PING: [PATCH] X86: Optimize access to globals in PIE with copy reloc

From: H.J. Lu <hjl.tools@gmail.com>

PING.


---------- Forwarded message ----------
From: H.J. Lu <hjl.tools@gmail.com>
Date: Wed, Jul 1, 2015 at 5:11 AM
Subject: [PATCH] X86: Optimize access to globals in PIE with copy reloc
To: gcc-patches@gcc.gnu.org


Normally, with PIE, GCC accesses globals that are extern to the module
using GOT.  This is two instructions, one to get the address of the global
from GOT and the other to get the value.  Examples:

---
extern int a_glob;
int
main ()
{
  return a_glob;
}
---

With PIE, the generated code accesses global via GOT using two memory
loads:

        movq    a_glob@GOTPCREL(%rip), %rax
        movl    (%rax), %eax

for 64-bit or

        movl    a_glob@GOT(%ecx), %eax
        movl    (%eax), %eax

for 32-bit.

Some experiments on google and SPEC CPU benchmarks show that the extra
instruction affects performance by 1% to 5%.

Solution - Copy Relocations:

When the linker supports copy relocations, GCC can always assume that
the global will be defined in the executable.  For globals that are
truly extern (come from shared objects), the linker will create copy
relocations and have them defined in the executable.  Result is that
no global access needs to go through GOT and hence improves performance.
We can generate

        movl    a_glob(%rip), %eax

for 64-bit and

        movl    a_glob@GOTOFF(%eax), %eax

for 32-bit.  This optimization only applies to undefined non-weak
non-TLS global data.  Undefined weak global or TLS data access still
must go through GOT.

This patch reverts legitimate_pic_address_disp_p change made in revision
218397, which only applies to x86-64.  Instead, this patch updates
targetm.binds_local_p to indicate if undefined non-weak non-TLS global
data is defined locally in PIE.  It also introduces a new target hook,
binds_tls_local_p to distinguish TLS variable from non-TLS variable.  By
default, binds_tls_local_p is the same as binds_local_p which assumes
TLS variable.

This patch checks if 32-bit and 64-bit linkers support PIE with copy
reloc at configure time.  64-bit linker is enabled in binutils 2.25
and 32-bit linker is enabled in binutils 2.26.  This optimization
is enabled only if the linker support is available.

Since copy relocation in PIE is incompatible with DSO created by
-Wl,-Bsymbolic, this patch also adds a new option, -fsymbolic, which
controls how references to global symbols are bound.  The -fsymbolic
option binds references to global symbols to the local definitions
and external references globally.  It avoids copy relocations in PIE
and optimizes global symbol references in shared library created
by -Wl,-Bsymbolic.

gcc/

        PR target/65846
        PR target/65886
        * configure.ac (HAVE_LD_PIE_COPYRELOC): Renamed to ...
        (HAVE_LD_X86_64_PIE_COPYRELOC): This.
        (HAVE_LD_386_PIE_COPYRELOC): New.   Defined to 1 if Linux/ia32
        linker supports PIE with copy reloc.
        * output.h (default_binds_tls_local_p): New.
        (default_binds_local_p_3): Add 2 bool arguments.
        * target.def (binds_tls_local_p): New target hook.
        * varasm.c (decl_default_tls_model): Replace targetm.binds_local_p
        with targetm.binds_tls_local_p.
        (default_binds_local_p_3): Add a bool argument to indicate TLS
        variable and a bool argument to indicate if an undefined non-TLS
        non-weak data is local.  Double check TLS variable.  If an
        undefined non-TLS non-weak data is local, treat it as defined
        locally.
        (default_binds_local_p): Pass true and false to
        default_binds_local_p_3.
        (default_binds_local_p_2): Likewise.
        (default_binds_local_p_1): Likewise.
        (default_binds_tls_local_p): New.
        * config.in: Regenerated.
        * configure: Likewise.
        * doc/tm.texi: Likewise.
        * config/i386/i386.c (legitimate_pic_address_disp_p): Don't
        check HAVE_LD_PIE_COPYRELOC here.
        (ix86_binds_local): New.
        (ix86_binds_tls_local_p): Likewise.
        (ix86_binds_local_p): Use it.
        (TARGET_BINDS_TLS_LOCAL_P): New.
        * doc/tm.texi.in (TARGET_BINDS_TLS_LOCAL_P): New hook.

gcc/testsuite/

        PR target/65846
        PR target/65886
        * gcc.target/i386/pie-copyrelocs-1.c: Updated for ia32.
        * gcc.target/i386/pie-copyrelocs-2.c: Likewise.
        * gcc.target/i386/pie-copyrelocs-3.c: Likewise.
        * gcc.target/i386/pie-copyrelocs-4.c: Likewise.
        * gcc.target/i386/pr32219-9.c: Likewise.
        * gcc.target/i386/pr32219-10.c: New file.
        * gcc.target/i386/pr65886-1.c: Likewise.
        * gcc.target/i386/pr65886-2.c: Likewise.
        * gcc.target/i386/pr65886-3.c: Likewise.
        * gcc.target/i386/pr65886-4.c: Likewise.
        * gcc.target/i386/pr65886-4.c: Likewise.
        * gcc.target/i386/pr65886-5.c: Likewise.

        * lib/target-supports.exp (check_effective_target_pie_copyreloc):
        Check HAVE_LD_X86_64_PIE_COPYRELOC and HAVE_LD_386_PIE_COPYRELOC
        instead of HAVE_LD_X86_64_PIE_COPYRELOC.
---
 gcc/common.opt                                   |  4 ++
 gcc/config.in                                    | 18 ++++---
 gcc/config/i386/i386.c                           | 44 ++++++++++-----
 gcc/configure                                    | 68 ++++++++++++++++++++---
 gcc/configure.ac                                 | 64 +++++++++++++++++++---
 gcc/doc/invoke.texi                              | 12 ++++-
 gcc/doc/tm.texi                                  | 10 ++++
 gcc/doc/tm.texi.in                               |  2 +
 gcc/output.h                                     |  4 +-
 gcc/target.def                                   | 14 +++++
 gcc/testsuite/gcc.target/i386/pie-copyrelocs-1.c |  4 +-
 gcc/testsuite/gcc.target/i386/pie-copyrelocs-2.c |  4 +-
 gcc/testsuite/gcc.target/i386/pie-copyrelocs-3.c |  2 +-
 gcc/testsuite/gcc.target/i386/pie-copyrelocs-4.c |  4 +-
 gcc/testsuite/gcc.target/i386/pr32219-10.c       | 16 ++++++
 gcc/testsuite/gcc.target/i386/pr32219-9.c        |  2 +
 gcc/testsuite/gcc.target/i386/pr65886-1.c        | 17 ++++++
 gcc/testsuite/gcc.target/i386/pr65886-2.c        | 15 ++++++
 gcc/testsuite/gcc.target/i386/pr65886-3.c        | 16 ++++++
 gcc/testsuite/gcc.target/i386/pr65886-4.c        | 15 ++++++
 gcc/testsuite/gcc.target/i386/pr65886-5.c        | 18 +++++++
 gcc/testsuite/gcc.target/i386/pr65886-6.c        | 18 +++++++
 gcc/testsuite/lib/target-supports.exp            | 10 +++-
 gcc/varasm.c                                     | 69 ++++++++++++++++++------
 24 files changed, 393 insertions(+), 57 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/i386/pr32219-10.c
 create mode 100644 gcc/testsuite/gcc.target/i386/pr65886-1.c
 create mode 100644 gcc/testsuite/gcc.target/i386/pr65886-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/pr65886-3.c
 create mode 100644 gcc/testsuite/gcc.target/i386/pr65886-4.c
 create mode 100644 gcc/testsuite/gcc.target/i386/pr65886-5.c
 create mode 100644 gcc/testsuite/gcc.target/i386/pr65886-6.c

   if (!DECL_P (exp))
@@ -6851,10 +6855,33 @@ default_binds_local_p_3 (const_tree exp, bool
shlib, bool weak_dominate,
                              || (!in_lto_p
                                  && DECL_INITIAL (exp) == error_mark_node)));

-  /* A non-external variable is defined locally only if it isn't
-     uninitialized COMMON variable or common_local_p is true.  */
-  bool defined_locally = (!DECL_EXTERNAL (exp)
-                         && (!uninited_common || common_local_p));
+  /* If TLS is false, double check if EXP is a TLS variable.  */
+  if (!tls)
+    tls = TREE_CODE (exp) == VAR_DECL && DECL_THREAD_LOCAL_P (exp);
+
+  bool defined_locally;
+  if (!DECL_EXTERNAL (exp))
+    {
+      /* When a variable isn't defined externally, it is defined locally
+        only if it isn't uninitialized COMMON variable or common_local_p
+        is true and external references aren't bound globally.  */
+      defined_locally = (!uninited_common
+                        || (common_local_p && !flag_symbolic));
+
+    }
+  else
+    {
+      /* An undefined variable is defined locally only if it is a
+        non-TLS non-weak data variable, external references aren't
+        bound globally and linker can guarantee it will be defined
+        locally.  */
+      defined_locally = (!tls
+                        && data_local_p
+                        && !flag_symbolic
+                        && TREE_CODE (exp) != FUNCTION_DECL
+                        && !DECL_WEAK (exp));
+    }
+
   if (symtab_node *node = symtab_node::get (exp))
     {
       if (node->in_other_partition)
@@ -6864,7 +6891,7 @@ default_binds_local_p_3 (const_tree exp, bool
shlib, bool weak_dominate,
       else if (resolution_local_p (node->resolution))
        resolved_locally = true;
     }
-  if (defined_locally && weak_dominate && !shlib)
+  if (defined_locally && weak_dominate && (!shlib || flag_symbolic))
     resolved_locally = true;

   /* Undefined weak symbols are never defined locally.  */
@@ -6877,13 +6904,15 @@ default_binds_local_p_3 (const_tree exp, bool
shlib, bool weak_dominate,
   if (DECL_VISIBILITY (exp) != VISIBILITY_DEFAULT
       && (TREE_CODE (exp) == FUNCTION_DECL
          || !extern_protected_data
+         || flag_symbolic
          || DECL_VISIBILITY (exp) != VISIBILITY_PROTECTED)
       && (DECL_VISIBILITY_SPECIFIED (exp) || defined_locally))
     return true;

   /* If PIC, then assume that any global name can be overridden by
-     symbols resolved from other modules.  */
-  if (shlib)
+     symbols resolved from other modules unless global references are
+     bound locally  */
+  if (shlib && !flag_symbolic)
     return false;

   /* Variables defined outside this object might not be local.  */
@@ -6910,7 +6939,16 @@ default_binds_local_p_3 (const_tree exp, bool
shlib, bool weak_dominate,
 bool
 default_binds_local_p (const_tree exp)
 {
-  return default_binds_local_p_3 (exp, flag_shlib != 0, true, false, false);
+  return default_binds_local_p_3 (exp, true, flag_shlib != 0, true,
+                                 false, false, false);
+}
+
+/* Similar to default_binds_local_p, but for TLS variable.  */
+
+bool
+default_binds_tls_local_p (const_tree exp)
+{
+  return targetm.binds_local_p (exp);
 }

 /* Similar to default_binds_local_p, but common symbol may be local.  */
@@ -6918,14 +6956,15 @@ default_binds_local_p (const_tree exp)
 bool
 default_binds_local_p_2 (const_tree exp)
 {
-  return default_binds_local_p_3 (exp, flag_shlib != 0, true, false,
-                                 !flag_pic);
+  return default_binds_local_p_3 (exp, true, flag_shlib != 0, true,
+                                 false, !flag_pic, false);
 }

 bool
 default_binds_local_p_1 (const_tree exp, int shlib)
 {
-  return default_binds_local_p_3 (exp, shlib != 0, false, false, false);
+  return default_binds_local_p_3 (exp, true, shlib != 0, false, false,
+                                 false, false);
 }

 /* Return true when references to DECL must bind to current definition in
--
2.4.3


PING: [PATCH] X86: Optimize access to globals in PIE with copy reloc

Commit Message

Patch