Thanks

Resolve nits.

if (foo)

bar;

I also removed the tests as suggested by @springerm.

Remove the marker; add filterFn lambda.

+1 for the lambda.
Is it OK to keep the markers for testing only though? Otherwise I don't know how I'd test this.
Thanks!

In D136519#3876652, @cota wrote:

We should later also fix the Utilities typo in the file name.

Can you please provide a rationale for the rename in the commit message? Or a link to a previous discussion on this topic. Thanks!

We should later also fix the Utilities typo in the file name.

In D134762#3821046, @kuhar wrote:

Based on internal feedback, I want to split this patch into a few more gradual changes:

1. (This revision) Rename the dialect and paths, but add an alias for the old dialect name
2. Remove the alias
3. Rename passes/conversions (landed a few days later so that folks have time to catch up)

Does this work for other folks?

@rengolin: AFAIK we do not want dependency cycles in the build system (Cmake or otherwise). I wrongly thought I was reporting one such case.

@akuegel is right, my apologies. I wrongly assumed that in Cmake, ArithDialect depends on DialectUtils, just like it (unnecessarily) does in Bazel. I confirm that removing that superfluous dependence fixes the cycle in Bazel.

This change introduces a circular build dependence: DialectUtils <- ArithmeticUtils <- ArithDialect <- DialectUtils.

Rebase on current HEAD to hopefully avoid unrelated BUILD failures.

Context: https://reviews.llvm.org/D120380

Capture options.sectionMemoryMapper by value instead of by reference

In D112736#3094954, @silvas wrote:

1. I would just like Math dialect transforms to not depend on target-specific dialects. So concretely just RsqrtOpApproximation pattern would move to x86vector/Transform/, since it is the target specific one.
2. I agree, removing Polynomial would be good.

In D112729#3094170, @ezhulenev wrote:

Included this fix in: https://reviews.llvm.org/D112736

Thanks for this! Just some nits below, feel free to ignore.

• Re-indented test/Dialect/Math/polynomial-approximation-avx2.mlir as suggested by pifon2a.

• Remove stray comments from a previous iteration
• Rename flag from avx2 to enableAvx2/enable-avx2
• Document the enableAvx2 bool in the Options struct
• Use FileCheck --check-prefix; remove the now-redundant _avx2.mlir file
• Fix coding style issues

Add missing newline's at EOFs

After discussing with Mehdi and Aart, I'm withdrawing this change in favor of https://reviews.llvm.org/D111461

Thanks Tobias for the quick reply. Can you please commit this change? I don't yet have commit access to the repo.