linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Marc Zyngier <marc.zyngier@arm.com>,
	Will Deacon <will@kernel.org>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH AUTOSEL 5.2 38/76] arm64: Force SSBS on context switch
Date: Fri,  2 Aug 2019 09:19:12 -0400	[thread overview]
Message-ID: <20190802131951.11600-38-sashal@kernel.org> (raw)
In-Reply-To: <20190802131951.11600-1-sashal@kernel.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: Sasha Levin <sashal@kernel.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 fd5b1a4efc70e..844e2964b0f5e 100644
--- a/arch/arm64/include/asm/processor.h
+++ b/arch/arm64/include/asm/processor.h
@@ -193,6 +193,16 @@ static inline void start_thread_common(struct pt_regs *regs, unsigned long pc)
 		regs->pmr_save = GIC_PRIO_IRQON;
 }
 
+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)
 {
@@ -200,7 +210,7 @@ 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;
 }
@@ -219,7 +229,7 @@ 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 6a869d9f304f7..b0c859ca63201 100644
--- a/arch/arm64/kernel/process.c
+++ b/arch/arm64/kernel/process.c
@@ -398,7 +398,7 @@ 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);
 
 		if (system_uses_irq_prio_masking())
 			childregs->pmr_save = GIC_PRIO_IRQON;
@@ -442,6 +442,32 @@ 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.
@@ -471,6 +497,7 @@ __notrace_funcgraph struct task_struct *__switch_to(struct task_struct *prev,
 	entry_task_switch(next);
 	uao_thread_switch(next);
 	ptrauth_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:[~2019-08-02 13:21 UTC|newest]

Thread overview: 78+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-02 13:18 [PATCH AUTOSEL 5.2 01/76] powerpc: fix off by one in max_zone_pfn initialization for ZONE_DMA Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 02/76] netfilter: nfnetlink: avoid deadlock due to synchronous request_module Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 03/76] vfio-ccw: Set pa_nr to 0 if memory allocation fails for pa_iova_pfn Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 04/76] vfio-ccw: Don't call cp_free if we are processing a channel program Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 05/76] netfilter: Fix rpfilter dropping vrf packets by mistake Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 06/76] netfilter: nf_tables: fix module autoload for redir Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 07/76] netfilter: conntrack: always store window size un-scaled Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 08/76] netfilter: nft_hash: fix symhash with modulus one Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 09/76] scripts/sphinx-pre-install: fix script for RHEL/CentOS Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 10/76] scripts/sphinx-pre-install: don't use LaTeX with CentOS 7 Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 11/76] scripts/sphinx-pre-install: fix latexmk dependencies Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 12/76] rq-qos: don't reset has_sleepers on spurious wakeups Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 13/76] rq-qos: set ourself TASK_UNINTERRUPTIBLE after we schedule Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 14/76] rq-qos: use a mb for got_token Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 15/76] netfilter: nf_tables: Support auto-loading for inet nat Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 16/76] drm/amd/display: No audio endpoint for Dell MST display Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 17/76] drm/amd/display: Clock does not lower in Updateplanes Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 18/76] drm/amd/display: Wait for backlight programming completion in set backlight level Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 19/76] drm/amd/display: fix DMCU hang when going into Modern Standby Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 20/76] drm/amd/display: use encoder's engine id to find matched free audio device Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 21/76] drm/amd/display: put back front end initialization sequence Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 22/76] drm/amd/display: allocate 4 ddc engines for RV2 Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 23/76] drm/amd/display: Fix dc_create failure handling and 666 color depths Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 24/76] drm/amd/display: Only enable audio if speaker allocation exists Sasha Levin
2019-08-02 13:18 ` [PATCH AUTOSEL 5.2 25/76] drm/amd/display: Increase size of audios array Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 26/76] iscsi_ibft: make ISCSI_IBFT dependson ACPI instead of ISCSI_IBFT_FIND Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 27/76] nl80211: fix NL80211_HE_MAX_CAPABILITY_LEN Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 28/76] mac80211: fix possible memory leak in ieee80211_assign_beacon Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 29/76] mac80211: don't warn about CW params when not using them Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 30/76] allocate_flower_entry: should check for null deref Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 31/76] hwmon: (occ) Fix division by zero issue Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 32/76] hwmon: (nct6775) Fix register address and added missed tolerance for nct6106 Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 33/76] ARM: dts: imx6ul: fix clock frequency property name of I2C buses Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 34/76] x86/mm: Check for pfn instead of page in vmalloc_sync_one() Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 35/76] x86/mm: Sync also unmappings in vmalloc_sync_all() Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 36/76] mm/vmalloc: Sync unmappings in __purge_vmap_area_lazy() Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 37/76] powerpc/papr_scm: Force a scm-unbind if initial scm-bind fails Sasha Levin
2019-08-02 13:19 ` Sasha Levin [this message]
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 39/76] arm64: entry: SP Alignment Fault doesn't write to FAR_EL1 Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 40/76] iommu/vt-d: Check if domain->pgd was allocated Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 41/76] drm/msm/dpu: Correct dpu encoder spinlock initialization Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 42/76] drm/msm: stop abusing dma_map/unmap for cache Sasha Levin
2019-08-03  0:14   ` Rob Clark
2019-08-14  2:03     ` Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 43/76] drm: silence variable 'conn' set but not used Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 44/76] arm64: dts: imx8mm: Correct SAI3 RXC/TXFS pin's mux option #1 Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 45/76] arm64: dts: imx8mq: fix SAI compatible Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 46/76] cpufreq/pasemi: fix use-after-free in pas_cpufreq_cpu_init() Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 47/76] s390/qdio: add sanity checks to the fast-requeue path Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 48/76] ALSA: compress: Fix regression on compressed capture streams Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 49/76] ALSA: compress: Prevent bypasses of set_params Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 50/76] ALSA: compress: Don't allow paritial drain operations on capture streams Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 51/76] ALSA: compress: Be more restrictive about when a drain is allowed Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 52/76] perf script: Fix off by one in brstackinsn IPC computation Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 53/76] perf tools: Fix proper buffer size for feature processing Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 54/76] perf stat: Fix segfault for event group in repeat mode Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 55/76] perf session: Fix loading of compressed data split across adjacent records Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 56/76] perf probe: Avoid calling freeing routine multiple times for same pointer Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 57/76] drbd: dynamically allocate shash descriptor Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 58/76] ACPI/IORT: Fix off-by-one check in iort_dev_find_its_id() Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 59/76] nvme: ignore subnqn for ADATA SX6000LNP Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 60/76] nvme: fix memory leak caused by incorrect subsystem free Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 61/76] nvme: fix multipath crash when ANA is deactivated Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 62/76] ARM: davinci: fix sleep.S build error on ARMv4 Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 63/76] ARM: dts: bcm: bcm47094: add missing #cells for mdio-bus-mux Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 64/76] scsi: megaraid_sas: fix panic on loading firmware crashdump Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 65/76] scsi: ibmvfc: fix WARN_ON during event pool release Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 66/76] scsi: scsi_dh_alua: always use a 2 second delay before retrying RTPG Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 67/76] test_firmware: fix a memory leak bug Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 68/76] sched/fair: Don't free p->numa_faults with concurrent readers Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 69/76] sched/fair: Use RCU accessors consistently for ->numa_group Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 70/76] tty/ldsem, locking/rwsem: Add missing ACQUIRE to read_failed sleep loop Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 71/76] perf/x86/intel: Fix SLOTS PEBS event constraint Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 72/76] perf/x86/intel: Fix invalid Bit 13 for Icelake MSR_OFFCORE_RSP_x register Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 73/76] perf/x86: Apply more accurate check on hypervisor platform Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 74/76] perf/core: Fix creating kernel counters for PMUs that override event->cpu Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 75/76] s390/dma: provide proper ARCH_ZONE_DMA_BITS value Sasha Levin
2019-08-02 13:19 ` [PATCH AUTOSEL 5.2 76/76] gen_compile_commands: lower the entry count threshold Sasha Levin

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=20190802131951.11600-38-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=marc.zyngier@arm.com \
    --cc=stable@vger.kernel.org \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).