Implement PowerPC64 .localentry for value 1

This adds support for ".localentry 1", a new st_other
STO_PPC64_LOCAL_MASK encoding that signifies a function with a single
entry point like ".localentry 0", but unlike a ".localentry 0"
function does not preserve r2.

include/
	* elf/ppc64.h: Specify byte offset to local entry for values
	of two to six in STO_PPC64_LOCAL_MASK.  Clarify r2 return
	value for such functions when entering via global entry point.
	Specify meaning of a value of one in STO_PPC64_LOCAL_MASK.
bfd/
	* elf64-ppc.c (ppc64_elf_size_stubs): Use a ppc_stub_long_branch_r2off
	for calls to symbols with STO_PPC64_LOCAL_MASK bits set to 1.
gas/
	* config/tc-ppc.c (ppc_elf_localentry): Allow .localentry values
	of 1 and 7 to directly set value into STO_PPC64_LOCAL_MASK bits.
ld/testsuite/
	* ld-powerpc/elfv2.s: Add .localentry f5,1 testcase.
	* ld-powerpc/elfv2exe.d: Update.
	* ld-powerpc/elfv2so.d: Update.

10 files changed, 106 insertions, 31 deletions

diff --git a/bfd/ChangeLog b/bfd/ChangeLog
index e0e1a46..a7dfe1a 100644
--- a/bfd/ChangeLog
+++ b/bfd/ChangeLog
@@ -1,3 +1,8 @@
+2018-07-26  Alan Modra  <amodra@gmail.com>
+
+	* elf64-ppc.c (ppc64_elf_size_stubs): Use a ppc_stub_long_branch_r2off
+	for calls to symbols with STO_PPC64_LOCAL_MASK bits set to 1.
+
 2018-07-25  Alan Modra  <amodra@gmail.com>
 
 	* elf32-ppc.c (ppc_elf_relax_section): Ignore common or undef locals.
diff --git a/bfd/elf64-ppc.c b/bfd/elf64-ppc.c
index 6b293ce..5cd3b8d 100644
--- a/bfd/elf64-ppc.c
+++ b/bfd/elf64-ppc.c
@@ -12862,12 +12862,15 @@ ppc64_elf_size_stubs (struct bfd_link_info *info)
 			 _init and _fini functions, it may be that a
 			 call to what looks like a local sym is in
 			 fact a call needing a TOC adjustment.  */
-		      if (code_sec != NULL
-			  && code_sec->output_section != NULL
-			  && (htab->sec_info[code_sec->id].toc_off
-			      != htab->sec_info[section->id].toc_off)
-			  && (code_sec->has_toc_reloc
-			      || code_sec->makes_toc_func_call))
+		      if ((code_sec != NULL
+			   && code_sec->output_section != NULL
+			   && (htab->sec_info[code_sec->id].toc_off
+			       != htab->sec_info[section->id].toc_off)
+			   && (code_sec->has_toc_reloc
+			       || code_sec->makes_toc_func_call))
+			  || (((hash ? hash->elf.other : sym->st_other)
+			       & STO_PPC64_LOCAL_MASK)
+			      == 1 << STO_PPC64_LOCAL_BIT))
 			stub_type = ppc_stub_long_branch_r2off;
 		    }
 
diff --git a/gas/ChangeLog b/gas/ChangeLog
index e0de16e..fa8a7c5 100644
--- a/gas/ChangeLog
+++ b/gas/ChangeLog
@@ -1,3 +1,8 @@
+2018-07-26  Alan Modra  <amodra@gmail.com>
+
+	* config/tc-ppc.c (ppc_elf_localentry): Allow .localentry values
+	of 1 and 7 to directly set value into STO_PPC64_LOCAL_MASK bits.
+
 2018-07-25  H.J. Lu  <hongjiu.lu@intel.com>
 
 	* config/tc-i386.c (Broadcast_Operation): Add bytes.
diff --git a/gas/config/tc-ppc.c b/gas/config/tc-ppc.c
index 92e5eb5..4388b25 100644
--- a/gas/config/tc-ppc.c
+++ b/gas/config/tc-ppc.c
@@ -2378,13 +2378,23 @@ ppc_elf_localentry (int ignore ATTRIBUTE_UNUSED)
   if (resolve_expression (&exp)
       && exp.X_op == O_constant)
     {
-      unsigned char encoded = PPC64_SET_LOCAL_ENTRY_OFFSET (exp.X_add_number);
+      unsigned int encoded, ok;
 
-      if (exp.X_add_number != (offsetT) PPC64_LOCAL_ENTRY_OFFSET (encoded))
-        as_bad (_(".localentry expression for `%s' "
-		  "is not a valid power of 2"), S_GET_NAME (sym));
+      ok = 1;
+      if (exp.X_add_number == 1 || exp.X_add_number == 7)
+	encoded = exp.X_add_number << STO_PPC64_LOCAL_BIT;
       else
 	{
+	  encoded = PPC64_SET_LOCAL_ENTRY_OFFSET (exp.X_add_number);
+	  if (exp.X_add_number != (offsetT) PPC64_LOCAL_ENTRY_OFFSET (encoded))
+	    {
+	      as_bad (_(".localentry expression for `%s' "
+			"is not a valid power of 2"), S_GET_NAME (sym));
+	      ok = 0;
+	    }
+	}
+      if (ok)
+	{
 	  bfdsym = symbol_get_bfdsym (sym);
 	  elfsym = elf_symbol_from (bfd_asymbol_bfd (bfdsym), bfdsym);
 	  gas_assert (elfsym);
diff --git a/include/ChangeLog b/include/ChangeLog
index c408eb7..8558368 100644
--- a/include/ChangeLog
+++ b/include/ChangeLog
@@ -1,3 +1,10 @@
+2018-07-26  Alan Modra  <amodra@gmail.com>
+
+	* elf/ppc64.h: Specify byte offset to local entry for values
+	of two to six in STO_PPC64_LOCAL_MASK.  Clarify r2 return
+	value for such functions when entering via global entry point.
+	Specify meaning of a value of one in STO_PPC64_LOCAL_MASK.
+
 2018-07-24  Alan Modra  <amodra@gmail.com>
 
 	PR 23430
diff --git a/include/elf/ppc64.h b/include/elf/ppc64.h
index 501aebc..387c1d6 100644
--- a/include/elf/ppc64.h
+++ b/include/elf/ppc64.h
@@ -198,13 +198,15 @@ END_RELOC_NUMBERS (R_PPC64_max)
 #define EF_PPC64_ABI	3
 
 /* The ELFv2 ABI uses three bits in the symbol st_other field of a
-   function definition to specify the number of instructions between a
+   function definition to specify the number of bytes between a
    function's global entry point and local entry point.
+   Values of two to six specify powers of two from four to sixty four
+   bytes.  For such functions:
    The global entry point is used when it is necessary to set up the
    toc pointer (r2) for the function.  Callers must enter the global
    entry point with r12 set to the global entry point address.  On
-   return from the function, r2 may have a different value to that
-   which it had on entry.
+   return from the function r2 will contain the toc pointer for the
+   function.
    The local entry point is used when r2 is known to already be valid
    for the function.  There is no requirement on r12 when using the
    local entry point, and on return r2 will contain the same value as
@@ -212,7 +214,9 @@ END_RELOC_NUMBERS (R_PPC64_max)
    A value of zero in these bits means that the function has a single
    entry point with no requirement on r12 or r2, and that on return r2
    will contain the same value as at entry.
-   Values of one and seven are reserved.  */
+   A value of one means that the function has a single entry point
+   with no requirement on r12 or r2, and that r2 is *not* preserved.
+   A value of seven is reserved.  */
 #define STO_PPC64_LOCAL_BIT		5
 #define STO_PPC64_LOCAL_MASK		(7 << STO_PPC64_LOCAL_BIT)
 
diff --git a/ld/ChangeLog b/ld/ChangeLog
index 0b3058b..e1692f1 100644
--- a/ld/ChangeLog
+++ b/ld/ChangeLog
@@ -1,3 +1,9 @@
+2018-07-26  Alan Modra  <amodra@gmail.com>
+
+	* ld-powerpc/elfv2.s: Add .localentry f5,1 testcase.
+	* ld-powerpc/elfv2exe.d: Update.
+	* ld-powerpc/elfv2so.d: Update.
+
 2018-07-25  Alan Modra  <amodra@gmail.com>
 
 	* testsuite/ld-powerpc/big.s: New file.
diff --git a/ld/testsuite/ld-powerpc/elfv2.s b/ld/testsuite/ld-powerpc/elfv2.s
index c2a4c3b..4fedbb7 100644
--- a/ld/testsuite/ld-powerpc/elfv2.s
+++ b/ld/testsuite/ld-powerpc/elfv2.s
@@ -25,8 +25,17 @@ f1:
  nop
  bl f4
  nop
+ bl f5
+ nop
  ld 0,48(1)
  addi 1,1,32
  mtlr 0
  blr
  .size f1,.-f1
+
+ .globl f5
+ .type f5,@function
+f5:
+ .localentry f5,1
+ blr
+ .size f5,.-f5
diff --git a/ld/testsuite/ld-powerpc/elfv2exe.d b/ld/testsuite/ld-powerpc/elfv2exe.d
index 77bf6e2..d08e600 100644
--- a/ld/testsuite/ld-powerpc/elfv2exe.d
+++ b/ld/testsuite/ld-powerpc/elfv2exe.d
@@ -9,17 +9,22 @@ Disassembly of section \.text:
 
 0+100000c0 <.*\.plt_branch\.f4>:
 .*:	(3d 82 ff ff|ff ff 82 3d) 	addis   r12,r2,-1
-.*:	(e9 8c 7f 28|28 7f 8c e9) 	ld      r12,32552\(r12\)
+.*:	(e9 8c 7f 58|58 7f 8c e9) 	ld      r12,32600\(r12\)
 .*:	(7d 89 03 a6|a6 03 89 7d) 	mtctr   r12
 .*:	(4e 80 04 20|20 04 80 4e) 	bctr
 
 0+100000d0 <.*\.plt_branch\.f2>:
 .*:	(3d 82 ff ff|ff ff 82 3d) 	addis   r12,r2,-1
-.*:	(e9 8c 7f 30|30 7f 8c e9) 	ld      r12,32560\(r12\)
+.*:	(e9 8c 7f 60|60 7f 8c e9) 	ld      r12,32608\(r12\)
 .*:	(7d 89 03 a6|a6 03 89 7d) 	mtctr   r12
 .*:	(4e 80 04 20|20 04 80 4e) 	bctr
 
-0+100000e0 <(f1|_start)>:
+0+100000e0 <.*\.long_branch_r2off\.f5>:
+.*:	(f8 41 00 18|18 00 41 f8) 	std     r2,24\(r1\)
+.*:	(48 00 00 6c|6c 00 00 48) 	b       .* <f5>
+	\.\.\.
+
+0+10000100 <(f1|_start)>:
 .*:	(3c 40 10 02|02 10 40 3c) 	lis     r2,4098
 .*:	(38 42 82 00|00 82 42 38) 	addi    r2,r2,-32256
 .*:	(7c 08 02 a6|a6 02 08 7c) 	mflr    r0
@@ -27,14 +32,19 @@ Disassembly of section \.text:
 .*:	(f8 01 00 30|30 00 01 f8) 	std     r0,48\(r1\)
 .*:	(4b ff ff f5|f5 ff ff 4b) 	bl      .* <(f1|_start)\+0x8>
 .*:	(e8 62 80 08|08 80 62 e8) 	ld      r3,-32760\(r2\)
-.*:	(4b ff ff d5|d5 ff ff 4b) 	bl      .*\.plt_branch\.f2>
+.*:	(4b .. .. ..|.. .. .. 4b) 	bl      .*\.plt_branch\.f2>
 .*:	(60 00 00 00|00 00 00 60) 	nop
 .*:	(e8 62 80 10|10 80 62 e8) 	ld      r3,-32752\(r2\)
-.*:	(48 00 87 81|81 87 00 48) 	bl      10008888 <f3>
+.*:	(48 .. .. ..|.. .. .. 48) 	bl      10008888 <f3>
 .*:	(60 00 00 00|00 00 00 60) 	nop
-.*:	(4b ff ff b1|b1 ff ff 4b) 	bl      .*\.plt_branch\.f4>
+.*:	(4b .. .. ..|.. .. .. 4b) 	bl      .*\.plt_branch\.f4>
 .*:	(60 00 00 00|00 00 00 60) 	nop
+.*:	(4b .. .. ..|.. .. .. 4b) 	bl      .*\.long_branch_r2off\.f5>
+.*:	(e8 41 00 18|18 00 41 e8) 	ld      r2,24\(r1\)
 .*:	(e8 01 00 30|30 00 01 e8) 	ld      r0,48\(r1\)
 .*:	(38 21 00 20|20 00 21 38) 	addi    r1,r1,32
 .*:	(7c 08 03 a6|a6 03 08 7c) 	mtlr    r0
 .*:	(4e 80 00 20|20 00 80 4e) 	blr
+
+0+10000150 <f5>:
+.*:	(4e 80 00 20|20 00 80 4e) 	blr
diff --git a/ld/testsuite/ld-powerpc/elfv2so.d b/ld/testsuite/ld-powerpc/elfv2so.d
index 5f198a9..0853f8a 100644
--- a/ld/testsuite/ld-powerpc/elfv2so.d
+++ b/ld/testsuite/ld-powerpc/elfv2so.d
@@ -9,13 +9,20 @@ Disassembly of section \.text:
 
 .* <.*\.plt_call\.f4>:
 .*:	(f8 41 00 18|18 00 41 f8) 	std     r2,24\(r1\)
-.*:	(e9 82 80 38|38 80 82 e9) 	ld      r12,-32712\(r2\)
+.*:	(e9 82 80 40|40 80 82 e9) 	ld      r12,-32704\(r2\)
 .*:	(7d 89 03 a6|a6 03 89 7d) 	mtctr   r12
 .*:	(4e 80 04 20|20 04 80 4e) 	bctr
 	\.\.\.
 
 .* <.*\.plt_call\.f3>:
 .*:	(f8 41 00 18|18 00 41 f8) 	std     r2,24\(r1\)
+.*:	(e9 82 80 30|30 80 82 e9) 	ld      r12,-32720\(r2\)
+.*:	(7d 89 03 a6|a6 03 89 7d) 	mtctr   r12
+.*:	(4e 80 04 20|20 04 80 4e) 	bctr
+	\.\.\.
+
+.* <.*\.plt_call\.f5>:
+.*:	(f8 41 00 18|18 00 41 f8) 	std     r2,24\(r1\)
 .*:	(e9 82 80 28|28 80 82 e9) 	ld      r12,-32728\(r2\)
 .*:	(7d 89 03 a6|a6 03 89 7d) 	mtctr   r12
 .*:	(4e 80 04 20|20 04 80 4e) 	bctr
@@ -23,14 +30,14 @@ Disassembly of section \.text:
 
 .* <.*\.plt_call\.f1>:
 .*:	(f8 41 00 18|18 00 41 f8) 	std     r2,24\(r1\)
-.*:	(e9 82 80 40|40 80 82 e9) 	ld      r12,-32704\(r2\)
+.*:	(e9 82 80 48|48 80 82 e9) 	ld      r12,-32696\(r2\)
 .*:	(7d 89 03 a6|a6 03 89 7d) 	mtctr   r12
 .*:	(4e 80 04 20|20 04 80 4e) 	bctr
 	\.\.\.
 
 .* <.*\.plt_call\.f2>:
 .*:	(f8 41 00 18|18 00 41 f8) 	std     r2,24\(r1\)
-.*:	(e9 82 80 30|30 80 82 e9) 	ld      r12,-32720\(r2\)
+.*:	(e9 82 80 38|38 80 82 e9) 	ld      r12,-32712\(r2\)
 .*:	(7d 89 03 a6|a6 03 89 7d) 	mtctr   r12
 .*:	(4e 80 04 20|20 04 80 4e) 	bctr
 	\.\.\.
@@ -41,19 +48,25 @@ Disassembly of section \.text:
 .*:	(7c 08 02 a6|a6 02 08 7c) 	mflr    r0
 .*:	(f8 21 ff e1|e1 ff 21 f8) 	stdu    r1,-32\(r1\)
 .*:	(f8 01 00 30|30 00 01 f8) 	std     r0,48\(r1\)
-.*:	(4b ff ff ad|ad ff ff 4b) 	bl      .*\.plt_call\.f1>
+.*:	(4b .. .. ..|.. .. .. 4b) 	bl      .*\.plt_call\.f1>
 .*:	(e8 62 80 08|08 80 62 e8) 	ld      r3,-32760\(r2\)
-.*:	(4b ff ff c5|c5 ff ff 4b) 	bl      .*\.plt_call\.f2>
+.*:	(4b .. .. ..|.. .. .. 4b) 	bl      .*\.plt_call\.f2>
 .*:	(e8 41 00 18|18 00 41 e8) 	ld      r2,24\(r1\)
 .*:	(e8 62 80 10|10 80 62 e8) 	ld      r3,-32752\(r2\)
-.*:	(4b ff ff 79|79 ff ff 4b) 	bl      .*\.plt_call\.f3>
+.*:	(4b .. .. ..|.. .. .. 4b) 	bl      .*\.plt_call\.f3>
+.*:	(e8 41 00 18|18 00 41 e8) 	ld      r2,24\(r1\)
+.*:	(4b .. .. ..|.. .. .. 4b) 	bl      .*\.plt_call\.f4>
 .*:	(e8 41 00 18|18 00 41 e8) 	ld      r2,24\(r1\)
-.*:	(4b ff ff 51|51 ff ff 4b) 	bl      .*\.plt_call\.f4>
+.*:	(4b .. .. ..|.. .. .. 4b) 	bl      .*\.plt_call\.f5>
 .*:	(e8 41 00 18|18 00 41 e8) 	ld      r2,24\(r1\)
 .*:	(e8 01 00 30|30 00 01 e8) 	ld      r0,48\(r1\)
 .*:	(38 21 00 20|20 00 21 38) 	addi    r1,r1,32
 .*:	(7c 08 03 a6|a6 03 08 7c) 	mtlr    r0
 .*:	(4e 80 00 20|20 00 80 4e) 	blr
+
+.* <f5>:
+.*:	(4e 80 00 20|20 00 80 4e) 	blr
+.*:	(60 00 00 00|00 00 00 60) 	nop
 .*
 .*
 
@@ -73,14 +86,17 @@ Disassembly of section \.text:
 .*:	(e9 6b 00 08|08 00 6b e9) 	ld      r11,8\(r11\)
 .*:	(4e 80 04 20|20 04 80 4e) 	bctr
 
-.* <f3@plt>:
+.* <f5@plt>:
 .*:	(4b ff ff c8|c8 ff ff 4b) 	b       .* <__glink_PLTresolve>
 
-.* <f2@plt>:
+.* <f3@plt>:
 .*:	(4b ff ff c4|c4 ff ff 4b) 	b       .* <__glink_PLTresolve>
 
-.* <f4@plt>:
+.* <f2@plt>:
 .*:	(4b ff ff c0|c0 ff ff 4b) 	b       .* <__glink_PLTresolve>
 
-.* <f1@plt>:
+.* <f4@plt>:
 .*:	(4b ff ff bc|bc ff ff 4b) 	b       .* <__glink_PLTresolve>
+
+.* <f1@plt>:
+.*:	(4b ff ff b8|b8 ff ff 4b) 	b       .* <__glink_PLTresolve>