[L0] Only Override max allocation limits given env #1245
Merged
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
Contributor
nrspruit
commented
Jan 11, 2024
nrspruit
commented
- Change the defaults from always allowing > 4GB allocations to making the user have to request > 4GB allocation support given the max allocation allowed on that system is less than 4GB.
- This ensures performance is maintained on systems that dont handle > 4GB allocations natively and avoids breaking Ahead of Time (AOT) binaries that were built without > 4GB resource support.
- By setting UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=1 , the L0 Adapter will force the modules to be built with stateless or > 4GB support and will allow for the allocations to exceed the max single allocation size limit for that device.
nrspruit
force-pushed
the
enable_relaxed_alloc
branch
from
b5ea595
to
b8f6137
Compare
Codecov ReportAll modified and coverable lines are covered by tests ✅
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## main #1245 +/- ##
==========================================
- Coverage 15.80% 15.80% -0.01%
==========================================
Files 223 223
Lines 31481 31481
Branches 3558 3558
==========================================
- Hits 4977 4975 -2
- Misses 26454 26455 +1
- Partials 50 51 +1 ☔ View full report in Codecov by Sentry. |
nrspruit
force-pushed
the
enable_relaxed_alloc
branch
from
b8f6137
to
ff29eed
Compare
pbalcer
reviewed
Jan 12, 2024
nrspruit
force-pushed
the
enable_relaxed_alloc
branch
from
ff29eed
to
0af43e5
Compare
pbalcer
approved these changes
Jan 12, 2024
pbalcer
added
level-zero
nrspruit
force-pushed
the
enable_relaxed_alloc
branch
from
0af43e5
to
aaf71cd
Compare
nrspruit
commented
Jan 12, 2024
- Change the defaults from always allowing > 4GB allocations to making the user have to request > 4GB allocation support given the max allocation allowed on that system is less than 4GB. - This ensures performance is maintained on systems that dont handle > 4GB allocations natively and avoids breaking Ahead of Time (AOT) binaries that were built without > 4GB resource support. - By setting UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS=1 , the L0 Adapter will force the modules to be built with stateless or > 4GB support and will allow for the allocations to exceed the max single allocation size limit for that device. Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
nrspruit
force-pushed
the
enable_relaxed_alloc
branch
from
aaf71cd
to
0ebaca6
Compare
|
Pre-merge checks have passed here: intel/llvm#12375 This change should be ready to be merged to main, then cherry-picked to 0.8.x release. @kbenzie , please merge this when able. |
8 tasks
kbenzie
added a commit
to kbenzie/unified-runtime
that referenced
this pull request
Jan 17, 2024
[L0] Only Override max allocation limits given env
kbenzie
added a commit
to kbenzie/unified-runtime
that referenced
this pull request
Jan 22, 2024
[L0] Only Override max allocation limits given env
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.