[flang-commits] [PATCH] D101081: [flang] (NFC) Document Fortran feature history

[Peter Klausler via Phabricator via flang-commits]

klausler added inline comments.


================
Comment at: flang/docs/FortranFeatureHistory.md:11-14
+```eval_rst
+.. contents::
+   :local:
+```
----------------
PeteSteinfeld wrote:
> This looks bogus.
It sure does, but all of our documentation files have this boilerplate.


================
Comment at: flang/docs/FortranFeatureHistory.md:60
+* EXTERNAL subprograms for use as actual arguments
+* &666 label actual arg (for assigned GO TO alternate return)
+* alternate RETURN, ENTRY
----------------
PeteSteinfeld wrote:
> Is "&666" really what you mean here?
Yes, it's what a label looks like when passed as an alternate return actual argument.


================
Comment at: flang/docs/FortranFeatureHistory.md:65
+## Unknown, but were in place by FORTRAN IV at the latest:
+* no INT**REAL (or D.P.); promotion on [+-*/] only REAL -> COMPLEX or D.P.
+* trig and transcendentals are externals, not intrinsics
----------------
PeteSteinfeld wrote:
> It looks like the "**" is not getting rendered as you intended in the formatted markdown.
I'll convert to English.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D101081/new/

https://reviews.llvm.org/D101081