[PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Wen Yang]

From: Peter Zijlstra <peterz@infradead.org>

commit ab3e9c0b75dcb13e9254ef68caa7f15bc66b6471 upstream.

CAT has happened, WBINDV is bad (even before CAT blowing away the
entire cache on a multi-core platform wasn't nice), try not to use it
ever.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Dave Hansen <dave.hansen@intel.com>
Cc: Bin Yang <bin.yang@intel.com>
Cc: Mark Gross <mark.gross@intel.com>
Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
Cc: <stable@vger.kernel.org> # 4.19.x
Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
---
 arch/x86/mm/pageattr.c | 18 ++----------------
 1 file changed, 2 insertions(+), 16 deletions(-)

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 101f3ad0d6ad..ab87da7a6043 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -239,26 +239,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
 			    int in_flags, struct page **pages)
 {
 	unsigned int i, level;
-#ifdef CONFIG_PREEMPT
-	/*
-	 * Avoid wbinvd() because it causes latencies on all CPUs,
-	 * regardless of any CPU isolation that may be in effect.
-	 *
-	 * This should be extended for CAT enabled systems independent of
-	 * PREEMPT because wbinvd() does not respect the CAT partitions and
-	 * this is exposed to unpriviledged users through the graphics
-	 * subsystem.
-	 */
-	unsigned long do_wbinvd = 0;
-#else
-	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
-#endif
 
 	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
 
-	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
+	flush_tlb_all();
 
-	if (!cache || do_wbinvd)
+	if (!cache)
 		return;
 
 	/*
-- 
2.19.1.6.gb485710b

Re: [PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Greg Kroah-Hartman]

On Mon, Jul 04, 2022 at 04:28:17PM +0800, Wen Yang wrote:
> From: Peter Zijlstra <peterz@infradead.org>
> 
> commit ab3e9c0b75dcb13e9254ef68caa7f15bc66b6471 upstream.

This commit id is not in Linus's tree :(

Re: [PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Wen Yang]

在 2022/7/4 下午7:05, Greg Kroah-Hartman 写道:
> On Mon, Jul 04, 2022 at 04:28:17PM +0800, Wen Yang wrote:
>> From: Peter Zijlstra <peterz@infradead.org>
>>
>> commit ab3e9c0b75dcb13e9254ef68caa7f15bc66b6471 upstream.
> 
> This commit id is not in Linus's tree :(


Sorry, accidentally added the internal id.

It is this commit id in Linus' tree:
ddd07b750382adc2b78fdfbec47af8a6e0d8ef37

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ddd07b750382adc2b78fdfbec47af8a6e0d8ef37


--
Best wishes,
Wen

Re: [PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Greg Kroah-Hartman]

On Mon, Jul 04, 2022 at 10:29:57PM +0800, Wen Yang wrote:
> 
> 
> 在 2022/7/4 下午7:05, Greg Kroah-Hartman 写道:
> > On Mon, Jul 04, 2022 at 04:28:17PM +0800, Wen Yang wrote:
> > > From: Peter Zijlstra <peterz@infradead.org>
> > > 
> > > commit ab3e9c0b75dcb13e9254ef68caa7f15bc66b6471 upstream.
> > 
> > This commit id is not in Linus's tree :(
> 
> 
> Sorry, accidentally added the internal id.
> 
> It is this commit id in Linus' tree:
> ddd07b750382adc2b78fdfbec47af8a6e0d8ef37
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ddd07b750382adc2b78fdfbec47af8a6e0d8ef37

Please fix up the patch and resend it with this information.

thanks,

greg k-h

[PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Wen Yang]

From: Peter Zijlstra <peterz@infradead.org>

commit ddd07b750382adc2b78fdfbec47af8a6e0d8ef37 upstream.

CAT has happened, WBINDV is bad (even before CAT blowing away the
entire cache on a multi-core platform wasn't nice), try not to use it
ever.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Dave Hansen <dave.hansen@intel.com>
Cc: Bin Yang <bin.yang@intel.com>
Cc: Mark Gross <mark.gross@intel.com>
Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
Cc: <stable@vger.kernel.org> # 4.19.x
Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
---
 arch/x86/mm/pageattr.c | 18 ++----------------
 1 file changed, 2 insertions(+), 16 deletions(-)

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 101f3ad0d6ad..ab87da7a6043 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -239,26 +239,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
 			    int in_flags, struct page **pages)
 {
 	unsigned int i, level;
-#ifdef CONFIG_PREEMPT
-	/*
-	 * Avoid wbinvd() because it causes latencies on all CPUs,
-	 * regardless of any CPU isolation that may be in effect.
-	 *
-	 * This should be extended for CAT enabled systems independent of
-	 * PREEMPT because wbinvd() does not respect the CAT partitions and
-	 * this is exposed to unpriviledged users through the graphics
-	 * subsystem.
-	 */
-	unsigned long do_wbinvd = 0;
-#else
-	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
-#endif
 
 	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
 
-	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
+	flush_tlb_all();
 
-	if (!cache || do_wbinvd)
+	if (!cache)
 		return;
 
 	/*
-- 
2.19.1.6.gb485710b

Re: [PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Greg Kroah-Hartman]

On Mon, Jul 04, 2022 at 11:45:08PM +0800, Wen Yang wrote:
> From: Peter Zijlstra <peterz@infradead.org>
> 
> commit ddd07b750382adc2b78fdfbec47af8a6e0d8ef37 upstream.
> 
> CAT has happened, WBINDV is bad (even before CAT blowing away the
> entire cache on a multi-core platform wasn't nice), try not to use it
> ever.
> 
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
> Reviewed-by: Dave Hansen <dave.hansen@intel.com>
> Cc: Bin Yang <bin.yang@intel.com>
> Cc: Mark Gross <mark.gross@intel.com>
> Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
> Cc: <stable@vger.kernel.org> # 4.19.x
> Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
> ---
>  arch/x86/mm/pageattr.c | 18 ++----------------
>  1 file changed, 2 insertions(+), 16 deletions(-)
> 
> diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
> index 101f3ad0d6ad..ab87da7a6043 100644
> --- a/arch/x86/mm/pageattr.c
> +++ b/arch/x86/mm/pageattr.c
> @@ -239,26 +239,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
>  			    int in_flags, struct page **pages)
>  {
>  	unsigned int i, level;
> -#ifdef CONFIG_PREEMPT
> -	/*
> -	 * Avoid wbinvd() because it causes latencies on all CPUs,
> -	 * regardless of any CPU isolation that may be in effect.
> -	 *
> -	 * This should be extended for CAT enabled systems independent of
> -	 * PREEMPT because wbinvd() does not respect the CAT partitions and
> -	 * this is exposed to unpriviledged users through the graphics
> -	 * subsystem.
> -	 */
> -	unsigned long do_wbinvd = 0;
> -#else
> -	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
> -#endif
>  
>  	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
>  
> -	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
> +	flush_tlb_all();
>  
> -	if (!cache || do_wbinvd)
> +	if (!cache)
>  		return;
>  
>  	/*
> -- 
> 2.19.1.6.gb485710b
> 

Why is this needed on 4.19.y?  What problem does it solve, it looks only
like an optimization, not a bugfix.

And if it's a bugfix, why only 4.19.y, why not older kernels too?

We need more information here please.

thanks,

greg k-h

Re: [PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Wen Yang]

在 2022/7/4 下午11:59, Greg Kroah-Hartman 写道:
> On Mon, Jul 04, 2022 at 11:45:08PM +0800, Wen Yang wrote:
>> From: Peter Zijlstra <peterz@infradead.org>
>>
>> commit ddd07b750382adc2b78fdfbec47af8a6e0d8ef37 upstream.
>>
>> CAT has happened, WBINDV is bad (even before CAT blowing away the
>> entire cache on a multi-core platform wasn't nice), try not to use it
>> ever.
>>
>> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
>> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
>> Reviewed-by: Dave Hansen <dave.hansen@intel.com>
>> Cc: Bin Yang <bin.yang@intel.com>
>> Cc: Mark Gross <mark.gross@intel.com>
>> Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
>> Cc: <stable@vger.kernel.org> # 4.19.x
>> Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
>> ---
>>   arch/x86/mm/pageattr.c | 18 ++----------------
>>   1 file changed, 2 insertions(+), 16 deletions(-)
>>
>> diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
>> index 101f3ad0d6ad..ab87da7a6043 100644
>> --- a/arch/x86/mm/pageattr.c
>> +++ b/arch/x86/mm/pageattr.c
>> @@ -239,26 +239,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
>>   			    int in_flags, struct page **pages)
>>   {
>>   	unsigned int i, level;
>> -#ifdef CONFIG_PREEMPT
>> -	/*
>> -	 * Avoid wbinvd() because it causes latencies on all CPUs,
>> -	 * regardless of any CPU isolation that may be in effect.
>> -	 *
>> -	 * This should be extended for CAT enabled systems independent of
>> -	 * PREEMPT because wbinvd() does not respect the CAT partitions and
>> -	 * this is exposed to unpriviledged users through the graphics
>> -	 * subsystem.
>> -	 */
>> -	unsigned long do_wbinvd = 0;
>> -#else
>> -	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
>> -#endif
>>   
>>   	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
>>   
>> -	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
>> +	flush_tlb_all();
>>   
>> -	if (!cache || do_wbinvd)
>> +	if (!cache)
>>   		return;
>>   
>>   	/*
>> -- 
>> 2.19.1.6.gb485710b
>>
> 
> Why is this needed on 4.19.y?  What problem does it solve, it looks only
> like an optimization, not a bugfix.
> 
> And if it's a bugfix, why only 4.19.y, why not older kernels too?
> 
> We need more information here please.
> 

On a 128-core Intel(R) Xeon(R) Platinum 8369B CPU @ 2.90GHz server, when 
the user program frequently calls nv_alloc_system_pages to allocate 
large memory, it often causes a delay of about 200 milliseconds for the 
entire system. In this way, other latency-sensitive tasks on this system 
are heavily impacted, causing stability issues in large-scale clusters 
as well.

nv_alloc_system_pages
-> _set_memory_array
-> change_page_attr_set_clr
-> cpa_flush_array
-> on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);


This patch can be directly merged into the 4.19 kernel to solve this 
problem, and most of the machines in our production environment are 4.19 
kernels.

We're also happy to apply it to the 4.14 and 4.9 kernels, and send the 
corresponding patches soon, although there are very few such servers in 
our production clusters.


--
Best wishes,
Wen

Re: [PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Dave Hansen]

On 7/4/22 20:45, Wen Yang wrote:> On a 128-core Intel(R) Xeon(R)
Platinum 8369B CPU @ 2.90GHz server,
> when the user program frequently calls nv_alloc_system_pages to
> allocate large memory,

We seem to be repeating the same conversation we had back in November of
last year:

> https://lore.kernel.org/all/9c415df9-9575-8217-03e9-a6bbf20a491a@linux.alibaba.com/T/#u

It's the binary nvidia driver doing unusual stuff again.  Please talk to
the folks you got that driver from.

Re: [PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Greg Kroah-Hartman]

On Tue, Jul 05, 2022 at 11:45:29AM +0800, Wen Yang wrote:
> 
> 
> 在 2022/7/4 下午11:59, Greg Kroah-Hartman 写道:
> > On Mon, Jul 04, 2022 at 11:45:08PM +0800, Wen Yang wrote:
> > > From: Peter Zijlstra <peterz@infradead.org>
> > > 
> > > commit ddd07b750382adc2b78fdfbec47af8a6e0d8ef37 upstream.
> > > 
> > > CAT has happened, WBINDV is bad (even before CAT blowing away the
> > > entire cache on a multi-core platform wasn't nice), try not to use it
> > > ever.
> > > 
> > > Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> > > Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
> > > Reviewed-by: Dave Hansen <dave.hansen@intel.com>
> > > Cc: Bin Yang <bin.yang@intel.com>
> > > Cc: Mark Gross <mark.gross@intel.com>
> > > Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
> > > Cc: <stable@vger.kernel.org> # 4.19.x
> > > Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
> > > ---
> > >   arch/x86/mm/pageattr.c | 18 ++----------------
> > >   1 file changed, 2 insertions(+), 16 deletions(-)
> > > 
> > > diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
> > > index 101f3ad0d6ad..ab87da7a6043 100644
> > > --- a/arch/x86/mm/pageattr.c
> > > +++ b/arch/x86/mm/pageattr.c
> > > @@ -239,26 +239,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
> > >   			    int in_flags, struct page **pages)
> > >   {
> > >   	unsigned int i, level;
> > > -#ifdef CONFIG_PREEMPT
> > > -	/*
> > > -	 * Avoid wbinvd() because it causes latencies on all CPUs,
> > > -	 * regardless of any CPU isolation that may be in effect.
> > > -	 *
> > > -	 * This should be extended for CAT enabled systems independent of
> > > -	 * PREEMPT because wbinvd() does not respect the CAT partitions and
> > > -	 * this is exposed to unpriviledged users through the graphics
> > > -	 * subsystem.
> > > -	 */
> > > -	unsigned long do_wbinvd = 0;
> > > -#else
> > > -	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
> > > -#endif
> > >   	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
> > > -	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
> > > +	flush_tlb_all();
> > > -	if (!cache || do_wbinvd)
> > > +	if (!cache)
> > >   		return;
> > >   	/*
> > > -- 
> > > 2.19.1.6.gb485710b
> > > 
> > 
> > Why is this needed on 4.19.y?  What problem does it solve, it looks only
> > like an optimization, not a bugfix.
> > 
> > And if it's a bugfix, why only 4.19.y, why not older kernels too?
> > 
> > We need more information here please.
> > 
> 
> On a 128-core Intel(R) Xeon(R) Platinum 8369B CPU @ 2.90GHz server, when the
> user program frequently calls nv_alloc_system_pages to allocate large
> memory, it often causes a delay of about 200 milliseconds for the entire
> system. In this way, other latency-sensitive tasks on this system are
> heavily impacted, causing stability issues in large-scale clusters as well.
> 
> nv_alloc_system_pages
> -> _set_memory_array
> -> change_page_attr_set_clr
> -> cpa_flush_array
> -> on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
> 
> 
> This patch can be directly merged into the 4.19 kernel to solve this
> problem, and most of the machines in our production environment are 4.19
> kernels.

Ah.  So what has changed from last year when I rejected this then:
	https://lore.kernel.org/all/9c415df9-9575-8217-03e9-a6bbf20a491a@linux.alibaba.com/T/#m06369d080fa97eda3dd6a8eaf54a8ca2d430b3ab

Please do not try to submit previously-rejected patches, that is very
disingenuous.

greg k-h

[PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Wen Yang]

From: Peter Zijlstra <peterz@infradead.org>

commit ddd07b750382adc2b78fdfbec47af8a6e0d8ef37 upstream.

CAT has happened, WBINDV is bad (even before CAT blowing away the
entire cache on a multi-core platform wasn't nice), try not to use it
ever.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Dave Hansen <dave.hansen@intel.com>
Cc: Bin Yang <bin.yang@intel.com>
Cc: Mark Gross <mark.gross@intel.com>
Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
Cc: <stable@vger.kernel.org> # 4.19.x
Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
---
 arch/x86/mm/pageattr.c | 18 ++----------------
 1 file changed, 2 insertions(+), 16 deletions(-)

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 101f3ad0d6ad..ab87da7a6043 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -239,26 +239,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
 			    int in_flags, struct page **pages)
 {
 	unsigned int i, level;
-#ifdef CONFIG_PREEMPT
-	/*
-	 * Avoid wbinvd() because it causes latencies on all CPUs,
-	 * regardless of any CPU isolation that may be in effect.
-	 *
-	 * This should be extended for CAT enabled systems independent of
-	 * PREEMPT because wbinvd() does not respect the CAT partitions and
-	 * this is exposed to unpriviledged users through the graphics
-	 * subsystem.
-	 */
-	unsigned long do_wbinvd = 0;
-#else
-	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
-#endif
 
 	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
 
-	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
+	flush_tlb_all();
 
-	if (!cache || do_wbinvd)
+	if (!cache)
 		return;
 
 	/*
-- 
2.19.1.6.gb485710b

[PATCH 4.14] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Wen Yang]

From: Peter Zijlstra <peterz@infradead.org>

commit ddd07b750382adc2b78fdfbec47af8a6e0d8ef37 upstream.

CAT has happened, WBINDV is bad (even before CAT blowing away the
entire cache on a multi-core platform wasn't nice), try not to use it
ever.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Dave Hansen <dave.hansen@intel.com>
Cc: Bin Yang <bin.yang@intel.com>
Cc: Mark Gross <mark.gross@intel.com>
Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
Cc: <stable@vger.kernel.org> # 4.14.x: d2479a3: x86/pti: Fix boot problems from Global-bit setting
Cc: <stable@vger.kernel.org> # 4.14.x
Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
---
 arch/x86/mm/pageattr.c | 18 ++----------------
 1 file changed, 2 insertions(+), 16 deletions(-)

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 101f3ad0d6ad..ab87da7a6043 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -239,26 +239,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
 			    int in_flags, struct page **pages)
 {
 	unsigned int i, level;
-#ifdef CONFIG_PREEMPT
-	/*
-	 * Avoid wbinvd() because it causes latencies on all CPUs,
-	 * regardless of any CPU isolation that may be in effect.
-	 *
-	 * This should be extended for CAT enabled systems independent of
-	 * PREEMPT because wbinvd() does not respect the CAT partitions and
-	 * this is exposed to unpriviledged users through the graphics
-	 * subsystem.
-	 */
-	unsigned long do_wbinvd = 0;
-#else
-	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
-#endif
 
 	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
 
-	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
+	flush_tlb_all();
 
-	if (!cache || do_wbinvd)
+	if (!cache)
 		return;
 
 	/*
-- 
2.19.1.6.gb485710b

[PATCH 4.9] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Wen Yang]

From: Wen Yang <simon.wy@alibaba-inc.com>

commit ddd07b750382adc2b78fdfbec47af8a6e0d8ef37 upstream.

CAT has happened, WBINDV is bad (even before CAT blowing away the
entire cache on a multi-core platform wasn't nice), try not to use it
ever.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Dave Hansen <dave.hansen@intel.com>
Cc: Bin Yang <bin.yang@intel.com>
Cc: Mark Gross <mark.gross@intel.com>
Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
Cc: <stable@vger.kernel.org> # 4.9.x: d2479a3: x86/pti: Fix boot problems from Global-bit setting
Cc: <stable@vger.kernel.org> # 4.9.x
Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
---
 arch/x86/mm/pageattr.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index a586faeab0e5..e1b5b603eb48 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -215,13 +215,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
 			    int in_flags, struct page **pages)
 {
 	unsigned int i, level;
-	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
 
 	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
 
-	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
+	flush_tlb_all();
 
-	if (!cache || do_wbinvd)
+	if (!cache)
 		return;
 
 	/*
-- 
2.19.1.6.gb485710b

Re: [PATCH 4.19] x86/mm/cpa: Unconditionally avoid WBINDV when we can

[Greg Kroah-Hartman]

On Tue, Jul 05, 2022 at 02:02:07PM +0800, Wen Yang wrote:
> From: Peter Zijlstra <peterz@infradead.org>
> 
> commit ddd07b750382adc2b78fdfbec47af8a6e0d8ef37 upstream.
> 
> CAT has happened, WBINDV is bad (even before CAT blowing away the
> entire cache on a multi-core platform wasn't nice), try not to use it
> ever.
> 
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
> Reviewed-by: Dave Hansen <dave.hansen@intel.com>
> Cc: Bin Yang <bin.yang@intel.com>
> Cc: Mark Gross <mark.gross@intel.com>
> Link: https://lkml.kernel.org/r/20180919085947.933674526@infradead.org
> Cc: <stable@vger.kernel.org> # 4.19.x
> Signed-off-by: Wen Yang <wenyang@linux.alibaba.com>
> ---
>  arch/x86/mm/pageattr.c | 18 ++----------------
>  1 file changed, 2 insertions(+), 16 deletions(-)
> 
> diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
> index 101f3ad0d6ad..ab87da7a6043 100644
> --- a/arch/x86/mm/pageattr.c
> +++ b/arch/x86/mm/pageattr.c
> @@ -239,26 +239,12 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
>  			    int in_flags, struct page **pages)
>  {
>  	unsigned int i, level;
> -#ifdef CONFIG_PREEMPT
> -	/*
> -	 * Avoid wbinvd() because it causes latencies on all CPUs,
> -	 * regardless of any CPU isolation that may be in effect.
> -	 *
> -	 * This should be extended for CAT enabled systems independent of
> -	 * PREEMPT because wbinvd() does not respect the CAT partitions and
> -	 * this is exposed to unpriviledged users through the graphics
> -	 * subsystem.
> -	 */
> -	unsigned long do_wbinvd = 0;
> -#else
> -	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
> -#endif
>  
>  	BUG_ON(irqs_disabled() && !early_boot_irqs_disabled);
>  
> -	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
> +	flush_tlb_all();
>  
> -	if (!cache || do_wbinvd)
> +	if (!cache)
>  		return;
>  
>  	/*
> -- 
> 2.19.1.6.gb485710b
> 

As stated before, I am not allowed to accept this due to it being a
requirement of a closed source kernel module.  Please do not keep
resending it hoping we will forget or I will have to add this to my
email filters.

thanks,

greg k-h