public inbox for binutils@sourceware.org
 help / color / mirror / Atom feed
From: "H.J. Lu" <hjl.tools@gmail.com>
To: Alan Modra <amodra@gmail.com>
Cc: Binutils <binutils@sourceware.org>
Subject: Re: [PATCH] PR ld/21402: Skip PIE indirect5 and indirect6 tests on i386
Date: Mon, 24 Apr 2017 21:43:00 -0000	[thread overview]
Message-ID: <CAMe9rOpzNQs9OEogG1arCjSYnfLEy7V1Heoizfascnk=CYeALQ@mail.gmail.com> (raw)
In-Reply-To: <20170424040944.GN24006@bubble.grove.modra.org>

[-- Attachment #1: Type: text/plain, Size: 3424 bytes --]

On Sun, Apr 23, 2017 at 9:09 PM, Alan Modra <amodra@gmail.com> wrote:
> On Fri, Apr 21, 2017 at 11:37:34AM -0700, H.J. Lu wrote:
>> @@ -2863,10 +2867,14 @@ elf_i386_allocate_dynrelocs (struct elf_link_hash_entry *h, void *inf)
>>        int tls_type = elf_i386_hash_entry(h)->tls_type;
>>
>>        /* Make sure this symbol is output as a dynamic symbol.
>> -      Undefined weak syms won't yet be marked as dynamic.  */
>> +      Undefined weak syms won't yet be marked as dynamic.
>> +      PR ld/21402: If this symbol isn't undefined weak symbol,
>> +      don't make it dynamic in PIE.   */
>>        if (h->dynindx == -1
>>         && !h->forced_local
>> -       && !resolved_to_zero)
>> +       && !resolved_to_zero
>> +       && (h->root.type == bfd_link_hash_undefweak
>> +           || !bfd_link_pie (info)))
>>       {
>>         if (! bfd_elf_link_record_dynamic_symbol (info, h))
>>           return FALSE;
>
> This doesn't look correct to me.  Sure, it fixes the particular
> testcase, but why do you make non-pie non-undefweak symbols dynamic
> here?  I believe that only undefweak symbols should be made dynamic
> here, and in all the other places in allocate_dynrelocs.
>
> x86_64 too, but fixing allocate_dynrelocs for x86_64 causes no-plt-1e
> to fail due to lack of a relative GOT dynamic reloc on
>         cmpq    func@GOTPCREL(%rip), %rax
> in no-plt-check1.S.  Seems like x86_64 would need something like your
> change to elf32-i386.c to emit relative GOT dynamic relocs in
> relocate_section.  (For powerpc I decided to only handle plt and copy
> relocs in finish_dynamic_symbol.  GOT dynamic relocs are all handled
> in relocate_section as otherwise there is a lot of duplication and IMO
> less comprehensible code.)

This sounds a goo idea.  I will keep it in mind.

>> @@ -3818,6 +3826,7 @@ elf_i386_relocate_section (bfd *output_bfd,
>>        bfd_vma st_size;
>>        asection *resolved_plt;
>>        bfd_boolean resolved_to_zero;
>> +      bfd_boolean relative_reloc;
>>
>>        r_type = ELF32_R_TYPE (rel->r_info);
>>        if (r_type == R_386_GNU_VTINHERIT
>> @@ -4268,6 +4277,7 @@ r_386_got32:
>>         if (htab->elf.sgot == NULL)
>>           abort ();
>>
>> +       relative_reloc = FALSE;
>>         if (h != NULL)
>>           {
>>             bfd_boolean dyn;
>> @@ -4301,6 +4311,17 @@ r_386_got32:
>>                     bfd_put_32 (output_bfd, relocation,
>>                                 htab->elf.sgot->contents + off);
>>                     h->got.offset |= 1;
>> +
>> +                   if (h->dynindx == -1
>> +                       && !h->forced_local
>> +                       && h->root.type != bfd_link_hash_undefweak
>> +                       && bfd_link_pie (info))
>> +                     {
>> +                       /* PR ld/21402: If this symbol isn't dynamic
>> +                          in PIE, generate R_386_RELATIVE here.  */
>> +                       eh->no_finish_dynamic_symbol = 1;
>> +                       relative_reloc = TRUE;
>> +                     }
>>                   }
>>               }
>>             else
>
> This also looks odd.  Surely you need a relative reloc when pic, not
> pie?  It seems to me you need one any time you won't be calling
> finish_dynamic_symbol (which is where x86 sets most dynamic GOT
> relocs) and have a symbol that doesn't resolve to zero.
>

I checked in this patch for x86-64.

Thanks.

-- 
H.J.

[-- Attachment #2: 0001-x86-64-Force-symbol-dynamic-if-it-isn-t-undefined-we.patch --]
[-- Type: text/x-patch, Size: 8053 bytes --]

From 9e9821ddd80c0d0b3dda54d34cc8867f256d4583 Mon Sep 17 00:00:00 2001
From: "H.J. Lu" <hjl.tools@gmail.com>
Date: Mon, 24 Apr 2017 14:37:12 -0700
Subject: [PATCH] x86-64: Force symbol dynamic if it isn't undefined weak

Force symbol dynamic if it isn't undefined weak.  Generate relative
relocation for GOT reference against non-dynamic symbol in PIC to
avoid unnecessary dynamic symbols.

bfd/

	* elf64-x86-64.c (elf_x86_64_link_hash_entry): Add
	no_finish_dynamic_symbol.
	(elf_x86_64_link_hash_newfunc): Set no_finish_dynamic_symbol to
	0.
	(elf_x86_64_allocate_dynrelocs): If a symbol isn't undefined
	weak symbol, don't make it dynamic.
	(elf_x86_64_relocate_section): If a symbol isn't dynamic in PIC,
	set no_finish_dynamic_symbol and generate R_X86_64_RELATIVE
	relocation for GOT reference.
	(elf_x86_64_finish_dynamic_symbol): Abort if
	no_finish_dynamic_symbol isn't 0.

ld/

	* testsuite/ld-x86-64/no-plt.exp: Also check no-plt-1e.nd.
	* testsuite/ld-x86-64/no-plt-1e.nd: New file.
---
 bfd/ChangeLog                       | 14 ++++++++
 bfd/elf64-x86-64.c                  | 66 +++++++++++++++++++++++++------------
 ld/ChangeLog                        |  5 +++
 ld/testsuite/ld-x86-64/no-plt-1e.nd |  7 ++++
 ld/testsuite/ld-x86-64/no-plt.exp   |  2 +-
 5 files changed, 72 insertions(+), 22 deletions(-)
 create mode 100644 ld/testsuite/ld-x86-64/no-plt-1e.nd

diff --git a/bfd/ChangeLog b/bfd/ChangeLog
index 54e2f0d..b902a76 100644
--- a/bfd/ChangeLog
+++ b/bfd/ChangeLog
@@ -1,5 +1,19 @@
 2017-04-24  H.J. Lu  <hongjiu.lu@intel.com>
 
+	* elf64-x86-64.c (elf_x86_64_link_hash_entry): Add
+	no_finish_dynamic_symbol.
+	(elf_x86_64_link_hash_newfunc): Set no_finish_dynamic_symbol to
+	0.
+	(elf_x86_64_allocate_dynrelocs): If a symbol isn't undefined
+	weak symbol, don't make it dynamic.
+	(elf_x86_64_relocate_section): If a symbol isn't dynamic in PIC,
+	set no_finish_dynamic_symbol and generate R_X86_64_RELATIVE
+	relocation for GOT reference.
+	(elf_x86_64_finish_dynamic_symbol): Abort if
+	no_finish_dynamic_symbol isn't 0.
+
+2017-04-24  H.J. Lu  <hongjiu.lu@intel.com>
+
 	PR ld/21402
 	* elf32-i386.c (elf_i386_allocate_dynrelocs): If a symbol isn't
 	undefined weak symbol, don't make it dynamic.
diff --git a/bfd/elf64-x86-64.c b/bfd/elf64-x86-64.c
index 746b847..965ceb2 100644
--- a/bfd/elf64-x86-64.c
+++ b/bfd/elf64-x86-64.c
@@ -867,6 +867,9 @@ struct elf_x86_64_link_hash_entry
   /* TRUE if symbol has non-GOT/non-PLT relocations in text sections.  */
   unsigned int has_non_got_reloc : 1;
 
+  /* Don't call finish_dynamic_symbol on this symbol.  */
+  unsigned int no_finish_dynamic_symbol : 1;
+
   /* 0: symbol isn't __tls_get_addr.
      1: symbol is __tls_get_addr.
      2: symbol is unknown.  */
@@ -1022,6 +1025,7 @@ elf_x86_64_link_hash_newfunc (struct bfd_hash_entry *entry,
       eh->has_bnd_reloc = 0;
       eh->has_got_reloc = 0;
       eh->has_non_got_reloc = 0;
+      eh->no_finish_dynamic_symbol = 0;
       eh->tls_get_addr = 2;
       eh->func_pointer_refcount = 0;
       eh->plt_bnd.offset = (bfd_vma) -1;
@@ -3218,7 +3222,8 @@ elf_x86_64_allocate_dynrelocs (struct elf_link_hash_entry *h, void * inf)
 	 Undefined weak syms won't yet be marked as dynamic.  */
       if (h->dynindx == -1
 	  && !h->forced_local
-	  && !resolved_to_zero)
+	  && !resolved_to_zero
+	  && h->root.type == bfd_link_hash_undefweak)
 	{
 	  if (! bfd_elf_link_record_dynamic_symbol (info, h))
 	    return FALSE;
@@ -3338,7 +3343,8 @@ elf_x86_64_allocate_dynrelocs (struct elf_link_hash_entry *h, void * inf)
 	 Undefined weak syms won't yet be marked as dynamic.  */
       if (h->dynindx == -1
 	  && !h->forced_local
-	  && !resolved_to_zero)
+	  && !resolved_to_zero
+	  && h->root.type == bfd_link_hash_undefweak)
 	{
 	  if (! bfd_elf_link_record_dynamic_symbol (info, h))
 	    return FALSE;
@@ -3475,6 +3481,7 @@ elf_x86_64_allocate_dynrelocs (struct elf_link_hash_entry *h, void * inf)
 	  if (h->dynindx == -1
 	      && ! h->forced_local
 	      && ! resolved_to_zero
+	      && h->root.type == bfd_link_hash_undefweak
 	      && ! bfd_elf_link_record_dynamic_symbol (info, h))
 	    return FALSE;
 
@@ -4270,6 +4277,7 @@ elf_x86_64_relocate_section (bfd *output_bfd,
       asection *base_got, *resolved_plt;
       bfd_vma st_size;
       bfd_boolean resolved_to_zero;
+      bfd_boolean relative_reloc;
 
       r_type = ELF32_R_TYPE (rel->r_info);
       if (r_type == (int) R_X86_64_GNU_VTINHERIT
@@ -4637,6 +4645,7 @@ do_ifunc_pointer:
 	  if (htab->elf.sgot == NULL)
 	    abort ();
 
+	  relative_reloc = FALSE;
 	  if (h != NULL)
 	    {
 	      bfd_boolean dyn;
@@ -4683,6 +4692,17 @@ do_ifunc_pointer:
 		      /* Note that this is harmless for the GOTPLT64 case,
 			 as -1 | 1 still is -1.  */
 		      h->got.offset |= 1;
+
+		      if (h->dynindx == -1
+			  && !h->forced_local
+			  && h->root.type != bfd_link_hash_undefweak
+			  && bfd_link_pic (info))
+			{
+			  /* If this symbol isn't dynamic in PIC,
+			     generate R_X86_64_RELATIVE here.  */
+			  eh->no_finish_dynamic_symbol = 1;
+			  relative_reloc = TRUE;
+			}
 		    }
 		}
 	      else
@@ -4704,30 +4724,32 @@ do_ifunc_pointer:
 		{
 		  bfd_put_64 (output_bfd, relocation,
 			      base_got->contents + off);
+		  local_got_offsets[r_symndx] |= 1;
 
 		  if (bfd_link_pic (info))
-		    {
-		      asection *s;
-		      Elf_Internal_Rela outrel;
-
-		      /* We need to generate a R_X86_64_RELATIVE reloc
-			 for the dynamic linker.  */
-		      s = htab->elf.srelgot;
-		      if (s == NULL)
-			abort ();
-
-		      outrel.r_offset = (base_got->output_section->vma
-					 + base_got->output_offset
-					 + off);
-		      outrel.r_info = htab->r_info (0, R_X86_64_RELATIVE);
-		      outrel.r_addend = relocation;
-		      elf_append_rela (output_bfd, s, &outrel);
-		    }
-
-		  local_got_offsets[r_symndx] |= 1;
+		    relative_reloc = TRUE;
 		}
 	    }
 
+	  if (relative_reloc)
+	    {
+	      asection *s;
+	      Elf_Internal_Rela outrel;
+
+	      /* We need to generate a R_X86_64_RELATIVE reloc
+		 for the dynamic linker.  */
+	      s = htab->elf.srelgot;
+	      if (s == NULL)
+		abort ();
+
+	      outrel.r_offset = (base_got->output_section->vma
+				 + base_got->output_offset
+				 + off);
+	      outrel.r_info = htab->r_info (0, R_X86_64_RELATIVE);
+	      outrel.r_addend = relocation;
+	      elf_append_rela (output_bfd, s, &outrel);
+	    }
+
 	  if (off >= (bfd_vma) -2)
 	    abort ();
 
@@ -5797,6 +5819,8 @@ elf_x86_64_finish_dynamic_symbol (bfd *output_bfd,
 	  : get_elf_x86_64_backend_data (output_bfd));
 
   eh = (struct elf_x86_64_link_hash_entry *) h;
+  if (eh->no_finish_dynamic_symbol)
+    abort ();
 
   /* We keep PLT/GOT entries without dynamic PLT/GOT relocations for
      resolved undefined weak symbols in executable so that their
diff --git a/ld/ChangeLog b/ld/ChangeLog
index d40086e..ed3797e 100644
--- a/ld/ChangeLog
+++ b/ld/ChangeLog
@@ -1,5 +1,10 @@
 2017-04-24  H.J. Lu  <hongjiu.lu@intel.com>
 
+	* testsuite/ld-x86-64/no-plt.exp: Also check no-plt-1e.nd.
+	* testsuite/ld-x86-64/no-plt-1e.nd: New file.
+
+2017-04-24  H.J. Lu  <hongjiu.lu@intel.com>
+
 	* testsuite/ld-i386/pr12570a.d: Skip for nacl targets.
 	* testsuite/ld-i386/pr12570b.d: Likewise.
 
diff --git a/ld/testsuite/ld-x86-64/no-plt-1e.nd b/ld/testsuite/ld-x86-64/no-plt-1e.nd
new file mode 100644
index 0000000..c96760c
--- /dev/null
+++ b/ld/testsuite/ld-x86-64/no-plt-1e.nd
@@ -0,0 +1,7 @@
+#nm: -g -D
+#target: x86_64-*-*
+
+#failif
+#...
+[0-9]+ +T +func
+#...
diff --git a/ld/testsuite/ld-x86-64/no-plt.exp b/ld/testsuite/ld-x86-64/no-plt.exp
index 5b5c817..9cecd03 100644
--- a/ld/testsuite/ld-x86-64/no-plt.exp
+++ b/ld/testsuite/ld-x86-64/no-plt.exp
@@ -111,7 +111,7 @@ run_cc_link_tests [list \
 	 tmpdir/no-plt-func1.o tmpdir/no-plt-extern1.o" \
 	"" \
 	{dummy.s} \
-	{{readelf -Wr no-plt-1e.rd} {objdump -dwrj.text no-plt-1e.dd}} \
+	{{readelf -Wr no-plt-1e.rd} {nm -gD no-plt-1e.nd} {objdump -dwrj.text no-plt-1e.dd}} \
 	"no-plt-1e" \
     ] \
     [list \
-- 
2.9.3


      parent reply	other threads:[~2017-04-24 21:43 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-20 19:41 H.J. Lu
2017-04-21  1:09 ` Alan Modra
2017-04-21 18:37   ` H.J. Lu
2017-04-24  4:10     ` Alan Modra
2017-04-24 20:51       ` H.J. Lu
2017-04-24 21:43       ` H.J. Lu [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAMe9rOpzNQs9OEogG1arCjSYnfLEy7V1Heoizfascnk=CYeALQ@mail.gmail.com' \
    --to=hjl.tools@gmail.com \
    --cc=amodra@gmail.com \
    --cc=binutils@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).