2018-09-27 05:12:02

by Chunyan Zhang

[permalink] [raw]
Subject: [RESEND PATCH] ARM: prevent tracing IPI_CPU_BACKTRACE

From: Arnd Bergmann <[email protected]>

When function tracing for IPIs is enabled, we get a warning for an
overflow of the ipi_types array with the IPI_CPU_BACKTRACE type
as triggered by raise_nmi():

arch/arm/kernel/smp.c: In function 'raise_nmi':
arch/arm/kernel/smp.c:489:2: error: array subscript is above array bounds [-Werror=array-bounds]
trace_ipi_raise(target, ipi_types[ipinr]);

This is a correct warning as we actually overflow the array here.

This patch raise_nmi() to call __smp_cross_call() instead of
smp_cross_call(), to avoid calling into ftrace. For clarification,
I'm also adding a two new code comments describing how this one
is special.

The warning appears to have shown up after patch e7273ff49acf
("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI"), which
changed the number assignment from '15' to '8', but as far as I can
tell has existed since the IPI tracepoints were first introduced.
If we decide to backport this patch to stable kernels, we probably
need to backport e7273ff49acf as well.

Resubmiting this patch is because that I found coverity is complaining
the issue this patch fixed, and also I got the traces like below:
"ipi_raise: target_mask=00000001 (machine_suspend)" which actually was
the TPS of suspend_resume[1] rather than ipi_raise.

[1]
https://elixir.bootlin.com/linux/latest/source/kernel/power/suspend.c#L80

Signed-off-by: Arnd Bergmann <[email protected]>
Fixes: e7273ff49acf ("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI")
Fixes: 365ec7b17327 ("ARM: add IPI tracepoints") # v3.17
Signed-off-by: Chunyan Zhang <[email protected]>
---
arch/arm/include/asm/hardirq.h | 1 +
arch/arm/kernel/smp.c | 6 +++++-
2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/arch/arm/include/asm/hardirq.h b/arch/arm/include/asm/hardirq.h
index cba23ea..7a88f16 100644
--- a/arch/arm/include/asm/hardirq.h
+++ b/arch/arm/include/asm/hardirq.h
@@ -6,6 +6,7 @@
#include <linux/threads.h>
#include <asm/irq.h>

+/* number of IPIS _not_ including IPI_CPU_BACKTRACE */
#define NR_IPI 7

typedef struct {
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 0978282..123be77 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -75,6 +75,10 @@ enum ipi_msg_type {
IPI_CPU_STOP,
IPI_IRQ_WORK,
IPI_COMPLETION,
+ /*
+ * CPU_BACKTRACE is special and not included in NR_IPI
+ * or tracable with trace_ipi_*
+ */
IPI_CPU_BACKTRACE,
/*
* SGI8-15 can be reserved by secure firmware, and thus may
@@ -755,7 +759,7 @@ core_initcall(register_cpufreq_notifier);

static void raise_nmi(cpumask_t *mask)
{
- smp_cross_call(mask, IPI_CPU_BACKTRACE);
+ _smp_cross_call(mask, IPI_CPU_BACKTRACE);
}

void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
--
2.7.4



2018-10-09 02:21:40

by Chunyan Zhang

[permalink] [raw]
Subject: Re: [RESEND PATCH] ARM: prevent tracing IPI_CPU_BACKTRACE

Hi All,

If there's no comments, should I submit this patch on RMK's Patch system?

Thanks,
Chunyan

On 27 September 2018 at 11:41, Chunyan Zhang <[email protected]> wrote:
> From: Arnd Bergmann <[email protected]>
>
> When function tracing for IPIs is enabled, we get a warning for an
> overflow of the ipi_types array with the IPI_CPU_BACKTRACE type
> as triggered by raise_nmi():
>
> arch/arm/kernel/smp.c: In function 'raise_nmi':
> arch/arm/kernel/smp.c:489:2: error: array subscript is above array bounds [-Werror=array-bounds]
> trace_ipi_raise(target, ipi_types[ipinr]);
>
> This is a correct warning as we actually overflow the array here.
>
> This patch raise_nmi() to call __smp_cross_call() instead of
> smp_cross_call(), to avoid calling into ftrace. For clarification,
> I'm also adding a two new code comments describing how this one
> is special.
>
> The warning appears to have shown up after patch e7273ff49acf
> ("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI"), which
> changed the number assignment from '15' to '8', but as far as I can
> tell has existed since the IPI tracepoints were first introduced.
> If we decide to backport this patch to stable kernels, we probably
> need to backport e7273ff49acf as well.
>
> Resubmiting this patch is because that I found coverity is complaining
> the issue this patch fixed, and also I got the traces like below:
> "ipi_raise: target_mask=00000001 (machine_suspend)" which actually was
> the TPS of suspend_resume[1] rather than ipi_raise.
>
> [1]
> https://elixir.bootlin.com/linux/latest/source/kernel/power/suspend.c#L80
>
> Signed-off-by: Arnd Bergmann <[email protected]>
> Fixes: e7273ff49acf ("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI")
> Fixes: 365ec7b17327 ("ARM: add IPI tracepoints") # v3.17
> Signed-off-by: Chunyan Zhang <[email protected]>
> ---
> arch/arm/include/asm/hardirq.h | 1 +
> arch/arm/kernel/smp.c | 6 +++++-
> 2 files changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm/include/asm/hardirq.h b/arch/arm/include/asm/hardirq.h
> index cba23ea..7a88f16 100644
> --- a/arch/arm/include/asm/hardirq.h
> +++ b/arch/arm/include/asm/hardirq.h
> @@ -6,6 +6,7 @@
> #include <linux/threads.h>
> #include <asm/irq.h>
>
> +/* number of IPIS _not_ including IPI_CPU_BACKTRACE */
> #define NR_IPI 7
>
> typedef struct {
> diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
> index 0978282..123be77 100644
> --- a/arch/arm/kernel/smp.c
> +++ b/arch/arm/kernel/smp.c
> @@ -75,6 +75,10 @@ enum ipi_msg_type {
> IPI_CPU_STOP,
> IPI_IRQ_WORK,
> IPI_COMPLETION,
> + /*
> + * CPU_BACKTRACE is special and not included in NR_IPI
> + * or tracable with trace_ipi_*
> + */
> IPI_CPU_BACKTRACE,
> /*
> * SGI8-15 can be reserved by secure firmware, and thus may
> @@ -755,7 +759,7 @@ core_initcall(register_cpufreq_notifier);
>
> static void raise_nmi(cpumask_t *mask)
> {
> - smp_cross_call(mask, IPI_CPU_BACKTRACE);
> + _smp_cross_call(mask, IPI_CPU_BACKTRACE);
> }
>
> void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
> --
> 2.7.4
>

2018-10-18 03:47:51

by Chunyan Zhang

[permalink] [raw]
Subject: Re: [RESEND PATCH] ARM: prevent tracing IPI_CPU_BACKTRACE

On Thu, 27 Sep 2018 at 11:42, Chunyan Zhang <[email protected]> wrote:
>
> From: Arnd Bergmann <[email protected]>
>
> When function tracing for IPIs is enabled, we get a warning for an
> overflow of the ipi_types array with the IPI_CPU_BACKTRACE type
> as triggered by raise_nmi():
>
> arch/arm/kernel/smp.c: In function 'raise_nmi':
> arch/arm/kernel/smp.c:489:2: error: array subscript is above array bounds [-Werror=array-bounds]
> trace_ipi_raise(target, ipi_types[ipinr]);
>
> This is a correct warning as we actually overflow the array here.
>
> This patch raise_nmi() to call __smp_cross_call() instead of
> smp_cross_call(), to avoid calling into ftrace. For clarification,
> I'm also adding a two new code comments describing how this one
> is special.
>
> The warning appears to have shown up after patch e7273ff49acf
> ("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI"), which
> changed the number assignment from '15' to '8', but as far as I can
> tell has existed since the IPI tracepoints were first introduced.
> If we decide to backport this patch to stable kernels, we probably
> need to backport e7273ff49acf as well.
>
> Resubmiting this patch is because that I found coverity is complaining
> the issue this patch fixed, and also I got the traces like below:
> "ipi_raise: target_mask=00000001 (machine_suspend)" which actually was
> the TPS of suspend_resume[1] rather than ipi_raise.
>
> [1]
> https://elixir.bootlin.com/linux/latest/source/kernel/power/suspend.c#L80
>
> Signed-off-by: Arnd Bergmann <[email protected]>
> Fixes: e7273ff49acf ("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI")
> Fixes: 365ec7b17327 ("ARM: add IPI tracepoints") # v3.17
> Signed-off-by: Chunyan Zhang <[email protected]>
> ---
> arch/arm/include/asm/hardirq.h | 1 +
> arch/arm/kernel/smp.c | 6 +++++-
> 2 files changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm/include/asm/hardirq.h b/arch/arm/include/asm/hardirq.h
> index cba23ea..7a88f16 100644
> --- a/arch/arm/include/asm/hardirq.h
> +++ b/arch/arm/include/asm/hardirq.h
> @@ -6,6 +6,7 @@
> #include <linux/threads.h>
> #include <asm/irq.h>
>
> +/* number of IPIS _not_ including IPI_CPU_BACKTRACE */
> #define NR_IPI 7
>
> typedef struct {
> diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
> index 0978282..123be77 100644
> --- a/arch/arm/kernel/smp.c
> +++ b/arch/arm/kernel/smp.c
> @@ -75,6 +75,10 @@ enum ipi_msg_type {
> IPI_CPU_STOP,
> IPI_IRQ_WORK,
> IPI_COMPLETION,
> + /*
> + * CPU_BACKTRACE is special and not included in NR_IPI
> + * or tracable with trace_ipi_*
> + */
> IPI_CPU_BACKTRACE,
> /*
> * SGI8-15 can be reserved by secure firmware, and thus may
> @@ -755,7 +759,7 @@ core_initcall(register_cpufreq_notifier);
>
> static void raise_nmi(cpumask_t *mask)
> {
> - smp_cross_call(mask, IPI_CPU_BACKTRACE);
> + _smp_cross_call(mask, IPI_CPU_BACKTRACE);

Here should be double underline, i.e.

> + __smp_cross_call(mask, IPI_CPU_BACKTRACE);

> }
>
> void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
> --
> 2.7.4
>

2018-10-18 08:26:05

by Chunyan Zhang

[permalink] [raw]
Subject: [PATCH] ARM: prevent tracing IPI_CPU_BACKTRACE

From: Arnd Bergmann <[email protected]>

When function tracing for IPIs is enabled, we get a warning for an
overflow of the ipi_types array with the IPI_CPU_BACKTRACE type
as triggered by raise_nmi():

arch/arm/kernel/smp.c: In function 'raise_nmi':
arch/arm/kernel/smp.c:489:2: error: array subscript is above array bounds [-Werror=array-bounds]
trace_ipi_raise(target, ipi_types[ipinr]);

This is a correct warning as we actually overflow the array here.

This patch raise_nmi() to call __smp_cross_call() instead of
smp_cross_call(), to avoid calling into ftrace. For clarification,
I'm also adding a two new code comments describing how this one
is special.

The warning appears to have shown up after patch e7273ff49acf
("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI"), which
changed the number assignment from '15' to '8', but as far as I can
tell has existed since the IPI tracepoints were first introduced.
If we decide to backport this patch to stable kernels, we probably
need to backport e7273ff49acf as well.

Resubmitting this patch is because that I found coverity is complaining
the issue this patch fixed, and also I got the traces like below:
"ipi_raise: target_mask=00000001 (machine_suspend)" which actually was
the TPS of suspend_resume[1] rather that ipi_raise.

[1]
https://elixir.bootlin.com/linux/latest/source/kernel/power/suspend.c#L80

Signed-off-by: Arnd Bergmann <[email protected]>
Fixes: e7273ff49acf ("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI")
Fixes: 365ec7b17327 ("ARM: add IPI tracepoints") # v3.17
Signed-off-by: Chunyan Zhang <[email protected]>
---
arch/arm/include/asm/hardirq.h | 1 +
arch/arm/kernel/smp.c | 6 +++++-
2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/arch/arm/include/asm/hardirq.h b/arch/arm/include/asm/hardirq.h
index cba23ea..7a88f16 100644
--- a/arch/arm/include/asm/hardirq.h
+++ b/arch/arm/include/asm/hardirq.h
@@ -6,6 +6,7 @@
#include <linux/threads.h>
#include <asm/irq.h>

+/* number of IPIS _not_ including IPI_CPU_BACKTRACE */
#define NR_IPI 7

typedef struct {
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 0978282..ddd48e2 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -75,6 +75,10 @@ enum ipi_msg_type {
IPI_CPU_STOP,
IPI_IRQ_WORK,
IPI_COMPLETION,
+ /*
+ * CPU_BACKTRACE is special and not included in NR_IPI
+ * or tracable with trace_ipi_*
+ */
IPI_CPU_BACKTRACE,
/*
* SGI8-15 can be reserved by secure firmware, and thus may
@@ -755,7 +759,7 @@ core_initcall(register_cpufreq_notifier);

static void raise_nmi(cpumask_t *mask)
{
- smp_cross_call(mask, IPI_CPU_BACKTRACE);
+ __smp_cross_call(mask, IPI_CPU_BACKTRACE);
}

void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
--
2.7.4


2019-01-23 16:50:37

by Julien Thierry

[permalink] [raw]
Subject: Re: [PATCH] ARM: prevent tracing IPI_CPU_BACKTRACE

Hi,

On 18/10/2018 09:23, Chunyan Zhang wrote:
> From: Arnd Bergmann <[email protected]>
>
> When function tracing for IPIs is enabled, we get a warning for an
> overflow of the ipi_types array with the IPI_CPU_BACKTRACE type
> as triggered by raise_nmi():
>
> arch/arm/kernel/smp.c: In function 'raise_nmi':
> arch/arm/kernel/smp.c:489:2: error: array subscript is above array bounds [-Werror=array-bounds]
> trace_ipi_raise(target, ipi_types[ipinr]);
>
> This is a correct warning as we actually overflow the array here.
>
> This patch raise_nmi() to call __smp_cross_call() instead of
> smp_cross_call(), to avoid calling into ftrace. For clarification,
> I'm also adding a two new code comments describing how this one
> is special.
>
> The warning appears to have shown up after patch e7273ff49acf
> ("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI"), which
> changed the number assignment from '15' to '8', but as far as I can
> tell has existed since the IPI tracepoints were first introduced.
> If we decide to backport this patch to stable kernels, we probably
> need to backport e7273ff49acf as well.
>
> Resubmitting this patch is because that I found coverity is complaining
> the issue this patch fixed, and also I got the traces like below:
> "ipi_raise: target_mask=00000001 (machine_suspend)" which actually was
> the TPS of suspend_resume[1] rather that ipi_raise.
>

What's the status on this patch? it does not seem to be in v5.0-rc3 nor
linux-next and looks like an actual issue.

Also, I'm assuming the patch is avoiding the tracing of
IPI_CPU_BACKTRACE because the tracing is not NMI safe. It would be good
to have a comment about that either near NR_IPI or ipi_types stating why
this IPI is excluded.

Thanks,

> [1]
> https://elixir.bootlin.com/linux/latest/source/kernel/power/suspend.c#L80
>
> Signed-off-by: Arnd Bergmann <[email protected]>
> Fixes: e7273ff49acf ("ARM: 8488/1: Make IPI_CPU_BACKTRACE a "non-secure" SGI")
> Fixes: 365ec7b17327 ("ARM: add IPI tracepoints") # v3.17
> Signed-off-by: Chunyan Zhang <[email protected]>
> ---
> arch/arm/include/asm/hardirq.h | 1 +
> arch/arm/kernel/smp.c | 6 +++++-
> 2 files changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm/include/asm/hardirq.h b/arch/arm/include/asm/hardirq.h
> index cba23ea..7a88f16 100644
> --- a/arch/arm/include/asm/hardirq.h
> +++ b/arch/arm/include/asm/hardirq.h
> @@ -6,6 +6,7 @@
> #include <linux/threads.h>
> #include <asm/irq.h>
>
> +/* number of IPIS _not_ including IPI_CPU_BACKTRACE */
> #define NR_IPI 7
>
> typedef struct {
> diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
> index 0978282..ddd48e2 100644
> --- a/arch/arm/kernel/smp.c
> +++ b/arch/arm/kernel/smp.c
> @@ -75,6 +75,10 @@ enum ipi_msg_type {
> IPI_CPU_STOP,
> IPI_IRQ_WORK,
> IPI_COMPLETION,
> + /*
> + * CPU_BACKTRACE is special and not included in NR_IPI
> + * or tracable with trace_ipi_*
> + */
> IPI_CPU_BACKTRACE,
> /*
> * SGI8-15 can be reserved by secure firmware, and thus may
> @@ -755,7 +759,7 @@ core_initcall(register_cpufreq_notifier);
>
> static void raise_nmi(cpumask_t *mask)
> {
> - smp_cross_call(mask, IPI_CPU_BACKTRACE);
> + __smp_cross_call(mask, IPI_CPU_BACKTRACE);
> }
>
> void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
>

--
Julien Thierry