[PATCH 0/2] x86, microcode: __ref / __refdata cleanups

[PATCH 0/2] x86, microcode: __ref / __refdata cleanups

[Mathias Krause]

Hi Boris,

two small cleanups for wrongly annotated variables / functions. They
used to require that annotation for __cpuinit but as that one is gone
since v3.11 we can drop the __ref / __refdata annotation, too.

Please apply.


Mathias Krause (2):
  x86, microcode: Drop bogus __refdata annotation of cpu notifier
  x86, microcode: Drop bogus __ref annotation of show_saved_mc()

 arch/x86/kernel/cpu/microcode/core.c        |    2 +-
 arch/x86/kernel/cpu/microcode/intel_early.c |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

-- 
1.7.10.4

[PATCH 1/2] x86, microcode: Drop bogus __refdata annotation of cpu notifier

[Mathias Krause]

The __cpuinit annotation was dropped from mc_cpu_callback() in commit
148f9bb87745 ("x86: delete __cpuinit usage from all x86 files"),
vanishing the need for the __refdata annotation of mc_cpu_notifier.

Signed-off-by: Mathias Krause <minipli@googlemail.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
---
 arch/x86/kernel/cpu/microcode/core.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c
index 6236a54a63f4..532026d48096 100644
--- a/arch/x86/kernel/cpu/microcode/core.c
+++ b/arch/x86/kernel/cpu/microcode/core.c
@@ -460,7 +460,7 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
 	return NOTIFY_OK;
 }
 
-static struct notifier_block __refdata mc_cpu_notifier = {
+static struct notifier_block mc_cpu_notifier = {
 	.notifier_call	= mc_cpu_callback,
 };
 
-- 
1.7.10.4

[PATCH 2/2] x86, microcode: Drop bogus __ref annotation of show_saved_mc()

[Mathias Krause]

show_saved_mc() does not reference any .init / .exit sections. Drop the
bogus annotation.

Signed-off-by: Mathias Krause <minipli@googlemail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
---
Test build with '#define DEBUG'.

 arch/x86/kernel/cpu/microcode/intel_early.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/cpu/microcode/intel_early.c b/arch/x86/kernel/cpu/microcode/intel_early.c
index 8187b7247d1c..37ea89c11520 100644
--- a/arch/x86/kernel/cpu/microcode/intel_early.c
+++ b/arch/x86/kernel/cpu/microcode/intel_early.c
@@ -390,7 +390,7 @@ static int collect_cpu_info_early(struct ucode_cpu_info *uci)
 }
 
 #ifdef DEBUG
-static void __ref show_saved_mc(void)
+static void show_saved_mc(void)
 {
 	int i, j;
 	unsigned int sig, pf, rev, total_size, data_size, date;
-- 
1.7.10.4

Re: [PATCH 0/2] x86, microcode: __ref / __refdata cleanups

[Ingo Molnar]

* Mathias Krause <minipli@googlemail.com> wrote:

> Hi Boris,
> 
> two small cleanups for wrongly annotated variables / functions. They
> used to require that annotation for __cpuinit but as that one is gone
> since v3.11 we can drop the __ref / __refdata annotation, too.
> 
> Please apply.
> 
> 
> Mathias Krause (2):
>   x86, microcode: Drop bogus __refdata annotation of cpu notifier
>   x86, microcode: Drop bogus __ref annotation of show_saved_mc()
> 
>  arch/x86/kernel/cpu/microcode/core.c        |    2 +-
>  arch/x86/kernel/cpu/microcode/intel_early.c |    2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)

Please merge this into the single x86 patch of the previous series that gets rid 
of __ref/__refdata. There's no need to create unnecessary churn.

Thanks,

	Ingo

Re: [PATCH 0/2] x86, microcode: __ref / __refdata cleanups

[Mathias Krause]

On 20 July 2015 at 09:40, Ingo Molnar <mingo@kernel.org> wrote:
> * Mathias Krause <minipli@googlemail.com> wrote:
>>  arch/x86/kernel/cpu/microcode/core.c        |    2 +-
>>  arch/x86/kernel/cpu/microcode/intel_early.c |    2 +-
>>  2 files changed, 2 insertions(+), 2 deletions(-)
>
> Please merge this into the single x86 patch of the previous series that gets rid
> of __ref/__refdata. There's no need to create unnecessary churn.

Will do. Just thought it may make sense to split the series due to
different maintainers (x86 vs. microcode).

What's your opinion on [1], btw? Should patch 2 go through the tip tree as well?

[1] https://lkml.org/lkml/2015/7/19/214


Thanks,
Mathias