kernel.git/arch/x86/kernel/smp.c, branch linux-4.14.y

x86/reboot: Disable SVM, not just VMX, when stopping CPUs

2023-03-11T15:26:45Z

commit a2b07fa7b93321c059af0c6d492cc9a4f1e390aa upstream. Disable SVM and more importantly force GIF=1 when halting a CPU or rebooting the machine. Similar to VMX, SVM allows software to block INITs via CLGI, and thus can be problematic for a crash/reboot. The window for failure is smaller with SVM as INIT is only blocked while GIF=0, i.e. between CLGI and STGI, but the window does exist. Fixes: fba4f472b33a ("x86/reboot: Turn off KVM when halting a CPU") Cc: stable@vger.kernel.org Reviewed-by: Thomas Gleixner Link: https://lore.kernel.org/r/20221130233650.1404148-5-seanjc@google.com Signed-off-by: Sean Christopherson Signed-off-by: Greg Kroah-Hartman

x86/reboot: Always use NMI fallback when shutdown via reboot vector IPI fails

2019-10-05T10:47:46Z

[ Upstream commit 747d5a1bf293dcb33af755a6d285d41b8c1ea010 ] A reboot request sends an IPI via the reboot vector and waits for all other CPUs to stop. If one or more CPUs are in critical regions with interrupts disabled then the IPI is not handled on those CPUs and the shutdown hangs if native_stop_other_cpus() is called with the wait argument set. Such a situation can happen when one CPU was stopped within a lock held section and another CPU is trying to acquire that lock with interrupts disabled. There are other scenarios which can cause such a lockup as well. In theory the shutdown should be attempted by an NMI IPI after the timeout period elapsed. Though the wait loop after sending the reboot vector IPI prevents this. It checks the wait request argument and the timeout. If wait is set, which is true for sys_reboot() then it won't fall through to the NMI shutdown method after the timeout period has finished. This was an oversight when the NMI shutdown mechanism was added to handle the 'reboot IPI is not working' situation. The mechanism was added to deal with stuck panic shutdowns, which do not have the wait request set, so the 'wait request' case was probably not considered. Remove the wait check from the post reboot vector IPI wait loop and enforce that the wait loop in the NMI fallback path is invoked even if NMI IPIs are disabled or the registration of the NMI handler fails. That second wait loop will then hang if not all CPUs shutdown and the wait argument is set. [ tglx: Avoid the hard to parse line break in the NMI fallback path, add comments and massage the changelog ] Fixes: 7d007d21e539 ("x86/reboot: Use NMI to assist in shutting down if IRQ fails") Signed-off-by: Grzegorz Halat Signed-off-by: Thomas Gleixner Cc: Don Zickus Link: https://lkml.kernel.org/r/20190628122813.15500-1-ghalat@redhat.com Signed-off-by: Sasha Levin

x86/irq: Let interrupt handlers set kvm_cpu_l1tf_flush_l1d

2018-08-15T16:12:58Z

commit ffcba43ff66c7dab34ec700debd491d2a4d319b4 upstream The last missing piece to having vmx_l1d_flush() take interrupts after VMEXIT into account is to set the kvm_cpu_l1tf_flush_l1d per-cpu flag on irq entry. Issue calls to kvm_set_cpu_l1tf_flush_l1d() from entering_irq(), ipi_entering_ack_irq(), smp_reschedule_interrupt() and uv_bau_message_interrupt(). Suggested-by: Paolo Bonzini Signed-off-by: Nicolai Stange Signed-off-by: Thomas Gleixner Signed-off-by: Greg Kroah-Hartman

x86/tracing: Disentangle pagefault and resched IPI tracing key

2017-08-29T09:42:29Z

The pagefault and the resched IPI handler are the only ones where it is worth to optimize the code further in case tracepoints are disabled. But it makes no sense to have a single static key for both. Seperate the static keys so the facilities are handled seperately. Signed-off-by: Thomas Gleixner Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Steven Rostedt Link: http://lkml.kernel.org/r/20170828064957.536699116@linutronix.de Signed-off-by: Ingo Molnar

x86/smp: Use static key for reschedule interrupt tracing

2017-08-29T09:42:27Z

It's worth to avoid the extra irq_enter()/irq_exit() pair in the case that the reschedule interrupt tracepoints are disabled. Use the static key which indicates that exception tracing is enabled. For now this key is global. It will be optimized in a later step. Signed-off-by: Thomas Gleixner Reviewed-by: Steven Rostedt (VMware) Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20170828064957.299808677@linutronix.de Signed-off-by: Ingo Molnar

x86/smp: Remove pointless duplicated interrupt code

2017-08-29T09:42:27Z

Two NOP5s are really a good tradeoff vs. the unholy IDT switching mess, which duplicates code all over the place. The rescheduling interrupt gets optimized in a later step. Make the ordering of function call and statistics increment the same as in other places. Calculate stats first, then do the function call. Signed-off-by: Thomas Gleixner Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Steven Rostedt Link: http://lkml.kernel.org/r/20170828064957.222101344@linutronix.de Signed-off-by: Ingo Molnar

Merge branch 'x86-boot-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2017-05-02T03:51:12Z

Pull x86 boot updates from Ingo Molnar: "The biggest changes in this cycle were: - reworking of the e820 code: separate in-kernel and boot-ABI data structures and apply a whole range of cleanups to the kernel side. No change in functionality. - enable KASLR by default: it's used by all major distros and it's out of the experimental stage as well. - ... misc fixes and cleanups" * 'x86-boot-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (63 commits) x86/KASLR: Fix kexec kernel boot crash when KASLR randomization fails x86/reboot: Turn off KVM when halting a CPU x86/boot: Fix BSS corruption/overwrite bug in early x86 kernel startup x86: Enable KASLR by default boot/param: Move next_arg() function to lib/cmdline.c for later reuse x86/boot: Fix Sparse warning by including required header file x86/boot/64: Rename start_cpu() x86/xen: Update e820 table handling to the new core x86 E820 code x86/boot: Fix pr_debug() API braindamage xen, x86/headers: Add dependency to x86/boot/e820: Simplify e820__update_table() x86/boot/e820: Separate the E820 ABI structures from the in-kernel structures x86/boot/e820: Fix and clean up e820_type switch() statements x86/boot/e820: Rename the remaining E820 APIs to the e820__*() prefix x86/boot/e820: Remove unnecessary #include's x86/boot/e820: Rename e820_mark_nosave_regions() to e820__register_nosave_regions() x86/boot/e820: Rename e820_reserve_resources*() to e820__reserve_resources*() x86/boot/e820: Use bool in query APIs x86/boot/e820: Document e820__reserve_setup_data() x86/boot/e820: Clean up __e820__update_table() et al ...

sched/x86: Update reschedule warning text

2017-04-20T08:14:30Z

Modify the reschedule warning to output the offline CPU number and use a better debug message. Signed-off-by: Prarit Bhargava Cc: Andrew Morton Cc: Daniel Bristot de Oliveira Cc: Hidehiro Kawai Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Steven Rostedt (VMware) Cc: Thomas Gleixner Cc: Wanpeng Li Link: http://lkml.kernel.org/r/1492518305-3808-1-git-send-email-prarit@redhat.com [ Tweaked the warning message. ] Signed-off-by: Ingo Molnar

x86/reboot: Turn off KVM when halting a CPU

2017-04-20T08:06:57Z

A CPU in VMX root mode will ignore INIT signals and will fail to bring up the APs after reboot. Therefore, on a panic we disable VMX on all CPUs before rebooting or triggering kdump. Do this when halting the machine as well, in case a firmware-level reboot does not perform a cold reset for all processors. Without doing this, rebooting the host may hang. Signed-off-by: Tiantian Feng Signed-off-by: Xishi Qiu [ Rewritten commit message. ] Signed-off-by: Paolo Bonzini Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: kvm@vger.kernel.org Link: http://lkml.kernel.org/r/20170419161839.30550-1-pbonzini@redhat.com Signed-off-by: Ingo Molnar

x86/irq, trace: Add __irq_entry annotation to x86's platform IRQ handlers

2017-01-05T07:58:49Z

This patch adds the __irq_entry annotation to the default x86 platform IRQ handlers. ftrace's function_graph tracer uses the __irq_entry annotation to notify the entry and return of IRQ handlers. For example, before the patch: 354549.667252 | 3) d..1 | default_idle_call() { 354549.667252 | 3) d..1 | arch_cpu_idle() { 354549.667253 | 3) d..1 | default_idle() { 354549.696886 | 3) d..1 | smp_trace_reschedule_interrupt() { 354549.696886 | 3) d..1 | irq_enter() { 354549.696886 | 3) d..1 | rcu_irq_enter() { After the patch: 366416.254476 | 3) d..1 | arch_cpu_idle() { 366416.254476 | 3) d..1 | default_idle() { 366416.261566 | 3) d..1 ==========> | 366416.261566 | 3) d..1 | smp_trace_reschedule_interrupt() { 366416.261566 | 3) d..1 | irq_enter() { 366416.261566 | 3) d..1 | rcu_irq_enter() { KASAN also uses this annotation. The smp_apic_timer_interrupt() was already annotated. Signed-off-by: Daniel Bristot de Oliveira Acked-by: Steven Rostedt (VMware) Cc: Aaron Lu Cc: Andrew Morton Cc: Baoquan He Cc: Borislav Petkov Cc: Claudio Fontana Cc: Denys Vlasenko Cc: Dou Liyang Cc: Gu Zheng Cc: Hidehiro Kawai Cc: Linus Torvalds Cc: Nicolai Stange Cc: Peter Zijlstra (Intel) Cc: Steven Rostedt Cc: Thomas Gleixner Cc: Tony Luck Cc: Wanpeng Li Cc: linux-edac@vger.kernel.org Link: http://lkml.kernel.org/r/059fdf437c2f0c09b13c18c8fe4e69999d3ffe69.1483528431.git.bristot@redhat.com Signed-off-by: Ingo Molnar