NVIDIA/open-gpu-kernel-modules
1
0
Fork 0
mirror of https://github.com/NVIDIA/open-gpu-kernel-modules.git synced 2026-02-05 15:49:58 +00:00
Code Issues Packages Projects Releases Wiki Activity

570.86.15

Browse Source
This commit is contained in:
Bernhard Stoeckner
2025-01-27 19:36:56 +01:00
parent 9d0b0414a5
commit 54d69484da
1166 changed files with 318863 additions and 182687 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
kernel-open/nvidia-uvm/uvm_processors.c
View File

@@ -137,3 +137,26 @@ bool uvm_processor_has_memory(uvm_processor_id_t id)
return uvm_gpu_get(id)->mem_info.size > 0;
}
// Extract a portion of the processor mask in order to compute the sub-processor
// mask for a particular parent GPU id.
// We use direct bitmap operations in order to avoid having to allocate a new
// uvm_processor_mask_t, which could fail.
//
// The size of the 'word' bitmap is two words in order to cover IDs that span
// a word size due to the offset caused by the CPU ID.
uvm_sub_processor_mask_t uvm_sub_processor_mask_from_processor_mask(const uvm_processor_mask_t *mask,
uvm_parent_gpu_id_t parent_id)
{
uvm_sub_processor_mask_t sub_processor_mask;
uvm_gpu_id_t gpu_id = uvm_gpu_id_from_parent_gpu_id(parent_id);
size_t word_index = uvm_id_value(gpu_id) / BITS_PER_LONG;
DECLARE_BITMAP(words, BITS_PER_LONG * 2);
BUILD_BUG_ON(BITS_PER_LONG < UVM_PARENT_ID_MAX_SUB_PROCESSORS);
bitmap_copy(words, mask->bitmap + word_index, BITS_PER_LONG * 2);
bitmap_shift_right(words, words, uvm_id_value(gpu_id) % BITS_PER_LONG, BITS_PER_LONG * 2);
sub_processor_mask.bitmap = words[0] & ((1 << UVM_PARENT_ID_MAX_SUB_PROCESSORS) - 1);
return sub_processor_mask;
}