Skip to contentSkip to conversation

RFC: introduce CONFIG_INIT_ALL_MEMORY

PATCH Superseded Alexander Potapenko 6 participants 12 messages Apr 8, 2019 – Apr 9, 2019

Subsystems (5)

v3 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

v3 cover RFC: introduce CONFIG_INIT_ALL_MEMORY

This patch is a part of a bigger initiative to allow initializing heap/stack memory in the Linux kernels by default.
The rationale behind doing so is to reduce the severity of bugs caused by using uninitialized memory.

Over the last two years KMSAN (https://github.com/google/kmsan/) has found more than a hundred bugs running in a really moderate setup (orders of magnitude less CPU/months than KASAN). Some of those bugs led to information leaks if uninitialized memory was copied to the userspace, other could cause DoS because of subverted control flow.
A lot more bugs remain uncovered, so we want to provide the distros and OS vendors with a last resort measure to mitigate such bugs.

Our plan is to introduce configuration flags to force initialization of stack and heap variables with a fixed pattern.
This is going to render information leaks inefficient (as we'll only leak pattern data) and make uses of uninitialized values in conditions more deterministic and discoverable.

The stack instrumentation part is based on Clang's -ftrivial-auto-var-init (see https://reviews.llvm.org/D54604 ; there's also a GCC feature request for a similar flag: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87210)
or GCC's -fplugin-arg-structleak_plugin-byref-all
The heap initialization part is compiler-agnostic and is based on the existing CONFIG_SLUB_DEBUG and CONFIG_PAGE_POISONING.

Alexander Potapenko (2):
initmem: introduce CONFIG_INIT_ALL_MEMORY and CONFIG_INIT_ALL_STACK
initmem: introduce CONFIG_INIT_ALL_HEAP

 Makefile                 |  3 ++-
 mm/page_poison.c         |  5 +++++
 mm/slub.c                |  2 ++
 scripts/Makefile.initmem | 10 ++++++++++
 security/Kconfig         |  1 +
 security/Kconfig.initmem | 40 ++++++++++++++++++++++++++++++++++++++++
 6 files changed, 60 insertions(+), 1 deletion(-)
 create mode 100644 scripts/Makefile.initmem
 create mode 100644 security/Kconfig.initmem
Signature · 1 line
2.21.0.392.gf8f6787159e-goog
#1 lore raw

PATCH v3 1/2 initmem: introduce CONFIG_INIT_ALL_MEMORY and CONFIG_INIT_ALL_STACK

CONFIG_INIT_ALL_MEMORY is going to be an umbrella config for options that force heap and stack initialization.
The rationale behind doing so is to reduce the severity of bugs caused by using uninitialized memory.

CONFIG_INIT_ALL_STACK turns on stack initialization based on -ftrivial-auto-var-init in Clang builds and on -fplugin-arg-structleak_plugin-byref-all in GCC builds.

-ftrivial-auto-var-init is a Clang flag that provides trivial initializers for uninitialized local variables, variable fields and padding.

It has three possible values:
pattern - uninitialized locals are filled with a fixed pattern

    (mostly 0xAA on 64-bit platforms, see https://reviews.llvm.org/D54604
    for more details) likely to cause crashes when uninitialized value is
    used;
  zero (it's still debated whether this flag makes it to the official
    Clang release) - uninitialized locals are filled with zeroes;

uninitialized (default) - uninitialized locals are left intact.

The proposed config builds the kernel with
-ftrivial-auto-var-init=pattern.

Developers have the possibility to opt-out of this feature on a per-variable basis by using __attribute__((uninitialized)).

For GCC builds, CONFIG_INIT_ALL_STACK is simply wired up to CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. No opt-out is possible at the moment.

Cc Masahiro Yamada <[email protected]>
Cc James Morris <[email protected]>
Cc "Serge E. Hallyn" <[email protected]>
Cc Nick Desaulniers <[email protected]>
Cc Kostya Serebryany <[email protected]>
Cc Dmitry Vyukov <[email protected]>
Cc Kees Cook <[email protected]>
Cc Sandeep Patil <[email protected]>

---
v2:
- addressed Kees Cook's comments: added GCC support
v3: addressed Masahiro Yamada's comments:
- dropped per-file opt-out mechanism
- fixed GCC_PLUGINS dependencies

---
 Makefile                 |  3 ++-
 scripts/Makefile.initmem | 10 ++++++++++
 security/Kconfig         |  1 +
 security/Kconfig.initmem | 29 +++++++++++++++++++++++++++++
 4 files changed, 42 insertions(+), 1 deletion(-)
 create mode 100644 scripts/Makefile.initmem
 create mode 100644 security/Kconfig.initmem
Makefile +2 -1
index f070e0d65186..028ca37878fd 100644
@@ -448,7 +448,7 @@ export HOSTCXX KBUILD_HOSTCXXFLAGS LDFLAGS_MODULE CHECK CHECKFLAGS
 
 export KBUILD_CPPFLAGS NOSTDINC_FLAGS LINUXINCLUDE OBJCOPYFLAGS KBUILD_LDFLAGS
 export KBUILD_CFLAGS CFLAGS_KERNEL CFLAGS_MODULE
-export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE CFLAGS_UBSAN
+export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE CFLAGS_UBSAN CFLAGS_INITMEM
 export KBUILD_AFLAGS AFLAGS_KERNEL AFLAGS_MODULE
 export KBUILD_AFLAGS_MODULE KBUILD_CFLAGS_MODULE KBUILD_LDFLAGS_MODULE
 export KBUILD_AFLAGS_KERNEL KBUILD_CFLAGS_KERNEL
@@ -840,6 +840,7 @@ KBUILD_ARFLAGS := $(call ar-option,D)
 include scripts/Makefile.kasan
 include scripts/Makefile.extrawarn
 include scripts/Makefile.ubsan
+include scripts/Makefile.initmem
 
 # Add any arch overrides and user supplied CPPFLAGS, AFLAGS and CFLAGS as the
 # last assignments
scripts/Makefile.initmem new file +10 -0
new file mode 100644
index 000000000000..a6253d78fe35
@@ -0,0 +1,10 @@
+ifdef CONFIG_INIT_ALL_STACK
+
+# Clang's -ftrivial-auto-var-init=pattern flag initializes the
+# uninitialized parts of local variables (including fields and padding)
+# with a fixed pattern (0xAA in most cases).
+ifdef CONFIG_CC_HAS_AUTO_VAR_INIT
+  CFLAGS_INITMEM := -ftrivial-auto-var-init=pattern
+endif
+
+endif
security/Kconfig +1 -0
index e4fe2f3c2c65..cc12a39424dd 100644
@@ -230,6 +230,7 @@ config STATIC_USERMODEHELPER_PATH
 	  If you wish for all usermode helper programs to be disabled,
 	  specify an empty string here (i.e. "").
 
+source "security/Kconfig.initmem"
 source "security/selinux/Kconfig"
 source "security/smack/Kconfig"
 source "security/tomoyo/Kconfig"
security/Kconfig.initmem new file +29 -0
new file mode 100644
index 000000000000..5e49a55382ad
@@ -0,0 +1,29 @@
+menu "Initialize all memory"
+
+config CC_HAS_AUTO_VAR_INIT
+	def_bool $(cc-option,-ftrivial-auto-var-init=pattern)
+
+config INIT_ALL_MEMORY
+	bool "Initialize all memory"
+	default n
+	help
+	  Enforce memory initialization to mitigate infoleaks and make
+	  the control-flow bugs depending on uninitialized values more
+	  deterministic.
+
+if INIT_ALL_MEMORY
+
+config INIT_ALL_STACK
+	bool "Initialize all stack"
+	depends on INIT_ALL_MEMORY
+	depends on CC_HAS_AUTO_VAR_INIT || (HAVE_GCC_PLUGINS && PLUGIN_HOSTCC != "")
+	select GCC_PLUGINS if !CC_HAS_AUTO_VAR_INIT
+	select GCC_PLUGIN_STRUCTLEAK if !CC_HAS_AUTO_VAR_INIT
+	select GCC_PLUGIN_STRUCTLEAK_BYREF_ALL if !CC_HAS_AUTO_VAR_INIT
+	default y
+	help
+	  Initialize uninitialized stack data with a fixed pattern
+	  (0x00 in GCC, 0xAA in Clang).
+
+endif # INIT_ALL_MEMORY
+endmenu
Signature · 1 line
2.21.0.392.gf8f6787159e-goog
#8 lore raw

PATCH v3 2/2 initmem: introduce CONFIG_INIT_ALL_HEAP

In reply to Alexander Potapenko

This config option enables CONFIG_SLUB_DEBUG and CONFIG_PAGE_POISONING without the need to pass any boot parameters.

No performance optimizations are done at the moment to reduce double initialization of memory regions.

Cc Masahiro Yamada <[email protected]>
Cc James Morris <[email protected]>
Cc "Serge E. Hallyn" <[email protected]>
Cc Nick Desaulniers <[email protected]>
Cc Kostya Serebryany <[email protected]>
Cc Dmitry Vyukov <[email protected]>
Cc Kees Cook <[email protected]>
Cc Sandeep Patil <[email protected]>

---
v3:
- addressed comments by Masahiro Yamada (Kconfig fixes)

---
 mm/page_poison.c         |  5 +++++
 mm/slub.c                |  2 ++
 security/Kconfig.initmem | 11 +++++++++++
 3 files changed, 18 insertions(+)
mm/page_poison.c +5 -0
index 21d4f97cb49b..a1985f33f635 100644
@@ -12,9 +12,14 @@ static bool want_page_poisoning __read_mostly;
 
 static int __init early_page_poison_param(char *buf)
 {
+#ifdef CONFIG_INIT_ALL_HEAP
+	want_page_poisoning = true;
+	return 0;
+#else
 	if (!buf)
 		return -EINVAL;
 	return strtobool(buf, &want_page_poisoning);
+#endif
 }
 early_param("page_poison", early_page_poison_param);
 
mm/slub.c +2 -0
index 1b08fbcb7e61..00e0197d3f35 100644
@@ -1287,6 +1287,8 @@ static int __init setup_slub_debug(char *str)
 	if (*str == ',')
 		slub_debug_slabs = str + 1;
 out:
+	if (IS_ENABLED(CONFIG_INIT_ALL_HEAP))
+		slub_debug |= SLAB_POISON;
 	return 1;
 }
 
security/Kconfig.initmem +11 -0
index 5e49a55382ad..37cc10a2eeb5 100644
@@ -13,6 +13,17 @@ config INIT_ALL_MEMORY
 
 if INIT_ALL_MEMORY
 
+config INIT_ALL_HEAP
+	bool "Initialize all heap"
+	depends on INIT_ALL_MEMORY
+	select PAGE_POISONING
+	select PAGE_POISONING_NO_SANITY
+	select PAGE_POISONING_ZERO
+	select SLUB_DEBUG if SLUB
+	default y
+	help
+	  Enable page poisoning and SLUB poisoning by default.
+
 config INIT_ALL_STACK
 	bool "Initialize all stack"
 	depends on INIT_ALL_MEMORY
Signature · 1 line
2.21.0.392.gf8f6787159e-goog
#9 lore raw

Re: [PATCH v3 2/2] initmem: introduce CONFIG_INIT_ALL_HEAP

Quoted 5 lines · Alexander Potapenko

This config option enables CONFIG_SLUB_DEBUG and CONFIG_PAGE_POISONING without the need to pass any boot parameters.

No performance optimizations are done at the moment to reduce double initialization of memory regions.

[...]

Quoted 24 lines
mm/page_poison.c +5 -0
index 21d4f97cb49b..a1985f33f635 100644
@@ -12,9 +12,14 @@ static bool want_page_poisoning __read_mostly;
 
 static int __init early_page_poison_param(char *buf)
 {
+#ifdef CONFIG_INIT_ALL_HEAP
+       want_page_poisoning = true;
+       return 0;
+#else
        if (!buf)
                return -EINVAL;
        return strtobool(buf, &want_page_poisoning);
+#endif
 }
 early_param("page_poison", early_page_poison_param);
 
mm/slub.c +0 -0
index 1b08fbcb7e61..00e0197d3f35 100644

mm/slub.c static int __init setup_slub_debug(char *str) · line 1293

Quoted 7 lines
mm/slub.c +2 -0
@@ -1287,6 +1287,8 @@ static int __init setup_slub_debug(char *str)
        if (*str == ',')
                slub_debug_slabs = str + 1;
 out:
+       if (IS_ENABLED(CONFIG_INIT_ALL_HEAP))
+               slub_debug |= SLAB_POISON;
        return 1;
 }

I don't understand how this is supposed to work. As far as I can tell, the "slub_debug |= SLAB_POISON;" only happens if you actually pass in a "slub_debug" boot parameter? Same thing for "want_page_poisoning = true;".

Also, didn't Laura suggest in
https://www.openwall.com/lists/kernel-hardening/2019/04/08/4 that a different approach might be more sensible to reduce the performance hit?

#2 lore raw

Re: [PATCH v3 1/2] initmem: introduce CONFIG_INIT_ALL_MEMORY and CONFIG_INIT_ALL_STACK

In reply to Alexander Potapenko
Quoted 14 lines · Alexander Potapenko
security/Kconfig.initmem new file +8 -0
new file mode 100644
index 000000000000..5e49a55382ad
@@ -0,0 +1,29 @@
+menu "Initialize all memory"
+
+config CC_HAS_AUTO_VAR_INIT
+	def_bool $(cc-option,-ftrivial-auto-var-init=pattern)
+
+config INIT_ALL_MEMORY
+	bool "Initialize all memory"
+	default n

security/Kconfig.initmem line 18

Quoted 10 lines
security/Kconfig.initmem new file +10 -0
+	help
+	  Enforce memory initialization to mitigate infoleaks and make
+	  the control-flow bugs depending on uninitialized values more
+	  deterministic.
+
+if INIT_ALL_MEMORY
+
+config INIT_ALL_STACK
+	bool "Initialize all stack"
+	depends on INIT_ALL_MEMORY

This "depends on" is redundant with the "if INIT_ALL_MEMORY" above it.

Similar comment for patch 2/2 and INIT_ALL_HEAP.

Quoted 11 lines
security/Kconfig.initmem new file +11 -0
+	depends on CC_HAS_AUTO_VAR_INIT || (HAVE_GCC_PLUGINS && PLUGIN_HOSTCC != "")
+	select GCC_PLUGINS if !CC_HAS_AUTO_VAR_INIT
+	select GCC_PLUGIN_STRUCTLEAK if !CC_HAS_AUTO_VAR_INIT
+	select GCC_PLUGIN_STRUCTLEAK_BYREF_ALL if !CC_HAS_AUTO_VAR_INIT
+	default y
+	help
+	  Initialize uninitialized stack data with a fixed pattern
+	  (0x00 in GCC, 0xAA in Clang).
+
+endif # INIT_ALL_MEMORY
+endmenu
Signature · 1 line
~Randy
#3 lore raw

Re: [PATCH v3 1/2] initmem: introduce CONFIG_INIT_ALL_MEMORY and CONFIG_INIT_ALL_STACK

Quoted 20 lines · Randy Dunlap
Quoted 18 lines · Alexander Potapenko
security/Kconfig.initmem new file +12 -0
new file mode 100644
index 000000000000..5e49a55382ad
@@ -0,0 +1,29 @@
+menu "Initialize all memory"
+
+config CC_HAS_AUTO_VAR_INIT
+     def_bool $(cc-option,-ftrivial-auto-var-init=pattern)
+
+config INIT_ALL_MEMORY
+     bool "Initialize all memory"
+     default n
+     help
+       Enforce memory initialization to mitigate infoleaks and make
+       the control-flow bugs depending on uninitialized values more
+       deterministic.
Quoted 8 lines

security/Kconfig.initmem line 18

Quoted 6 lines
security/Kconfig.initmem new file +6 -0
+
+if INIT_ALL_MEMORY
+
+config INIT_ALL_STACK
+     bool "Initialize all stack"
+     depends on INIT_ALL_MEMORY

This "depends on" is redundant with the "if INIT_ALL_MEMORY" above it.

Ack, thank you!

Quoted 17 lines

Similar comment for patch 2/2 and INIT_ALL_HEAP.

security/Kconfig.initmem line 29

Quoted 11 lines
security/Kconfig.initmem new file +11 -0
+     depends on CC_HAS_AUTO_VAR_INIT || (HAVE_GCC_PLUGINS && PLUGIN_HOSTCC != "")
+     select GCC_PLUGINS if !CC_HAS_AUTO_VAR_INIT
+     select GCC_PLUGIN_STRUCTLEAK if !CC_HAS_AUTO_VAR_INIT
+     select GCC_PLUGIN_STRUCTLEAK_BYREF_ALL if !CC_HAS_AUTO_VAR_INIT
+     default y
+     help
+       Initialize uninitialized stack data with a fixed pattern
+       (0x00 in GCC, 0xAA in Clang).
+
+endif # INIT_ALL_MEMORY
+endmenu

--
~Randy

Signature · 10 lines
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Paul Manicle, Halimah DeLaine Prado
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg
#4 lore raw

Re: [PATCH v3 1/2] initmem: introduce CONFIG_INIT_ALL_MEMORY and CONFIG_INIT_ALL_STACK

In reply to Alexander Potapenko
Quoted 87 lines · Alexander Potapenko

CONFIG_INIT_ALL_MEMORY is going to be an umbrella config for options that force heap and stack initialization.
The rationale behind doing so is to reduce the severity of bugs caused by using uninitialized memory.

CONFIG_INIT_ALL_STACK turns on stack initialization based on

(quoted diff) +0 -3
-ftrivial-auto-var-init in Clang builds and on
-fplugin-arg-structleak_plugin-byref-all in GCC builds.
 
-ftrivial-auto-var-init is a Clang flag that provides trivial

initializers for uninitialized local variables, variable fields and padding.

It has three possible values:
pattern - uninitialized locals are filled with a fixed pattern

    (mostly 0xAA on 64-bit platforms, see https://reviews.llvm.org/D54604
    for more details) likely to cause crashes when uninitialized value is
    used;
  zero (it's still debated whether this flag makes it to the official
    Clang release) - uninitialized locals are filled with zeroes;

uninitialized (default) - uninitialized locals are left intact.

The proposed config builds the kernel with
-ftrivial-auto-var-init=pattern.

Developers have the possibility to opt-out of this feature on a per-variable basis by using __attribute__((uninitialized)).

For GCC builds, CONFIG_INIT_ALL_STACK is simply wired up to CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. No opt-out is possible at the moment.

Cc Masahiro Yamada <[email protected]>
Cc James Morris <[email protected]>
Cc "Serge E. Hallyn" <[email protected]>
Cc Nick Desaulniers <[email protected]>
Cc Kostya Serebryany <[email protected]>
Cc Dmitry Vyukov <[email protected]>
Cc Kees Cook <[email protected]>
Cc Sandeep Patil <[email protected]>

---
v2:
- addressed Kees Cook's comments: added GCC support
v3: addressed Masahiro Yamada's comments:
- dropped per-file opt-out mechanism
- fixed GCC_PLUGINS dependencies

---
 Makefile                 |  3 ++-
 scripts/Makefile.initmem | 10 ++++++++++
 security/Kconfig         |  1 +
 security/Kconfig.initmem | 29 +++++++++++++++++++++++++++++
 4 files changed, 42 insertions(+), 1 deletion(-)
 create mode 100644 scripts/Makefile.initmem
 create mode 100644 security/Kconfig.initmem
Makefile +2 -1
index f070e0d65186..028ca37878fd 100644
@@ -448,7 +448,7 @@ export HOSTCXX KBUILD_HOSTCXXFLAGS LDFLAGS_MODULE CHECK CHECKFLAGS
 
 export KBUILD_CPPFLAGS NOSTDINC_FLAGS LINUXINCLUDE OBJCOPYFLAGS KBUILD_LDFLAGS
 export KBUILD_CFLAGS CFLAGS_KERNEL CFLAGS_MODULE
-export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE CFLAGS_UBSAN
+export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE CFLAGS_UBSAN CFLAGS_INITMEM
 export KBUILD_AFLAGS AFLAGS_KERNEL AFLAGS_MODULE
 export KBUILD_AFLAGS_MODULE KBUILD_CFLAGS_MODULE KBUILD_LDFLAGS_MODULE
 export KBUILD_AFLAGS_KERNEL KBUILD_CFLAGS_KERNEL
@@ -840,6 +840,7 @@ KBUILD_ARFLAGS := $(call ar-option,D)
 include scripts/Makefile.kasan
 include scripts/Makefile.extrawarn
 include scripts/Makefile.ubsan
+include scripts/Makefile.initmem
 
 # Add any arch overrides and user supplied CPPFLAGS, AFLAGS and CFLAGS as the
 # last assignments
scripts/Makefile.initmem new file +0 -0
new file mode 100644
index 000000000000..a6253d78fe35

scripts/Makefile.initmem line 10

Quoted 10 lines
scripts/Makefile.initmem new file +10 -0
@@ -0,0 +1,10 @@
+ifdef CONFIG_INIT_ALL_STACK
+
+# Clang's -ftrivial-auto-var-init=pattern flag initializes the
+# uninitialized parts of local variables (including fields and padding)
+# with a fixed pattern (0xAA in most cases).
+ifdef CONFIG_CC_HAS_AUTO_VAR_INIT
+  CFLAGS_INITMEM := -ftrivial-auto-var-init=pattern
+endif
+
+endif

Now that you dropped per-file control,
why is this necessary?
Also, CFLAGS_INITMEM is not wired up to anything, so this patch has no effect.

I think you can add the following to top Makefile, and that's it.

ifdef CONFIG_AUTO_VAR_INIT
KBUILD_CFLAGS += -ftrivial-auto-var-init=pattern endif

Quoted 35 lines
security/Kconfig +1 -0
index e4fe2f3c2c65..cc12a39424dd 100644
@@ -230,6 +230,7 @@ config STATIC_USERMODEHELPER_PATH
          If you wish for all usermode helper programs to be disabled,
          specify an empty string here (i.e. "").
 
+source "security/Kconfig.initmem"
 source "security/selinux/Kconfig"
 source "security/smack/Kconfig"
 source "security/tomoyo/Kconfig"
security/Kconfig.initmem new file +17 -0
new file mode 100644
index 000000000000..5e49a55382ad
@@ -0,0 +1,29 @@
+menu "Initialize all memory"
+
+config CC_HAS_AUTO_VAR_INIT
+       def_bool $(cc-option,-ftrivial-auto-var-init=pattern)
+
+config INIT_ALL_MEMORY
+       bool "Initialize all memory"
+       default n
+       help
+         Enforce memory initialization to mitigate infoleaks and make
+         the control-flow bugs depending on uninitialized values more
+         deterministic.
+
+if INIT_ALL_MEMORY
+
+config INIT_ALL_STACK
+       bool "Initialize all stack"

security/Kconfig.initmem line 27

Quoted 10 lines
security/Kconfig.initmem new file +10 -0
+       depends on INIT_ALL_MEMORY
+       depends on CC_HAS_AUTO_VAR_INIT || (HAVE_GCC_PLUGINS && PLUGIN_HOSTCC != "")
+       select GCC_PLUGINS if !CC_HAS_AUTO_VAR_INIT
+       select GCC_PLUGIN_STRUCTLEAK if !CC_HAS_AUTO_VAR_INIT
+       select GCC_PLUGIN_STRUCTLEAK_BYREF_ALL if !CC_HAS_AUTO_VAR_INIT
+       default y
+       help
+         Initialize uninitialized stack data with a fixed pattern
+         (0x00 in GCC, 0xAA in Clang).
+

I am not sure about the benefit of these umbrella CONFIGs since these make 'depends on' and 'select' complicated. I will leave it to Kees, though.

Quoted 5 lines
security/Kconfig.initmem new file +2 -1
+endif # INIT_ALL_MEMORY
+endmenu
--

2.21.0.392.gf8f6787159e-goog

Signature · 2 lines
Best Regards
Masahiro Yamada
#5 lore raw

Re: [PATCH v3 1/2] initmem: introduce CONFIG_INIT_ALL_MEMORY and CONFIG_INIT_ALL_STACK

Quoted 96 lines · Masahiro Yamada
Quoted 94 lines · Alexander Potapenko

CONFIG_INIT_ALL_MEMORY is going to be an umbrella config for options that force heap and stack initialization.
The rationale behind doing so is to reduce the severity of bugs caused by using uninitialized memory.

CONFIG_INIT_ALL_STACK turns on stack initialization based on

(quoted diff) +0 -3
-ftrivial-auto-var-init in Clang builds and on
-fplugin-arg-structleak_plugin-byref-all in GCC builds.
 
-ftrivial-auto-var-init is a Clang flag that provides trivial

initializers for uninitialized local variables, variable fields and padding.

It has three possible values:
pattern - uninitialized locals are filled with a fixed pattern

    (mostly 0xAA on 64-bit platforms, see https://reviews.llvm.org/D54604
    for more details) likely to cause crashes when uninitialized value is
    used;
  zero (it's still debated whether this flag makes it to the official
    Clang release) - uninitialized locals are filled with zeroes;

uninitialized (default) - uninitialized locals are left intact.

The proposed config builds the kernel with
-ftrivial-auto-var-init=pattern.

Developers have the possibility to opt-out of this feature on a per-variable basis by using __attribute__((uninitialized)).

For GCC builds, CONFIG_INIT_ALL_STACK is simply wired up to CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. No opt-out is possible at the moment.

Cc Masahiro Yamada <[email protected]>
Cc James Morris <[email protected]>
Cc "Serge E. Hallyn" <[email protected]>
Cc Nick Desaulniers <[email protected]>
Cc Kostya Serebryany <[email protected]>
Cc Dmitry Vyukov <[email protected]>
Cc Kees Cook <[email protected]>
Cc Sandeep Patil <[email protected]>

---
v2:
- addressed Kees Cook's comments: added GCC support
v3: addressed Masahiro Yamada's comments:
- dropped per-file opt-out mechanism
- fixed GCC_PLUGINS dependencies

---
 Makefile                 |  3 ++-
 scripts/Makefile.initmem | 10 ++++++++++
 security/Kconfig         |  1 +
 security/Kconfig.initmem | 29 +++++++++++++++++++++++++++++
 4 files changed, 42 insertions(+), 1 deletion(-)
 create mode 100644 scripts/Makefile.initmem
 create mode 100644 security/Kconfig.initmem
Makefile +2 -1
index f070e0d65186..028ca37878fd 100644
@@ -448,7 +448,7 @@ export HOSTCXX KBUILD_HOSTCXXFLAGS LDFLAGS_MODULE CHECK CHECKFLAGS
 
 export KBUILD_CPPFLAGS NOSTDINC_FLAGS LINUXINCLUDE OBJCOPYFLAGS KBUILD_LDFLAGS
 export KBUILD_CFLAGS CFLAGS_KERNEL CFLAGS_MODULE
-export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE CFLAGS_UBSAN
+export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE CFLAGS_UBSAN CFLAGS_INITMEM
 export KBUILD_AFLAGS AFLAGS_KERNEL AFLAGS_MODULE
 export KBUILD_AFLAGS_MODULE KBUILD_CFLAGS_MODULE KBUILD_LDFLAGS_MODULE
 export KBUILD_AFLAGS_KERNEL KBUILD_CFLAGS_KERNEL
@@ -840,6 +840,7 @@ KBUILD_ARFLAGS := $(call ar-option,D)
 include scripts/Makefile.kasan
 include scripts/Makefile.extrawarn
 include scripts/Makefile.ubsan
+include scripts/Makefile.initmem
 
 # Add any arch overrides and user supplied CPPFLAGS, AFLAGS and CFLAGS as the
 # last assignments
scripts/Makefile.initmem new file +7 -0
new file mode 100644
index 000000000000..a6253d78fe35
@@ -0,0 +1,10 @@
+ifdef CONFIG_INIT_ALL_STACK
+
+# Clang's -ftrivial-auto-var-init=pattern flag initializes the
+# uninitialized parts of local variables (including fields and padding)
+# with a fixed pattern (0xAA in most cases).
+ifdef CONFIG_CC_HAS_AUTO_VAR_INIT
+  CFLAGS_INITMEM := -ftrivial-auto-var-init=pattern
Quoted 8 lines

scripts/Makefile.initmem line 10

Quoted 3 lines
scripts/Makefile.initmem new file +3 -0
+endif
+
+endif

Now that you dropped per-file control,
why is this necessary?
Also, CFLAGS_INITMEM is not wired up to anything, so this patch has no effect.

There's CFLAGS_INITMEM in the top Makefile, so I suppose it should be working. But I agree there's no point in adding extra files for a single build flag.

Quoted 73 lines

I think you can add the following to top Makefile, and that's it.

ifdef CONFIG_AUTO_VAR_INIT
KBUILD_CFLAGS += -ftrivial-auto-var-init=pattern endif

Quoted 35 lines
security/Kconfig +1 -0
index e4fe2f3c2c65..cc12a39424dd 100644
@@ -230,6 +230,7 @@ config STATIC_USERMODEHELPER_PATH
          If you wish for all usermode helper programs to be disabled,
          specify an empty string here (i.e. "").
 
+source "security/Kconfig.initmem"
 source "security/selinux/Kconfig"
 source "security/smack/Kconfig"
 source "security/tomoyo/Kconfig"
security/Kconfig.initmem new file +17 -0
new file mode 100644
index 000000000000..5e49a55382ad
@@ -0,0 +1,29 @@
+menu "Initialize all memory"
+
+config CC_HAS_AUTO_VAR_INIT
+       def_bool $(cc-option,-ftrivial-auto-var-init=pattern)
+
+config INIT_ALL_MEMORY
+       bool "Initialize all memory"
+       default n
+       help
+         Enforce memory initialization to mitigate infoleaks and make
+         the control-flow bugs depending on uninitialized values more
+         deterministic.
+
+if INIT_ALL_MEMORY
+
+config INIT_ALL_STACK
+       bool "Initialize all stack"

security/Kconfig.initmem line 27

Quoted 10 lines
security/Kconfig.initmem new file +10 -0
+       depends on INIT_ALL_MEMORY
+       depends on CC_HAS_AUTO_VAR_INIT || (HAVE_GCC_PLUGINS && PLUGIN_HOSTCC != "")
+       select GCC_PLUGINS if !CC_HAS_AUTO_VAR_INIT
+       select GCC_PLUGIN_STRUCTLEAK if !CC_HAS_AUTO_VAR_INIT
+       select GCC_PLUGIN_STRUCTLEAK_BYREF_ALL if !CC_HAS_AUTO_VAR_INIT
+       default y
+       help
+         Initialize uninitialized stack data with a fixed pattern
+         (0x00 in GCC, 0xAA in Clang).
+

I am not sure about the benefit of these umbrella CONFIGs since these make 'depends on' and 'select' complicated. I will leave it to Kees, though.

Quoted 5 lines
security/Kconfig.initmem new file +2 -1
+endif # INIT_ALL_MEMORY
+endmenu
--

2.21.0.392.gf8f6787159e-goog

--
Best Regards
Masahiro Yamada

Signature · 10 lines
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Paul Manicle, Halimah DeLaine Prado
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg
#6 lore raw

Re: [PATCH v3 1/2] initmem: introduce CONFIG_INIT_ALL_MEMORY and CONFIG_INIT_ALL_STACK

Quoted 107 lines · Alexander Potapenko
Quoted 96 lines · Masahiro Yamada
Quoted 94 lines · Alexander Potapenko

CONFIG_INIT_ALL_MEMORY is going to be an umbrella config for options that force heap and stack initialization.
The rationale behind doing so is to reduce the severity of bugs caused by using uninitialized memory.

CONFIG_INIT_ALL_STACK turns on stack initialization based on

(quoted diff) +0 -3
-ftrivial-auto-var-init in Clang builds and on
-fplugin-arg-structleak_plugin-byref-all in GCC builds.
 
-ftrivial-auto-var-init is a Clang flag that provides trivial

initializers for uninitialized local variables, variable fields and padding.

It has three possible values:
pattern - uninitialized locals are filled with a fixed pattern

    (mostly 0xAA on 64-bit platforms, see https://reviews.llvm.org/D54604
    for more details) likely to cause crashes when uninitialized value is
    used;
  zero (it's still debated whether this flag makes it to the official
    Clang release) - uninitialized locals are filled with zeroes;

uninitialized (default) - uninitialized locals are left intact.

The proposed config builds the kernel with
-ftrivial-auto-var-init=pattern.

Developers have the possibility to opt-out of this feature on a per-variable basis by using __attribute__((uninitialized)).

For GCC builds, CONFIG_INIT_ALL_STACK is simply wired up to CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. No opt-out is possible at the moment.

Cc Masahiro Yamada <[email protected]>
Cc James Morris <[email protected]>
Cc "Serge E. Hallyn" <[email protected]>
Cc Nick Desaulniers <[email protected]>
Cc Kostya Serebryany <[email protected]>
Cc Dmitry Vyukov <[email protected]>
Cc Kees Cook <[email protected]>
Cc Sandeep Patil <[email protected]>

---
v2:
- addressed Kees Cook's comments: added GCC support
v3: addressed Masahiro Yamada's comments:
- dropped per-file opt-out mechanism
- fixed GCC_PLUGINS dependencies

---
 Makefile                 |  3 ++-
 scripts/Makefile.initmem | 10 ++++++++++
 security/Kconfig         |  1 +
 security/Kconfig.initmem | 29 +++++++++++++++++++++++++++++
 4 files changed, 42 insertions(+), 1 deletion(-)
 create mode 100644 scripts/Makefile.initmem
 create mode 100644 security/Kconfig.initmem
Makefile +2 -1
index f070e0d65186..028ca37878fd 100644
@@ -448,7 +448,7 @@ export HOSTCXX KBUILD_HOSTCXXFLAGS LDFLAGS_MODULE CHECK CHECKFLAGS
 
 export KBUILD_CPPFLAGS NOSTDINC_FLAGS LINUXINCLUDE OBJCOPYFLAGS KBUILD_LDFLAGS
 export KBUILD_CFLAGS CFLAGS_KERNEL CFLAGS_MODULE
-export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE CFLAGS_UBSAN
+export CFLAGS_KASAN CFLAGS_KASAN_NOSANITIZE CFLAGS_UBSAN CFLAGS_INITMEM
 export KBUILD_AFLAGS AFLAGS_KERNEL AFLAGS_MODULE
 export KBUILD_AFLAGS_MODULE KBUILD_CFLAGS_MODULE KBUILD_LDFLAGS_MODULE
 export KBUILD_AFLAGS_KERNEL KBUILD_CFLAGS_KERNEL
@@ -840,6 +840,7 @@ KBUILD_ARFLAGS := $(call ar-option,D)
 include scripts/Makefile.kasan
 include scripts/Makefile.extrawarn
 include scripts/Makefile.ubsan
+include scripts/Makefile.initmem
 
 # Add any arch overrides and user supplied CPPFLAGS, AFLAGS and CFLAGS as the
 # last assignments
scripts/Makefile.initmem new file +7 -0
new file mode 100644
index 000000000000..a6253d78fe35
@@ -0,0 +1,10 @@
+ifdef CONFIG_INIT_ALL_STACK
+
+# Clang's -ftrivial-auto-var-init=pattern flag initializes the
+# uninitialized parts of local variables (including fields and padding)
+# with a fixed pattern (0xAA in most cases).
+ifdef CONFIG_CC_HAS_AUTO_VAR_INIT
+  CFLAGS_INITMEM := -ftrivial-auto-var-init=pattern
Quoted 8 lines

scripts/Makefile.initmem line 10

Quoted 3 lines
scripts/Makefile.initmem new file +3 -0
+endif
+
+endif

Now that you dropped per-file control,
why is this necessary?
Also, CFLAGS_INITMEM is not wired up to anything, so this patch has no effect.

Quoted 1 line

There's CFLAGS_INITMEM in the top Makefile, so I suppose it should be working.

Ah, now I see. Yes, I was too hasty with this patchset.

Quoted 88 lines

But I agree there's no point in adding extra files for a single build flag.

Quoted 63 lines

I think you can add the following to top Makefile, and that's it.

ifdef CONFIG_AUTO_VAR_INIT
KBUILD_CFLAGS += -ftrivial-auto-var-init=pattern endif

Quoted 35 lines
security/Kconfig +1 -0
index e4fe2f3c2c65..cc12a39424dd 100644
@@ -230,6 +230,7 @@ config STATIC_USERMODEHELPER_PATH
          If you wish for all usermode helper programs to be disabled,
          specify an empty string here (i.e. "").
 
+source "security/Kconfig.initmem"
 source "security/selinux/Kconfig"
 source "security/smack/Kconfig"
 source "security/tomoyo/Kconfig"
security/Kconfig.initmem new file +17 -0
new file mode 100644
index 000000000000..5e49a55382ad
@@ -0,0 +1,29 @@
+menu "Initialize all memory"
+
+config CC_HAS_AUTO_VAR_INIT
+       def_bool $(cc-option,-ftrivial-auto-var-init=pattern)
+
+config INIT_ALL_MEMORY
+       bool "Initialize all memory"
+       default n
+       help
+         Enforce memory initialization to mitigate infoleaks and make
+         the control-flow bugs depending on uninitialized values more
+         deterministic.
+
+if INIT_ALL_MEMORY
+
+config INIT_ALL_STACK
+       bool "Initialize all stack"

security/Kconfig.initmem line 27

Quoted 10 lines
security/Kconfig.initmem new file +10 -0
+       depends on INIT_ALL_MEMORY
+       depends on CC_HAS_AUTO_VAR_INIT || (HAVE_GCC_PLUGINS && PLUGIN_HOSTCC != "")
+       select GCC_PLUGINS if !CC_HAS_AUTO_VAR_INIT
+       select GCC_PLUGIN_STRUCTLEAK if !CC_HAS_AUTO_VAR_INIT
+       select GCC_PLUGIN_STRUCTLEAK_BYREF_ALL if !CC_HAS_AUTO_VAR_INIT
+       default y
+       help
+         Initialize uninitialized stack data with a fixed pattern
+         (0x00 in GCC, 0xAA in Clang).
+

I am not sure about the benefit of these umbrella CONFIGs since these make 'depends on' and 'select' complicated. I will leave it to Kees, though.

Quoted 10 lines
Quoted 5 lines
security/Kconfig.initmem new file +2 -1
+endif # INIT_ALL_MEMORY
+endmenu
--

2.21.0.392.gf8f6787159e-goog

--
Best Regards
Masahiro Yamada

--
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Paul Manicle, Halimah DeLaine Prado Registergericht und -nummer: Hamburg, HRB 86891 Sitz der Gesellschaft: Hamburg

Signature · 10 lines
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Paul Manicle, Halimah DeLaine Prado
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg
#10 lore raw

Re: [PATCH v3 2/2] initmem: introduce CONFIG_INIT_ALL_HEAP

In reply to Alexander Potapenko
Quoted 35 lines · Alexander Potapenko

This config option enables CONFIG_SLUB_DEBUG and CONFIG_PAGE_POISONING without the need to pass any boot parameters.

No performance optimizations are done at the moment to reduce double initialization of memory regions.

Cc Masahiro Yamada <[email protected]>
Cc James Morris <[email protected]>
Cc "Serge E. Hallyn" <[email protected]>
Cc Nick Desaulniers <[email protected]>
Cc Kostya Serebryany <[email protected]>
Cc Dmitry Vyukov <[email protected]>
Cc Kees Cook <[email protected]>
Cc Sandeep Patil <[email protected]>

---
v3:
- addressed comments by Masahiro Yamada (Kconfig fixes)

---
 mm/page_poison.c         |  5 +++++
 mm/slub.c                |  2 ++
 security/Kconfig.initmem | 11 +++++++++++
 3 files changed, 18 insertions(+)
mm/page_poison.c +0 -0
index 21d4f97cb49b..a1985f33f635 100644
@@ -12,9 +12,14 @@ static bool want_page_poisoning __read_mostly;
 
 static int __init early_page_poison_param(char *buf)
 {

mm/page_poison.c static bool want_page_poisoning __read_mostly; · line 24

Quoted 10 lines
mm/page_poison.c +5 -0
+#ifdef CONFIG_INIT_ALL_HEAP
+	want_page_poisoning = true;
+	return 0;
+#else
 	if (!buf)
 		return -EINVAL;
 	return strtobool(buf, &want_page_poisoning);
+#endif
 }
 early_param("page_poison", early_page_poison_param);

IIUC this is only called if page_poison is passed on the command line, so want_page_poisoning won't be set automatically unless that's passed.

Presumably you want to initialize it at definition with:

static bool want_page_poisoning __read_mostly = IS_ENABLED(CONFIG_INIT_ALL_HEAP);

with:

#ifndef CONFIG_INIT_ALL_HEAP
static int __init early_page_poison_param(char *buf)

{
	...
}

early_param("page_poison", early_page_poison_param); #endif

... so that it can't be disabled?

Thanks,
Mark.

#11 lore raw

Re: [PATCH v3 2/2] initmem: introduce CONFIG_INIT_ALL_HEAP

Quoted 51 lines · Mark Rutland
Quoted 35 lines · Alexander Potapenko

This config option enables CONFIG_SLUB_DEBUG and CONFIG_PAGE_POISONING without the need to pass any boot parameters.

No performance optimizations are done at the moment to reduce double initialization of memory regions.

Cc Masahiro Yamada <[email protected]>
Cc James Morris <[email protected]>
Cc "Serge E. Hallyn" <[email protected]>
Cc Nick Desaulniers <[email protected]>
Cc Kostya Serebryany <[email protected]>
Cc Dmitry Vyukov <[email protected]>
Cc Kees Cook <[email protected]>
Cc Sandeep Patil <[email protected]>

---
v3:
- addressed comments by Masahiro Yamada (Kconfig fixes)

---
 mm/page_poison.c         |  5 +++++
 mm/slub.c                |  2 ++
 security/Kconfig.initmem | 11 +++++++++++
 3 files changed, 18 insertions(+)
mm/page_poison.c +0 -0
index 21d4f97cb49b..a1985f33f635 100644
@@ -12,9 +12,14 @@ static bool want_page_poisoning __read_mostly;
 
 static int __init early_page_poison_param(char *buf)
 {

mm/page_poison.c static bool want_page_poisoning __read_mostly; · line 24

Quoted 10 lines
mm/page_poison.c +5 -0
+#ifdef CONFIG_INIT_ALL_HEAP
+     want_page_poisoning = true;
+     return 0;
+#else
      if (!buf)
              return -EINVAL;
      return strtobool(buf, &want_page_poisoning);
+#endif
 }
 early_param("page_poison", early_page_poison_param);

IIUC this is only called if page_poison is passed on the command line, so want_page_poisoning won't be set automatically unless that's passed.

Quoted 3 lines

Presumably you want to initialize it at definition with:

static bool want_page_poisoning __read_mostly = IS_ENABLED(CONFIG_INIT_ALL_HEAP);

Yes, I've misunderstood how boot parameters work.
Thanks for the suggestions!
We'd better look into wiping allocations without POISON_SLAB though.

Quoted 14 lines

with:

#ifndef CONFIG_INIT_ALL_HEAP
static int __init early_page_poison_param(char *buf)

{
        ...
}

early_param("page_poison", early_page_poison_param); #endif

... so that it can't be disabled?

Thanks,
Mark.

Signature · 10 lines
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Paul Manicle, Halimah DeLaine Prado
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg
#7 lore raw

Re: [PATCH v3 1/2] initmem: introduce CONFIG_INIT_ALL_MEMORY and CONFIG_INIT_ALL_STACK

In reply to Masahiro Yamada
Quoted 8 lines · Masahiro Yamada
Quoted 8 lines
security/Kconfig.initmem new file +8 -0
+config INIT_ALL_STACK
+       bool "Initialize all stack"
+       depends on INIT_ALL_MEMORY
+       depends on CC_HAS_AUTO_VAR_INIT || (HAVE_GCC_PLUGINS && PLUGIN_HOSTCC != "")
+       select GCC_PLUGINS if !CC_HAS_AUTO_VAR_INIT
+       select GCC_PLUGIN_STRUCTLEAK if !CC_HAS_AUTO_VAR_INIT
+       select GCC_PLUGIN_STRUCTLEAK_BYREF_ALL if !CC_HAS_AUTO_VAR_INIT
+       default y
Quoted 8 lines

security/Kconfig.initmem line 27

Quoted 4 lines
security/Kconfig.initmem new file +4 -0
+       help
+         Initialize uninitialized stack data with a fixed pattern
+         (0x00 in GCC, 0xAA in Clang).
+

I am not sure about the benefit of these umbrella CONFIGs since these make 'depends on' and 'select' complicated. I will leave it to Kees, though.

Yeah, I think this could use some more cleanup to get a single config that represents that if gcc plugins are available (instead of having it split across two configs now). Additionally I think this needs refactoring against the current menu choices for GCC_PLUGIN_STRUCTLEAK. Let me send a proposed patch in a few hours...

Signature · 1 line
Kees Cook