Add DPEX_OPT, INLINE_THRESHOLD config and do not use numba's OPT #1158
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
diptorupd
reviewed
Oct 5, 2023
fcharras
reviewed
Oct 5, 2023
numba_dpex/core/codegen.py
Outdated
| # Make optimization level depending on config.OPT variable | ||
| pmb.opt_level = config.OPT | ||
| # Make optimization level depending on config.DPEX_OPT variable | ||
| pmb.opt_level = config.DPEX_OPT | ||
|
|
||
| pmb.disable_unit_at_a_time = False | ||
| pmb.inlining_threshold = 2 |
There was a problem hiding this comment.
TY for investigating those issues. Also while investigating for #1152 we've also found that pmb.inlining_threshold = 2 causes miscompilations. Could it be disabled by default ?
There was a problem hiding this comment.
FYI, opt3 works fine one devcloud with threshold being removed and fails on opt2 with theshold. However I have an environment that fails with opt3 without threshold.
ZzEeKkAa
force-pushed
the
feature/introduce_dpex_opt_config
branch
2 times, most recently
from
e112e88 to
45aca12
Compare
ZzEeKkAa
changed the title
diptorupd
reviewed
Oct 5, 2023
ZzEeKkAa
force-pushed
the
feature/introduce_dpex_opt_config
branch
from
45aca12 to
cbc2911
Compare
ZzEeKkAa
force-pushed
the
feature/introduce_dpex_opt_config
branch
from
cbc2911 to
e70a0ff
Compare
diptorupd
approved these changes
Oct 6, 2023
github-actions bot
added a commit
that referenced
this pull request
Oct 6, 2023
…config Add DPEX_OPT config and do not use numba's OPT 511241f
This was referenced Oct 6, 2023
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Outdated suggestions cannot be applied.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.
It appears that level 3 optimization breaks kernel execution, so we need to set default value to 2 (which seems to work acceptable). This is a quick fix for #1152 and further investigation needed.