CPU2017 Flag Description
Quanta Cloud Technology QuantaGrid D44N-1U (2.25 GHz,AMD EPYC 9754)
Test sponsored by Quanta Computer Inc.
Compilers: AMD Optimizing C/C++ Compiler Suite
Base Compiler Invocation
-
- clang
![[user]](https://www.spec.org/auto/cpu2017/images/user.png)
- CC, LD
-
clang is a C compiler which encompasses preprocessing, parsing, optimization, code generation, assembly, and linking.
Depending on which high-level mode setting is passed, Clang will stop before doing a full link.
-
- flang
- FC, LD
-
flang is a Fortran compiler which encompasses parsing, optimization, code generation, assembly, and linking. Depending on
which high-level mode setting is passed, Flang will stop before doing a full link.
-
- flang
- FC, LD
-
flang is a Fortran compiler which encompasses parsing, optimization, code generation, assembly, and linking. Depending on
which high-level mode setting is passed, Flang will stop before doing a full link.
-
- clang
- CC
-
clang is a C compiler which encompasses preprocessing, parsing, optimization, code generation, assembly, and linking.
Depending on which high-level mode setting is passed, Clang will stop before doing a full link.
-
- clang++
- CXX, LD
-
clang++ C++ compiler which encompasses preprocessing, parsing, optimization, code generation, assembly, and linking.
Depending on which high-level mode setting is passed, Clang will stop before doing a full link.
-
- clang
- CC
-
clang is a C compiler which encompasses preprocessing, parsing, optimization, code generation, assembly, and linking.
Depending on which high-level mode setting is passed, Clang will stop before doing a full link.
-
- flang
- FC
-
flang is a Fortran compiler which encompasses parsing, optimization, code generation, assembly, and linking. Depending on
which high-level mode setting is passed, Flang will stop before doing a full link.
Base Portability Flags
-
![[suite]](https://www.spec.org/auto/cpu2017/images/suite.png)
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
-
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
-
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
-
- EXTRA_PORTABILITY
-
This option is used to indicate that the host system's integers are 32-bits
wide, and longs and pointers are 64-bits wide. Not all benchmarks
recognize this macro, but the preferred practice for data model selection
applies the flags to all benchmarks; this flag description is a placeholder
for those benchmarks that do not recognize this macro.
Base Optimization Flags
-
- -m64
- CC, LD
-
Generates code for a 64-bit environment. The 64-bit environment sets int to 32 bits and long and
pointer to 64 bits and generates code for AMD's x86-64 architecture. The compiler generates AMD64, INTEL64,
x86-64 64-bit ABI. The default on a 32-bit host is 32-bit ABI. The default on a 64-bit host is 64-bit ABI if the target
platform specified is 64-bit, otherwise the default is 32-bit.
-
-
-
- -O3
- COPTIMIZE
-
Like -O2, except that it enables optimizations that take longer to perform or that may generate larger code (in
an attempt to make the program run faster).
If multiple "O" options are used, with or without level numbers, the last such option is the one that is effective.
- Includes:
-
- -march=znver4
- COPTIMIZE
-
Specify that Clang should generate code for a specific processor family member and later. For example, if you specify
-march=znver1, the compiler is allowed to generate instructions that are valid on AMD Zen processors, but
which may not exist on earlier products. -march=znver4 enables AVX 512 ISA for Genoa (znver4) processors.
-
-
- -ffast-math
- COPTIMIZE
-
Enables a range of optimizations that provide faster, though sometimes less precise, mathematical operations that may
not conform to the IEEE-754 specifications. When this option is specified, the __STDC_IEC_559__ macro is
ignored even if set by the system headers.
-
-
-
- -fstruct-layout=7
- COPTIMIZE
-
Analyzes the whole program to determine if the structures in the code can be peeled, if dead or redundant fields can be deleted, and if
the pointer or integer fields in the structure can be compressed. If feasible, this optimization transforms the code to
enable these improvements. This transformation is likely to improve cache utilization and memory bandwidth. It is expected
to improve the scalability of programs executed on multiple cores.
This is effective only under flto as the whole program analysis is required to perform this optimization. You can choose
different levels of aggressiveness with which this optimization can be applied to your application; with 1 being the least
aggressive and 7 being the most aggressive level.
Possible values:
- fstruct-layout=0: disables structure peeling (default).
- fstruct-layout=1: enables structure peeling.
- fstruct-layout=2: enables structure peeling and selectively compresses self-referential pointers in these
structures to 32-bit pointers wherever safe.
- fstruct-layout=3: enables structure peeling and selectively compresses self-referential pointers in these
structures to 16-bit pointers wherever safe.
- fstruct-layout=4: enables structure peeling, pointer compression as in level 2 and further enables
compression of structure fields which are of 64-bit integer type to 32-bit integer type. This is performed under a
strict safety check.
- fstruct-layout=5: enables structure peeling, pointer compression as in level 3 and further enables compression
of structure fields which are of 64-bit integer type to 32-bit integer type. This is performed under a strict safety
check.
- fstruct-layout=6: enables structure peeling, pointer compression as in level 2 and further enables compression
of structure fields which are of type 64-bit integer type to 16-bit integer type. This is performed under a strict
safety check.
- fstruct-layout=7: enables structure peeling, pointer compression as in level 3 and further enables compression
of structure fields which are of type 64-bit integer type to 16-bit integer type. This is performed under a strict
safety check.
- fstruct-layout=8: enables structure peeling, pointer compression, 64 bit integer type compression
as in level 6 and creates optimal ordering of peeled structure fields which could improve runtime performance.
- fstruct-layout=9: enables structure peeling, pointer compression, 64 bit integer type compression
as in level 7 and creates optimal ordering of peeled structure fields which could improve runtime performance.
Note:
fstruct-layout=4 and fstruct-layout=5 are derived from fstruct-layout=2 and fstruct-layout=3 respectively with the added
feature of safe compression of 64-bit integer fields to 32-bit integer fields in structures. Going from
fstruct-layout=4 to fstruct-layout=5 may result in higher performance if the pointer values are such that the pointers
can be compressed to 16-bits.
fstruct-layout=6 and fstruct-layout=7 are derived from fstruct-layout=2 and fstructlayout=3 respectively, with the
added feature of safe compression of 64 bit integer fields to 16 bit integer in structures. Going from fstruct-layout=6
to fstruct-layout=7 may result in higher performance if the pointer values are such that the pointers can be
compressed to 16-bits.
-
-
-
-
-
-
-
- -zopt
- COPTIMIZE
-
This option enables a subset of scalar, vector and loop transformations including improved variants of loop invariant code motion, SLP and loop vectorizations, loop-fusion, loop-interchange, loop-unswitch, loop tiling and loop distribution.
-
-
-
-
-
-
- -m64
- FC, LD
-
Generates code for a 64-bit environment. The 64-bit environment sets int to 32 bits and long and
pointer to 64 bits and generates code for AMD's x86-64 architecture. The compiler generates AMD64, INTEL64,
x86-64 64-bit ABI. The default on a 32-bit host is 32-bit ABI. The default on a 64-bit host is 64-bit ABI if the target
platform specified is 64-bit, otherwise the default is 32-bit.
-
-
-
-
-
- -O3
- FOPTIMIZE
-
Like -O2, except that it enables optimizations that take longer to perform or that may generate larger code (in
an attempt to make the program run faster).
If multiple "O" options are used, with or without level numbers, the last such option is the one that is effective.
- Includes:
-
- -march=znver4
- FOPTIMIZE
-
Specify that Clang should generate code for a specific processor family member and later. For example, if you specify
-march=znver1, the compiler is allowed to generate instructions that are valid on AMD Zen processors, but
which may not exist on earlier products. -march=znver4 enables AVX 512 ISA for Genoa (znver4) processors.
-
-
- -ffast-math
- FOPTIMIZE
-
Enables a range of optimizations that provide faster, though sometimes less precise, mathematical operations that may
not conform to the IEEE-754 specifications. When this option is specified, the __STDC_IEC_559__ macro is
ignored even if set by the system headers.
-
-
-
- -Mrecursive
- FOPTIMIZE
-
Allocate local variables on the stack, thus allowing recursion.
SAVEd, data-initialized, or namelist members are always allocated
statically, regardless of the setting of this switch.
-
-
-
-
- -zopt
- FOPTIMIZE
-
This option enables a subset of scalar, vector and loop transformations including improved variants of loop invariant code motion, SLP and loop vectorizations, loop-fusion, loop-interchange, loop-unswitch, loop tiling and loop distribution.
-
-
-
-
-
-
- -m64
- CC, FC, LD
-
Generates code for a 64-bit environment. The 64-bit environment sets int to 32 bits and long and
pointer to 64 bits and generates code for AMD's x86-64 architecture. The compiler generates AMD64, INTEL64,
x86-64 64-bit ABI. The default on a 32-bit host is 32-bit ABI. The default on a 64-bit host is 64-bit ABI if the target
platform specified is 64-bit, otherwise the default is 32-bit.
-
-
-
-
- -O3
- COPTIMIZE, FOPTIMIZE
-
Like -O2, except that it enables optimizations that take longer to perform or that may generate larger code (in
an attempt to make the program run faster).
If multiple "O" options are used, with or without level numbers, the last such option is the one that is effective.
- Includes:
-
- -march=znver4
- COPTIMIZE, FOPTIMIZE
-
Specify that Clang should generate code for a specific processor family member and later. For example, if you specify
-march=znver1, the compiler is allowed to generate instructions that are valid on AMD Zen processors, but
which may not exist on earlier products. -march=znver4 enables AVX 512 ISA for Genoa (znver4) processors.
-
-
- -ffast-math
- COPTIMIZE, FOPTIMIZE
-
Enables a range of optimizations that provide faster, though sometimes less precise, mathematical operations that may
not conform to the IEEE-754 specifications. When this option is specified, the __STDC_IEC_559__ macro is
ignored even if set by the system headers.
-
-
-
- -fstruct-layout=7
- COPTIMIZE
-
Analyzes the whole program to determine if the structures in the code can be peeled, if dead or redundant fields can be deleted, and if
the pointer or integer fields in the structure can be compressed. If feasible, this optimization transforms the code to
enable these improvements. This transformation is likely to improve cache utilization and memory bandwidth. It is expected
to improve the scalability of programs executed on multiple cores.
This is effective only under flto as the whole program analysis is required to perform this optimization. You can choose
different levels of aggressiveness with which this optimization can be applied to your application; with 1 being the least
aggressive and 7 being the most aggressive level.
Possible values:
- fstruct-layout=0: disables structure peeling (default).
- fstruct-layout=1: enables structure peeling.
- fstruct-layout=2: enables structure peeling and selectively compresses self-referential pointers in these
structures to 32-bit pointers wherever safe.
- fstruct-layout=3: enables structure peeling and selectively compresses self-referential pointers in these
structures to 16-bit pointers wherever safe.
- fstruct-layout=4: enables structure peeling, pointer compression as in level 2 and further enables
compression of structure fields which are of 64-bit integer type to 32-bit integer type. This is performed under a
strict safety check.
- fstruct-layout=5: enables structure peeling, pointer compression as in level 3 and further enables compression
of structure fields which are of 64-bit integer type to 32-bit integer type. This is performed under a strict safety
check.
- fstruct-layout=6: enables structure peeling, pointer compression as in level 2 and further enables compression
of structure fields which are of type 64-bit integer type to 16-bit integer type. This is performed under a strict
safety check.
- fstruct-layout=7: enables structure peeling, pointer compression as in level 3 and further enables compression
of structure fields which are of type 64-bit integer type to 16-bit integer type. This is performed under a strict
safety check.
- fstruct-layout=8: enables structure peeling, pointer compression, 64 bit integer type compression
as in level 6 and creates optimal ordering of peeled structure fields which could improve runtime performance.
- fstruct-layout=9: enables structure peeling, pointer compression, 64 bit integer type compression
as in level 7 and creates optimal ordering of peeled structure fields which could improve runtime performance.
Note:
fstruct-layout=4 and fstruct-layout=5 are derived from fstruct-layout=2 and fstruct-layout=3 respectively with the added
feature of safe compression of 64-bit integer fields to 32-bit integer fields in structures. Going from
fstruct-layout=4 to fstruct-layout=5 may result in higher performance if the pointer values are such that the pointers
can be compressed to 16-bits.
fstruct-layout=6 and fstruct-layout=7 are derived from fstruct-layout=2 and fstructlayout=3 respectively, with the
added feature of safe compression of 64 bit integer fields to 16 bit integer in structures. Going from fstruct-layout=6
to fstruct-layout=7 may result in higher performance if the pointer values are such that the pointers can be
compressed to 16-bits.
-
-
-
-
-
-
-
- -zopt
- COPTIMIZE, FOPTIMIZE
-
This option enables a subset of scalar, vector and loop transformations including improved variants of loop invariant code motion, SLP and loop vectorizations, loop-fusion, loop-interchange, loop-unswitch, loop tiling and loop distribution.
-
- -Mrecursive
- FOPTIMIZE
-
Allocate local variables on the stack, thus allowing recursion.
SAVEd, data-initialized, or namelist members are always allocated
statically, regardless of the setting of this switch.
-
-
-
-
-
-
-
-
- -m64
- CC, CXX, FC, LD
-
Generates code for a 64-bit environment. The 64-bit environment sets int to 32 bits and long and
pointer to 64 bits and generates code for AMD's x86-64 architecture. The compiler generates AMD64, INTEL64,
x86-64 64-bit ABI. The default on a 32-bit host is 32-bit ABI. The default on a 64-bit host is 64-bit ABI if the target
platform specified is 64-bit, otherwise the default is 32-bit.
-
-
-
-
- -O3
- COPTIMIZE, CXXOPTIMIZE, FOPTIMIZE
-
Like -O2, except that it enables optimizations that take longer to perform or that may generate larger code (in
an attempt to make the program run faster).
If multiple "O" options are used, with or without level numbers, the last such option is the one that is effective.
- Includes:
-
- -march=znver4
- COPTIMIZE, CXXOPTIMIZE, FOPTIMIZE
-
Specify that Clang should generate code for a specific processor family member and later. For example, if you specify
-march=znver1, the compiler is allowed to generate instructions that are valid on AMD Zen processors, but
which may not exist on earlier products. -march=znver4 enables AVX 512 ISA for Genoa (znver4) processors.
-
-
- -ffast-math
- COPTIMIZE, CXXOPTIMIZE, FOPTIMIZE
-
Enables a range of optimizations that provide faster, though sometimes less precise, mathematical operations that may
not conform to the IEEE-754 specifications. When this option is specified, the __STDC_IEC_559__ macro is
ignored even if set by the system headers.
-
-
-
- -fstruct-layout=7
- COPTIMIZE
-
Analyzes the whole program to determine if the structures in the code can be peeled, if dead or redundant fields can be deleted, and if
the pointer or integer fields in the structure can be compressed. If feasible, this optimization transforms the code to
enable these improvements. This transformation is likely to improve cache utilization and memory bandwidth. It is expected
to improve the scalability of programs executed on multiple cores.
This is effective only under flto as the whole program analysis is required to perform this optimization. You can choose
different levels of aggressiveness with which this optimization can be applied to your application; with 1 being the least
aggressive and 7 being the most aggressive level.
Possible values:
- fstruct-layout=0: disables structure peeling (default).
- fstruct-layout=1: enables structure peeling.
- fstruct-layout=2: enables structure peeling and selectively compresses self-referential pointers in these
structures to 32-bit pointers wherever safe.
- fstruct-layout=3: enables structure peeling and selectively compresses self-referential pointers in these
structures to 16-bit pointers wherever safe.
- fstruct-layout=4: enables structure peeling, pointer compression as in level 2 and further enables
compression of structure fields which are of 64-bit integer type to 32-bit integer type. This is performed under a
strict safety check.
- fstruct-layout=5: enables structure peeling, pointer compression as in level 3 and further enables compression
of structure fields which are of 64-bit integer type to 32-bit integer type. This is performed under a strict safety
check.
- fstruct-layout=6: enables structure peeling, pointer compression as in level 2 and further enables compression
of structure fields which are of type 64-bit integer type to 16-bit integer type. This is performed under a strict
safety check.
- fstruct-layout=7: enables structure peeling, pointer compression as in level 3 and further enables compression
of structure fields which are of type 64-bit integer type to 16-bit integer type. This is performed under a strict
safety check.
- fstruct-layout=8: enables structure peeling, pointer compression, 64 bit integer type compression
as in level 6 and creates optimal ordering of peeled structure fields which could improve runtime performance.
- fstruct-layout=9: enables structure peeling, pointer compression, 64 bit integer type compression
as in level 7 and creates optimal ordering of peeled structure fields which could improve runtime performance.
Note:
fstruct-layout=4 and fstruct-layout=5 are derived from fstruct-layout=2 and fstruct-layout=3 respectively with the added
feature of safe compression of 64-bit integer fields to 32-bit integer fields in structures. Going from
fstruct-layout=4 to fstruct-layout=5 may result in higher performance if the pointer values are such that the pointers
can be compressed to 16-bits.
fstruct-layout=6 and fstruct-layout=7 are derived from fstruct-layout=2 and fstructlayout=3 respectively, with the
added feature of safe compression of 64 bit integer fields to 16 bit integer in structures. Going from fstruct-layout=6
to fstruct-layout=7 may result in higher performance if the pointer values are such that the pointers can be
compressed to 16-bits.
-
-
-
-
-
-
-
- -zopt
- COPTIMIZE, CXXOPTIMIZE, FOPTIMIZE
-
This option enables a subset of scalar, vector and loop transformations including improved variants of loop invariant code motion, SLP and loop vectorizations, loop-fusion, loop-interchange, loop-unswitch, loop tiling and loop distribution.
-
-
-
-
- -Mrecursive
- FOPTIMIZE
-
Allocate local variables on the stack, thus allowing recursion.
SAVEd, data-initialized, or namelist members are always allocated
statically, regardless of the setting of this switch.
-
-
-
-
-
-
-
Implicitly Included Flags
This section contains descriptions of flags that were included implicitly
by other flags, but which do not have a permanent home at SPEC.
-
- -O2
-
Moderate level of optimization which enables most optimizations. This is the default when no "-O" option is
specified, or if no value is specified (i.e. "-O").
If multiple "O" options are used, with or without level numbers, the last such option is the one that is effective.
- Includes:
-
- -O1
-
Somewhere between -O0 and -O2.
If multiple "O" options are used, with or without level numbers, the last such option is the one that is effective.
Using numactl to bind processes and memory to cores
For multi-copy runs or single copy runs on systems with multiple sockets, it is advantageous to bind a process to a
particular core. Otherwise, the OS may arbitrarily move your process from one core to another. This can affect
performance. To help, SPEC allows the use of a "submit" command where users can specify a utility to use to bind
processes. We have found the utility 'numactl' to be the best choice.
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. The numactl flag "--physcpubind" specifies
which core(s) to bind the process. "-l" instructs numactl to keep a process's memory on the
local node while "-m" specifies which node(s) to place a process's memory. For full details on using
numactl, please refer to your Linux documentation, 'man numactl'
Note that some older versions of numactl incorrectly interpret application arguments as its own. For
example, with the command "numactl --physcpubind=0 -l a.out -m a", numactl will interpret
a.out's "-m" option as its own "-m" option. To work around this problem, we put
the command to be run in a shell script and then run the shell script using numactl. For example:
"echo 'a.out -m a' > run.sh ; numactl --physcpubind=0 bash run.sh"
numactl --interleave=all runcpu
numactl --interleave=all runcpu executes the SPEC CPU command runcpu so that memory is consumed across NUMA nodes rather than consumed from a single node. This helps prevent local node out-of-memory conditions which can occur when runcpu is executed without interleaving.
For full details on using numactl, please refer to your Linux documentation, 'man numactl'
Transparent Huge Pages (THP)
THP is an abstraction layer that automates most aspects of creating, managing,
and using huge pages. It is designed to hide much of the complexity in using
huge pages from system administrators and developers. Huge pages
increase the memory page size from 4 kilobytes to 2 megabytes. This provides
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 huge pages being allocated, the kernel will assign
smaller 4k pages instead. Most recent Linux OS releases have THP enabled by default.
THP usage is controlled by the sysfs setting /sys/kernel/mm/transparent_hugepage/enabled.
Possible values:
- never: entirely disable THP usage.
- madvise: enable THP usage only inside regions marked MADV_HUGEPAGE using madvise(3).
- always: enable THP usage system-wide. This is the default.
The SPEC CPU benchmark codes themselves never explicitly request huge pages, as the mechanism to do that is OS-specific
and can change over time. Libraries such as amdalloc which are used by the benchmarks may explicitly request huge pages,
and use of such libraries can make the "madvise" setting relevant and useful.
When no huge pages are immediately available and one is requested, how the system handles the request for THP creation is
controlled by the sysfs setting /sys/kernel/mm/transparent_hugepage/defrag.
Possible values:
- never: if no THP are available to satisfy a request, do not attempt to make any.
- defer: an allocation requesting THP when none are available gets normal pages while requesting THP creation in the
background.
- defer+madvise: acts like "always", but only for allocations in regions marked MADV_HUGEPAGE using madvise(3); for all
other regions it's like "defer".
- madvise: acts like "always", but only for allocations in regions marked MADV_HUGEPAGE using madvise(3). This is the
default.
- always: an allocation requesting THP when none are available will stall until some are made.
An application that "always" requests THP often can benefit from waiting for an allocation until those huge pages can be assembled.
For more information see the Linux transparent hugepage documentation.
ulimit -s <n>
Sets the stack size to n kbytes, or unlimited to allow the stack size to grow without limit.
ulimit -l <n>
Sets the maximum size of memory that may be locked into physical memory.
powersave -f (on SuSE)
Makes the powersave daemon set the CPUs to the highest supported frequency.
/etc/init.d/cpuspeed stop (on Red Hat)
Disables the cpu frequency scaling program in order to set the CPUs to the highest supported frequency.
LD_LIBRARY_PATH
An environment variable that indicates the location in the filesystem of bundled libraries to use when running the
benchmark binaries.
LIBOMP_NUM_HIDDEN_HELPER_THREADS
target nowait is supported via hidden helper task, which is a task not bound to any parallel region. A hidden helper team with a number of threads is created when the first hidden helper task is encountered.
The number of threads can be configured via the environment variable LIBOMP_NUM_HIDDEN_HELPER_THREADS. The default is 8. If LIBOMP_NUM_HIDDEN_HELPER_THREADS is 0, the hidden helper task is disabled and support falls back to a regular OpenMP task. The hidden helper task can also be disabled by setting the environment variable LIBOMP_USE_HIDDEN_HELPER_TASK=OFF.
sysctl -w vm.dirty_ratio=8
Limits dirty cache to 8% of memory.
sysctl -w vm.swappiness=1
Limits swap usage to minimum necessary.
sysctl -w vm.zone_reclaim_mode=1
Frees local node memory first to avoid remote memory usage.
kernel/numa_balancing
This OS setting controls automatic NUMA balancing on memory mapping and process placement.
NUMA balancing incurs overhead for no benefit on workloads that are already bound to NUMA nodes.
Possible settings:
- 0: disables this feature
- 1: enables the feature (this is the default)
For more information see the numa_balancing entry in the
Linux sysctl documentation.
kernel/randomize_va_space (ASLR)
This setting can be used to select the type of process address space
randomization. Defaults differ based on whether the architecture supports
ASLR, whether the kernel was built with the CONFIG_COMPAT_BRK
option or not, or the kernel boot options used.
Possible settings:
- 0 - Turn the process address space randomization off. This is the default for architectures that do not support
this feature anyway, and kernels that are booted with the "
norandmaps" parameter.
- 1 - Randomize addresses of mmap base, stack, and VDSO pages.
This is the default if the
CONFIG_COMPAT_BRK option is enabled at kernel build time.
- 2 - Additionally enable heap randomization. This is the default if
CONFIG_COMPAT_BRK is
disabled.
Disabling ASLR can make process execution more deterministic and runtimes more consistent.
For more information see the randomize_va_space entry in the
Linux sysctl documentation.
vm/drop_caches
The two commands are equivalent:
echo 3> /proc/sys/vm/drop_caches
and
sysctl -w vm.drop_caches=3
Both must be run as root.
The commands are used to free up the filesystem page cache, dentries, and inodes.
Possible settings:
- 1 - Clear pagecache
- 2 - Clear dentries and inodes
- 3 - Clear pagecache, dentries, and inodes
MALLOC_CONF
The amdalloc library is a variant of jemalloc library. The amdalloc
library has tunable parameters, many of which may be changed at run-time via several mechanisms, one of which
is the MALLOC_CONF environment variable. Other methods, as well as the order in which they're referenced,
are detailed in the jemalloc documentation's TUNING section.
The options that can be tuned at run-time are everything in the jemalloc documentation's
MALLCTL NAMESPACE section that begins with
"opt.".
The options that may be encountered in SPEC CPU 2017 results are detailed here:
retain:true - Causes unused virtual memory to
be retained for later reuse rather than discarding it. This is the default for 64-bit Linux.oversize_threshold:0 - The threshold in bytes
of which allocation requests are considered oversize. The default threshold is 8MiB.thp:never - Attempts to never utilize huge pages
by using MADV_NOHUGEPAGE on all mappings. This option has no effect except when THP is set to
"madvise".
PGHPF_ZMEM
An environment variable used to initialize the allocated memory. Setting PGHPF_ZMEM to "Yes" has the effect of
initializing all allocated memory to zero.
GOMP_CPU_AFFINITY
This environment variable is used to set the thread affinity for threads spawned by OpenMP.
OMP_DYNAMIC
This environment variable is defined as part of the OpenMP standard.
Setting it to "false" prevents the OpenMP runtime from dynamically adjusting the number of threads to use for parallel
execution.
For more information, see chapter 4 ("Environment Variables") in the
OpenMP 4.5 Specification.
OMP_SCHEDULE
This environment variable is defined as part of the OpenMP standard.
Setting it to "static" causes loop iterations to be assigned to threads in round-robin fashion in the order of the thread
number.
For more information, see chapter 4 ("Environment Variables") in the
OpenMP 4.5 Specification.
OMP_STACKSIZE
This environment variable is defined as part of the OpenMP standard and controls the size of the stack for threads created
by OpenMP.
For more information, see chapter 4 ("Environment Variables") in the
OpenMP 4.5 Specification.
OMP_THREAD_LIMIT
This environment variable is defined as part of the OpenMP standard and limits the maximum number of OpenMP threads that
can be created.
For more information, see chapter 4 ("Environment Variables") in the
OpenMP 4.5 Specification.
OS Tuning
ulimit:
Used to set user limits of system-wide resources. Provides control over resources available to the shell and processes started by it. Some common ulimit commands may include:
- ulimit -s [n | unlimited]: Set the stack size to n kbytes, or unlimited to allow the stack size to grow without limit.
- ulimit -l (number): Set the maximum size that can be locked into memory.
Performance Governors (Linux):
In-kernel CPU frequency governors are pre-configured power schemes for the CPU. The CPUfreq governors use P-states to change frequencies and lower power consumption. The dynamic governors can switch between CPU frequencies, based on CPU utilization to allow for power savings while not sacrificing performance.
Other options beside a generic performance governor can be set, such as the Performance governor and Powersave governor:
--governor , -g
The governor defines the power characteristics of the system CPU, which in turn affects CPU performance. Each governor has its own unique behavior, purpose, and suitability in terms of workload.
On many Linux systems one can set the governor for all CPUs through the cpupower utility with following commands:
- "cpupower frequency-set -g performance"
Tuning Kernel parameters:
The following Linux Kernel parameters were tuned to better optimize performance of some areas of the system:
- dirty_background_ratio: Set through "echo 40 > /proc/sys/vm/dirty_background_ratio". This setting can help Linux disk caching and performance by setting the percentage of system memory that can be filled with dirty pages.
- dirty_ratio: Set through "echo 40 > /proc/sys/vm/dirty_ratio". This setting is the absolute maximum amount of system memory that can be filled with dirty pages before everything must get committed to disk.
- swappiness: The swappiness value can range from 1 to 100. A value of 100 will cause the kernel to swap out inactive processes frequently in favor of file system performance, resulting in large disk cache sizes. A value of 1 tells the kernel to only swap processes to disk if absolutely necessary. This can be set through a command like "echo 1 > /proc/sys/vm/swappiness"
- ksm/sleep_millisecs: Set through "echo 200 > /sys/kernel/mm/ksm/sleep_millisecs". This setting controls how many milliseconds the ksmd (KSM daeomn) should sleep before the next scan.
- khugepaged/scan_sleep_millisecs: Set through "echo 50000 > /sys/kernel/mm/transparent_hugepage/khugepaged/scan_sleep_millisecs". This setting controls how many milliseconds to wait in khugepaged is there is a hugepage allocation failure to throttle the next allocation attempt.
- numa_balancing: Disabled through "echo 0 > /proc/sys/kernel/numa_balancing". This feature will automatically migrate data on demand so memory nodes are aligned to the local CPU that is accessing data. Depending on the workload involved, enabling this can boost the performance if the workload performs well on NUMA hardware. If the workload is statically set to balance between nodes, then this service may not provide a benefit.
- Zone Reclaim Mode: Zone reclaim allows the reclaiming of pages from a zone if the number of free pages falls below a watermark even if other zones still have enough pages available. Reclaiming a page can be more beneficial than taking the performance penalties that are associated with allocating a page on a remote zone, especially for NUMA machines. To tell the kernel to free local node memory rather than grabbing free memory from remote nodes, use a command like "echo 1 > /proc/sys/vm/zone_reclaim_mode"
- file system page cache: The command "echo 3> /proc/sys/vm/drop_caches" is used to free slab objects and pagecache.
kernel.randomize_va_space (ASLR)
This setting can be used to select the type of process address space randomization. Defaults differ based on whether the architecture supports ASLR, whether the kernel was built with the CONFIG_COMPAT_BRK option or not, or the kernel boot options used.
Possible settings:
- 0: Turn process address space randomization off.
- 1: Randomize addresses of mmap base, stack, and VDSO pages.
- 2: Additionally randomize the heap. (This is probably the default.)
Disabling ASLR can make process execution more deterministic and runtimes more consistent.
For more information see the randomize_va_space entry in the "https://www.kernel.org/doc/Documentation/sysctl/kernel.txt" Linux sysctl documentation
Transparent Hugepages (THP)
THP is an abstraction layer that automates most aspects of creating, managing, and using huge pages. It is designed to hide much of the complexity in using huge pages from system administrators and developers. Huge pages increase the memory page size from 4 kilobytes to 2 megabytes. This provides 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. Most recent Linux OS releases have THP enabled by default.
THP usage is controlled by the sysfs setting /sys/kernel/mm/transparent_hugepage/enabled .
Possible values:
- never: entirely disable THP usage.
- madvise: enable THP usage only inside regions marked MADV_HUGEPAGE using madvise(3).
- always: enable THP usage system-wide. This is the default.
THP creation is controlled by the sysfs setting /sys/kernel/mm/transparent_hugepage/defrag.
Possible values:
- never: if no THP are available to satisfy a request, do not attempt to make any.
- defer: an allocation requesting THP when none are available get normal pages while requesting THP creation in the background.
- defer+madvise: acts like "always", but only for allocations in regions marked MADV_HUGEPAGE using madvise(3); for all other regions it's like "defer".
- madvise: acts like "always", but only for allocations in regions marked MADV_HUGEPAGE using madvise(3). This is the default.
- always: an allocation requesting THP when none are available will stall until some are made.
An application that "always" requests THP often can benefit from waiting for an allocation until those huge pages can be assembled.
For more information see the "https://www.kernel.org/doc/Documentation/vm/transhuge.txt" Linux transparent hugepage documentation.
- Determinism Control:
- This BIOS option allows user to choose AGESA determinism control. Available settings are:
-
- Manual: User can set customized determinism.
- Auto (Default setting): Use the fused determinism.
- Determinism Slider:
- Selects the determinism mode for the CPU:
-
- Auto (Default setting): Use default performance determinism settings.
- Power: Maximizes performance within the power limits defined by cTDP and PPT.
- Performance: Provides predictable performance across all processors of the same type.
- cTDP Control(Configurable TDP):
- TDP is an acronym for “Thermal Design Power.” TDP is the recommended target for power used when designing the cooling capacity for a server. EPYC processors are able to control this target power consumption within certain limits. This capability is referred to as “configurable TDP” or "cTDP." cTDP can be used to reduce power consumption for greater efficiency, or in some cases, increase power consumption above the default value to provide additional performance. cTDP is controlled using a BIOS option.
- The default EPYC cTDP value corresponds with the microprocessor’s nominal TDP. For the EPYC 7601, the default value is 180W. The default cTDP value is set at a good balance between performance and energy efficiency. The EPYC 7601 cTDP can be reduced as low as 165W, which will minimize the power consumption for the processor under load, but at the expense of peak performance. Increasing the EPYC 7601 cTDP to 200W will maximize peak performance by allowing the CPU to maintain higher dynamic clock speeds, but will make the microprocessor less energy efficient. Note that at maximum cTDP, the CPU thermal solution must be capable of dissipating at least 200W or the EPYC 7601 processor might engage in thermal throttling under load.
- The available cTDP ranges for each EPYC model are in the table below:
-
| Model |
Nominal TDP |
Minimum cTDP |
Maximum cTDP** |
| EPYC 9654 |
360 |
320 |
400 |
| EPYC 9654P |
360 |
320 |
400 |
| EPYC 9734 |
340 |
320 |
400 |
| EPYC 9754 |
360 |
320 |
400 |
- Package Power Limit (PPT) Control:
- Specifies the maximum power that each CPU package may consume in the system. The actual power limit is the maximum of the Package Power Limit and cTDP. Available settings are:
-
- Auto (Default setting): Use the fused processor PPT value.
- Manual: Let user specifies customized processor PPT value.
- NUMA nodes per socket (NPS):
- Non-Uniform Memory Architecture (NUMA) enables the CPU cores to access memory via NUMA domains / nodes. Users can specify the number of desired NUMA nodes per populated socket in the system:
-
- NPS0: Zero will attempt to interleave two CPU socket together.
- NPS1: Each physical processor is a NUMA node, and memory accesses are interleaved across all memory channels directly connected to the physical processor.
- NPS2: Each physical processor is two NUMA nodes, and memory accesses are interleaved across 4 memory channels.
- NPS4: Each physical processor is four NUMA nodes, and memory accesses are interleaved across 2 memory channels.
- Auto (Default setting): BIOS will use default NPS setting NPS1.
- ACPI SRAT L3 Cache as NUMA Domain:
- Enable the option to report each L3 cache as a NUMA domain to BIOS ACPI System Resource Affinity Table (SRAT):
-
- Disable: Do not report each L3 cache as a NUMA domain to the OS.
- Enable: Report each L3 cache as a NUMA domain to the OS.
- Auto (Default setting): BIOS will use default setting.
- SMT Control:
- Can be used to disable symmetric multithreading. To re-enable SMT, a POWER CYCLE is needed after selecting the 'Auto' option. WARNING - S3 is NOT SUPPORTED on systems where SMT is disabled.
-
- Disable: Single hardware thread per core.
- Auto (Default setting): Two hardware threads per core.
- L1 Stream HW Prefetcher:
- uses the history of L1 cache memory access patterns to fetch additional sequential lines in ascending or descending order.:
-
- auto: Default BIOS settings for general purpose.
- Enable: Enable L1 Stream HW Prefetcher.
- Disable: Disable L1 Stream HW Prefetcher.
- L2 Stream HW Prefetcher:
- uses the history of L2 cache memory access patterns to fetch additional sequential lines in ascending or descending order.
-
- auto: Default BIOS settings for general purpose.
- Enable: Enable L1 Stream HW Prefetcher.
- Disable: Disable L1 Stream HW Prefetcher.
- Memory interleaving:
- This setting allows interleaved memory accesses across multiple memory channels in each socket, providing higher memory bandwidth.
-
- 0 = Disable.
- 1 = Auto (Default setting).
- IOMMU:
- Enable: Enables the I/O Memory Management Unit (IOMMU), which extends the AMD64 system architecture by adding support for address translation and system memory access protection on DMA transfers from peripheral devices.
-
- 0 = Disable.
- 1 = Enable (Default setting).
- APBDIS:
- APBDIS is an IO Boost disable on uncore. For any system user that needs to block these uncore optimizations that are impacting base core clock speed, we are exposing a method to disable this behavior called APBDis. This locks the fabric clock to the non-boosted speeds. Available settings are:
-
- 0 = not APBDIS (mission mode).
- 1 = Enable APBDIS.
- Fixed SOC Pstate:
- Specifying a fixed SOC P-state. this option is available if APBDIS is enabled. Available settings are:
-
- P0: Highest-performing SOC P-state.
- P1: Next-highest-performing SOC P-state.
- P2: Minimum Infinity Fabric P-State.
- Auto (Default setting): Dynamic.
- ACPI CST C2 Latency:
- Enter in microseconds (decimal value).Larger C2 latency values will reduce the number of C2 transitions and reduce C2 residency. Fewer transitions can help when performance is sensitive to the latency of C2 entry and exit. Higher residency can improve performance by allowing higher frequency boost and reduce idle core power.
-
Last updated Dec. 07, 2023.
For questions about the meanings of these flags, please contact the tester.
For other inquiries, please contact info@spec.org
Copyright 2017-2024 Standard Performance Evaluation Corporation
Tested with SPEC CPU2017 v1.1.9.
Report generated on 2024-01-03 17:30:29 by SPEC CPU2017 flags formatter v5178.
![[benchmark]](https://www.spec.org/auto/cpu2017/images/benchmark.png)