linux-hardening - [RFC PATCH 09/11] x86/FineIBT: Add FINEIBT

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 19 Apr 2022 17:42:39 -0700
From:   joao@...rdrivepizza.com
To:     linux-kernel@...r.kernel.org, linux-hardening@...r.kernel.org
Cc:     joao@...rdrivepizza.com, peterz@...radead.org, jpoimboe@...hat.com,
        andrew.cooper3@...rix.com, keescook@...omium.org,
        samitolvanen@...gle.com, mark.rutland@....com, hjl.tools@...il.com,
        alyssa.milburn@...ux.intel.com, ndesaulniers@...gle.com,
        gabriel.gomes@...ux.intel.com, rick.p.edgecombe@...el.com
Subject: [RFC PATCH 09/11] x86/FineIBT: Add FINEIBT_TEST module

From: Joao Moreira <joao@...rdrivepizza.com>

Adds a module that on load will call a function directly ensuring that
FineIBT fixes for module relocations are working as expected. Next the
module invokes another function indirectly, with a wrong hash into R11,
causing a violation to be triggered (and the __fineibt_handler to be
invoked).

Signed-off-by: Joao Moreira <joao@...rdrivepizza.com>
---
 arch/x86/Kconfig.debug         |  5 +++++
 arch/x86/kernel/Makefile       |  1 +
 arch/x86/kernel/fineibt_test.c | 39 ++++++++++++++++++++++++++++++++++
 3 files changed, 45 insertions(+)
 create mode 100644 arch/x86/kernel/fineibt_test.c

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index d2463dd912c1..4a5617c2470d 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -284,3 +284,8 @@ config X86_CET_TEST
 	depends on m
 	depends on X86_KERNEL_IBT
 	tristate "in-kernel CET testing module"
+
+config X86_FINEIBT_TEST
+	depends on m
+	depends on X86_KERNEL_FINEIBT
+	tristate "in-kernel FineIBT testing module"
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index a82bcd14bd40..5d7f39f3d909 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -150,6 +150,7 @@ obj-$(CONFIG_UNWINDER_GUESS)		+= unwind_guess.o
 obj-$(CONFIG_AMD_MEM_ENCRYPT)		+= sev.o
 obj-$(CONFIG_X86_KERNEL_FINEIBT)	+= fineibt.o
 obj-$(CONFIG_X86_CET_TEST)		+= cet_test.o
+obj-$(CONFIG_X86_FINEIBT_TEST)		+= fineibt_test.o
 
 ###
 # 64 bit specific files
diff --git a/arch/x86/kernel/fineibt_test.c b/arch/x86/kernel/fineibt_test.c
new file mode 100644
index 000000000000..c8cbff6208f8
--- /dev/null
+++ b/arch/x86/kernel/fineibt_test.c
@@ -0,0 +1,39 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/module.h>
+
+void __fineibt_debug(void);
+
+void fineibt_foo(void) {
+  pr_info("FineIBT: dmesg should show a FineIBT violation message.\n");
+}
+
+void fineibt_bar(void) {
+  pr_info("FineIBT: this first one should run smoothly.\n");
+}
+
+static int fineibt_test_init(void)
+{
+  pr_info("FineIBT test\n");
+
+  __fineibt_debug();
+
+  asm volatile(
+    "call fineibt_bar\n"
+    "lea fineibt_foo(%%rip), %%rax\n"
+    "mov $0xdeadbeef, %%r11\n"
+    "call *%%rax\n"
+    /* this should trigger the handler because the hash is wrong */
+    ::: "rax"
+  );
+  return 0;
+}
+
+static void fineibt_test_exit(void)
+{
+}
+
+module_init(fineibt_test_init);
+module_exit(fineibt_test_exit);
+
+MODULE_LICENSE("GPL v2");
-- 
2.35.1