[GOLD] powerpc DT_RELACOUNT

Message ID ZJWDWhpXisjeWfPg@squeak.grove.modra.org
State Accepted
Headers
Series [GOLD] powerpc DT_RELACOUNT |

Checks

Context Check Description
snail/binutils-gdb-check success Github commit url

Commit Message

Alan Modra June 23, 2023, 11:34 a.m. UTC
  DT_RELACOUNT was calculated incorrectly, and relative relocs not
sorted as they should be to the start of .rela.dyn, due to adding one
particular class of dynamic reloc using the wrong "add" method.

	* powerpc.cc (Target_powerpc::Scan::global): Add relative
	dyn relocs for ADDR64 and similar using add_global_relative.
  

Patch

diff --git a/gold/powerpc.cc b/gold/powerpc.cc
index aa2d9a18787..0c15e1b3721 100644
--- a/gold/powerpc.cc
+++ b/gold/powerpc.cc
@@ -9005,9 +9005,14 @@  Target_powerpc<size, big_endian>::Scan::global(
 		  = target->rela_dyn_section(symtab, layout, is_ifunc);
 		unsigned int dynrel = (is_ifunc ? elfcpp::R_POWERPC_IRELATIVE
 				       : elfcpp::R_POWERPC_RELATIVE);
-		rela_dyn->add_symbolless_global_addend(
+		// Use the "add" method that marks the reloc as being
+		// relative.  This is proper here and in other places
+		// that add IRELATIVE relocs because those relocs go
+		// into a separate section that isn't sorted, so it
+		// doesn't matter that they are marked is_relative.
+		rela_dyn->add_global_relative(
 		    gsym, dynrel, output_section, object, data_shndx,
-		    reloc.get_r_offset(), reloc.get_r_addend());
+		    reloc.get_r_offset(), reloc.get_r_addend(), false);
 	      }
 	    else
 	      {