[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20220713213133.455599-1-keescook@chromium.org>
Date: Wed, 13 Jul 2022 14:31:33 -0700
From: Kees Cook <keescook@...omium.org>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Kees Cook <keescook@...omium.org>,
kernel test robot <oliver.sang@...el.com>, x86@...nel.org,
Josh Poimboeuf <jpoimboe@...nel.org>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
Dave Hansen <dave.hansen@...ux.intel.com>,
"H. Peter Anvin" <hpa@...or.com>, Arnd Bergmann <arnd@...db.de>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
linux-kernel@...r.kernel.org, linux-hardening@...r.kernel.org
Subject: [PATCH] x86: Allow for exclusions in checking RETHUNK
LKDTM builds a "just return" function that lives in .rodata, but this
creates problems when validating alternatives in the face of RETHUNK.
Export RETHUNK_CFLAGS so they can be disabled for the LKDTM function,
and ask objtool to ignore this function. (Use of STACK_FRAME_NON_STANDARD
here seems to generate a non-.rela section, that needed to be adjusted.)
Reported-by: kernel test robot <oliver.sang@...el.com>
Link: https://lore.kernel.org/lkml/Ys58BxHxoDZ7rfpr@xsang-OptiPlex-9020/
Debugged-by: Peter Zijlstra <peterz@...radead.org>
Fixes: ee88d363d156 ("x86,static_call: Use alternative RET encoding")
Cc: Peter Zijlstra <peterz@...radead.org>
Cc: x86@...nel.org
Signed-off-by: Kees Cook <keescook@...omium.org>
---
arch/x86/Makefile | 1 +
drivers/misc/lkdtm/Makefile | 2 +-
drivers/misc/lkdtm/rodata.c | 4 ++++
tools/objtool/check.c | 4 +++-
4 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 1f40dad30d50..7854685c5f25 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -27,6 +27,7 @@ RETHUNK_CFLAGS := -mfunction-return=thunk-extern
RETPOLINE_CFLAGS += $(RETHUNK_CFLAGS)
endif
+export RETHUNK_CFLAGS
export RETPOLINE_CFLAGS
export RETPOLINE_VDSO_CFLAGS
diff --git a/drivers/misc/lkdtm/Makefile b/drivers/misc/lkdtm/Makefile
index 2e0aa74ac185..fd96ac1617f7 100644
--- a/drivers/misc/lkdtm/Makefile
+++ b/drivers/misc/lkdtm/Makefile
@@ -16,7 +16,7 @@ lkdtm-$(CONFIG_PPC_64S_HASH_MMU) += powerpc.o
KASAN_SANITIZE_rodata.o := n
KASAN_SANITIZE_stackleak.o := n
KCOV_INSTRUMENT_rodata.o := n
-CFLAGS_REMOVE_rodata.o += $(CC_FLAGS_LTO)
+CFLAGS_REMOVE_rodata.o += $(CC_FLAGS_LTO) $(RETHUNK_CFLAGS)
OBJCOPYFLAGS :=
OBJCOPYFLAGS_rodata_objcopy.o := \
diff --git a/drivers/misc/lkdtm/rodata.c b/drivers/misc/lkdtm/rodata.c
index baacb876d1d9..708a2558a7ac 100644
--- a/drivers/misc/lkdtm/rodata.c
+++ b/drivers/misc/lkdtm/rodata.c
@@ -4,8 +4,12 @@
* (via objcopy tricks), to validate the non-executability of .rodata.
*/
#include "lkdtm.h"
+#include <linux/objtool.h>
void noinstr lkdtm_rodata_do_nothing(void)
{
/* Does nothing. We just want an architecture agnostic "return". */
}
+
+/* This is a lie, but given the objcopy, we need objtool to ignore it. */
+STACK_FRAME_NON_STANDARD(lkdtm_rodata_do_nothing);
diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index b341f8a8c7c5..c1b58a682ace 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -902,6 +902,8 @@ static void add_ignores(struct objtool_file *file)
struct reloc *reloc;
sec = find_section_by_name(file->elf, ".rela.discard.func_stack_frame_non_standard");
+ if (!sec)
+ sec = find_section_by_name(file->elf, ".discard.func_stack_frame_non_standard");
if (!sec)
return;
@@ -3719,7 +3721,7 @@ static int validate_retpoline(struct objtool_file *file)
insn->type != INSN_RETURN)
continue;
- if (insn->retpoline_safe)
+ if (insn->retpoline_safe || insn->ignore)
continue;
/*
--
2.32.0
Powered by blists - more mailing lists