Re: [PATCH v5 7/7] sched, smp: Trace smp callback causing an IPI
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
- To: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
- Subject: Re: [PATCH v5 7/7] sched, smp: Trace smp callback causing an IPI
- From: Valentin Schneider <vschneid@xxxxxxxxxx>
- Date: Thu, 23 Mar 2023 16:25:25 +0000
- Cc: linux-alpha@xxxxxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, linux-snps-arc@xxxxxxxxxxxxxxxxxxx, linux-arm-kernel@xxxxxxxxxxxxxxxxxxx, linux-csky@xxxxxxxxxxxxxxx, linux-hexagon@xxxxxxxxxxxxxxx, linux-ia64@xxxxxxxxxxxxxxx, loongarch@xxxxxxxxxxxxxxx, linux-mips@xxxxxxxxxxxxxxx, openrisc@xxxxxxxxxxxxxxxxxxxx, linux-parisc@xxxxxxxxxxxxxxx, linuxppc-dev@xxxxxxxxxxxxxxxx, linux-riscv@xxxxxxxxxxxxxxxxxxx, linux-s390@xxxxxxxxxxxxxxx, linux-sh@xxxxxxxxxxxxxxx, sparclinux@xxxxxxxxxxxxxxx, linux-xtensa@xxxxxxxxxxxxxxxx, x86@xxxxxxxxxx, "Paul E. McKenney" <paulmck@xxxxxxxxxx>, Steven Rostedt <rostedt@xxxxxxxxxxx>, Thomas Gleixner <tglx@xxxxxxxxxxxxx>, Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>, Juri Lelli <juri.lelli@xxxxxxxxxx>, Daniel Bristot de Oliveira <bristot@xxxxxxxxxx>, Marcelo Tosatti <mtosatti@xxxxxxxxxx>, Frederic Weisbecker <frederic@xxxxxxxxxx>, Ingo Molnar <mingo@xxxxxxxxxx>, Borislav Petkov <bp@xxxxxxxxx>, Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>, "H. Peter Anvin" <hpa@xxxxxxxxx>, Marc Zyngier <maz@xxxxxxxxxx>, Mark Rutland <mark.rutland@xxxxxxx>, Russell King <linux@xxxxxxxxxxxxxxx>, Nicholas Piggin <npiggin@xxxxxxxxx>, Guo Ren <guoren@xxxxxxxxxx>, "David S. Miller" <davem@xxxxxxxxxxxxx>
- In-reply-to: <20230322140434.GC2357380@hirez.programming.kicks-ass.net>
- References: <20230307143558.294354-1-vschneid@redhat.com> <20230307143558.294354-8-vschneid@redhat.com> <20230322095329.GS2017917@hirez.programming.kicks-ass.net> <xhsmhmt45c703.mognet@vschneid.remote.csb> <20230322140434.GC2357380@hirez.programming.kicks-ass.net>
On 22/03/23 15:04, Peter Zijlstra wrote:
> @@ -798,14 +794,20 @@ static void smp_call_function_many_cond(
> }
>
> /*
> + * Trace each smp_function_call_*() as an IPI, actual IPIs
> + * will be traced with func==generic_smp_call_function_single_ipi().
> + */
> + trace_ipi_send_cpumask(cfd->cpumask_ipi, _RET_IP_, func);
I just got a trace pointing out this can emit an event even though no IPI
is sent if e.g. the cond_func predicate filters all CPUs in the argument
mask:
ipi_send_cpumask: cpumask= callsite=on_each_cpu_cond_mask+0x3c callback=flush_tlb_func+0x0
Maybe something like so on top?
---
diff --git a/kernel/smp.c b/kernel/smp.c
index ba5478814e677..1dc452017d000 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -791,6 +791,8 @@ static void smp_call_function_many_cond(const struct cpumask *mask,
}
}
+ if (!nr_cpus)
+ goto local;
/*
* Trace each smp_function_call_*() as an IPI, actual IPIs
* will be traced with func==generic_smp_call_function_single_ipi().
@@ -804,10 +806,10 @@ static void smp_call_function_many_cond(const struct cpumask *mask,
*/
if (nr_cpus == 1)
send_call_function_single_ipi(last_cpu);
- else if (likely(nr_cpus > 1))
+ else
send_call_function_ipi_mask(cfd->cpumask_ipi);
}
-
+local:
if (run_local && (!cond_func || cond_func(this_cpu, info))) {
unsigned long flags;
[Index of Archives]
[Linux Kernel]
[Sparc Linux]
[DCCP]
[Linux ARM]
[Yosemite News]
[Linux SCSI]
[Linux x86_64]
[Linux for Ham Radio]