linux-kernel - [RFC][PATCH 4/8] jump_label, x86: Remove init NOP optimization

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]

Date:   Fri, 28 Jun 2019 12:21:17 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     x86@...nel.org, peterz@...radead.org, linux-kernel@...r.kernel.org
Cc:     Josh Poimboeuf <jpoimboe@...hat.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Masami Hiramatsu <mhiramat@...nel.org>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Jason Baron <jbaron@...mai.com>, Nadav Amit <namit@...are.com>,
        Andy Lutomirski <luto@...nel.org>,
        Eugeniy Paltsev <Eugeniy.Paltsev@...opsys.com>,
        Vineet Gupta <Vineet.Gupta1@...opsys.com>
Subject: [RFC][PATCH 4/8] jump_label, x86: Remove init NOP optimization

Instead of checking if the emitted (default) NOP is the ideal NOP, and
conditionally rewrite the NOP, just rewrite the NOP.

This shouldn't be a problem because init / module_load uses
text_poke_early() which is cheap and this saves us from having to go
figure out which NOP to compare against when we go variable size.

Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
---
 arch/x86/kernel/jump_label.c |   19 ++++---------------
 1 file changed, 4 insertions(+), 15 deletions(-)

--- a/arch/x86/kernel/jump_label.c
+++ b/arch/x86/kernel/jump_label.c
@@ -177,21 +177,10 @@ __init_or_module void arch_jump_label_tr
 				      enum jump_label_type type)
 {
 	/*
-	 * This function is called at boot up and when modules are
-	 * first loaded. Check if the default nop, the one that is
-	 * inserted at compile time, is the ideal nop. If it is, then
-	 * we do not need to update the nop, and we can leave it as is.
-	 * If it is not, then we need to update the nop to the ideal nop.
+	 * Rewrite the NOP on init / module-load to ensure we got the ideal
+	 * nop.  Don't bother with trying to figure out what size and what nop
+	 * it should be for now, simply do an unconditional rewrite.
 	 */
-	if (jlstate == JL_STATE_START) {
-		const unsigned char default_nop[] = { STATIC_KEY_INIT_NOP };
-		const unsigned char *ideal_nop = ideal_nops[NOP_ATOMIC5];
-
-		if (memcmp(ideal_nop, default_nop, 5) != 0)
-			jlstate = JL_STATE_UPDATE;
-		else
-			jlstate = JL_STATE_NO_UPDATE;
-	}
-	if (jlstate == JL_STATE_UPDATE)
+	if (jlstate == JL_STATE_UPDATE || jlstate == JL_STATE_START)
 		__jump_label_transform(entry, type, 1);
 }