linux-kernel - [RFC PATCH v2 5/6] x86/entry/pti: avoid setting CR3 when it's already correct

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:   Tue,  9 Jan 2018 13:56:19 +0100
From:   Willy Tarreau <w@....eu>
To:     linux-kernel@...r.kernel.org, x86@...nel.org
Cc:     Willy Tarreau <w@....eu>, Andy Lutomirski <luto@...nel.org>,
        Borislav Petkov <bp@...en8.de>,
        Brian Gerst <brgerst@...il.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Ingo Molnar <mingo@...nel.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        "H. Peter Anvin" <hpa@...or.com>, Kees Cook <keescook@...omium.org>
Subject: [RFC PATCH v2 5/6] x86/entry/pti: avoid setting CR3 when it's already correct

When entering the kernel with CR3 pointing to the kernel's PGD, there's
no need to set it again. This will avoid a TLB flush on syscalls for tasks
running with the kernel's PGD (see next patch).

Signed-off-by: Willy Tarreau <w@....eu>
Cc: Andy Lutomirski <luto@...nel.org>
Cc: Borislav Petkov <bp@...en8.de>
Cc: Brian Gerst <brgerst@...il.com>
Cc: Dave Hansen <dave.hansen@...ux.intel.com>
Cc: Ingo Molnar <mingo@...nel.org>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Peter Zijlstra <peterz@...radead.org>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Josh Poimboeuf <jpoimboe@...hat.com>
Cc: "H. Peter Anvin" <hpa@...or.com>
Cc: Kees Cook <keescook@...omium.org>

v2:
  - updated comments according to Ingo's suggestions
  - split the code to keep only the CR3 changes here
---
 arch/x86/entry/calling.h | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/arch/x86/entry/calling.h b/arch/x86/entry/calling.h
index 45a63e0..2c0d3b5 100644
--- a/arch/x86/entry/calling.h
+++ b/arch/x86/entry/calling.h
@@ -214,6 +214,11 @@
 .macro SWITCH_TO_KERNEL_CR3 scratch_reg:req
 	ALTERNATIVE "jmp .Lend_\@", "", X86_FEATURE_PTI
 	mov	%cr3, \scratch_reg
+
+	/* if we're already on the kernel PGD, we don't switch */
+	testq $(PTI_SWITCH_PGTABLES_MASK), \scratch_reg
+	jz .Lend_\@
+
 	ADJUST_KERNEL_CR3 \scratch_reg
 	mov	\scratch_reg, %cr3
 .Lend_\@:
@@ -262,6 +267,14 @@
 	ALTERNATIVE "jmp .Ldone_\@", "", X86_FEATURE_PTI
 	movq	%cr3, \scratch_reg
 	movq	\scratch_reg, \save_reg
+
+	/*
+	 * If we're already on the kernel PGD, we don't switch,
+	 * we just save the current CR3.
+	 */
+	testq $(PTI_SWITCH_PGTABLES_MASK), \scratch_reg
+	jz .Ldone_\@
+
 	/*
 	 * Is the "switch mask" all zero?  That means that both of
 	 * these are zero:
@@ -284,6 +297,13 @@
 .macro RESTORE_CR3 scratch_reg:req save_reg:req
 	ALTERNATIVE "jmp .Lend_\@", "", X86_FEATURE_PTI
 
+	/*
+	 * If we saved a kernel context on entry, we didn't switch the CR3,
+	 * so we don't need to restore it on the way out either:
+	 */
+	testq $(PTI_SWITCH_PGTABLES_MASK), \save_reg
+	jz .Lend_\@
+
 	ALTERNATIVE "jmp .Lwrcr3_\@", "", X86_FEATURE_PCID
 
 	/*
-- 
1.7.12.1