NVIDIA/nvbench
1
0
Fork 0
mirror of https://github.com/NVIDIA/nvbench.git synced 2026-03-14 20:27:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

Swapped data members m_sm_clock_rates and m_sm_clock_rate_accumulator

Browse Source 
This places all std::vector members together. Added default initialization
to all std::vector members, and all other members with default constructors.

Exceptions are references and nvbench::launch m_launch; member
This commit is contained in:
Oleksandr Pavlyk
2026-02-19 15:33:57 -06:00
parent 4da9f431c0
commit 731e0c2c30
1 changed files with 14 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
nvbench/detail/measure_cold.cuh
View File

@@ -85,27 +85,27 @@ protected:
nvbench::state &m_state;
nvbench::launch m_launch;
nvbench::cuda_timer m_cuda_timer;
nvbench::cpu_timer m_cpu_timer;
nvbench::cpu_timer m_walltime_timer;
nvbench::detail::l2flush m_l2flush;
nvbench::blocking_kernel m_blocker;
nvbench::cuda_timer m_cuda_timer{};
nvbench::cpu_timer m_cpu_timer{};
nvbench::cpu_timer m_walltime_timer{};
nvbench::detail::l2flush m_l2flush{};
nvbench::blocking_kernel m_blocker{};
nvbench::criterion_params m_criterion_params;
nvbench::criterion_params m_criterion_params{};
nvbench::stopping_criterion_base &m_stopping_criterion;
nvbench::detail::gpu_frequency m_gpu_frequency;
nvbench::detail::gpu_frequency m_gpu_frequency{};
bool m_disable_blocking_kernel{false};
bool m_run_once{false};
bool m_check_throttling;
bool m_check_throttling{true};
nvbench::int64_t m_min_samples{};
nvbench::float64_t m_skip_time{};
nvbench::float64_t m_timeout{};
nvbench::float32_t m_throttle_threshold; // [% of default SM clock rate]
nvbench::float32_t m_throttle_recovery_delay; // [seconds]
nvbench::float32_t m_throttle_threshold{}; // [% of default SM clock rate]
nvbench::float32_t m_throttle_recovery_delay{}; // [seconds]
// Dynamically increased when repeated throttling occurs
// without successfully recording a sample.
@@ -122,13 +122,13 @@ protected:
nvbench::float64_t m_max_cpu_time{};
nvbench::float64_t m_total_cpu_time{};
std::vector<nvbench::float64_t> m_sm_clock_rates{};
nvbench::float64_t m_sm_clock_rate_accumulator{};
std::vector<nvbench::float64_t> m_sm_clock_rates{};
std::vector<nvbench::float64_t> m_cuda_times;
std::vector<nvbench::float64_t> m_cpu_times;
std::vector<nvbench::float64_t> m_cuda_times{};
std::vector<nvbench::float64_t> m_cpu_times{};
bool m_max_time_exceeded{};
bool m_max_time_exceeded{false};
};
struct measure_cold_base::kernel_launch_timer