[PATCH] D141461: [lld-macho][nfc] Use alignToPowerOf2 instead of alignTo when possible

[Jez Ng via Phabricator via llvm-commits]

int3 created this revision.
int3 added a reviewer: lld-macho.
Herald added projects: lld-macho, All.
int3 requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

Skips the divide operation which is generally expensive. Not that it
matters in this diff, the code changed is not particularly hot, but just
for principle & consistency...


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D141461

Files:
  lld/MachO/Writer.cpp


Index: lld/MachO/Writer.cpp
===================================================================
--- lld/MachO/Writer.cpp
+++ lld/MachO/Writer.cpp
@@ -133,8 +133,8 @@
   LCSubFramework(StringRef umbrella) : umbrella(umbrella) {}
 
   uint32_t getSize() const override {
-    return alignTo(sizeof(sub_framework_command) + umbrella.size() + 1,
-                   target->wordSize);
+    return alignToPowerOf2(sizeof(sub_framework_command) + umbrella.size() + 1,
+                           target->wordSize);
   }
 
   void writeTo(uint8_t *buf) const override {
@@ -388,7 +388,8 @@
   explicit LCRPath(StringRef path) : path(path) {}
 
   uint32_t getSize() const override {
-    return alignTo(sizeof(rpath_command) + path.size() + 1, target->wordSize);
+    return alignToPowerOf2(sizeof(rpath_command) + path.size() + 1,
+                           target->wordSize);
   }
 
   void writeTo(uint8_t *buf) const override {
@@ -412,8 +413,8 @@
   explicit LCDyldEnv(StringRef name) : name(name) {}
 
   uint32_t getSize() const override {
-    return alignTo(sizeof(dyld_env_command) + name.size() + 1,
-                   target->wordSize);
+    return alignToPowerOf2(sizeof(dyld_env_command) + name.size() + 1,
+                           target->wordSize);
   }
 
   void writeTo(uint8_t *buf) const override {
@@ -1075,8 +1076,8 @@
     // `fileOff + fileSize == next segment fileOff`. So we call alignTo() before
     // (instead of after) computing fileSize to ensure that the segments are
     // contiguous. We handle addr / vmSize similarly for the same reason.
-    fileOff = alignTo(fileOff, pageSize);
-    addr = alignTo(addr, pageSize);
+    fileOff = alignToPowerOf2(fileOff, pageSize);
+    addr = alignToPowerOf2(addr, pageSize);
     seg->vmSize = addr - seg->addr;
     seg->fileSize = fileOff - seg->fileOff;
     seg->assignAddressesToStartEndSymbols();


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D141461.488088.patch
Type: text/x-patch
Size: 1878 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230111/c171dbb3/attachment.bin>