In TBX mode, the host could not write to host buffers after access from device
code due to the lack of a migration mechanism post-initial TBX upload.
Migration is unnecessary with real hardware, but required for TBX.
This patch introduces a new page fault manager type that extends the original
CPU fault manager, enabling automatic migration of host buffers in TBX mode.
Refactoring was necessary to avoid diamond inheritance, achieved by using a
template parameter as the base class for OS-specific fault managers.
Related-To: NEO-12268
Signed-off-by: Jack Myers <jack.myers@intel.com>
getBaseAddress was incorrectly returning pLeftBound which changes after
memory allocation.
Added baseAddress field to store and return initial address value.
Signed-off-by: Fabian Zwoliński <fabian.zwolinski@intel.com>
Explicitly delete unused functions of CheckIfExitCalled
and VfManagementHandleContext. Comply with the rule of five.
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
Leverage features of the mechanism to simplify implementation:
- The maximum number of possible cache-region reservations is a small
value known at compile-time
- Each reservation is unique (described by `CacheRegion`) so can have
a dedicated entry with either zero (free) or non-zero (reserved) value
So, there is no need for a dynamic collection (unordered_map here) to
keep track of reservations. A simple array is enough for that purpose.
Also, add some helper-code to enable array-indexing with the values of
`CacheRegion` enum.
Related-To: NEO-12837
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
Refactors the STATE_BASE_ADDRESS to align with the latest specification.
Removes redundant functionality for multiple GPU partial writes and
atomics.
Related-To: NEO-13147
Signed-off-by: Vysochyn, Illia <illia.vysochyn@intel.com>
getBaseAddress was incorrectly returning pLeftBound which changes after
memory allocation.
Added baseAddress field to store and return initial address value.
Signed-off-by: Fabian Zwoliński <fabian.zwolinski@intel.com>
This PR handles the situation in which a component
has reserved a front window space for itself in the external heap,
so that the Compute Runtime cannot access this area.
In such a situation, we perform the following steps:
1. reserve 4GB chunk in heapStandard
2. split our chunk into 2 parts: heapFrontWindow, heapRegular
3. from this point on, map all linearStream allocations in reserved 4GB
chunk
Patch applies to Windows and WSL.
Patch only applies when the bindless global allocator is enabled.
Related-To: HSD-16025889919
Signed-off-by: Fabian Zwoliński <fabian.zwolinski@intel.com>
Related-To: GSD-10253, GSD-9467, GSD-9381, NEO-11908
When EnableGlobalTimestampViaSubmission is set then
zeDeviceGetGlobalTimestamp uses immediate cmd submission
method to get GPU time.
Signed-off-by: Chandio, Bibrak Qamar <bibrak.qamar.chandio@intel.com>
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>
