Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Lai Jiangshan <laijs@linux.alibaba.com>
To:     linux-kernel@vger.kernel.org
Cc:     Lai Jiangshan <laijs@linux.alibaba.com>,
        Andy Lutomirski <luto@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>,
        Thomas Gleixner <tglx@linutronix.de>, x86@kernel.org,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Alexandre Chartre <alexandre.chartre@oracle.com>
Subject: [RFC PATCH 4/5] x86/hw_breakpoint: Prevent data breakpoints on user_pcid_flush_mask
Date:   Mon, 25 May 2020 14:51:01 +0000
Message-Id: <20200525145102.122557-5-laijs@linux.alibaba.com>
In-Reply-To: <20200525145102.122557-1-laijs@linux.alibaba.com>
References: <20200525145102.122557-1-laijs@linux.alibaba.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

The percpu user_pcid_flush_mask is used for CPU entry
(#DB/#NMI/#DF/#MCE -> paranoid_exit() -> RESTORE_CR3).
If a data breakpoint on it, it will cause an unwanted #DB.

There are some other percpu data used in CPU entry, but they are
either in already-protected cpu_tss_rw or are safe to trigger #DB
(espfix related).

Cc: Andy Lutomirski <luto@kernel.org>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: x86@kernel.org
Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
---
 arch/x86/kernel/hw_breakpoint.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c
index 7d3966b9aa12..9579bd6fb589 100644
--- a/arch/x86/kernel/hw_breakpoint.c
+++ b/arch/x86/kernel/hw_breakpoint.c
@@ -33,6 +33,7 @@
 #include <asm/debugreg.h>
 #include <asm/user.h>
 #include <asm/desc.h>
+#include <asm/tlbflush.h>
 
 /* Per cpu debug control register value */
 DEFINE_PER_CPU(unsigned long, cpu_dr7);
@@ -251,6 +252,8 @@ static inline bool within_cpu_entry(unsigned long addr, unsigned long end)
 		return true;
 
 	for_each_possible_cpu(cpu) {
+		unsigned short *mask_ptr;
+
 		/* The original rw GDT is being used after load_direct_gdt() */
 		if (within_area(addr, end, (unsigned long)get_cpu_gdt_rw(cpu),
 				GDT_SIZE))
@@ -268,6 +271,17 @@ static inline bool within_cpu_entry(unsigned long addr, unsigned long end)
 				(unsigned long)&per_cpu(cpu_tss_rw, cpu),
 				sizeof(struct tss_struct)))
 			return true;
+
+		/*
+		 * cpu_tlbstate.user_pcid_flush_mask is used for CPU entry
+		 * (#DB/#NMI/#DF/#MCE -> paranoid_exit() -> RESTORE_CR3).
+		 * If a data breakpoint on it, it will cause an unwanted
+		 * #DB.
+		 */
+		mask_ptr = &per_cpu(cpu_tlbstate.user_pcid_flush_mask, cpu);
+		if (within_area(addr, end, (unsigned long)mask_ptr,
+				sizeof(*mask_ptr)))
+			return true;
 	}
 
 	return false;
-- 
2.20.1