NVIDIA/open-gpu-kernel-modules
1
0
Fork 0
mirror of https://github.com/NVIDIA/open-gpu-kernel-modules.git synced 2026-02-03 15:04:39 +00:00
Code Issues Packages Projects Releases Wiki Activity

550.40.07

Browse Source
This commit is contained in:
Bernhard Stoeckner
2024-01-24 17:51:53 +01:00
parent bb2dac1f20
commit 91676d6628
1411 changed files with 261367 additions and 145959 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
kernel-open/nvidia-uvm/uvm_map_external.c
View File

@@ -1,5 +1,5 @@
/*******************************************************************************
Copyright (c) 2016-2022 NVIDIA Corporation
Copyright (c) 2016-2023 NVIDIA Corporation
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to
@@ -633,10 +633,17 @@ static NV_STATUS set_ext_gpu_map_location(uvm_ext_gpu_map_t *ext_gpu_map,
uvm_gpu_t *mapping_gpu,
const UvmGpuMemoryInfo *mem_info)
{
uvm_gpu_t *owning_gpu;
uvm_gpu_t *owning_gpu = NULL;
uvm_gpu_t *gpu;
if (mem_info->egm)
UVM_ASSERT(mem_info->sysmem);
// !mem_info->deviceDescendant && !mem_info->sysmem imply fabric allocation.
// !mem_info->deviceDescendant also means that mem_info->uuid is invalid. In
// this case the owning GPU is NULL, meaning that UVM is oblivious to the
// topology and relies on RM and/or the fabric manager (FM) for memory
// lifetime management and GPU ref counting.
if (!mem_info->deviceDescendant && !mem_info->sysmem) {
ext_gpu_map->owning_gpu = NULL;
ext_gpu_map->is_sysmem = false;
@@ -645,7 +652,17 @@ static NV_STATUS set_ext_gpu_map_location(uvm_ext_gpu_map_t *ext_gpu_map,
// This is a local or peer allocation, so the owning GPU must have been
// registered.
// This also checks for if EGM owning GPU is registered.
owning_gpu = uvm_va_space_get_gpu_by_uuid(va_space, &mem_info->uuid);
// TODO: Bug 4351121: RM will return the GI UUID, but
// uvm_va_space_get_gpu_by_uuid() currently matches on physical GPU UUIDs.
// Match on GI UUID until the UVM user level API has been updated to use
// the GI UUID.
for_each_va_space_gpu(gpu, va_space) {
if (uvm_uuid_eq(&gpu->uuid, &mem_info->uuid)) {
owning_gpu = gpu;
break;
}
}
if (!owning_gpu)
return NV_ERR_INVALID_DEVICE;
@@ -1343,7 +1360,9 @@ static NV_STATUS uvm_free(uvm_va_space_t *va_space, NvU64 base, NvU64 length)
{
uvm_va_range_t *va_range;
NV_STATUS status = NV_OK;
uvm_global_processor_mask_t retained_mask;
// TODO: Bug 4351121: retained_mask should be pre-allocated, not on the
// stack.
uvm_processor_mask_t retained_mask;
LIST_HEAD(deferred_free_list);
if (uvm_api_range_invalid_4k(base, length))
@@ -1379,14 +1398,14 @@ static NV_STATUS uvm_free(uvm_va_space_t *va_space, NvU64 base, NvU64 length)
// External ranges may have deferred free work, so the GPUs may have to
// be retained. Construct the mask of all the GPUs that need to be
// retained.
uvm_va_space_global_gpus_in_mask(va_space, &retained_mask, &va_range->external.mapped_gpus);
uvm_processor_mask_and(&retained_mask, &va_range->external.mapped_gpus, &va_space->registered_gpus);
}
uvm_va_range_destroy(va_range, &deferred_free_list);
// If there is deferred work, retain the required GPUs.
if (!list_empty(&deferred_free_list))
uvm_global_mask_retain(&retained_mask);
uvm_global_gpu_retain(&retained_mask);
out:
uvm_va_space_up_write(va_space);
@@ -1394,7 +1413,7 @@ out:
if (!list_empty(&deferred_free_list)) {
UVM_ASSERT(status == NV_OK);
uvm_deferred_free_object_list(&deferred_free_list);
uvm_global_mask_release(&retained_mask);
uvm_global_gpu_release(&retained_mask);
}
return status;