2016-10-06 11:27:49

by Markus Trippelsdorf

[permalink] [raw]
Subject: Bogus "APIC: NR_CPUS/possible_cpus limit of 4 reached" messages

On current trunk I get during boot:

[ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 4/0x84 ignored.
[ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 5/0x85 ignored.

I don't think these messages make much sense on a 4-core machine.

--
Markus


2016-10-06 11:48:31

by Alan Cox

[permalink] [raw]
Subject: Re: Bogus "APIC: NR_CPUS/possible_cpus limit of 4 reached" messages

On Thu, 6 Oct 2016 13:27:37 +0200
Markus Trippelsdorf <[email protected]> wrote:

> On current trunk I get during boot:
>
> [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 4/0x84 ignored.
> [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 5/0x85 ignored.
>
> I don't think these messages make much sense on a 4-core machine.
>

Four cores with or without hyperthreading ?

Alan

2016-10-06 11:52:43

by Markus Trippelsdorf

[permalink] [raw]
Subject: Re: Bogus "APIC: NR_CPUS/possible_cpus limit of 4 reached" messages

On 2016.10.06 at 12:48 +0100, One Thousand Gnomes wrote:
> On Thu, 6 Oct 2016 13:27:37 +0200
> Markus Trippelsdorf <[email protected]> wrote:
>
> > On current trunk I get during boot:
> >
> > [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 4/0x84 ignored.
> > [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 5/0x85 ignored.
> >
> > I don't think these messages make much sense on a 4-core machine.
> >
>
> Four cores with or without hyperthreading ?

Without. This is a rather old AMD machine (AMD Phenom II X4 955).

--
Markus

2016-10-07 13:26:44

by Markus Trippelsdorf

[permalink] [raw]
Subject: Re: Bogus "APIC: NR_CPUS/possible_cpus limit of 4 reached" messages

On 2016.10.06 at 13:52 +0200, Markus Trippelsdorf wrote:
> On 2016.10.06 at 12:48 +0100, One Thousand Gnomes wrote:
> > On Thu, 6 Oct 2016 13:27:37 +0200
> > Markus Trippelsdorf <[email protected]> wrote:
> >
> > > On current trunk I get during boot:
> > >
> > > [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 4/0x84 ignored.
> > > [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 5/0x85 ignored.
> > >
> > > I don't think these messages make much sense on a 4-core machine.
> > >
> >
> > Four cores with or without hyperthreading ?
>
> Without. This is a rather old AMD machine (AMD Phenom II X4 955).

CCing more people.

--
Markus

2016-10-07 13:58:40

by Thomas Gleixner

[permalink] [raw]
Subject: Re: Bogus "APIC: NR_CPUS/possible_cpus limit of 4 reached" messages

On Fri, 7 Oct 2016, Markus Trippelsdorf wrote:
> On 2016.10.06 at 13:52 +0200, Markus Trippelsdorf wrote:
> > On 2016.10.06 at 12:48 +0100, One Thousand Gnomes wrote:
> > > On Thu, 6 Oct 2016 13:27:37 +0200
> > > Markus Trippelsdorf <[email protected]> wrote:
> > >
> > > > On current trunk I get during boot:
> > > >
> > > > [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 4/0x84 ignored.
> > > > [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 5/0x85 ignored.
> > > >
> > > > I don't think these messages make much sense on a 4-core machine.
> > > >
> > >
> > > Four cores with or without hyperthreading ?
> >
> > Without. This is a rather old AMD machine (AMD Phenom II X4 955).

That's due to the recent cpuid -> nodeid changes. The patch below should
fix it.

Thanks,

tglx

diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index f266b8a92a9e..88c657b057e2 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -2128,9 +2128,11 @@ int __generic_processor_info(int apicid, int version, bool enabled)
if (num_processors >= nr_cpu_ids) {
int thiscpu = max + disabled_cpus;

- pr_warning(
- "APIC: NR_CPUS/possible_cpus limit of %i reached."
- " Processor %d/0x%x ignored.\n", max, thiscpu, apicid);
+ if (enabled) {
+ pr_warning("APIC: NR_CPUS/possible_cpus limit of %i "
+ "reached. Processor %d/0x%x ignored.\n",
+ max, thiscpu, apicid);
+ }

disabled_cpus++;
return -EINVAL;

2016-10-07 14:09:33

by Markus Trippelsdorf

[permalink] [raw]
Subject: Re: Bogus "APIC: NR_CPUS/possible_cpus limit of 4 reached" messages

On 2016.10.07 at 15:55 +0200, Thomas Gleixner wrote:
> On Fri, 7 Oct 2016, Markus Trippelsdorf wrote:
> > On 2016.10.06 at 13:52 +0200, Markus Trippelsdorf wrote:
> > > On 2016.10.06 at 12:48 +0100, One Thousand Gnomes wrote:
> > > > On Thu, 6 Oct 2016 13:27:37 +0200
> > > > Markus Trippelsdorf <[email protected]> wrote:
> > > >
> > > > > On current trunk I get during boot:
> > > > >
> > > > > [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 4/0x84 ignored.
> > > > > [ 0.000000] APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 5/0x85 ignored.
> > > > >
> > > > > I don't think these messages make much sense on a 4-core machine.
> > > > >
> > > >
> > > > Four cores with or without hyperthreading ?
> > >
> > > Without. This is a rather old AMD machine (AMD Phenom II X4 955).
>
> That's due to the recent cpuid -> nodeid changes. The patch below should
> fix it.

Indeed it does. Thanks for the quick fix.

--
Markus

Subject: [tip:x86/urgent] x86/apic: Prevent pointless warning messages

Commit-ID: df610d678893c85b82d3a68eea0d87dd4e03e615
Gitweb: http://git.kernel.org/tip/df610d678893c85b82d3a68eea0d87dd4e03e615
Author: Thomas Gleixner <[email protected]>
AuthorDate: Fri, 7 Oct 2016 15:55:13 +0200
Committer: Thomas Gleixner <[email protected]>
CommitDate: Sat, 8 Oct 2016 12:18:36 +0200

x86/apic: Prevent pointless warning messages

Markus reported that he sees new warnings:

APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 4/0x84 ignored.
APIC: NR_CPUS/possible_cpus limit of 4 reached. Processor 5/0x85 ignored.

This comes from the recent persistant cpuid - nodeid changes. The code
which emits the warning has been called prior to these changes only for
enabled processors. Now it's called for disabled processors as well to get
the possible cpu accounting correct. So if the kernel is compiled for the
number of actual available/enabled CPUs and the BIOS reports disabled CPUs
as well then the above warnings are printed.

That's a pointless exercise as it only makes sense if there are more CPUs
enabled than the kernel supports.

Nake the warning conditional on enabled processors so we are back to the
state before these changes.

Fixes: 8f54969dc8d6 ("x86/acpi: Introduce persistent storage for cpuid <-> apicid mapping")
Reported-and-tested-by: Markus Trippelsdorf <[email protected]>
Cc: One Thousand Gnomes <[email protected]>
Cc: Dou Liyang <[email protected]>
Cc: [email protected]
Cc: Gu Zheng <[email protected]>
Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1610071549330.19804@nanos
Signed-off-by: Thomas Gleixner <[email protected]>

---
arch/x86/kernel/apic/apic.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index f266b8a..88c657b 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -2128,9 +2128,11 @@ int __generic_processor_info(int apicid, int version, bool enabled)
if (num_processors >= nr_cpu_ids) {
int thiscpu = max + disabled_cpus;

- pr_warning(
- "APIC: NR_CPUS/possible_cpus limit of %i reached."
- " Processor %d/0x%x ignored.\n", max, thiscpu, apicid);
+ if (enabled) {
+ pr_warning("APIC: NR_CPUS/possible_cpus limit of %i "
+ "reached. Processor %d/0x%x ignored.\n",
+ max, thiscpu, apicid);
+ }

disabled_cpus++;
return -EINVAL;