel(R) processors with support for SSE 4.1 instructions. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> This option tells the compiler to generate code specialized for processors that support for SSE3 instructions. Code generated with these options should execute on any compatible, non-Intel processor with support for the SSE3 instruction set.

]]> Code is optimized for Intel Pentium M and compatible Intel processors. The resulting code may contain unconditional use of features that are not supported on other processors. This option also enables new optimizations in addition to Intel processor-specific optimizations including advanced data layout and code restructuring optimizations to improve memory accesses for Intel processors.

]]> Code is optimized for Intel Pentium 4 and compatible Intel processors; this is the default for Intel?EM64T systems. The resulting code may contain unconditional use of features that are not supported on other processors.

]]> Tells the auto-parallelizer to generate multithreaded code for loops that can be safely executed in parallel. To use this option, you must also specify option O2 or O3. The default numbers of threads spawned is equal to the number of processors detected in the system where the binary is compiled. Can be changed by setting the environment variable OMP_NUM_THREADS

]]> Runtime library for auto-parallelized code The use of -Qparallel to generate auto-parallelized code requires support libraries that are dynamically linked by default. Specifying libguide40.lib on the link line, statically links in libguide40.lib to allow auto-parallelized binaries to work on systems which do not have the dynamic version of this library installed.

]]> Optimizes for Intel Pentium 4 and compatible processors with Streaming SIMD Extensions 2 (SSE2).

]]> (disable/enable[default] -prec-div) -no-prec-div enables optimizations that give slightly less precise results than full IEEE division.

When you specify -no-prec-div along with some optimizations, such as -xN and -xB (Linux) or /QxN and /QxB (Windows), the compiler may change floating-point division computations into multiplication by the reciprocal of the denominator. For example, A/B is computed as A * (1/B) to improve the speed of the computation.

However, sometimes the value produced by this transformation is not as accurate as full IEEE division. When it is important to have fully precise IEEE division, do not use -no-prec-div. This will enable the default -prec-div and the result will be more accurate, with some loss of performance.

]]> Instrument program for profiling for the first phase of two-phase profile guided otimization. This instrumentation gathers information about a program's execution paths and data values but does not gather information from hardware performance counters. The profile instrumentation also gathers data for optimizations which are unique to profile-feedback optimization.

-profgen:threadsafe option collects profile guided optimization data with guards for threaded applications.

]]> Allows for tuning of application performance by setting the number of threads to use in a parallel region. It has a similar effect as environment variable OMP_NUM_THREADS. This option overrides the environment variable when both are specified. Ex: -par-num-threads=1 specifies one thread to use.

]]> Instructs the compiler to produce a profile-optimized executable and merges available dynamic information (.dyn) files into a pgopti.dpi file. If you perform multiple executions of the instrumented program, -prof-use merges the dynamic information files again and overwrites the previous pgopti.dpi file.
Without any other options, the current directory is searched for .dyn files

]]> Instructs the compiler to produce a profile-optimized executable and merges available dynamic information.

]]> Force linker to ignore multiple definitions Enable SmartHeap and/or other library usage by forcing the linker to ignore multiple definitions if present

]]> Specify build time link path for jemalloc 64bit built to support the CPU 2026 build. See jemalloc.net for more information.

]]> Specify build time link path for jemalloc 32bit built to support the CPU 2026 build. See jemalloc.net for more information.

]]> Linker toggle to specify jemalloc linker library. See jemalloc.net for more information.

]]> Build time link path for libraries supplied with the compiler (for example, the qkmalloc library).

]]> Linker toggle to specify qkmalloc linker library. See https://www.intel.com/content/www/us/en/docs/dpcpp-cpp-compiler/developer-guide-reference/2023-1/intel-s-memory-allocator-library.html for more information.

]]> Directory for 64-bit files Enable the use of the 64-bit compiler by passing the directory names for the library and include files

]]> Stack reserve amount set the stack reserve amount specified to the linker Enable/disable(DEFAULT) use of ANSI aliasing rules in optimizations; user asserts that the program adheres to these rules. Disable use of strict aliasing rules in optimizations. Enable/disable(DEFAULT) the compiler to generate prefetch instructions to prefetch data. Directs the compiler to inline calloc() calls as malloc()/memset() Enables the delayed parsing behavior. The compiler adds setup code in the C/C++/Fortran main function to enable optimal malloc algorithms:

n=0: Default, no changes to the malloc options. No call to mallopt() is made.
n=1: M_MMAP_MAX=2 and M_TRIM_THRESHOLD=0x10000000. Call mallopt with the two settings.
n=2: M_MMAP_MAX=2 and M_TRIM_THRESHOLD=0x40000000. Call mallopt with these two settings.
n=3: M_MMAP_MAX=0 and M_TRIM_THRESHOLD=-1. Call mallopt with these two settings. This will cause use of sbrk() calls instead of mmap() calls to get memory from the system.

The two parameters, M_MMAP_MAX and M_TRIM_THRESHOLD, are described below

Function: int mallopt (int param, int value) When calling mallopt, the param argument specifies the parameter to be set, and value the new value to be set. Possible choices for param, as defined in malloc.h, are:

M_TRIM_THRESHOLD This is the minimum size (in bytes) of the top-most, releasable chunk that will cause sbrk to be called with a negative argument in order to return memory to the system.
M_TOP_PAD This parameter determines the amount of extra memory to obtain from the system when a call to sbrk is required. It also specifies the number of bytes to retain when shrinking the heap by calling sbrk with a negative argument. This provides the necessary hysteresis in heap size such that excessive amounts of system calls can be avoided.
M_MMAP_THRESHOLD All chunks larger than this value are allocated outside the normal heap, using the mmap system call. This way it is guaranteed that the memory for these chunks can be returned to the system on free. Note that requests smaller than this threshold might still be allocated via mmap.
M_MMAP_MAX The maximum number of chunks to allocate with mmap. Setting this to zero disables all use of mmap.

]]> This option defines a level of zmm registers usage. The low setting causes the compiler to generate code with zmm registers very carefully, only when the gain from their usage is proven. The high setting causes the compiler to use much less restrictive heuristics for zmm code generation. Specifies the level of zmm registers usage. Possible values are:

low: Tells the compiler that the compiled program is unlikely to benefit from zmm registers usage. It specifies that the compiler should avoid using zmm registers unless it can prove the gain from their usage.
high: Tells the compiler to generate zmm code without restrictions.

Default:

The default is low when you specify [Q]xCORE-AVX512.
The default is high when you specify [Q]xCOMMON-AVX512.

]]> Enable cache/bandwidth optimization for stores under conditionals (within vector loops) This option tells the compiler to perform a conditional check in a vectorized loop. This checking avoids unnecessary stores and may improve performance by conserving bandwidth. Enable compiler to generate runtime control code for effective automatic parallelization. This option generates code to perform run-time checks for loops that have symbolic loop bounds. If the granularity of a loop is greater than the parallelization threshold, the loop will be executed in parallel. If you do not specify this option, the compiler may not parallelize loops with symbolic loop bounds if the compile-time granularity estimation of a loop can not ensure it is beneficial to parallelize the loop. Select the method that the register allocator uses to partition each routine into regions

routine - one region per routine
block - one region per block
trace - one region per trace
loop - one region per loop
default - compiler selects best option

]]> Specifies preferred vector width for auto-vectorization. Defaults to 'none' which allows target specific decisions. Multi-versioning is used for generating different versions of the loop based on run time dependence testing, alignment and checking for short/long trip counts. If this option is turned on, it will trigger more versioning at the expense of creating more overhead to check for pointer aliasing and scalar replacement. Make all local variables AUTOMATIC. Same as -automatic Enable more aggressive unrolling heuristics Specifies whether streaming stores are generated:

always - enable generation of streaming stores under the assumption that the application is memory bound

auto - compiler decides when streaming stores are used (DEFAULT)

never - disables generation of streaming stores

]]> Disables inline expansion of all intrinsic functions. Disables conformance to the ANSI C and IEEE 754 standards for floating-point arithmetic.

]]> Allows use of EBP as a general-purpose register in optimizations. This option enables most speed optimizations, but disables some that increase code size for a small speed benefit.

]]> This option enables global optimizations. Specifies the level of inline function expansion.

0 - Disables inlining of user-defined functions. Note that statement functions are always inlined.

1 - Enable inlining when an inline keyword or an inline attribute is specified. Also enables inlining according to the C++ language.

2 - Enable inlining of any function at the compiler's discretion.

]]> Specifies the level of inline function expansion.

Ob0 - Disables inlining of user-defined functions. Note that statement functions are always inlined.

Ob1 - Enable inlining when an inline keyword or an inline attribute is specified. Also enables inlining according to the C++ language.

Ob2 - Enable inlining of any function at the compiler's discretion.

]]> This option tells the compiler to separate functions into COMDATs for the linker.

]]> This option enables read only string-pooling optimization. This option enables read/write string-pooling optimization. This option disables stack-checking for routines with 4096 bytes of local variables and compiler temporaries.

]]> Include debug symbols. This option enables or disables the optimization for multiple adjacent gather/scatter type vector memory references.

]]> This option instructs compiler to align branches and fused branches on 32 byte boundaries

]]> This option instructs compiler to define the relative error, measured by the number of correct bits,for math library function results

]]>