Implements a Loop Fusion Transformation #493
Draft
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
kaushikcfd
force-pushed
the
loop_fusion
branch
from
6cfd454
to
21502fd
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
2 times, most recently
from
b390c5a
to
47d7fa8
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
2 times, most recently
from
5bef1aa
to
214e6b9
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
2 times, most recently
from
d844444
to
776ee64
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
4 times, most recently
from
acbf9a8
to
921ea56
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
2 times, most recently
from
a8dfaf0
to
07194cc
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
from
07194cc
to
a9cd2eb
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
from
a9cd2eb
to
b79ad3a
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
3 times, most recently
from
4c89630
to
a6171e2
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
2 times, most recently
from
371632d
to
d46bfd1
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
3 times, most recently
from
f47d987
to
4e97120
Compare
kaushikcfd
commented
Jun 21, 2022
| candidates = frozenset(candidates) | ||
| vng = kernel.get_var_name_generator() | ||
| fused_chunks = {} | ||
|
|
There was a problem hiding this comment.
Note to self: Add a sanity check here if candidates are not a part of kernel.
|
This could be rebased now that the prerequisite |
kaushikcfd
force-pushed
the
loop_fusion
branch
from
e1498ee
to
3d8e988
Compare
loopy/transform/loop_fusion.py
Outdated
Comment on lines
527
to
533
| raise LoopyError(f"'{iname}' and '{conflict_iname}' " | ||
| "cannot fused be fused as they can be nested " | ||
| "within one another.") |
There was a problem hiding this comment.
Does this need to return an error or could we just remove iname from candidates in this case? (cc illinois-ceesd/mirgecom#1073)
There was a problem hiding this comment.
The idea was that this needs to be an error and the caller passes in the candidates appropriately so that the loop nests can be fused with each other.
kaushikcfd
force-pushed
the
loop_fusion
branch
from
3d8e988
to
02697b6
Compare
kaushikcfd
force-pushed
the
loop_fusion
branch
2 times, most recently
from
38a6e02
to
b57eb9d
Compare
Comment on lines
+37
to
+42
| try: | ||
| import faulthandler | ||
| except ImportError: | ||
| pass | ||
| else: | ||
| faulthandler.enable() |
There was a problem hiding this comment.
Why would the import fail?
There was a problem hiding this comment.
faulthandler was added in 3.3. Loopy, at one point, targeted Python 2.x.
kaushikcfd
force-pushed
the
loop_fusion
branch
from
b57eb9d
to
a05f94d
Compare
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.
Draft because: