linux-next: manual merge of the percpu tree with the tip tree

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 638 bytes --]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/apic/nmi.c between commit
5f2b0ba4d94b3ac23cbc4b7f675d98eb677a760a ("x86, nmi_watchdog: Remove the
old nmi_watchdog") from the tip tree and commits
aa0be2afeed15edd653830ca87a3f173f08c23e6 ("x86: Use this_cpu_ops to
optimize code") and e5195e91d9b579ba9def00ea2e8b96189c0120f4 ("x86: Use
this_cpu_inc_return for nmi counter") from the percpu tree.

I just ignored the percpu changes to this file and removed it as the tip
tree did.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 638 bytes --]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/apic/nmi.c between commit
5f2b0ba4d94b3ac23cbc4b7f675d98eb677a760a ("x86, nmi_watchdog: Remove the
old nmi_watchdog") from the tip tree and commits
aa0be2afeed15edd653830ca87a3f173f08c23e6 ("x86: Use this_cpu_ops to
optimize code") and e5195e91d9b579ba9def00ea2e8b96189c0120f4 ("x86: Use
this_cpu_inc_return for nmi counter") from the percpu tree.

I just ignored the percpu changes to this file and removed it as the tip
tree did.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

Re: linux-next: manual merge of the percpu tree with the tip tree

[Frederic Weisbecker]

On Thu, Oct 09, 2014 at 09:15:12AM -0400, Tejun Heo wrote:
> On Thu, Oct 09, 2014 at 03:13:31PM +0200, Frederic Weisbecker wrote:
> > On Thu, Oct 09, 2014 at 03:50:18PM +1100, Stephen Rothwell wrote:
> > > Hi all,
> > > 
> > > Today's linux-next merge of the percpu tree got a conflict in
> > > kernel/irq_work.c between commit 76a33061b932 ("irq_work: Force raised
> > > irq work to run on irq work interrupt") from the tip tree and commit
> > > 22127e93c587 ("time: Replace __get_cpu_var uses") from the percpu tree.
> > > 
> > > I fixed it up (see below) and can carry the fix as necessary (no action
> > > is required).
> > > 
> > > -- 
> > > Cheers,
> > > Stephen Rothwell                    sfr@canb.auug.org.au
> > > 
> > > diff --cc kernel/irq_work.c
> > > index 385b85aded19,345d19edcdae..000000000000
> > > --- a/kernel/irq_work.c
> > > +++ b/kernel/irq_work.c
> > > @@@ -113,12 -113,10 +113,12 @@@ bool irq_work_needs_cpu(void
> > >   {
> > >   	struct llist_head *raised, *lazy;
> > >   
> > > - 	raised = &__get_cpu_var(raised_list);
> > > - 	lazy = &__get_cpu_var(lazy_list);
> > > + 	raised = this_cpu_ptr(&raised_list);
> > > + 	lazy = this_cpu_ptr(&lazy_list);
> > 
> > Ah thanks! The conflict is compile time rather than merge time, thanks
> > for spotting it!
> > 
> > Should we notify Linus about it? That's certainly something that should
> > be applied with the percpu tree.
> 
> I'm holding back percpu/for-3.18-consistent-ops till other trees are
> merged and collecting the conflicts.  I'll list them when sending the
> pull request.

Sounds good!

Thanks.

Re: linux-next: manual merge of the percpu tree with the tip tree

[Tejun Heo]

On Thu, Oct 09, 2014 at 03:13:31PM +0200, Frederic Weisbecker wrote:
> On Thu, Oct 09, 2014 at 03:50:18PM +1100, Stephen Rothwell wrote:
> > Hi all,
> > 
> > Today's linux-next merge of the percpu tree got a conflict in
> > kernel/irq_work.c between commit 76a33061b932 ("irq_work: Force raised
> > irq work to run on irq work interrupt") from the tip tree and commit
> > 22127e93c587 ("time: Replace __get_cpu_var uses") from the percpu tree.
> > 
> > I fixed it up (see below) and can carry the fix as necessary (no action
> > is required).
> > 
> > -- 
> > Cheers,
> > Stephen Rothwell                    sfr@canb.auug.org.au
> > 
> > diff --cc kernel/irq_work.c
> > index 385b85aded19,345d19edcdae..000000000000
> > --- a/kernel/irq_work.c
> > +++ b/kernel/irq_work.c
> > @@@ -113,12 -113,10 +113,12 @@@ bool irq_work_needs_cpu(void
> >   {
> >   	struct llist_head *raised, *lazy;
> >   
> > - 	raised = &__get_cpu_var(raised_list);
> > - 	lazy = &__get_cpu_var(lazy_list);
> > + 	raised = this_cpu_ptr(&raised_list);
> > + 	lazy = this_cpu_ptr(&lazy_list);
> 
> Ah thanks! The conflict is compile time rather than merge time, thanks
> for spotting it!
> 
> Should we notify Linus about it? That's certainly something that should
> be applied with the percpu tree.

I'm holding back percpu/for-3.18-consistent-ops till other trees are
merged and collecting the conflicts.  I'll list them when sending the
pull request.

Thanks.

-- 
tejun

Re: linux-next: manual merge of the percpu tree with the tip tree

[Frederic Weisbecker]

On Thu, Oct 09, 2014 at 03:50:18PM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> Today's linux-next merge of the percpu tree got a conflict in
> kernel/irq_work.c between commit 76a33061b932 ("irq_work: Force raised
> irq work to run on irq work interrupt") from the tip tree and commit
> 22127e93c587 ("time: Replace __get_cpu_var uses") from the percpu tree.
> 
> I fixed it up (see below) and can carry the fix as necessary (no action
> is required).
> 
> -- 
> Cheers,
> Stephen Rothwell                    sfr@canb.auug.org.au
> 
> diff --cc kernel/irq_work.c
> index 385b85aded19,345d19edcdae..000000000000
> --- a/kernel/irq_work.c
> +++ b/kernel/irq_work.c
> @@@ -113,12 -113,10 +113,12 @@@ bool irq_work_needs_cpu(void
>   {
>   	struct llist_head *raised, *lazy;
>   
> - 	raised = &__get_cpu_var(raised_list);
> - 	lazy = &__get_cpu_var(lazy_list);
> + 	raised = this_cpu_ptr(&raised_list);
> + 	lazy = this_cpu_ptr(&lazy_list);

Ah thanks! The conflict is compile time rather than merge time, thanks
for spotting it!

Should we notify Linus about it? That's certainly something that should
be applied with the percpu tree.

>  -	if (llist_empty(raised) && llist_empty(lazy))
>  -		return false;
>  +
>  +	if (llist_empty(raised) || arch_irq_work_has_interrupt())
>  +		if (llist_empty(lazy))
>  +			return false;
>   
>   	/* All work should have been flushed before going offline */
>   	WARN_ON_ONCE(cpu_is_offline(smp_processor_id()));

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1154 bytes --]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
kernel/irq_work.c between commit 76a33061b932 ("irq_work: Force raised
irq work to run on irq work interrupt") from the tip tree and commit
22127e93c587 ("time: Replace __get_cpu_var uses") from the percpu tree.

I fixed it up (see below) and can carry the fix as necessary (no action
is required).

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc kernel/irq_work.c
index 385b85aded19,345d19edcdae..000000000000
--- a/kernel/irq_work.c
+++ b/kernel/irq_work.c
@@@ -113,12 -113,10 +113,12 @@@ bool irq_work_needs_cpu(void
  {
  	struct llist_head *raised, *lazy;
  
- 	raised = &__get_cpu_var(raised_list);
- 	lazy = &__get_cpu_var(lazy_list);
+ 	raised = this_cpu_ptr(&raised_list);
+ 	lazy = this_cpu_ptr(&lazy_list);
 -	if (llist_empty(raised) && llist_empty(lazy))
 -		return false;
 +
 +	if (llist_empty(raised) || arch_irq_work_has_interrupt())
 +		if (llist_empty(lazy))
 +			return false;
  
  	/* All work should have been flushed before going offline */
  	WARN_ON_ONCE(cpu_is_offline(smp_processor_id()));

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: linux-next: manual merge of the percpu tree with the tip tree

[Alex Shi]

On 05/16/2012 03:36 PM, Stephen Rothwell wrote:

> Hi all,
> 
> Today's linux-next merge of the percpu tree got a conflict in
> arch/x86/mm/tlb.c between commit a6fca40f1d7f ("x86, tlb: Switch cr3 in
> leave_mm() only when needed") from the tip tree and commit c6ae41e7d469
> ("x86: replace percpu_xxx funcs with this_cpu_xxx") from the percpu tree.
> 
> I fixed it up (see below) and can carry the fix as necessary.


Thanks a lot!

My auto-scripts just watching upstream kernel, so missing this...

but happy to know the pain of code clean up is closing soon. ;)

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1226 bytes --]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/mm/tlb.c between commit a6fca40f1d7f ("x86, tlb: Switch cr3 in
leave_mm() only when needed") from the tip tree and commit c6ae41e7d469
("x86: replace percpu_xxx funcs with this_cpu_xxx") from the percpu tree.

I fixed it up (see below) and can carry the fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/mm/tlb.c
index 125bcad,3804471..0000000
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@@ -61,13 -61,11 +61,13 @@@ static DEFINE_PER_CPU_READ_MOSTLY(int, 
   */
  void leave_mm(int cpu)
  {
- 	struct mm_struct *active_mm = percpu_read(cpu_tlbstate.active_mm);
- 	if (percpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
++	struct mm_struct *active_mm = this_cpu_read(cpu_tlbstate.active_mm);
+ 	if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
  		BUG();
 -	cpumask_clear_cpu(cpu,
 -			  mm_cpumask(this_cpu_read(cpu_tlbstate.active_mm)));
 -	load_cr3(swapper_pg_dir);
 +	if (cpumask_test_cpu(cpu, mm_cpumask(active_mm))) {
 +		cpumask_clear_cpu(cpu, mm_cpumask(active_mm));
 +		load_cr3(swapper_pg_dir);
 +	}
  }
  EXPORT_SYMBOL_GPL(leave_mm);
  

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1226 bytes --]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/mm/tlb.c between commit a6fca40f1d7f ("x86, tlb: Switch cr3 in
leave_mm() only when needed") from the tip tree and commit c6ae41e7d469
("x86: replace percpu_xxx funcs with this_cpu_xxx") from the percpu tree.

I fixed it up (see below) and can carry the fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/mm/tlb.c
index 125bcad,3804471..0000000
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@@ -61,13 -61,11 +61,13 @@@ static DEFINE_PER_CPU_READ_MOSTLY(int, 
   */
  void leave_mm(int cpu)
  {
- 	struct mm_struct *active_mm = percpu_read(cpu_tlbstate.active_mm);
- 	if (percpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
++	struct mm_struct *active_mm = this_cpu_read(cpu_tlbstate.active_mm);
+ 	if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
  		BUG();
 -	cpumask_clear_cpu(cpu,
 -			  mm_cpumask(this_cpu_read(cpu_tlbstate.active_mm)));
 -	load_cr3(swapper_pg_dir);
 +	if (cpumask_test_cpu(cpu, mm_cpumask(active_mm))) {
 +		cpumask_clear_cpu(cpu, mm_cpumask(active_mm));
 +		load_cr3(swapper_pg_dir);
 +	}
  }
  EXPORT_SYMBOL_GPL(leave_mm);
  

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/include/asm/percpu.h between commit b1e7734f024c ("x86, percpu:
Use ASM_NOP4 instead of hardcoding P6_NOP4") from the tip tree and commit
5f55924deaa6 ("percpu: Avoid extra NOP in percpu_cmpxchg16b_double") from
the percpu tree.

I fixed it up (see below) and can carry the fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/include/asm/percpu.h
index 53278b0,d68fca6..0000000
--- a/arch/x86/include/asm/percpu.h
+++ b/arch/x86/include/asm/percpu.h
@@@ -509,6 -509,11 +509,11 @@@ do {									
   * it in software.  The address used in the cmpxchg16 instruction must be
   * aligned to a 16 byte boundary.
   */
+ #ifdef CONFIG_SMP
 -#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" P6_NOP3
++#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP3
+ #else
 -#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" P6_NOP2
++#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP2
+ #endif
  #define percpu_cmpxchg16b_double(pcp1, o1, o2, n1, n2)			\
  ({									\
  	char __ret;							\

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/include/asm/percpu.h between commit b1e7734f024c ("x86, percpu:
Use ASM_NOP4 instead of hardcoding P6_NOP4") from the tip tree and commit
5f55924deaa6 ("percpu: Avoid extra NOP in percpu_cmpxchg16b_double") from
the percpu tree.

I fixed it up (see below) and can carry the fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/include/asm/percpu.h
index 53278b0,d68fca6..0000000
--- a/arch/x86/include/asm/percpu.h
+++ b/arch/x86/include/asm/percpu.h
@@@ -509,6 -509,11 +509,11 @@@ do {									
   * it in software.  The address used in the cmpxchg16 instruction must be
   * aligned to a 16 byte boundary.
   */
+ #ifdef CONFIG_SMP
 -#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" P6_NOP3
++#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP3
+ #else
 -#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" P6_NOP2
++#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP2
+ #endif
  #define percpu_cmpxchg16b_double(pcp1, o1, o2, n1, n2)			\
  ({									\
  	char __ret;							\

Re: linux-next: manual merge of the percpu tree with the tip tree

[Tejun Heo]

On Fri, Feb 11, 2011 at 02:58:10PM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> Today's linux-next merge of the percpu tree got a conflict in
> arch/x86/kernel/vmlinux.lds.S between commit
> 94d1ac8b55799be10487fff9766cce6d6628462a ("x86: Reduce back the alignment
> of the per-CPU data section") from the tip tree and commit
> 19df0c2fef010e94e90df514aaf4e73f6b80145c ("percpu: align percpu
> readmostly subsection to cacheline") from the percpu tree.
> 
> I fixed it up (see below) and can carry the fix as necessary.

Thanks.  Looks good to me.

-- 
tejun

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/vmlinux.lds.S between commit
94d1ac8b55799be10487fff9766cce6d6628462a ("x86: Reduce back the alignment
of the per-CPU data section") from the tip tree and commit
19df0c2fef010e94e90df514aaf4e73f6b80145c ("percpu: align percpu
readmostly subsection to cacheline") from the percpu tree.

I fixed it up (see below) and can carry the fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/kernel/vmlinux.lds.S
index e9f7a3c,cef446f..0000000
--- a/arch/x86/kernel/vmlinux.lds.S
+++ b/arch/x86/kernel/vmlinux.lds.S
@@@ -305,7 -305,7 +305,7 @@@ SECTION
  	}
  
  #if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP)
- 	PERCPU(PAGE_SIZE)
 -	PERCPU(INTERNODE_CACHE_BYTES, THREAD_SIZE)
++	PERCPU(INTERNODE_CACHE_BYTES, PAGE_SIZE)
  #endif
  
  	. = ALIGN(PAGE_SIZE);

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/process.c between commit
25e41933b58777f2d020c3b0186b430ea004ec28 ("perf: Clean up power events by
introducing new, more generic ones") from the tip tree and commit
7b543a5334ff4ea2e3ad3b777fc23cdb8072a988 ("x86: Replace uses of
current_cpu_data with this_cpu ops") from the percpu tree.

Just context changes.  I fixed it up (see below) and can carry the fix as
necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/kernel/process.c
index c852041,dae1c07..0000000
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@@ -444,9 -444,8 +444,9 @@@ EXPORT_SYMBOL_GPL(cpu_idle_wait)
  void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
  {
  	trace_power_start(POWER_CSTATE, (ax>>4)+1, smp_processor_id());
 +	trace_cpu_idle((ax>>4)+1, smp_processor_id());
  	if (!need_resched()) {
- 		if (cpu_has(&current_cpu_data, X86_FEATURE_CLFLUSH_MONITOR))
+ 		if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLUSH_MONITOR))
  			clflush((void *)&current_thread_info()->flags);
  
  		__monitor((void *)&current_thread_info()->flags, 0, 0);
@@@ -461,8 -460,7 +461,8 @@@ static void mwait_idle(void
  {
  	if (!need_resched()) {
  		trace_power_start(POWER_CSTATE, 1, smp_processor_id());
 +		trace_cpu_idle(1, smp_processor_id());
- 		if (cpu_has(&current_cpu_data, X86_FEATURE_CLFLUSH_MONITOR))
+ 		if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLUSH_MONITOR))
  			clflush((void *)&current_thread_info()->flags);
  
  		__monitor((void *)&current_thread_info()->flags, 0, 0);

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/process.c between commit
25e41933b58777f2d020c3b0186b430ea004ec28 ("perf: Clean up power events by
introducing new, more generic ones") from the tip tree and commit
7b543a5334ff4ea2e3ad3b777fc23cdb8072a988 ("x86: Replace uses of
current_cpu_data with this_cpu ops") from the percpu tree.

Just context changes.  I fixed it up (see below) and can carry the fix as
necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/kernel/process.c
index c852041,dae1c07..0000000
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@@ -444,9 -444,8 +444,9 @@@ EXPORT_SYMBOL_GPL(cpu_idle_wait)
  void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
  {
  	trace_power_start(POWER_CSTATE, (ax>>4)+1, smp_processor_id());
 +	trace_cpu_idle((ax>>4)+1, smp_processor_id());
  	if (!need_resched()) {
- 		if (cpu_has(&current_cpu_data, X86_FEATURE_CLFLUSH_MONITOR))
+ 		if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLUSH_MONITOR))
  			clflush((void *)&current_thread_info()->flags);
  
  		__monitor((void *)&current_thread_info()->flags, 0, 0);
@@@ -461,8 -460,7 +461,8 @@@ static void mwait_idle(void
  {
  	if (!need_resched()) {
  		trace_power_start(POWER_CSTATE, 1, smp_processor_id());
 +		trace_cpu_idle(1, smp_processor_id());
- 		if (cpu_has(&current_cpu_data, X86_FEATURE_CLFLUSH_MONITOR))
+ 		if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLUSH_MONITOR))
  			clflush((void *)&current_thread_info()->flags);
  
  		__monitor((void *)&current_thread_info()->flags, 0, 0);

Re: linux-next: manual merge of the percpu tree with the tip tree

[Tejun Heo]

On Fri, Dec 31, 2010 at 01:55:46PM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> Today's linux-next merge of the percpu tree got a conflict in
> arch/x86/kernel/apic/x2apic_uv_x.c between commit
> d8850ba425d9823d3184bd52f065899dac4689f9 ("x86, UV: Fix the effect of
> extra bits in the hub nodeid register") from the tip tree and commit
> 0a3aee0da4402aa19b66e458038533c896fb80c6 ("x86: Use this_cpu_ops to
> optimize code") from the percpu tree.
> 
> I fixed it up (see below) and can carry the fix as necessary.

Thank you.  Looks good to me.

-- 
tejun

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/apic/x2apic_uv_x.c between commit
d8850ba425d9823d3184bd52f065899dac4689f9 ("x86, UV: Fix the effect of
extra bits in the hub nodeid register") from the tip tree and commit
0a3aee0da4402aa19b66e458038533c896fb80c6 ("x86: Use this_cpu_ops to
optimize code") from the percpu tree.

I fixed it up (see below) and can carry the fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/kernel/apic/x2apic_uv_x.c
index 2a3f2a7,26ec9a7..0000000
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@@ -120,8 -118,8 +120,8 @@@ static int __init uv_acpi_madt_oem_chec
  		else if (!strcmp(oem_table_id, "UVX"))
  			uv_system_type = UV_X2APIC;
  		else if (!strcmp(oem_table_id, "UVH")) {
- 			__get_cpu_var(x2apic_extra_bits) =
- 				pnodeid << uvh_apicid.s.pnode_shift;
+ 			__this_cpu_write(x2apic_extra_bits,
 -				nodeid << (uvh_apicid.s.pnode_shift - 1));
++				pnodeid << uvh_apicid.s.pnode_shift);
  			uv_system_type = UV_NON_UNIQUE_APIC;
  			uv_set_apicid_hibit();
  			return 1;

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/apic/x2apic_uv_x.c between commit
d8850ba425d9823d3184bd52f065899dac4689f9 ("x86, UV: Fix the effect of
extra bits in the hub nodeid register") from the tip tree and commit
0a3aee0da4402aa19b66e458038533c896fb80c6 ("x86: Use this_cpu_ops to
optimize code") from the percpu tree.

I fixed it up (see below) and can carry the fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/kernel/apic/x2apic_uv_x.c
index 2a3f2a7,26ec9a7..0000000
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@@ -120,8 -118,8 +120,8 @@@ static int __init uv_acpi_madt_oem_chec
  		else if (!strcmp(oem_table_id, "UVX"))
  			uv_system_type = UV_X2APIC;
  		else if (!strcmp(oem_table_id, "UVH")) {
- 			__get_cpu_var(x2apic_extra_bits) =
- 				pnodeid << uvh_apicid.s.pnode_shift;
+ 			__this_cpu_write(x2apic_extra_bits,
 -				nodeid << (uvh_apicid.s.pnode_shift - 1));
++				pnodeid << uvh_apicid.s.pnode_shift);
  			uv_system_type = UV_NON_UNIQUE_APIC;
  			uv_set_apicid_hibit();
  			return 1;

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 437 bytes --]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/apic/nmi.c between commit
5f2b0ba4d94b3ac23cbc4b7f675d98eb677a760a ("x86, nmi_watchdog: Remove the
old nmi_watchdog") from the tip tree and a couple of commits from the
percpu tree.

The former commit removes the file, so I did that.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 437 bytes --]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/apic/nmi.c between commit
5f2b0ba4d94b3ac23cbc4b7f675d98eb677a760a ("x86, nmi_watchdog: Remove the
old nmi_watchdog") from the tip tree and a couple of commits from the
percpu tree.

The former commit removes the file, so I did that.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

Re: linux-next: manual merge of the percpu tree with the tip tree

[Frederic Weisbecker]

On Wed, Feb 03, 2010 at 04:38:22PM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> Today's linux-next merge of the percpu tree got a conflict in
> include/linux/ftrace_event.h, include/trace/ftrace.h,
> kernel/trace/trace_kprobe.c, kernel/trace/trace_syscalls.c and
> kernel/trace/trace_event_profile.c between commit
> 430ad5a600a83956749307b13257c464c3826b55 ("perf: Factorize trace events
> raw sample buffer operations") from the tip tree and commit
> eeb721be6bc03fe37755e69ab5c3ba2fe9897fd9 ("percpu: add __percpu sparse
> annotations to trace") from the percpu tree.
> 
> I attempted to fix them up (see below) but someone should check the
> result carefully.  (The apparently unchanged files below used the tip
> tree versions.)
> 
> -- 
> Cheers,
> Stephen Rothwell                    sfr@canb.auug.org.au



I'm not very used to review patches on patches :)

But yeah that looks good to me.

Thanks!

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
include/linux/ftrace_event.h, include/trace/ftrace.h,
kernel/trace/trace_kprobe.c, kernel/trace/trace_syscalls.c and
kernel/trace/trace_event_profile.c between commit
430ad5a600a83956749307b13257c464c3826b55 ("perf: Factorize trace events
raw sample buffer operations") from the tip tree and commit
eeb721be6bc03fe37755e69ab5c3ba2fe9897fd9 ("percpu: add __percpu sparse
annotations to trace") from the percpu tree.

I attempted to fix them up (see below) but someone should check the
result carefully.  (The apparently unchanged files below used the tip
tree versions.)

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc include/linux/ftrace_event.h
index 6b7c444,72fccdd..0000000
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
diff --cc include/trace/ftrace.h
index fb2c5bd,210d421..0000000
--- a/include/trace/ftrace.h
+++ b/include/trace/ftrace.h
diff --cc kernel/trace/trace_event_profile.c
index f0d6930,4b16312..0000000
--- a/kernel/trace/trace_event_profile.c
+++ b/kernel/trace/trace_event_profile.c
@@@ -10,8 -9,11 +10,8 @@@
  #include "trace.h"
  
  
- static char *perf_trace_buf;
- static char *perf_trace_buf_nmi;
 -char __percpu *perf_trace_buf;
 -EXPORT_PER_CPU_SYMBOL_GPL(perf_trace_buf);
 -
 -char __percpu *perf_trace_buf_nmi;
 -EXPORT_PER_CPU_SYMBOL_GPL(perf_trace_buf_nmi);
++static char __percpu *perf_trace_buf;
++static char __percpu *perf_trace_buf_nmi;
  
  typedef typeof(char [FTRACE_MAX_PROFILE_SIZE]) perf_trace_t ;
  
@@@ -118,47 -120,3 +118,48 @@@ void ftrace_profile_disable(int event_i
  	}
  	mutex_unlock(&event_mutex);
  }
 +
 +__kprobes void *ftrace_perf_buf_prepare(int size, unsigned short type,
 +					int *rctxp, unsigned long *irq_flags)
 +{
 +	struct trace_entry *entry;
- 	char *trace_buf, *raw_data;
++	char __percpu *trace_buf;
++	char *raw_data;
 +	int pc, cpu;
 +
 +	pc = preempt_count();
 +
 +	/* Protect the per cpu buffer, begin the rcu read side */
 +	local_irq_save(*irq_flags);
 +
 +	*rctxp = perf_swevent_get_recursion_context();
 +	if (*rctxp < 0)
 +		goto err_recursion;
 +
 +	cpu = smp_processor_id();
 +
 +	if (in_nmi())
 +		trace_buf = rcu_dereference(perf_trace_buf_nmi);
 +	else
 +		trace_buf = rcu_dereference(perf_trace_buf);
 +
 +	if (!trace_buf)
 +		goto err;
 +
 +	raw_data = per_cpu_ptr(trace_buf, cpu);
 +
 +	/* zero the dead bytes from align to not leak stack to user */
 +	*(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL;
 +
 +	entry = (struct trace_entry *)raw_data;
 +	tracing_generic_entry_update(entry, *irq_flags, pc);
 +	entry->type = type;
 +
 +	return raw_data;
 +err:
 +	perf_swevent_put_recursion_context(*rctxp);
 +err_recursion:
 +	local_irq_restore(*irq_flags);
 +	return NULL;
 +}
 +EXPORT_SYMBOL_GPL(ftrace_perf_buf_prepare);
diff --cc kernel/trace/trace_kprobe.c
index 9d085d3,4567950..0000000
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
diff --cc kernel/trace/trace_syscalls.c
index 6cce6a8,9f7de51..0000000
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/cpu/intel_cacheinfo.c between commit
ebb682f522411abbe358059a256a8672ec0bd55b ("x86, AMD: Fix stale
cpuid4_info shared_map data in shared_cpu_map cpumasks") from the tip
tree and commit 0fe1e009541e925adc1748a605d8b66188e4b2ab ("percpu: make
percpu symbols in x86 unique") from the percpu tree.

Just context changes.  I fixed it up (see below) and can carry the fix as
necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/kernel/cpu/intel_cacheinfo.c
index 63ada17,55d2ef3..0000000
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@@ -511,15 -508,14 +511,15 @@@ static void __cpuinit cache_shared_cpu_
  	struct cpuinfo_x86 *c = &cpu_data(cpu);
  
  	if ((index == 3) && (c->x86_vendor == X86_VENDOR_AMD)) {
 -		struct cpuinfo_x86 *d;
 -		for_each_online_cpu(i) {
 +		for_each_cpu(i, c->llc_shared_map) {
- 			if (!per_cpu(cpuid4_info, i))
+ 			if (!per_cpu(ici_cpuid4_info, i))
  				continue;
 -			d = &cpu_data(i);
  			this_leaf = CPUID4_INFO_IDX(i, index);
 -			cpumask_copy(to_cpumask(this_leaf->shared_cpu_map),
 -				     d->llc_shared_map);
 +			for_each_cpu(sibling, c->llc_shared_map) {
 +				if (!cpu_online(sibling))
 +					continue;
 +				set_bit(sibling, this_leaf->shared_cpu_map);
 +			}
  		}
  		return;
  	}

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 556 bytes --]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/powerpc/platforms/pseries/hvCall.S between commit
c8cd093a6e9f96ea6b871576fd4e46d7c818bb89 ("powerpc: tracing: Add
hypervisor call tracepoints") from the tip tree and commit
dd17c8f72993f9461e9c19250e3f155d6d99df22 ("percpu: remove per_cpu__
prefix") from the percpu tree.

The former removed the code changed by the latter, so I just used the tip
tree version.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
kernel/sched.c between commit 84e9dabf6e6a78928c6a1a8ba235c9fb0908d0f8
("sched: Rename init_cfs_rq => init_tg_cfs_rq") from the tip tree and
commit b9bf3121af348d9255f1c917830fe8c2df52efcb ("percpu: use
DEFINE_PER_CPU_SHARED_ALIGNED()") from the percpu tree.

I fixed it up (see below) and can carry the fix as necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc kernel/sched.c
index 4061929,d3d7e76..0000000
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@@ -318,7 -318,7 +318,7 @@@ struct task_group root_task_group
  /* Default task group's sched entity on each cpu */
  static DEFINE_PER_CPU(struct sched_entity, init_sched_entity);
  /* Default task group's cfs_rq on each cpu */
- static DEFINE_PER_CPU(struct cfs_rq, init_tg_cfs_rq) ____cacheline_aligned_in_smp;
 -static DEFINE_PER_CPU_SHARED_ALIGNED(struct cfs_rq, init_cfs_rq);
++static DEFINE_PER_CPU_SHARED_ALIGNED(struct cfs_rq, init_tg_cfs_rq);
  #endif /* CONFIG_FAIR_GROUP_SCHED */
  
  #ifdef CONFIG_RT_GROUP_SCHED

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/cpu/perf_counter.c between commit
0406ca6d8e849d9dd027c8cb6791448e81411aef ("perf_counter: Ignore the nmi
call frames in the x86-64 backtraces") from the tip tree and commit
245b2e70eabd797932adb263a65da0bab3711753 ("percpu: clean up percpu
variable definitions") from the percpu tree.

Just context changes.  I fixed it up (see below) and can carry the fix as
necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/kernel/cpu/perf_counter.c
index 36c3dc7,13bd6d6..0000000
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@@ -1559,9 -1559,8 +1559,9 @@@ void callchain_store(struct perf_callch
  		entry->ip[entry->nr++] = ip;
  }
  
- static DEFINE_PER_CPU(struct perf_callchain_entry, irq_entry);
- static DEFINE_PER_CPU(struct perf_callchain_entry, nmi_entry);
+ static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
+ static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry);
 +static DEFINE_PER_CPU(int, in_nmi_frame);
  
  
  static void

linux-next: manual merge of the percpu tree with the tip tree

[Stephen Rothwell]

Hi all,

Today's linux-next merge of the percpu tree got a conflict in
arch/x86/kernel/cpu/perf_counter.c between commit
0406ca6d8e849d9dd027c8cb6791448e81411aef ("perf_counter: Ignore the nmi
call frames in the x86-64 backtraces") from the tip tree and commit
245b2e70eabd797932adb263a65da0bab3711753 ("percpu: clean up percpu
variable definitions") from the percpu tree.

Just context changes.  I fixed it up (see below) and can carry the fix as
necessary.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --cc arch/x86/kernel/cpu/perf_counter.c
index 36c3dc7,13bd6d6..0000000
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@@ -1559,9 -1559,8 +1559,9 @@@ void callchain_store(struct perf_callch
  		entry->ip[entry->nr++] = ip;
  }
  
- static DEFINE_PER_CPU(struct perf_callchain_entry, irq_entry);
- static DEFINE_PER_CPU(struct perf_callchain_entry, nmi_entry);
+ static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
+ static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry);
 +static DEFINE_PER_CPU(int, in_nmi_frame);
  
  
  static void