From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Will Deacon <will.deacon@arm.com>,
Catalin Marinas <catalin.marinas@arm.com>,
Marc Zyngier <marc.zyngier@arm.com>,
Russell King <rmk+kernel@arm.linux.org.uk>
Subject: [ 62/72] ARM: 7769/1: Cortex-A15: fix erratum 798181 implementation
Date: Thu, 18 Jul 2013 22:26:19 -0700 [thread overview]
Message-ID: <20130719052604.108594547@linuxfoundation.org> (raw)
In-Reply-To: <20130719052559.852627424@linuxfoundation.org>
3.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marc Zyngier <Marc.Zyngier@arm.com>
commit 0d0752bca1f9a91fb646647aa4abbb21156f316c upstream.
Looking into the active_asids array is not enough, as we also need
to look into the reserved_asids array (they both represent processes
that are currently running).
Also, not holding the ASID allocator lock is racy, as another CPU
could schedule that process and trigger a rollover, making the erratum
workaround miss an IPI.
Exposing this outside of context.c is a little ugly on the side, so
let's define a new entry point that the erratum workaround can call
to obtain the cpumask.
Acked-by: Will Deacon <will.deacon@arm.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
arch/arm/include/asm/mmu_context.h | 10 +++++++++-
arch/arm/kernel/smp_tlb.c | 18 ++----------------
arch/arm/mm/context.c | 29 ++++++++++++++++++++++++++++-
3 files changed, 39 insertions(+), 18 deletions(-)
--- a/arch/arm/include/asm/mmu_context.h
+++ b/arch/arm/include/asm/mmu_context.h
@@ -27,7 +27,15 @@ void __check_vmalloc_seq(struct mm_struc
void check_and_switch_context(struct mm_struct *mm, struct task_struct *tsk);
#define init_new_context(tsk,mm) ({ atomic64_set(&mm->context.id, 0); 0; })
-DECLARE_PER_CPU(atomic64_t, active_asids);
+#ifdef CONFIG_ARM_ERRATA_798181
+void a15_erratum_get_cpumask(int this_cpu, struct mm_struct *mm,
+ cpumask_t *mask);
+#else /* !CONFIG_ARM_ERRATA_798181 */
+static inline void a15_erratum_get_cpumask(int this_cpu, struct mm_struct *mm,
+ cpumask_t *mask)
+{
+}
+#endif /* CONFIG_ARM_ERRATA_798181 */
#else /* !CONFIG_CPU_HAS_ASID */
--- a/arch/arm/kernel/smp_tlb.c
+++ b/arch/arm/kernel/smp_tlb.c
@@ -103,7 +103,7 @@ static void broadcast_tlb_a15_erratum(vo
static void broadcast_tlb_mm_a15_erratum(struct mm_struct *mm)
{
- int cpu, this_cpu;
+ int this_cpu;
cpumask_t mask = { CPU_BITS_NONE };
if (!erratum_a15_798181())
@@ -111,21 +111,7 @@ static void broadcast_tlb_mm_a15_erratum
dummy_flush_tlb_a15_erratum();
this_cpu = get_cpu();
- for_each_online_cpu(cpu) {
- if (cpu == this_cpu)
- continue;
- /*
- * We only need to send an IPI if the other CPUs are running
- * the same ASID as the one being invalidated. There is no
- * need for locking around the active_asids check since the
- * switch_mm() function has at least one dmb() (as required by
- * this workaround) in case a context switch happens on
- * another CPU after the condition below.
- */
- if (atomic64_read(&mm->context.id) ==
- atomic64_read(&per_cpu(active_asids, cpu)))
- cpumask_set_cpu(cpu, &mask);
- }
+ a15_erratum_get_cpumask(this_cpu, mm, &mask);
smp_call_function_many(&mask, ipi_flush_tlb_a15_erratum, NULL, 1);
put_cpu();
}
--- a/arch/arm/mm/context.c
+++ b/arch/arm/mm/context.c
@@ -45,10 +45,37 @@ static DEFINE_RAW_SPINLOCK(cpu_asid_lock
static atomic64_t asid_generation = ATOMIC64_INIT(ASID_FIRST_VERSION);
static DECLARE_BITMAP(asid_map, NUM_USER_ASIDS);
-DEFINE_PER_CPU(atomic64_t, active_asids);
+static DEFINE_PER_CPU(atomic64_t, active_asids);
static DEFINE_PER_CPU(u64, reserved_asids);
static cpumask_t tlb_flush_pending;
+#ifdef CONFIG_ARM_ERRATA_798181
+void a15_erratum_get_cpumask(int this_cpu, struct mm_struct *mm,
+ cpumask_t *mask)
+{
+ int cpu;
+ unsigned long flags;
+ u64 context_id, asid;
+
+ raw_spin_lock_irqsave(&cpu_asid_lock, flags);
+ context_id = mm->context.id.counter;
+ for_each_online_cpu(cpu) {
+ if (cpu == this_cpu)
+ continue;
+ /*
+ * We only need to send an IPI if the other CPUs are
+ * running the same ASID as the one being invalidated.
+ */
+ asid = per_cpu(active_asids, cpu).counter;
+ if (asid == 0)
+ asid = per_cpu(reserved_asids, cpu);
+ if (context_id == asid)
+ cpumask_set_cpu(cpu, mask);
+ }
+ raw_spin_unlock_irqrestore(&cpu_asid_lock, flags);
+}
+#endif
+
#ifdef CONFIG_ARM_LPAE
static void cpu_set_reserved_ttbr0(void)
{
next prev parent reply other threads:[~2013-07-19 5:30 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-19 5:25 [ 00/72] 3.10.2-stable review Greg Kroah-Hartman
2013-07-19 5:25 ` [ 01/72] CIFS use sensible file nlink values if unprovided Greg Kroah-Hartman
2013-07-19 5:25 ` [ 02/72] CIFS: Fix a deadlock when a file is reopened Greg Kroah-Hartman
2013-07-19 5:25 ` [ 03/72] rtlwifi: rtl8192cu: Add new USB ID for TP-Link TL-WN8200ND Greg Kroah-Hartman
2013-07-19 5:25 ` [ 04/72] rtlwifi: rtl8723ae: Fix typo in firmware names Greg Kroah-Hartman
2013-07-19 5:25 ` [ 05/72] rtlwifi: rtl8192cu: Fix duplicate if test Greg Kroah-Hartman
2013-07-19 5:25 ` [ 06/72] jbd2: move superblock checksum calculation to jbd2_write_superblock() Greg Kroah-Hartman
2013-07-19 5:25 ` [ 07/72] jbd2: fix theoretical race in jbd2__journal_restart Greg Kroah-Hartman
2013-07-19 5:25 ` [ 08/72] ext4: fix corruption when online resizing a fs with 1K block size Greg Kroah-Hartman
2013-07-19 5:25 ` [ 09/72] ext3,ext4: dont mess with dir_file->f_pos in htree_dirblock_to_tree() Greg Kroah-Hartman
2013-07-19 5:25 ` [ 10/72] ext4: check error return from ext4_write_inline_data_end() Greg Kroah-Hartman
2013-07-19 5:25 ` [ 11/72] pch_uart: Add uart_clk selection for the MinnowBoard Greg Kroah-Hartman
2013-07-19 5:25 ` [ 12/72] USB: option,qcserial: move Novatel Gobi1K IDs to qcserial Greg Kroah-Hartman
2013-07-19 5:25 ` [ 13/72] usb: gadget: f_mass_storage: add missing memory barrier for thread_wakeup_needed Greg Kroah-Hartman
2013-07-19 5:25 ` [ 14/72] USB: ehci-omap: Tweak PHY initialization sequence Greg Kroah-Hartman
2013-07-19 5:25 ` [ 15/72] xhci: check for failed dma pool allocation Greg Kroah-Hartman
2013-07-19 5:25 ` [ 16/72] usb: host: xhci-plat: release mem region while removing module Greg Kroah-Hartman
2013-07-19 5:25 ` [ 17/72] drivers: hv: switch to use mb() instead of smp_mb() Greg Kroah-Hartman
2013-07-19 5:25 ` [ 18/72] pcmcia: at91_cf: fix gpio_get_value in at91_cf_get_status Greg Kroah-Hartman
2013-07-19 5:25 ` [ 19/72] cgroup: fix umount vs cgroup_event_remove() race Greg Kroah-Hartman
2013-07-19 5:25 ` [ 20/72] cgroup: fix RCU accesses to task->cgroups Greg Kroah-Hartman
2013-07-19 5:25 ` [ 21/72] parisc: document the shadow registers Greg Kroah-Hartman
2013-07-19 5:25 ` [ 22/72] parisc: Fix gcc miscompilation in pa_memcpy() Greg Kroah-Hartman
2013-07-19 5:25 ` [ 23/72] slab: fix init_lock_keys Greg Kroah-Hartman
2013-07-19 5:25 ` [ 24/72] parisc: Ensure volatile space register %sr1 is not clobbered Greg Kroah-Hartman
2013-07-19 5:25 ` [ 25/72] parisc: fix LMMIO mismatch between PAT length and MASK register Greg Kroah-Hartman
2013-07-19 5:25 ` [ 26/72] parisc: optimize mtsp(0,sr) inline assembly Greg Kroah-Hartman
2013-07-19 5:25 ` [ 27/72] x86, efi: retry ExitBootServices() on failure Greg Kroah-Hartman
2013-07-19 5:25 ` [ 28/72] xen/time: remove blocked time accounting from xen "clockchip" Greg Kroah-Hartman
2013-07-19 5:25 ` [ 29/72] xen/pcifront: Deal with toolstack missing XenbusStateClosing state Greg Kroah-Hartman
2013-07-19 5:25 ` [ 30/72] genirq: Fix can_request_irq() for IRQs without an action Greg Kroah-Hartman
2013-07-19 5:25 ` [ 31/72] drivers/rtc/rtc-rv3029c2.c: fix disabling AIE irq Greg Kroah-Hartman
2013-07-19 5:25 ` [ 32/72] ACPI: Add CMOS RTC Operation Region handler support Greg Kroah-Hartman
2013-07-19 5:25 ` [ 33/72] ACPI / EC: Add HP Folio 13 to ec_dmi_table in order to skip DSDT scan Greg Kroah-Hartman
2013-07-19 5:25 ` [ 34/72] ACPICA: Do not use extended sleep registers unless HW-reduced bit is set Greg Kroah-Hartman
2013-07-19 5:25 ` [ 35/72] ACPI / PM: Fix corner case in acpi_bus_update_power() Greg Kroah-Hartman
2013-07-19 5:25 ` [ 36/72] HID: apple: Add support for the 2013 Macbook Air Greg Kroah-Hartman
2013-07-19 5:25 ` [ 37/72] Input: bcm5974 - add support for the 2013 MacBook Air Greg Kroah-Hartman
2013-07-19 5:40 ` Dmitry Torokhov
2013-07-19 5:44 ` Greg Kroah-Hartman
2013-07-22 1:42 ` Greg Kroah-Hartman
2013-07-22 2:21 ` Dmitry Torokhov
2013-07-22 10:07 ` rydberg
2013-07-19 5:25 ` [ 38/72] arch: c6x: mm: include "asm/uaccess.h" to pass compiling Greg Kroah-Hartman
2013-07-19 5:25 ` [ 39/72] ocfs2: xattr: fix inlined xattr reflink Greg Kroah-Hartman
2013-07-19 5:25 ` [ 40/72] nbd: correct disconnect behavior Greg Kroah-Hartman
2013-07-19 5:25 ` [ 41/72] PCI: Finish SR-IOV VF setup before adding the device Greg Kroah-Hartman
2013-07-19 5:25 ` [ 42/72] PCI: Fix refcount issue in pci_create_root_bus() error recovery path Greg Kroah-Hartman
2013-07-19 5:26 ` [ 43/72] iwlwifi: pcie: fix race in queue unmapping Greg Kroah-Hartman
2013-07-19 5:26 ` [ 44/72] iwlwifi: pcie: wake the queue if stopped when being unmapped Greg Kroah-Hartman
2013-07-19 5:26 ` [ 45/72] ahci: Add AMD CZ SATA device ID Greg Kroah-Hartman
2013-07-19 5:26 ` [ 46/72] ahci: AHCI-mode SATA patch for Intel Coleto Creek DeviceIDs Greg Kroah-Hartman
2013-07-19 5:26 ` [ 47/72] ahci: remove pmp link online check in FBS EH Greg Kroah-Hartman
2013-07-19 5:26 ` [ 48/72] timer: Fix jiffies wrap behavior of round_jiffies_common() Greg Kroah-Hartman
2013-07-19 5:26 ` [ 49/72] Btrfs: fix estale with btrfs send Greg Kroah-Hartman
2013-07-19 5:26 ` [ 50/72] Btrfs: hold the tree mod lock in __tree_mod_log_rewind Greg Kroah-Hartman
2013-07-19 5:26 ` [ 51/72] Btrfs: only do the tree_mod_log_free_eb if this is our last ref Greg Kroah-Hartman
2013-07-19 5:26 ` [ 52/72] ext4: fix data offset overflow on 32-bit archs in ext4_inline_data_fiemap() Greg Kroah-Hartman
2013-07-19 5:26 ` [ 53/72] ext4: fix overflows in SEEK_HOLE, SEEK_DATA implementations Greg Kroah-Hartman
2013-07-19 5:26 ` [ 54/72] ext4: fix data offset overflow in ext4_xattr_fiemap() on 32-bit archs Greg Kroah-Hartman
2013-07-19 5:26 ` [ 55/72] ext4: fix overflow when counting used blocks on 32-bit architectures Greg Kroah-Hartman
2013-07-19 5:26 ` [ 56/72] ext4: fix ext4_get_group_number() Greg Kroah-Hartman
2013-07-19 5:26 ` [ 57/72] ext4: dont show usrquota/grpquota twice in /proc/mounts Greg Kroah-Hartman
2013-07-19 5:26 ` [ 58/72] ext4: dont allow ext4_free_blocks() to fail due to ENOMEM Greg Kroah-Hartman
2013-07-19 5:26 ` [ 59/72] ARM: 7765/1: perf: Record the user-mode PC in the call chain Greg Kroah-Hartman
2013-07-19 5:26 ` [ 60/72] ARM: 7767/1: let the ASID allocator handle suspended animation Greg Kroah-Hartman
2013-07-19 5:26 ` [ 61/72] ARM: 7768/1: prevent risks of out-of-bound access in ASID allocator Greg Kroah-Hartman
2013-07-19 5:26 ` Greg Kroah-Hartman [this message]
2013-07-19 5:26 ` [ 63/72] ARM: 7778/1: smp_twd: twd_update_frequency need be run on all online CPUs Greg Kroah-Hartman
2013-07-19 5:26 ` [ 64/72] ARM: dts: imx: cpus/cpu nodes dts updates Greg Kroah-Hartman
2013-07-19 5:26 ` [ 65/72] ARM: shmobile: r8a73a4: Fix resources for SCIFB0 Greg Kroah-Hartman
2013-07-19 5:26 ` [ 66/72] ARM: shmobile: emev2 GIO3 resource fix Greg Kroah-Hartman
2013-07-19 5:26 ` [ 67/72] ARM: mm: fix boot on SA1110 Assabet Greg Kroah-Hartman
2013-07-19 5:26 ` [ 68/72] drivers/dma/pl330.c: fix locking in pl330_free_chan_resources() Greg Kroah-Hartman
2013-07-19 5:26 ` [ 69/72] memcg, kmem: fix reference count handling on the error path Greg Kroah-Hartman
2013-07-19 5:26 ` [ 70/72] mm/memory-hotplug: fix lowmem count overflow when offline pages Greg Kroah-Hartman
2013-07-19 5:26 ` [ 71/72] Handle big endianness in NTLM (ntlmv2) authentication Greg Kroah-Hartman
2013-07-19 5:26 ` [ 72/72] UBIFS: correct mount message Greg Kroah-Hartman
2013-07-19 7:27 ` [ 00/72] 3.10.2-stable review Sven Joachim
2013-07-19 15:40 ` Greg Kroah-Hartman
2013-07-21 2:23 ` Satoru Takeuchi
2013-07-19 23:16 ` Shuah Khan
2013-07-19 23:16 ` Shuah Khan
2013-07-19 23:44 ` Greg Kroah-Hartman
2013-07-20 0:17 ` Stefan Lippers-Hollmann
2013-07-20 0:25 ` [please disregard] " Stefan Lippers-Hollmann
2013-07-21 7:56 ` Willy Tarreau
2013-07-21 15:50 ` Greg Kroah-Hartman
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=20130719052604.108594547@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=catalin.marinas@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=marc.zyngier@arm.com \
--cc=rmk+kernel@arm.linux.org.uk \
--cc=stable@vger.kernel.org \
--cc=will.deacon@arm.com \
/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.