All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: stable@vger.kernel.org
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>,
	Will Deacon <will@kernel.org>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Marc Zyngier <maz@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Jeremy Linton <jeremy.linton@arm.com>,
	Andre Przywara <andre.przywara@arm.com>,
	Alexandru Elisei <alexandru.elisei@arm.com>,
	Marc Zyngier <marc.zyngier@arm.com>
Subject: [PATCH for-stable-4.14 46/48] arm64: Force SSBS on context switch
Date: Thu, 24 Oct 2019 14:48:31 +0200	[thread overview]
Message-ID: <20191024124833.4158-47-ard.biesheuvel@linaro.org> (raw)
In-Reply-To: <20191024124833.4158-1-ard.biesheuvel@linaro.org>

From: Marc Zyngier <marc.zyngier@arm.com>

[ Upstream commit cbdf8a189a66001c36007bf0f5c975d0376c5c3a ]

On a CPU that doesn't support SSBS, PSTATE[12] is RES0.  In a system
where only some of the CPUs implement SSBS, we end-up losing track of
the SSBS bit across task migration.

To address this issue, let's force the SSBS bit on context switch.

Fixes: 8f04e8e6e29c ("arm64: ssbd: Add support for PSTATE.SSBS rather than trapping to EL3")
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
[will: inverted logic and added comments]
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
---
 arch/arm64/include/asm/processor.h | 14 ++++++++--
 arch/arm64/kernel/process.c        | 29 +++++++++++++++++++-
 2 files changed, 40 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/include/asm/processor.h b/arch/arm64/include/asm/processor.h
index ec1725c6df21..9eb95ab19924 100644
--- a/arch/arm64/include/asm/processor.h
+++ b/arch/arm64/include/asm/processor.h
@@ -148,6 +148,16 @@ static inline void start_thread_common(struct pt_regs *regs, unsigned long pc)
 	regs->pc = pc;
 }
 
+static inline void set_ssbs_bit(struct pt_regs *regs)
+{
+	regs->pstate |= PSR_SSBS_BIT;
+}
+
+static inline void set_compat_ssbs_bit(struct pt_regs *regs)
+{
+	regs->pstate |= PSR_AA32_SSBS_BIT;
+}
+
 static inline void start_thread(struct pt_regs *regs, unsigned long pc,
 				unsigned long sp)
 {
@@ -155,7 +165,7 @@ static inline void start_thread(struct pt_regs *regs, unsigned long pc,
 	regs->pstate = PSR_MODE_EL0t;
 
 	if (arm64_get_ssbd_state() != ARM64_SSBD_FORCE_ENABLE)
-		regs->pstate |= PSR_SSBS_BIT;
+		set_ssbs_bit(regs);
 
 	regs->sp = sp;
 }
@@ -174,7 +184,7 @@ static inline void compat_start_thread(struct pt_regs *regs, unsigned long pc,
 #endif
 
 	if (arm64_get_ssbd_state() != ARM64_SSBD_FORCE_ENABLE)
-		regs->pstate |= PSR_AA32_SSBS_BIT;
+		set_compat_ssbs_bit(regs);
 
 	regs->compat_sp = sp;
 }
diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c
index 532ad6be9c2b..243fd247d04e 100644
--- a/arch/arm64/kernel/process.c
+++ b/arch/arm64/kernel/process.c
@@ -298,7 +298,7 @@ int copy_thread(unsigned long clone_flags, unsigned long stack_start,
 			childregs->pstate |= PSR_UAO_BIT;
 
 		if (arm64_get_ssbd_state() == ARM64_SSBD_FORCE_DISABLE)
-			childregs->pstate |= PSR_SSBS_BIT;
+			set_ssbs_bit(childregs);
 
 		p->thread.cpu_context.x19 = stack_start;
 		p->thread.cpu_context.x20 = stk_sz;
@@ -339,6 +339,32 @@ void uao_thread_switch(struct task_struct *next)
 	}
 }
 
+/*
+ * Force SSBS state on context-switch, since it may be lost after migrating
+ * from a CPU which treats the bit as RES0 in a heterogeneous system.
+ */
+static void ssbs_thread_switch(struct task_struct *next)
+{
+	struct pt_regs *regs = task_pt_regs(next);
+
+	/*
+	 * Nothing to do for kernel threads, but 'regs' may be junk
+	 * (e.g. idle task) so check the flags and bail early.
+	 */
+	if (unlikely(next->flags & PF_KTHREAD))
+		return;
+
+	/* If the mitigation is enabled, then we leave SSBS clear. */
+	if ((arm64_get_ssbd_state() == ARM64_SSBD_FORCE_ENABLE) ||
+	    test_tsk_thread_flag(next, TIF_SSBD))
+		return;
+
+	if (compat_user_mode(regs))
+		set_compat_ssbs_bit(regs);
+	else if (user_mode(regs))
+		set_ssbs_bit(regs);
+}
+
 /*
  * We store our current task in sp_el0, which is clobbered by userspace. Keep a
  * shadow copy so that we can restore this upon entry from userspace.
@@ -367,6 +393,7 @@ __notrace_funcgraph struct task_struct *__switch_to(struct task_struct *prev,
 	contextidr_thread_switch(next);
 	entry_task_switch(next);
 	uao_thread_switch(next);
+	ssbs_thread_switch(next);
 
 	/*
 	 * Complete any pending TLB or cache maintenance on this CPU in case
-- 
2.20.1


  parent reply	other threads:[~2019-10-24 12:50 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-24 12:47 [PATCH for-stable-4.14 00/48] arm64 spec mitigation backports Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 01/48] arm64: sysreg: Move to use definitions for all the SCTLR bits Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 02/48] arm64: Expose support for optional ARMv8-A features Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 03/48] arm64: Fix the feature type for ID register fields Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 04/48] arm64: v8.4: Support for new floating point multiplication instructions Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 05/48] arm64: Documentation: cpu-feature-registers: Remove RES0 fields Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 06/48] arm64: Expose Arm v8.4 features Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 07/48] arm64: move SCTLR_EL{1,2} assertions to <asm/sysreg.h> Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 08/48] arm64: add PSR_AA32_* definitions Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 09/48] arm64: Introduce sysreg_clear_set() Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 10/48] arm64: capabilities: Update prototype for enable call back Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 11/48] arm64: capabilities: Move errata work around check on boot CPU Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 12/48] arm64: capabilities: Move errata processing code Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 13/48] arm64: capabilities: Prepare for fine grained capabilities Ard Biesheuvel
2019-10-24 12:47 ` [PATCH for-stable-4.14 14/48] arm64: capabilities: Add flags to handle the conflicts on late CPU Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 15/48] arm64: capabilities: Unify the verification Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 16/48] arm64: capabilities: Filter the entries based on a given mask Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 17/48] arm64: capabilities: Prepare for grouping features and errata work arounds Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 18/48] arm64: capabilities: Split the processing of " Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 19/48] arm64: capabilities: Allow features based on local CPU scope Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 20/48] arm64: capabilities: Group handling of features and errata workarounds Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 21/48] arm64: capabilities: Introduce weak features based on local CPU Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 22/48] arm64: capabilities: Restrict KPTI detection to boot-time CPUs Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 23/48] arm64: capabilities: Add support for features enabled early Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 24/48] arm64: capabilities: Change scope of VHE to Boot CPU feature Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 25/48] arm64: capabilities: Clean up midr range helpers Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 26/48] arm64: Add helpers for checking CPU MIDR against a range Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 27/48] arm64: Add MIDR encoding for Arm Cortex-A55 and Cortex-A35 Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 28/48] arm64: capabilities: Add support for checks based on a list of MIDRs Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 29/48] arm64: KVM: Use SMCCC_ARCH_WORKAROUND_1 for Falkor BP hardening Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 30/48] arm64: don't zero DIT on signal return Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 31/48] arm64: Get rid of __smccc_workaround_1_hvc_* Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 32/48] arm64: cpufeature: Detect SSBS and advertise to userspace Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 33/48] arm64: ssbd: Add support for PSTATE.SSBS rather than trapping to EL3 Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 34/48] KVM: arm64: Set SCTLR_EL2.DSSBS if SSBD is forcefully disabled and !vhe Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 35/48] arm64: fix SSBS sanitization Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 36/48] arm64: Add sysfs vulnerability show for spectre-v1 Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 37/48] arm64: add sysfs vulnerability show for meltdown Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 38/48] arm64: enable generic CPU vulnerabilites support Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 39/48] arm64: Always enable ssb vulnerability detection Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 40/48] arm64: Provide a command line to disable spectre_v2 mitigation Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 41/48] arm64: Advertise mitigation of Spectre-v2, or lack thereof Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 42/48] arm64: Always enable spectre-v2 vulnerability detection Ard Biesheuvel
2019-10-24 14:34   ` Alexandru Elisei
2019-10-24 14:37     ` Ard Biesheuvel
2019-10-25 15:25       ` Sasha Levin
2019-10-25 15:28         ` Ard Biesheuvel
2019-10-25 15:39           ` Ard Biesheuvel
2019-10-26  8:01             ` Greg KH
2019-10-26 15:40               ` Sasha Levin
2019-10-26 15:46                 ` Ard Biesheuvel
2019-10-27 13:39                   ` Greg KH
2019-10-27 17:39                     ` Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 43/48] arm64: add sysfs vulnerability show for spectre-v2 Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 44/48] arm64: add sysfs vulnerability show for speculative store bypass Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 45/48] arm64: ssbs: Don't treat CPUs with SSBS as unaffected by SSB Ard Biesheuvel
2019-10-24 12:48 ` Ard Biesheuvel [this message]
2019-10-24 12:48 ` [PATCH for-stable-4.14 47/48] arm64: Use firmware to detect CPUs that are not affected by Spectre-v2 Ard Biesheuvel
2019-10-24 12:48 ` [PATCH for-stable-4.14 48/48] arm64/speculation: Support 'mitigations=' cmdline option Ard Biesheuvel

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20191024124833.4158-47-ard.biesheuvel@linaro.org \
    --to=ard.biesheuvel@linaro.org \
    --cc=alexandru.elisei@arm.com \
    --cc=andre.przywara@arm.com \
    --cc=catalin.marinas@arm.com \
    --cc=jeremy.linton@arm.com \
    --cc=marc.zyngier@arm.com \
    --cc=mark.rutland@arm.com \
    --cc=maz@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=suzuki.poulose@arm.com \
    --cc=will@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.