intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-03 14:55:24 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: limit usm device reuse based on used memory

Browse Source 
Calculate available memory for usm device reuse based as (total device
memory - used memory) * fraction for reuse.

Use sys mem allocs for devices without local memory.

Related-To: NEO-12902

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
This commit is contained in:
Dominik Dabek
2024-12-04 07:07:13 +00:00
committed by Compute-Runtime-Automation
parent d4bfa0f758
commit 1252b10ba9
6 changed files with 132 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
shared/source/device/device.h
View File

@@ -181,7 +181,7 @@ class Device : public ReferenceTrackedObject<Device> {
void initializeRayTracing(uint32_t maxBvhLevels);
void allocateRTDispatchGlobals(uint32_t maxBvhLevels);
uint64_t getGlobalMemorySize(uint32_t deviceBitfield) const;
MOCKABLE_VIRTUAL uint64_t getGlobalMemorySize(uint32_t deviceBitfield) const;
const std::vector<SubDevice *> &getSubDevices() const { return subdevices; }
bool getUuid(std::array<uint8_t, ProductHelper::uuidSize> &uuid);
void generateUuid(std::array<uint8_t, ProductHelper::uuidSize> &uuid);