[llvm] [llvm][DebugInfo] Use formatv() instead of format() (PR #191308)

[Konrad Kleine via llvm-commits]

================
@@ -30,19 +32,20 @@ uint8_t DWARFDebugMacro::MacroHeader::getOffsetByteSize() const {
 
 void DWARFDebugMacro::MacroHeader::dumpMacroHeader(raw_ostream &OS) const {
   // FIXME: Add support for dumping opcode_operands_table
-  OS << format("macro header: version = 0x%04" PRIx16, Version)
-     << format(", flags = 0x%02" PRIx8, Flags)
+  OS << formatv("macro header: version = {0:x+4}", Version)
+     << formatv(", flags = {0:x+2}", Flags)
      << ", format = " << FormatString(getDwarfFormat());
   if (Flags & MACRO_DEBUG_LINE_OFFSET)
-    OS << format(", debug_line_offset = 0x%0*" PRIx64, 2 * getOffsetByteSize(),
-                 DebugLineOffset);
+    OS << formatv(", debug_line_offset = 0x{0:x-}",
----------------
kwk wrote:

That is correct. It is because of the filler. 

https://github.com/llvm/llvm-project/pull/191308