Merge pull request ocaml#13582 from tmcgilchrist/arm64_prefetch

Enable software prefetching support for ARM64, s390x, PPC64 and Riscv
NickBarnes · Feb 12, 2025 · 1e2d88b · 1e2d88b
2 parents 9fb92d9 + 1c073d5
commit 1e2d88b
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 4 deletions.
diff --git a/Changes b/Changes
@@ -92,7 +92,13 @@ Working version
   Jan Midtgaard)
 
 - #13736: Fix major GC pacing bug triggered by synchronous collections.
-  (Nick Barnes, review by ?)
+  (Nick Barnes, review by Damien Doligez and Tim McGilchrist)
+
+- #13582: Enable software prefetching support for ARM64, s390x, PPC64 and RiscV.
+  Used during GC marking and sweeping to speed up both operations by
+  prefetching data.
+  (Tim McGilchrist, review by Nick Barnes, Antonin Décimo,
+   Stephen Dolan and Miod Vallat)
 
 ### Code generation and optimizations:
 

diff --git a/runtime/caml/misc.h b/runtime/caml/misc.h
@@ -164,9 +164,7 @@ CAMLdeprecated_typedef(addr, char *);
 /* Prefetching */
 
 #ifdef CAML_INTERNALS
-#if (__has_builtin(__builtin_prefetch) || defined(__GNUC__)) && \
-    (defined(__i386__) || defined(__x86_64__) || \
-     defined(_M_IX86) || defined(_M_AMD64))
+#if (__has_builtin(__builtin_prefetch) || defined(__GNUC__))
 #define caml_prefetch(p) __builtin_prefetch((p), 1, 3)
 /* 1 = intent to write; 3 = all cache levels */
 #elif defined(_MSC_VER) && (defined(_M_IX86) || defined(_M_AMD64))