[GPU] Updated GPU cache size retrieval and refined closest_pow_of_2 #28059
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Details:
Existing method for cache size calculation was static and need continious
updates to the sku table which was already being missed for latest skus e.g DG2.
This update introduces a new member variable, max_global_cache_size,
to store the GPU's global cache size, obtained via the OpenCL property
CL_DEVICE_GLOBAL_MEM_CACHE_SIZE. The existing hard coded cache
calculations are removed. Additionally, the closest_pow_of_2 function
has been enhanced to return the nearest power of 2, favoring the upper
value if the input is within 30% of the range for the upper bound.
These changes improve memory management and ensure better utilization
of GPU resources towards bottle neck situations.
Tickets:
CVS-159076
Signed-off-by: Arshad Mehmood <[email protected]>
arshadlab
force-pushed
the
auto_batch_update
branch
from
564fd28 to
f651a9b
Compare
|
Verified with ADLS+A310E |
|
build_jenkins |
vladimir-paramuzov
approved these changes
Dec 17, 2024
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
11happy
pushed a commit
to 11happy/openvino
that referenced
this pull request
Dec 23, 2024
…penvinotoolkit#28059) Details: Existing method for cache size calculation was static and need continious updates to the sku table which was already being missed for latest skus e.g DG2. This update introduces a new member variable, max_global_cache_size, to store the GPU's global cache size, obtained via the OpenCL property CL_DEVICE_GLOBAL_MEM_CACHE_SIZE. The existing hard coded cache calculations are removed. Additionally, the closest_pow_of_2 function has been enhanced to return the nearest power of 2, favoring the upper value if the input is within 30% of the range for the upper bound. These changes improve memory management and ensure better utilization of GPU resources towards bottle neck situations. Tickets: CVS-159076 Signed-off-by: Arshad Mehmood <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Details:
Existing method for cache size calculation was static and need continious updates to the sku table which was already being missed for latest skus e.g DG2.
This update introduces a new member variable, max_global_cache_size, to store the GPU's global cache size, obtained via the OpenCL property CL_DEVICE_GLOBAL_MEM_CACHE_SIZE. The existing hard coded cache calculations are removed. Additionally, the closest_pow_of_2 function has been enhanced to return the nearest power of 2, favoring the upper value if the input is within 30% of the range for the upper bound. These changes improve memory management and ensure better utilization of GPU resources towards bottle neck situations.
Tickets:
CVS-159076