From: Thomas Gleixner <tglx@linutronix.de>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
Ingo Molnar <mingo@kernel.org>, Peter Anvin <hpa@zytor.com>,
Peter Zijlstra <peterz@infradead.org>,
Andy Lutomirski <luto@kernel.org>, Borislav Petkov <bp@alien8.de>
Subject: Re: [patch 14/41] x86/smp: Use static key for reschedule interrupt tracing
Date: Fri, 25 Aug 2017 22:07:33 +0200 (CEST) [thread overview]
Message-ID: <alpine.DEB.2.20.1708252202410.2124@nanos> (raw)
In-Reply-To: <20170825154016.2b692126@gandalf.local.home>
On Fri, 25 Aug 2017, Steven Rostedt wrote:
> On Fri, 25 Aug 2017 12:31:17 +0200
> Thomas Gleixner <tglx@linutronix.de> wrote:
>
>
> > -__visible void __irq_entry smp_trace_reschedule_interrupt(struct pt_regs *regs)
> > -{
> > - /*
> > - * Need to call irq_enter() before calling the trace point.
> > - * __smp_reschedule_interrupt() calls irq_enter/exit() too (in
> > - * scheduler_ipi(). This is OK, since those functions are allowed
> > - * to nest.
> > - */
> > - ipi_entering_ack_irq();
> > - trace_reschedule_entry(RESCHEDULE_VECTOR);
> > + if (trace_irqvectors_enabled()) {
> > + /*
> > + * scheduler_ipi() might call irq_enter() as well, but
> > + * nested calls are fine.
> > + */
> > + irq_enter();
> > + trace_reschedule_entry(RESCHEDULE_VECTOR);
> > + }
> > +
> > inc_irq_stat(irq_resched_count);
> > scheduler_ipi();
> > - trace_reschedule_exit(RESCHEDULE_VECTOR);
> > - exiting_irq();
> > +
> > + if (trace_irqvectors_enabled()) {
> > + trace_reschedule_exit(RESCHEDULE_VECTOR);
> > + irq_exit();
>
> This is racy. If the static key gets changed on one and not the other,
> we lose the irq_enter/irq_exit match.
Indeed. Good point. Fixed up version below.
Thanks,
tglx
8<---------------------
--- a/arch/x86/include/asm/hw_irq.h
+++ b/arch/x86/include/asm/hw_irq.h
@@ -48,7 +48,7 @@ extern asmlinkage void call_function_sin
#ifdef CONFIG_TRACING
/* Interrupt handlers registered during init_IRQ */
-extern void trace_reschedule_interrupt(void);
+#define trace_reschedule_interrupt reschedule_interrupt
#define trace_call_function_interrupt call_function_interrupt
#define trace_call_function_single_interrupt call_function_single_interrupt
#define trace_thermal_interrupt thermal_interrupt
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -254,37 +254,27 @@ static void native_stop_other_cpus(int w
}
/*
- * Reschedule call back.
+ * Reschedule call back. KVM uses this interrupt to force a cpu out of
+ * guest mode
*/
-static inline void __smp_reschedule_interrupt(void)
-{
- inc_irq_stat(irq_resched_count);
- scheduler_ipi();
-}
-
__visible void __irq_entry smp_reschedule_interrupt(struct pt_regs *regs)
{
ack_APIC_irq();
- __smp_reschedule_interrupt();
- /*
- * KVM uses this interrupt to force a cpu out of guest mode
- */
-}
-
-__visible void __irq_entry smp_trace_reschedule_interrupt(struct pt_regs *regs)
-{
- /*
- * Need to call irq_enter() before calling the trace point.
- * __smp_reschedule_interrupt() calls irq_enter/exit() too (in
- * scheduler_ipi(). This is OK, since those functions are allowed
- * to nest.
- */
- ipi_entering_ack_irq();
- trace_reschedule_entry(RESCHEDULE_VECTOR);
inc_irq_stat(irq_resched_count);
+
+ if (trace_irqvectors_enabled()) {
+ /*
+ * scheduler_ipi() might call irq_enter() as well, but
+ * nested calls are fine.
+ */
+ irq_enter();
+ trace_reschedule_entry(RESCHEDULE_VECTOR);
+ scheduler_ipi();
+ trace_reschedule_exit(RESCHEDULE_VECTOR);
+ irq_exit();
+ return;
+ }
scheduler_ipi();
- trace_reschedule_exit(RESCHEDULE_VECTOR);
- exiting_irq();
}
__visible void __irq_entry smp_call_function_interrupt(struct pt_regs *regs)
next prev parent reply other threads:[~2017-08-25 20:07 UTC|newest]
Thread overview: 69+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-25 10:31 [patch 00/41] x86: Cleanup IDT code Thomas Gleixner
2017-08-25 10:31 ` [patch 01/41] x86/irq: Remove vector_used_by_percpu_irq() Thomas Gleixner
2017-08-25 10:31 ` [patch 02/41] x86/irq: Unexport used_vectors Thomas Gleixner
2017-08-25 10:31 ` [patch 03/41] x86/irq: Move ifdeffery to header file Thomas Gleixner
2017-08-25 12:45 ` Thomas Gleixner
2017-08-25 10:31 ` [patch 04/41] x86/boot: Move EISA setup to a proper place Thomas Gleixner
2017-08-25 11:59 ` Thomas Gleixner
2017-08-25 10:31 ` [patch 05/41] x86/tracing: Introduce a static key for exception tracing Thomas Gleixner
2017-08-25 15:24 ` Steven Rostedt
2017-08-25 10:31 ` [patch 06/41] x86/traps: Simplify pagefault tracing logic Thomas Gleixner
2017-08-25 14:59 ` Andy Lutomirski
2017-08-25 15:10 ` Steven Rostedt
2017-08-25 15:11 ` Thomas Gleixner
2017-08-25 10:31 ` [patch 07/41] x86/apic: Remove the duplicated tracing version of local_timer_interrupt Thomas Gleixner
2017-08-25 15:29 ` Steven Rostedt
2017-08-25 10:31 ` [patch 08/41] x86/apic: Use this_cpu_ptr in local_timer_interrupt Thomas Gleixner
2017-08-25 10:31 ` [patch 09/41] x86/irq: Get rid of duplicated trace_x86_platform_ipi() code Thomas Gleixner
2017-08-25 15:41 ` Steven Rostedt
2017-08-25 10:31 ` [patch 10/41] x86/apic: Remove the duplicated tracing versions of interrupts Thomas Gleixner
2017-08-25 15:49 ` Steven Rostedt
2017-08-28 7:50 ` Peter Zijlstra
2017-08-25 10:31 ` [patch 11/41] x86/irqwork: Get rid of duplicated tracing interrupt code Thomas Gleixner
2017-08-25 15:55 ` Steven Rostedt
2017-08-25 10:31 ` [patch 12/41] x86/mce: Remove " Thomas Gleixner
2017-08-25 16:09 ` Steven Rostedt
2017-08-25 16:13 ` Thomas Gleixner
2017-08-25 10:31 ` [patch 13/41] x86/smp: Remove pointless duplicated " Thomas Gleixner
2017-08-25 18:24 ` Steven Rostedt
2017-08-25 10:31 ` [patch 14/41] x86/smp: Use static key for reschedule interrupt tracing Thomas Gleixner
2017-08-25 19:40 ` Steven Rostedt
2017-08-25 20:07 ` Thomas Gleixner [this message]
2017-08-25 21:10 ` Steven Rostedt
2017-08-25 10:31 ` [patch 15/41] x86/idt: Remove tracing idt completely Thomas Gleixner
2017-08-25 21:18 ` Steven Rostedt
2017-08-25 21:22 ` Thomas Gleixner
2017-08-25 21:29 ` Thomas Gleixner
2017-08-25 10:31 ` [patch 16/41] x86/tracing: Disentangle pagefault and resched IPI tracing key Thomas Gleixner
2017-08-25 10:31 ` [patch 17/41] x86/idt: Unify gate_struct handling for 32/64bit Thomas Gleixner
2017-08-25 10:31 ` [patch 18/41] x86/percpu: Use static initializer for GDT entry Thomas Gleixner
2017-08-25 10:31 ` [patch 19/41] x86/fpu: Use bitfield accessors for desc_struct Thomas Gleixner
2017-08-25 10:31 ` [patch 20/41] x86: Replace access to desc_struct:a/b fields Thomas Gleixner
2017-08-25 11:14 ` Juergen Gross
2017-08-25 13:56 ` Boris Ostrovsky
2017-08-25 14:02 ` Thomas Gleixner
2017-08-25 14:16 ` Steven Rostedt
2017-08-25 10:31 ` [patch 21/41] x86/gdt: Use bitfields for initialization Thomas Gleixner
2017-08-25 10:31 ` [patch 22/41] x86/ldttss: Cleanup 32bit descriptors Thomas Gleixner
2017-08-25 10:31 ` [patch 23/41] x86/idt: Create file for IDT related code Thomas Gleixner
2017-08-25 10:31 ` [patch 24/41] x86/idt: Move 32bit idt_descr to C code Thomas Gleixner
2017-08-25 10:31 ` [patch 25/41] x86/idt: Remove unused set_trap_gate() Thomas Gleixner
2017-08-25 10:31 ` [patch 26/41] x86/idt: Consolidate IDT invalidation Thomas Gleixner
2017-08-25 10:31 ` [patch 27/41] x86/idt: Move early IDT handler setup to IDT code Thomas Gleixner
2017-08-25 10:31 ` [patch 28/41] x86/idt: Move early IDT setup out of 32bit asm Thomas Gleixner
2017-08-25 10:31 ` [patch 29/41] x86/idt: Prepare for table based init Thomas Gleixner
2017-08-25 10:31 ` [patch 30/41] x86/idt: Switch early trap init to IDT tables Thomas Gleixner
2017-08-25 14:57 ` Andy Lutomirski
2017-08-25 10:31 ` [patch 31/41] x86/idt: Move debug stack init to table based Thomas Gleixner
2017-08-25 10:31 ` [patch 32/41] x86/idt: Move ist stack based traps to table init Thomas Gleixner
2017-08-25 10:31 ` [patch 33/41] x86/idt: Move regular trap init to tables Thomas Gleixner
2017-08-25 10:31 ` [patch 34/41] x86/idt: Move APIC gate initialization " Thomas Gleixner
2017-08-25 10:31 ` [patch 35/41] x86/idt: Move interrupt gate initialization to IDT code Thomas Gleixner
2017-08-25 10:31 ` [patch 36/41] x86/idt: Remove unused functions/inlines Thomas Gleixner
2017-08-25 10:31 ` [patch 37/41] x86/irq: Get rid of the first_system_vector bogisity Thomas Gleixner
2017-08-25 10:31 ` [patch 38/41] x86/idt: Deinline setup functions Thomas Gleixner
2017-08-25 10:31 ` [patch 39/41] x86/idt: Simplify alloc_intr_gate Thomas Gleixner
2017-08-25 11:16 ` Juergen Gross
2017-08-25 10:31 ` [patch 40/41] x86/idt: Hide set_intr_gate() Thomas Gleixner
2017-08-25 10:56 ` Paolo Bonzini
2017-08-25 10:31 ` [patch 41/41] x86/irq: Remove duplicated used_vectors definition Thomas Gleixner
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=alpine.DEB.2.20.1708252202410.2124@nanos \
--to=tglx@linutronix.de \
--cc=bp@alien8.de \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.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
all inboxes | Powered by JetHome®