Limit memory size to available memory

Message ID 20210630133313.908246-1-hjl.tools@gmail.com
State New
Headers show
Series
  • Limit memory size to available memory
Related show

Commit Message

H.J. Lu via Binutils June 30, 2021, 1:33 p.m.
When link_info.keep_memory is true, linker caches the relocation
information and symbol tables of input files in memory.  When there
are many input files with many relocations, we may run out of memory.
This patch limits the allocated memory size to available memory or half
of the address space if available memory is unknown.

bfd/

	PR ld/18028
	* bfd.c (bfd): Add alloc_size.
	* elf-bfd.h (_bfd_elf_link_info_read_relocs): New.
	* elf32-i386.c (elf_i386_check_relocs): Use _bfd_link_keep_memory.
	Update cache_size.
	* elf64-x86-64.c (elf_x86_64_check_relocs): Likewise.
	* elflink.c (_bfd_elf_link_read_relocs): Renamed to ...
	(_bfd_elf_link_info_read_relocs): This.  Update cache_size.
	(_bfd_elf_link_read_relocs): New.
	(_bfd_elf_link_check_relocs): Call _bfd_elf_link_info_read_relocs
	instead of _bfd_elf_link_read_relocs.
	(elf_link_add_object_symbols): Likewise.
	(elf_link_input_bfd): Likewise.
	(init_reloc_cookie_rels): Likewise.
	(init_reloc_cookie): Update cache_size.  Call
	_bfd_elf_link_info_read_relocs instead of
	_bfd_elf_link_read_relocs.
	(link_info_ok): New.
	(elf_gc_smash_unused_vtentry_relocs): Updated.  Call
	_bfd_elf_link_info_read_relocs instead of
	_bfd_elf_link_read_relocs.
	(bfd_elf_gc_sections): Use link_info_ok.  Pass &link_info_ok
	to elf_gc_smash_unused_vtentry_relocs.
	* libbfd-in.h (_bfd_link_keep_memory): New.
	* linker.c (_bfd_link_keep_memory): New.
	* opncls.c (bfd_alloc): Update alloc_size.
	* bfd-in2.h: Regenerated.
	* libbfd.h: Likewise.

include/

	PR ld/18028
	* bfdlink.h (bfd_link_info): Add cache_size and max_alloc_size.

ld/

	PR ld/18028
	* configure.ac: Check <sys/sysinfo.h> and get_phys_pages.
	* ldmain.c: Include <sys/sysinfo.h> if available.
	(main): Set link_info.max_alloc_size to available memory or half
	of the address space if available memory is unknown.
	* config.in: Regenerated.
	* configure: Likewise.
---
 bfd/bfd-in2.h      |  3 ++
 bfd/bfd.c          |  3 ++
 bfd/elf-bfd.h      |  3 ++
 bfd/elf32-i386.c   |  3 +-
 bfd/elf64-x86-64.c |  3 +-
 bfd/elflink.c      | 90 ++++++++++++++++++++++++++++++++++------------
 bfd/libbfd-in.h    |  3 ++
 bfd/libbfd.h       |  3 ++
 bfd/linker.c       | 33 +++++++++++++++++
 bfd/opncls.c       |  2 ++
 include/bfdlink.h  |  7 ++++
 ld/config.in       |  6 ++++
 ld/configure       |  5 +--
 ld/configure.ac    |  5 +--
 ld/ldmain.c        | 13 +++++++
 15 files changed, 153 insertions(+), 29 deletions(-)

-- 
2.31.1

Patch

diff --git a/bfd/bfd-in2.h b/bfd/bfd-in2.h
index 57b3c453649..1c5ca22c50e 100644
--- a/bfd/bfd-in2.h
+++ b/bfd/bfd-in2.h
@@ -6711,6 +6711,9 @@  struct bfd
      be used only for archive elements.  */
   int archive_pass;
 
+  /* The total size of memory from bfd_alloc.  */
+  bfd_size_type alloc_size;
+
   /* Stuff only useful for object files:
      The start address.  */
   bfd_vma start_address;
diff --git a/bfd/bfd.c b/bfd/bfd.c
index 6e5d3397d15..a956866639f 100644
--- a/bfd/bfd.c
+++ b/bfd/bfd.c
@@ -280,6 +280,9 @@  CODE_FRAGMENT
 .     be used only for archive elements.  *}
 .  int archive_pass;
 .
+.  {* The total size of memory from bfd_alloc.  *}
+.  bfd_size_type alloc_size;
+.
 .  {* Stuff only useful for object files:
 .     The start address.  *}
 .  bfd_vma start_address;
diff --git a/bfd/elf-bfd.h b/bfd/elf-bfd.h
index 8f985ab8917..65c08ca9d4a 100644
--- a/bfd/elf-bfd.h
+++ b/bfd/elf-bfd.h
@@ -2482,6 +2482,9 @@  extern char *_bfd_elfcore_strndup
 
 extern Elf_Internal_Rela *_bfd_elf_link_read_relocs
   (bfd *, asection *, void *, Elf_Internal_Rela *, bool);
+extern Elf_Internal_Rela *_bfd_elf_link_info_read_relocs
+  (bfd *, struct bfd_link_info *, asection *, void *, Elf_Internal_Rela *,
+   bool);
 
 extern bool _bfd_elf_link_output_relocs
   (bfd *, asection *, Elf_Internal_Shdr *, Elf_Internal_Rela *,
diff --git a/bfd/elf32-i386.c b/bfd/elf32-i386.c
index cf7cd076b17..1898ba3e33a 100644
--- a/bfd/elf32-i386.c
+++ b/bfd/elf32-i386.c
@@ -1915,13 +1915,14 @@  elf_i386_check_relocs (bfd *abfd,
 
   if (elf_section_data (sec)->this_hdr.contents != contents)
     {
-      if (!converted && !info->keep_memory)
+      if (!converted && !_bfd_link_keep_memory (info))
 	free (contents);
       else
 	{
 	  /* Cache the section contents for elf_link_input_bfd if any
 	     load is converted or --no-keep-memory isn't used.  */
 	  elf_section_data (sec)->this_hdr.contents = contents;
+	  info->cache_size += sec->size;
 	}
     }
 
diff --git a/bfd/elf64-x86-64.c b/bfd/elf64-x86-64.c
index 98fb88113c0..4c55c5999c3 100644
--- a/bfd/elf64-x86-64.c
+++ b/bfd/elf64-x86-64.c
@@ -2365,13 +2365,14 @@  elf_x86_64_check_relocs (bfd *abfd, struct bfd_link_info *info,
 
   if (elf_section_data (sec)->this_hdr.contents != contents)
     {
-      if (!converted && !info->keep_memory)
+      if (!converted && !_bfd_link_keep_memory (info))
 	free (contents);
       else
 	{
 	  /* Cache the section contents for elf_link_input_bfd if any
 	     load is converted or --no-keep-memory isn't used.  */
 	  elf_section_data (sec)->this_hdr.contents = contents;
+	  info->cache_size += sec->size;
 	}
     }
 
diff --git a/bfd/elflink.c b/bfd/elflink.c
index 9a05208253c..003c9546b03 100644
--- a/bfd/elflink.c
+++ b/bfd/elflink.c
@@ -2619,14 +2619,16 @@  elf_link_read_relocs_from_section (bfd *abfd,
    according to the KEEP_MEMORY argument.  If O has two relocation
    sections (both REL and RELA relocations), then the REL_HDR
    relocations will appear first in INTERNAL_RELOCS, followed by the
-   RELA_HDR relocations.  */
+   RELA_HDR relocations.  If INFO isn't NULL and KEEP_MEMORY is true,
+   update cache_size.  */
 
 Elf_Internal_Rela *
-_bfd_elf_link_read_relocs (bfd *abfd,
-			   asection *o,
-			   void *external_relocs,
-			   Elf_Internal_Rela *internal_relocs,
-			   bool keep_memory)
+_bfd_elf_link_info_read_relocs (bfd *abfd,
+				struct bfd_link_info *info,
+				asection *o,
+				void *external_relocs,
+				Elf_Internal_Rela *internal_relocs,
+				bool keep_memory)
 {
   void *alloc1 = NULL;
   Elf_Internal_Rela *alloc2 = NULL;
@@ -2646,7 +2648,11 @@  _bfd_elf_link_read_relocs (bfd *abfd,
 
       size = (bfd_size_type) o->reloc_count * sizeof (Elf_Internal_Rela);
       if (keep_memory)
-	internal_relocs = alloc2 = (Elf_Internal_Rela *) bfd_alloc (abfd, size);
+	{
+	  internal_relocs = alloc2 = (Elf_Internal_Rela *) bfd_alloc (abfd, size);
+	  if (info)
+	    info->cache_size += size;
+	}
       else
 	internal_relocs = alloc2 = (Elf_Internal_Rela *) bfd_malloc (size);
       if (internal_relocs == NULL)
@@ -2710,6 +2716,22 @@  _bfd_elf_link_read_relocs (bfd *abfd,
   return NULL;
 }
 
+/* This is similar to _bfd_elf_link_info_read_relocs, except for that
+   NULL is passed to _bfd_elf_link_info_read_relocs for pointer to
+   struct bfd_link_info.  */
+
+Elf_Internal_Rela *
+_bfd_elf_link_read_relocs (bfd *abfd,
+			   asection *o,
+			   void *external_relocs,
+			   Elf_Internal_Rela *internal_relocs,
+			   bool keep_memory)
+{
+  return _bfd_elf_link_info_read_relocs (abfd, NULL, o, external_relocs,
+					 internal_relocs, keep_memory);
+
+}
+
 /* Compute the size of, and allocate space for, REL_HDR which is the
    section header for a section containing relocations for O.  */
 
@@ -4026,8 +4048,10 @@  _bfd_elf_link_check_relocs (bfd *abfd, struct bfd_link_info *info)
 	      || bfd_is_abs_section (o->output_section))
 	    continue;
 
-	  internal_relocs = _bfd_elf_link_read_relocs (abfd, o, NULL, NULL,
-						       info->keep_memory);
+	  internal_relocs = _bfd_elf_link_info_read_relocs (abfd, info,
+							    o, NULL,
+							    NULL,
+							    _bfd_link_keep_memory (info));
 	  if (internal_relocs == NULL)
 	    return false;
 
@@ -5356,9 +5380,10 @@  elf_link_add_object_symbols (bfd *abfd, struct bfd_link_info *info)
 		  && (s->flags & SEC_DEBUGGING) != 0))
 	    continue;
 
-	  internal_relocs = _bfd_elf_link_read_relocs (abfd, s, NULL,
-						       NULL,
-						       info->keep_memory);
+	  internal_relocs = _bfd_elf_link_info_read_relocs (abfd, info,
+							    s, NULL,
+							    NULL,
+							    _bfd_link_keep_memory (info));
 	  if (internal_relocs == NULL)
 	    goto error_free_vers;
 
@@ -11186,8 +11211,10 @@  elf_link_input_bfd (struct elf_final_link_info *flinfo, bfd *input_bfd)
 
 	  /* Get the swapped relocs.  */
 	  internal_relocs
-	    = _bfd_elf_link_read_relocs (input_bfd, o, flinfo->external_relocs,
-					 flinfo->internal_relocs, false);
+	    = _bfd_elf_link_info_read_relocs (input_bfd, flinfo->info, o,
+					      flinfo->external_relocs,
+					      flinfo->internal_relocs,
+					      false);
 	  if (internal_relocs == NULL
 	      && o->reloc_count > 0)
 	    return false;
@@ -13279,8 +13306,12 @@  init_reloc_cookie (struct elf_reloc_cookie *cookie,
 	  info->callbacks->einfo (_("%P%X: can not read symbols: %E\n"));
 	  return false;
 	}
-      if (info->keep_memory)
-	symtab_hdr->contents = (bfd_byte *) cookie->locsyms;
+      if (_bfd_link_keep_memory (info) )
+	{
+	  symtab_hdr->contents = (bfd_byte *) cookie->locsyms;
+	  info->cache_size += (cookie->locsymcount
+			       * sizeof (Elf_External_Sym_Shndx));
+	}
     }
   return true;
 }
@@ -13312,8 +13343,9 @@  init_reloc_cookie_rels (struct elf_reloc_cookie *cookie,
     }
   else
     {
-      cookie->rels = _bfd_elf_link_read_relocs (abfd, sec, NULL, NULL,
-						info->keep_memory);
+      cookie->rels = _bfd_elf_link_info_read_relocs (abfd, info, sec,
+						     NULL, NULL,
+						     _bfd_link_keep_memory (info));
       if (cookie->rels == NULL)
 	return false;
       cookie->rel = cookie->rels;
@@ -13881,14 +13913,22 @@  elf_gc_propagate_vtable_entries_used (struct elf_link_hash_entry *h, void *okp)
   return true;
 }
 
+struct link_info_ok
+{
+  struct bfd_link_info *info;
+  bool ok;
+};
+
 static bool
-elf_gc_smash_unused_vtentry_relocs (struct elf_link_hash_entry *h, void *okp)
+elf_gc_smash_unused_vtentry_relocs (struct elf_link_hash_entry *h,
+				    void *ptr)
 {
   asection *sec;
   bfd_vma hstart, hend;
   Elf_Internal_Rela *relstart, *relend, *rel;
   const struct elf_backend_data *bed;
   unsigned int log_file_align;
+  struct link_info_ok *info = (struct link_info_ok *) ptr;
 
   /* Take care of both those symbols that do not describe vtables as
      well as those that are not loaded.  */
@@ -13904,9 +13944,10 @@  elf_gc_smash_unused_vtentry_relocs (struct elf_link_hash_entry *h, void *okp)
   hstart = h->root.u.def.value;
   hend = hstart + h->size;
 
-  relstart = _bfd_elf_link_read_relocs (sec->owner, sec, NULL, NULL, true);
+  relstart = _bfd_elf_link_info_read_relocs (sec->owner, info->info,
+					     sec, NULL, NULL, true);
   if (!relstart)
-    return *(bool *) okp = false;
+    return info->ok = false;
   bed = get_elf_backend_data (sec->owner);
   log_file_align = bed->s->log_file_align;
 
@@ -14029,6 +14070,7 @@  bfd_elf_gc_sections (bfd *abfd, struct bfd_link_info *info)
   elf_gc_mark_hook_fn gc_mark_hook;
   const struct elf_backend_data *bed = get_elf_backend_data (abfd);
   struct elf_link_hash_table *htab;
+  struct link_info_ok info_ok;
 
   if (!bed->can_gc_sections
       || !is_elf_hash_table (info->hash))
@@ -14070,8 +14112,10 @@  bfd_elf_gc_sections (bfd *abfd, struct bfd_link_info *info)
     return false;
 
   /* Kill the vtable relocations that were not used.  */
-  elf_link_hash_traverse (htab, elf_gc_smash_unused_vtentry_relocs, &ok);
-  if (!ok)
+  info_ok.info = info;
+  info_ok.ok = true;
+  elf_link_hash_traverse (htab, elf_gc_smash_unused_vtentry_relocs, &info_ok);
+  if (!info_ok.ok)
     return false;
 
   /* Mark dynamically referenced symbols.  */
diff --git a/bfd/libbfd-in.h b/bfd/libbfd-in.h
index 1f7e22186ec..89d2997aedf 100644
--- a/bfd/libbfd-in.h
+++ b/bfd/libbfd-in.h
@@ -894,6 +894,9 @@  extern bfd_byte * _bfd_write_unsigned_leb128
 
 extern struct bfd_link_info *_bfd_get_link_info (bfd *);
 
+extern bool _bfd_link_keep_memory (struct bfd_link_info *)
+  ATTRIBUTE_HIDDEN;
+
 #if GCC_VERSION >= 7000
 #define _bfd_mul_overflow(a, b, res) __builtin_mul_overflow (a, b, res)
 #else
diff --git a/bfd/libbfd.h b/bfd/libbfd.h
index c37ddc03cfd..69496411622 100644
--- a/bfd/libbfd.h
+++ b/bfd/libbfd.h
@@ -899,6 +899,9 @@  extern bfd_byte * _bfd_write_unsigned_leb128
 
 extern struct bfd_link_info *_bfd_get_link_info (bfd *);
 
+extern bool _bfd_link_keep_memory (struct bfd_link_info *)
+  ATTRIBUTE_HIDDEN;
+
 #if GCC_VERSION >= 7000
 #define _bfd_mul_overflow(a, b, res) __builtin_mul_overflow (a, b, res)
 #else
diff --git a/bfd/linker.c b/bfd/linker.c
index c350cd3fd15..eeb21555145 100644
--- a/bfd/linker.c
+++ b/bfd/linker.c
@@ -3535,3 +3535,36 @@  _bfd_nolink_bfd_define_start_stop (struct bfd_link_info *info ATTRIBUTE_UNUSED,
 {
   return (struct bfd_link_hash_entry *) _bfd_ptr_bfd_null_error (sec->owner);
 }
+
+bool
+_bfd_link_keep_memory (struct bfd_link_info * info)
+{
+  bfd *abfd;
+  bfd_size_type size;
+
+  if (!info->keep_memory)
+    return false;
+
+  /* Keep allocated memory size below limit only for 32-bit hosts.  */
+  if (sizeof (void *) > 4)
+    return true;
+
+  abfd = info->input_bfds;
+  size = info->cache_size;
+  do
+    {
+      if (size >= info->max_alloc_size)
+	{
+	  /* Over the limit.  Reduce the memory usage.  */
+	  info->keep_memory = false;
+	  return false;
+	}
+      if (!abfd)
+	break;
+      size += abfd->alloc_size;
+      abfd = abfd->link.next;
+    }
+  while (1);
+
+  return true;
+}
diff --git a/bfd/opncls.c b/bfd/opncls.c
index 74df4c2f518..30c5b8db5f7 100644
--- a/bfd/opncls.c
+++ b/bfd/opncls.c
@@ -1030,6 +1030,8 @@  bfd_alloc (bfd *abfd, bfd_size_type size)
   ret = objalloc_alloc ((struct objalloc *) abfd->memory, ul_size);
   if (ret == NULL)
     bfd_set_error (bfd_error_no_memory);
+  else
+    abfd->alloc_size += size;
   return ret;
 }
 
diff --git a/include/bfdlink.h b/include/bfdlink.h
index 7f1b12dbf37..32242a7e2aa 100644
--- a/include/bfdlink.h
+++ b/include/bfdlink.h
@@ -682,6 +682,13 @@  struct bfd_link_info
 
   /* The version information.  */
   struct bfd_elf_version_tree *version_info;
+
+  /* Size of cache.  Backend can use it to keep strace cache size.   */
+  bfd_size_type cache_size;
+
+  /* The maximum size of allocated memory.  Backend can use cache_size
+     and and max_alloc_size to decide if keep_memory should be honored.  */
+  bfd_size_type max_alloc_size;
 };
 
 /* Some forward-definitions used by some callbacks.  */
diff --git a/ld/config.in b/ld/config.in
index 26d55a00d47..22b71d039fa 100644
--- a/ld/config.in
+++ b/ld/config.in
@@ -79,6 +79,9 @@ 
 /* Define to 1 if you have the `getpagesize' function. */
 #undef HAVE_GETPAGESIZE
 
+/* Define to 1 if you have the `get_phys_pages' function. */
+#undef HAVE_GET_PHYS_PAGES
+
 /* Define to 1 if you have the `glob' function. */
 #undef HAVE_GLOB
 
@@ -139,6 +142,9 @@ 
 /* Define to 1 if you have the <sys/stat.h> header file. */
 #undef HAVE_SYS_STAT_H
 
+/* Define to 1 if you have the <sys/sysinfo.h> header file. */
+#undef HAVE_SYS_SYSINFO_H
+
 /* Define to 1 if you have the <sys/time.h> header file. */
 #undef HAVE_SYS_TIME_H
 
diff --git a/ld/configure b/ld/configure
index d8f8b2de18e..7c24cfe8e3f 100755
--- a/ld/configure
+++ b/ld/configure
@@ -16274,7 +16274,7 @@  do_compare="$gcc_cv_prog_cmp_skip"
 # Besides those, we need to check anything used in ld/ not in C99.
 for ac_header in fcntl.h elf-hints.h limits.h inttypes.h stdint.h \
 		 sys/file.h sys/mman.h sys/param.h sys/stat.h sys/time.h \
-		 sys/types.h unistd.h
+		 sys/types.h unistd.h sys/sysinfo.h
 do :
   as_ac_Header=`$as_echo "ac_cv_header_$ac_header" | $as_tr_sh`
 ac_fn_c_check_header_mongrel "$LINENO" "$ac_header" "$as_ac_Header" "$ac_includes_default"
@@ -16287,7 +16287,8 @@  fi
 
 done
 
-for ac_func in close glob lseek mkstemp open realpath sbrk waitpid
+for ac_func in close glob lseek mkstemp open realpath sbrk waitpid \
+	       get_phys_pages
 do :
   as_ac_var=`$as_echo "ac_cv_func_$ac_func" | $as_tr_sh`
 ac_fn_c_check_func "$LINENO" "$ac_func" "$as_ac_var"
diff --git a/ld/configure.ac b/ld/configure.ac
index 8ea97c43cd4..f7d86669ada 100644
--- a/ld/configure.ac
+++ b/ld/configure.ac
@@ -312,8 +312,9 @@  AC_SUBST(NATIVE_LIB_DIRS)
 # Besides those, we need to check anything used in ld/ not in C99.
 AC_CHECK_HEADERS(fcntl.h elf-hints.h limits.h inttypes.h stdint.h \
 		 sys/file.h sys/mman.h sys/param.h sys/stat.h sys/time.h \
-		 sys/types.h unistd.h)
-AC_CHECK_FUNCS(close glob lseek mkstemp open realpath sbrk waitpid)
+		 sys/types.h unistd.h sys/sysinfo.h)
+AC_CHECK_FUNCS(close glob lseek mkstemp open realpath sbrk waitpid \
+	       get_phys_pages)
 
 BFD_BINARY_FOPEN
 
diff --git a/ld/ldmain.c b/ld/ldmain.c
index 42660eb9a3c..93c29019a44 100644
--- a/ld/ldmain.c
+++ b/ld/ldmain.c
@@ -52,6 +52,11 @@ 
 
 #include <string.h>
 
+#if defined(HAVE_SYS_SYSINFO_H) && defined (HAVE_GET_PHYS_PAGES) \
+    && defined(HAVE_GETPAGESIZE)
+#include <sys/sysinfo.h>
+#endif
+
 #ifndef TARGET_SYSTEM_ROOT
 #define TARGET_SYSTEM_ROOT ""
 #endif
@@ -331,6 +336,14 @@  main (int argc, char **argv)
 
   link_info.allow_undefined_version = true;
   link_info.keep_memory = true;
+#if defined(HAVE_SYS_SYSINFO_H) && defined (HAVE_GET_PHYS_PAGES) \
+    && defined(HAVE_GETPAGESIZE)
+  /* Limit the allocated memory size to available memory.  */
+  link_info.max_alloc_size = get_phys_pages () * getpagesize ();
+#else
+  /* Limit the allocated memory size to half of the address space.  */
+  link_info.max_alloc_size = ((uintptr_t) (void *) -1) / 2;
+#endif
   link_info.combreloc = true;
   link_info.strip_discarded = true;
   link_info.prohibit_multiple_definition_absolute = false;