/llvm-project/third-party/benchmark/docs/ |
H A D | reducing_variance.md | 3 <a name="disabling-cpu-frequency-scaling" /> 10 ***WARNING*** CPU scaling is enabled, the benchmark real time measurements may be noisy and will in… 13 you might want to disable the CPU frequency scaling while running the 45 above](user_guide#disabling-cpu-frequency-scaling) is not the only source 52 2. CPU scaling features that run on the CPU, like Intel's Turbo Boost and 72 above](user_guide#disabling-cpu-frequency-scaling). 98 scaling](https://wiki.archlinux.org/title/CPU_frequency_scaling) page.
|
H A D | user_guide.md | 65 [Disabling CPU Frequency Scaling](#disabling-cpu-frequency-scaling) 1277 <a name="disabling-cpu-frequency-scaling" /> 1284 ***WARNING*** CPU scaling is enabled, the benchmark real time measurements may 1288 you might want to disable the CPU frequency scaling while running the
|
/llvm-project/polly/lib/External/isl/test_inputs/codegen/ |
H A D | isolate1.st | 1 # Check that the isolate option is adjusted by schedule space scaling
|
H A D | isolate2.st | 1 # Check that the isolate option is adjusted by schedule space scaling
|
/llvm-project/libc/benchmarks/ |
H A D | RATIONALE.md | 95 scaling](https://en.wikipedia.org/wiki/Dynamic_frequency_scaling) can kick 178 ### Effect of dynamic frequency scaling 181 scaling](https://en.wikipedia.org/wiki/Dynamic_frequency_scaling). In so-called 188 **Decision: When benchmarking we want to make sure the dynamic frequency scaling 190 events are not impacted by frequency scaling.** 228 We reuse some parts of Google Benchmark (detection of frequency scaling, CPU
|
H A D | LibcBenchmark.cpp | 18 if (CpuInfo.scaling == benchmark::CPUInfo::ENABLED) in checkRequirements()
|
/llvm-project/third-party/benchmark/src/ |
H A D | sysinfo.cc | 649 double GetCPUCyclesPerSecond(CPUInfo::Scaling scaling) { in GetCPUCyclesPerSecond() argument 652 (void)scaling; in GetCPUCyclesPerSecond() 667 || (scaling == CPUInfo::Scaling::DISABLED && in GetCPUCyclesPerSecond() 868 scaling(CpuScaling(num_cpus)), in CPUInfo() 869 cycles_per_second(GetCPUCyclesPerSecond(scaling)), in CPUInfo()
|
H A D | reporter.cc | 80 if (CPUInfo::Scaling::ENABLED == info.scaling) { in PrintBasicContext()
|
H A D | json_reporter.cc | 136 if (CPUInfo::Scaling::UNKNOWN != info.scaling) { in ReportContext() 139 info.scaling == CPUInfo::Scaling::ENABLED ? true : false) in ReportContext()
|
/llvm-project/llvm/test/CodeGen/X86/ |
H A D | large-gep-scale.ll | 4 ; After scaling, this type doesn't fit in memory. Codegen should generate
|
H A D | avoid_complex_am.ll | 24 ; The induction variable should carry the scaling factor: 1.
|
H A D | pr32282.ll | 5 ; Check for assert in foldMaskAndShiftToScale due to out of range mask scaling.
|
/llvm-project/llvm/test/CodeGen/Thumb/ |
H A D | addr-modes.ll | 9 ; The Thumb1 target addressing modes don't support scaling.
|
/llvm-project/llvm/test/CodeGen/AArch64/ |
H A D | arm64-scaled_iv.ll | 22 ; The induction variable should carry the scaling factor: 1 * 8 = 8.
|
/llvm-project/mlir/test/Dialect/SPIRV/IR/ |
H A D | matrix-ops.mlir | 64 // expected-error @+1 {{input matrix components' type and scaling value must have the same type}} 72 // expected-error @+1 {{input matrix components' type and scaling value must have the same type}}
|
/llvm-project/libcxx/test/std/containers/views/mdspan/ |
H A D | CustomTestLayouts.h | 245 constexpr mapping(const extents_type& ext, index_type offset = 0, index_type scaling = 1) noexcept in extents_() 246 : extents_(ext), offset_(offset), scaling_(scaling) {}
|
/llvm-project/llvm/docs/ |
H A D | Benchmarking.rst | 26 * Disable frequency scaling, turbo boost and address space
|
/llvm-project/mlir/test/Integration/Dialect/SparseTensor/CPU/ |
H A D | sparse_scale.mlir | 66 // and then calls the sparse scaling kernel with the sparse tensor
|
/llvm-project/llvm/test/Transforms/SampleProfile/ |
H A D | norepeated-icp-4.ll | 11 ; call will be kept during the scaling.
|
/llvm-project/compiler-rt/test/profile/Linux/ |
H A D | instrprof-instr-suppl.test | 35 // scaling factor of a function is computed based on instrPGO profiles and
|
/llvm-project/libc/src/math/generic/ |
H A D | log1p.cpp | 981 double scaling = FPBits_t(s_u).get_val() + EXPONENT_CORRECTION[s_u == 0]; global() variable
|
/llvm-project/llvm/test/MC/Disassembler/Hexagon/ |
H A D | xtype_fp.txt | 104 # Floating point fused multiply-add with scaling
|
/llvm-project/mlir/include/mlir/Dialect/Tosa/IR/ |
H A D | TosaOpBase.td | 93 // The scaling of their accumulator output is done using an explicit
|
/llvm-project/llvm/lib/Target/AArch64/ |
H A D | AArch64SchedPredicates.td | 48 // Check for scaling in the register offset addressing mode.
|
/llvm-project/llvm/test/Transforms/LoopVectorize/AArch64/ |
H A D | predication_costs.ll | 8 ; and block probability scaling. Note that the functionality being tested is
|