[llvm] 2deffca - [LangRef] Fix language around strictfp semantics (#213879)

[via llvm-commits]

Author: Ramkumar Ramachandra
Date: 2026-08-04T12:21:06Z
New Revision: 2deffca50638d88b6df28316db2e6d3f617413da

URL: https://github.com/llvm/llvm-project/commit/2deffca50638d88b6df28316db2e6d3f617413da
DIFF: https://github.com/llvm/llvm-project/commit/2deffca50638d88b6df28316db2e6d3f617413da.diff

LOG: [LangRef] Fix language around strictfp semantics (#213879)

The LangRef wording changed in 15bb4a97a7 ([IR] Make semantics of
strictfp consistent v2, #211769) is unfortunately incorrect. There are
some open questions around its semantics, but a simple revert should do
for the moment.

Reported-by: Ömer Sinan Ağacan <omer at osa1.net>

Added: 
    

Modified: 
    llvm/docs/LangRef.md

Removed: 
    


################################################################################
diff  --git a/llvm/docs/LangRef.md b/llvm/docs/LangRef.md
index 1fc6ea1a7ed9d..843a2d626cd06 100644
--- a/llvm/docs/LangRef.md
+++ b/llvm/docs/LangRef.md
@@ -2800,8 +2800,7 @@ fn -> other_fn -> other_fn ; fn is norecurse
     optimizations that require assumptions about the floating-point rounding
     mode or that might alter the state of floating-point status flags that
     might otherwise be set or cleared by calling this function. LLVM will
-    not introduce any new floating-point instructions that may trap. All
-    function definitions that contain strictfp calls must be marked strictfp.
+    not introduce any new floating-point instructions that may trap.
 
 (denormal_fpenv)=
 
@@ -26741,6 +26740,9 @@ point intrinsics must have the `strictfp` attribute either on the
 calling instruction or on the declaration or definition of the function
 being called.
 
+All function *definitions* that use constrained floating point intrinsics
+must have the `strictfp` attribute.
+
 #### '`llvm.experimental.constrained.fadd`' Intrinsic
 
 ##### Syntax: