performance: change buffer type for new coherency model

Related-To: NEO-11882

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
This commit is contained in:
Szymon Morek
2024-07-02 07:50:14 +00:00
committed by Compute-Runtime-Automation
parent b4903a9a0f
commit c979495265
10 changed files with 143 additions and 52 deletions

View File

@@ -208,7 +208,8 @@ class Buffer : public MemObj {
uint32_t rootDeviceIndex,
bool forceCopyHostPtr);
static AllocationType getGraphicsAllocationTypeAndCompressionPreference(const MemoryProperties &properties,
bool &compressionEnabled, bool localMemoryEnabled);
bool &compressionEnabled, bool localMemoryEnabled,
bool isNewCoherencyModelSupported);
static bool isReadOnlyMemoryPermittedByFlags(const MemoryProperties &properties);
void transferData(void *dst, void *src, size_t copySize, size_t copyOffset);