frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 12:54:11 +08:00
Code Issues Packages Projects Releases Wiki Activity

Remove unused code in CUDAAllocatorConfig (#165136)

Browse Source 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165136
Approved by: https://github.com/Skylion007
ghstack dependencies: #165129, #165131, #165135
This commit is contained in:
Yu, Guangye
2025-10-14 13:29:35 +00:00
committed by PyTorch MergeBot
parent 608a6d4a26
commit 515b5ff539
2 changed files with 22 additions and 235 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
c10/cuda/CUDAAllocatorConfig.h
View File

@ -79,11 +79,10 @@ class C10_CUDA_API CUDAAllocatorConfig {
return 128;
}
// This is used to round-up allocation size to nearest power of 2 divisions.
// More description below in function roundup_power2_next_division
// As an example, if we want 4 divisions between 2's power, this can be done
// using env variable: PYTORCH_CUDA_ALLOC_CONF=roundup_power2_divisions:4
static size_t roundup_power2_divisions(size_t size);
static size_t roundup_power2_divisions(size_t size) {
return c10::CachingAllocator::AcceleratorAllocatorConfig::
roundup_power2_divisions(size);
}
static std::vector<size_t> roundup_power2_divisions() {
return c10::CachingAllocator::AcceleratorAllocatorConfig::
@ -152,16 +151,6 @@ class C10_CUDA_API CUDAAllocatorConfig {
const std::vector<std::string>& config,
size_t i,
const char c);
size_t parseMaxSplitSize(const std::vector<std::string>& config, size_t i);
size_t parseMaxNonSplitRoundingSize(
const std::vector<std::string>& config,
size_t i);
size_t parseGarbageCollectionThreshold(
const std::vector<std::string>& config,
size_t i);
size_t parseRoundUpPower2Divisions(
const std::vector<std::string>& config,
size_t i);
size_t parseAllocatorConfig(
const std::vector<std::string>& config,
size_t i,
@ -175,28 +164,17 @@ class C10_CUDA_API CUDAAllocatorConfig {
size_t parsePinnedReserveSegmentSize(
const std::vector<std::string>& config,
size_t i);
size_t parsePinnedUseBackgroundThreads(
const std::vector<std::string>& config,
size_t i);
size_t parseGraphCaptureRecordStreamReuse(
const std::vector<std::string>& config,
size_t i);
std::atomic<size_t> m_max_split_size;
std::atomic<size_t> m_max_non_split_rounding_size;
std::vector<size_t> m_roundup_power2_divisions;
std::atomic<double> m_garbage_collection_threshold;
std::atomic<size_t> m_pinned_num_register_threads;
std::atomic<size_t> m_pinned_reserve_segment_size_mb;
std::atomic<bool> m_expandable_segments;
std::atomic<Expandable_Segments_Handle_Type>
m_expandable_segments_handle_type;
std::atomic<bool> m_release_lock_on_cudamalloc;
std::atomic<bool> m_pinned_use_cuda_host_register;
std::atomic<bool> m_graph_capture_record_stream_reuse;
std::atomic<bool> m_pinned_use_background_threads;
std::string m_last_allocator_settings;
std::mutex m_last_allocator_settings_mutex;
};
// Keep this for backwards compatibility