linux-kernel - [tip:x86/process] x86/process: Correct and optimize TIF

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 for Android: free password hash cracker in your pocket

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

Date:   Sat, 11 Mar 2017 03:49:49 -0800
From:   tip-bot for Kyle Huey <tipbot@...or.com>
To:     linux-tip-commits@...r.kernel.org
Cc:     me@...ehuey.com, tglx@...utronix.de, khuey@...ehuey.com,
        linux-kernel@...r.kernel.org, peterz@...radead.org,
        luto@...nel.org, hpa@...or.com, mingo@...nel.org
Subject: [tip:x86/process] x86/process: Correct and optimize TIF_BLOCKSTEP
 switch

Commit-ID:  b9894a2f5bd18b1691cb6872c9afe32b148d0132
Gitweb:     http://git.kernel.org/tip/b9894a2f5bd18b1691cb6872c9afe32b148d0132
Author:     Kyle Huey <me@...ehuey.com>
AuthorDate: Tue, 14 Feb 2017 00:11:03 -0800
Committer:  Thomas Gleixner <tglx@...utronix.de>
CommitDate: Sat, 11 Mar 2017 12:45:18 +0100

x86/process: Correct and optimize TIF_BLOCKSTEP switch

The debug control MSR is "highly magical" as the blockstep bit can be
cleared by hardware under not well documented circumstances.

So a task switch relying on the bit set by the previous task (according to
the previous tasks thread flags) can trip over this and not update the flag
for the next task.

To fix this its required to handle DEBUGCTLMSR_BTF when either the previous
or the next or both tasks have the TIF_BLOCKSTEP flag set.

While at it avoid branching within the TIF_BLOCKSTEP case and evaluating
boot_cpu_data twice in kernels without CONFIG_X86_DEBUGCTLMSR.

x86_64: arch/x86/kernel/process.o
text	data	bss	dec	 hex
3024    8577    16      11617    2d61	Before
3008	8577	16	11601	 2d51	After

i386: No change

[ tglx: Made the shift value explicit, use a local variable to make the
code readable and massaged changelog]

Originally-by: Thomas Gleixner <tglx@...utronix.de>
Signed-off-by: Kyle Huey <khuey@...ehuey.com>
Cc: Peter Zijlstra <peterz@...radead.org>
Cc: Andy Lutomirski <luto@...nel.org>
Link: http://lkml.kernel.org/r/20170214081104.9244-3-khuey@kylehuey.com
Signed-off-by: Thomas Gleixner <tglx@...utronix.de>

---
 arch/x86/include/asm/msr-index.h |  1 +
 arch/x86/kernel/process.c        | 12 +++++++-----
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h
index d8b5f8a..4c928f3 100644
--- a/arch/x86/include/asm/msr-index.h
+++ b/arch/x86/include/asm/msr-index.h
@@ -127,6 +127,7 @@
 
 /* DEBUGCTLMSR bits (others vary by model): */
 #define DEBUGCTLMSR_LBR			(1UL <<  0) /* last branch recording */
+#define DEBUGCTLMSR_BTF_SHIFT		1
 #define DEBUGCTLMSR_BTF			(1UL <<  1) /* single-step on branches */
 #define DEBUGCTLMSR_TR			(1UL <<  6)
 #define DEBUGCTLMSR_BTS			(1UL <<  7)
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index ea9ea25..83fa3cb 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -222,13 +222,15 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 
 	propagate_user_return_notify(prev_p, next_p);
 
-	if ((tifp ^ tifn) & _TIF_BLOCKSTEP) {
-		unsigned long debugctl = get_debugctlmsr();
+	if ((tifp & _TIF_BLOCKSTEP || tifn & _TIF_BLOCKSTEP) &&
+	    arch_has_block_step()) {
+		unsigned long debugctl, msk;
 
+		rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
 		debugctl &= ~DEBUGCTLMSR_BTF;
-		if (tifn & _TIF_BLOCKSTEP)
-			debugctl |= DEBUGCTLMSR_BTF;
-		update_debugctlmsr(debugctl);
+		msk = tifn & _TIF_BLOCKSTEP;
+		debugctl |= (msk >> TIF_BLOCKSTEP) << DEBUGCTLMSR_BTF_SHIFT;
+		wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
 	}
 
 	if ((tifp ^ tifn) & _TIF_NOTSC) {