linux-kernel - [PATCH] static_call,x86: Robustify trampoline patching

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 linux-cve-announce 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]
Message-ID: <20211030074758.GT174703@worktop.programming.kicks-ass.net>
Date:   Sat, 30 Oct 2021 09:47:58 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Sami Tolvanen <samitolvanen@...gle.com>
Cc:     Ard Biesheuvel <ardb@...nel.org>,
        Mark Rutland <mark.rutland@....com>, X86 ML <x86@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        Nathan Chancellor <nathan@...nel.org>,
        Nick Desaulniers <ndesaulniers@...gle.com>,
        Sedat Dilek <sedat.dilek@...il.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        linux-hardening@...r.kernel.org,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        llvm@...ts.linux.dev
Subject: [PATCH] static_call,x86: Robustify trampoline patching

On Fri, Oct 29, 2021 at 10:03:24PM +0200, Peter Zijlstra wrote:

> So I had a bit of a peek at what clang generates:
> 
>     3fa4:       48 c7 c7 00 00 00 00    mov    $0x0,%rdi        3fa7: R_X86_64_32S      __SCK__x86_pmu_handle_irq
>     3fab:       48 c7 c6 00 00 00 00    mov    $0x0,%rsi        3fae: R_X86_64_32S      __SCT__x86_pmu_handle_irq.cfi_jt
>     3fb2:       e8 00 00 00 00          call   3fb7 <init_hw_perf_events+0x1dc> 3fb3: R_X86_64_PLT32    __static_call_update-0x4
> 
> So this then gives the trampoline jump table entry to
> __static_call_update(), with the result that it will rewrite the
> jump-table entry, not the trampoline!
> 
> Now it so happens that the trampoline looks *exactly* like the
> jump-table entry (one jmp.d32 instruction), so in that regards it'll
> again 'work'.
> 
> But this is all really, as in *really*, wrong. And I'm really sad I'm
> the one to have to discover this, even though I've mentioned
> static_call()s being tricky in previous reviews.

The below makes the clang-cfi build properly sick:

[    0.000000] trampoline signature fail
[    0.000000] ------------[ cut here ]------------
[    0.000000] kernel BUG at arch/x86/kernel/static_call.c:65!

---
Subject: static_call,x86: Robustify trampoline patching

Add a few signature bytes after the static call trampoline and verify
those bytes match before patching the trampoline. This avoids patching
random other JMPs (such as CFI jump-table entries) instead.

These bytes decode as:

   d:   53                      push   %rbx
   e:   43 54                   rex.XB push %r12

And happen to spell "SCT".

Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
---
 arch/x86/include/asm/static_call.h |  1 +
 arch/x86/kernel/static_call.c      | 14 ++++++++++----
 tools/objtool/check.c              |  3 +++
 3 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/arch/x86/include/asm/static_call.h b/arch/x86/include/asm/static_call.h
index cbb67b6030f9..39ebe0511869 100644
--- a/arch/x86/include/asm/static_call.h
+++ b/arch/x86/include/asm/static_call.h
@@ -27,6 +27,7 @@
 	    ".globl " STATIC_CALL_TRAMP_STR(name) "		\n"	\
 	    STATIC_CALL_TRAMP_STR(name) ":			\n"	\
 	    insns "						\n"	\
+	    ".byte 0x53, 0x43, 0x54				\n"	\
 	    ".type " STATIC_CALL_TRAMP_STR(name) ", @function	\n"	\
 	    ".size " STATIC_CALL_TRAMP_STR(name) ", . - " STATIC_CALL_TRAMP_STR(name) " \n" \
 	    ".popsection					\n")
diff --git a/arch/x86/kernel/static_call.c b/arch/x86/kernel/static_call.c
index ea028e736831..9c407a33a774 100644
--- a/arch/x86/kernel/static_call.c
+++ b/arch/x86/kernel/static_call.c
@@ -56,10 +56,15 @@ static void __ref __static_call_transform(void *insn, enum insn_type type, void
 	text_poke_bp(insn, code, size, emulate);
 }
 
-static void __static_call_validate(void *insn, bool tail)
+static void __static_call_validate(void *insn, bool tail, bool tramp)
 {
 	u8 opcode = *(u8 *)insn;
 
+	if (tramp && memcmp(insn+5, "SCT", 3)) {
+		pr_err("trampoline signature fail");
+		BUG();
+	}
+
 	if (tail) {
 		if (opcode == JMP32_INSN_OPCODE ||
 		    opcode == RET_INSN_OPCODE)
@@ -74,7 +79,8 @@ static void __static_call_validate(void *insn, bool tail)
 	/*
 	 * If we ever trigger this, our text is corrupt, we'll probably not live long.
 	 */
-	WARN_ONCE(1, "unexpected static_call insn opcode 0x%x at %pS\n", opcode, insn);
+	pr_err("unexpected static_call insn opcode 0x%x at %pS\n", opcode, insn);
+	BUG();
 }
 
 static inline enum insn_type __sc_insn(bool null, bool tail)
@@ -97,12 +103,12 @@ void arch_static_call_transform(void *site, void *tramp, void *func, bool tail)
 	mutex_lock(&text_mutex);
 
 	if (tramp) {
-		__static_call_validate(tramp, true);
+		__static_call_validate(tramp, true, true);
 		__static_call_transform(tramp, __sc_insn(!func, true), func);
 	}
 
 	if (IS_ENABLED(CONFIG_HAVE_STATIC_CALL_INLINE) && site) {
-		__static_call_validate(site, tail);
+		__static_call_validate(site, tail, false);
 		__static_call_transform(site, __sc_insn(!func, tail), func);
 	}
 
diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index fb3f251ea021..6d5951113d53 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -3310,6 +3310,9 @@ static bool ignore_unreachable_insn(struct objtool_file *file, struct instructio
 	if (!insn->func)
 		return false;
 
+	if (insn->func->static_call_tramp)
+		return true;
+
 	/*
 	 * CONFIG_UBSAN_TRAP inserts a UD2 when it sees
 	 * __builtin_unreachable().  The BUG() macro has an unreachable() after