- User Since
- Jan 6 2015, 1:11 PM (326 w, 6 d)
Sun, Apr 11
Thu, Apr 8
Looks fine to me except for use of C++-14 feature which might cause build issues. So LGTM apart from that issue. Please make sure no bots are broken
Wed, Apr 7
Mon, Apr 5
Drop "Linalg" in populateLinalgFoldUnitExtentDimsPattern.
Fri, Apr 2
Just a thought if we want to just rename them to F32. But this path is on deprecation, so I dont really care.
Thu, Apr 1
Wed, Mar 31
LGTM-ing this based on the strides being checked ahead of time (instead of in the loop)
Thanks. Just a few minor nits, this time around.
Mon, Mar 29
It is very specific use case :) .
I am fine with this since it is probably a stop-gap solution. Ill let @nicolasvasilache decide if this lives in core.
Treat as LGTM from me, but just dont want to stamp it till I see what Nicolas has to say.
Remove unnecessary flag from test.
Sun, Mar 28
@nicolasvasilache did you mean to approve the patch?
Sat, Mar 27
Fri, Mar 26
Still, it seems we aggressively use reshape to drop 1 dimensions even in the absence of function boundaries.
Maybe a solution would be to have extra canonicalizations:
- subtensor + reshape -> rank-reducing subtensor
- reshape + subtensor_insert -> rank-expanding subtensor_insert
Wed, Mar 24
LGTM. Just a few renaming nits.
THis is doing something different. It is reducing the rank of both the source and the dest, so
Actually fix the test
Tue, Mar 23
Fix test error.
Abandoning this for now (see https://llvm.discourse.group/t/tensor-to-memref-and-its-folding/3033)
Mostly looks fine to me, but probably someone with more context on these (like Nicolas or Uday) might have comments.
This looks fine to me too. I dont know if @nicolasvasilache has comments here.
Mon, Mar 22
LGTM after minor nit is addressed
Nice! Mostly looks fine to me.
Fri, Mar 19
Thu, Mar 18
Going back to the implementation which updated the interface of
InferShapedTypeInterface but dropping the use of OpFoldResult.
Mon, Mar 15
Wondering if there is a problem if we have the struct all the time. One of the issues on SPIR-V side is that we cannot do pointer casts. So a memref_cast where the operand is from something that doesnt need layout to something that does need layout would be not possible support. In general, there is no information of where the memref came from, so we might end up in situations where we cannot lower such memref_cast. If there is no reason to not have a struct, it seems more uniform to have a struct by default for all memref types.
Mar 11 2021
Probably need a test that fails as well and verify the error message. You can use -verify-diagnostics on the command line, and match the error using // expected-error . See other tests that have this.
Document the interface scope better.
Mar 10 2021
Fix build file.
@herhut Had some offline discussions with Jacques. For the time being moving the interface into Linalg. We can unify them once we iron out all the API of the interface. I am happy to pick that up based on what is finally converged upon.
Moving the interface into Linalg for now till can converge on using InferShapedTypeOpInterface
Mar 9 2021
Mar 8 2021
Rebase and remove unused methods.
Awesome! Thanks Lei for tracking this down! LGTM
Rebase and address concerns from offline conversation
Mar 4 2021
Looks good. Just one nit.
I folded things into a single patch. (Was planning to do that, sorry wasnt explicit before).