[PATCH] D128899: [LoongArch] Add codegen support for fpround, fpextend and converting between signed integer and floating-point

[Gong LingQin via Phabricator via llvm-commits]

gonglingqin added a comment.

In D128899#3626549 <https://reviews.llvm.org/D128899#3626549>, @xen0n wrote:

> Still not quite sure about the necessity of the `_LA64` nodes/pats (is it again because of tblgen intricacies?) but the rest looks okay.

Yes, we want to reduce the intricacies of tblgen and use the name to identify this node only for LA64.



================
Comment at: llvm/lib/Target/LoongArch/LoongArchISelLowering.h:42
+
+  FTINT_LA,
+
----------------
xen0n wrote:
> I'm not quite sure about what the "LA" suffix means. Aren't we obviously LoongArch? Or is this meaning "L(64-bit integer), Any FP precision"? This seems likely until I find both "w" and "l" `ftintrz`'s are pattern-matched from this.
I had hoped to use "LA" for "LoongArch". Thanks for your reminding, I found this suffix redundant for the FTINT node, I will remove it.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D128899/new/

https://reviews.llvm.org/D128899