[PATCH] x86, msr: fix rdmsrl_safe_on_cpu()

[PATCH] x86, msr: fix rdmsrl_safe_on_cpu()

[Eric Dumazet]

When changing rdmsr_safe_on_cpu() to schedule, I missed that
__rdmsr_safe_on_cpu() was also used by rdmsrl_safe_on_cpu()

Lets make rdmsrl_safe_on_cpu() a wrapper instead of copy/pasting
the code I added for the completion handling.

Fixes: 07cde313b2d2 ("x86/msr: Allow rdmsr_safe_on_cpu() to schedule")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
---
 arch/x86/lib/msr-smp.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/arch/x86/lib/msr-smp.c b/arch/x86/lib/msr-smp.c
index 761ba062afdaf7f7d0603ed94ed6cc3e46b37f76..fee8b9c0520c9954e3cf9093332e1a9b53009b8b 100644
--- a/arch/x86/lib/msr-smp.c
+++ b/arch/x86/lib/msr-smp.c
@@ -225,16 +225,13 @@ EXPORT_SYMBOL(wrmsrl_safe_on_cpu);
 
 int rdmsrl_safe_on_cpu(unsigned int cpu, u32 msr_no, u64 *q)
 {
+	u32 low, high;
 	int err;
-	struct msr_info rv;
 
-	memset(&rv, 0, sizeof(rv));
+	err = rdmsr_safe_on_cpu(cpu, msr_no, &low, &high);
+	*q = (u64)high << 32 | low;
 
-	rv.msr_no = msr_no;
-	err = smp_call_function_single(cpu, __rdmsr_safe_on_cpu, &rv, 1);
-	*q = rv.reg.q;
-
-	return err ? err : rv.err;
+	return err;
 }
 EXPORT_SYMBOL(rdmsrl_safe_on_cpu);
 
-- 
2.17.0.rc1.321.gba9d0f2565-goog

[tip:x86/cleanups] x86/msr: Make rdmsrl_safe_on_cpu() scheduling safe as well

[tip-bot for Eric Dumazet]

Commit-ID:  9b9a51354cae933f5640b5bb73bbcd32f989122f
Gitweb:     https://git.kernel.org/tip/9b9a51354cae933f5640b5bb73bbcd32f989122f
Author:     Eric Dumazet <edumazet@google.com>
AuthorDate: Tue, 27 Mar 2018 20:22:33 -0700
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Wed, 28 Mar 2018 10:34:13 +0200

x86/msr: Make rdmsrl_safe_on_cpu() scheduling safe as well

When changing rdmsr_safe_on_cpu() to schedule, it was missed that
__rdmsr_safe_on_cpu() was also used by rdmsrl_safe_on_cpu()

Make rdmsrl_safe_on_cpu() a wrapper instead of copy/pasting the code which
was added for the completion handling.

Fixes: 07cde313b2d2 ("x86/msr: Allow rdmsr_safe_on_cpu() to schedule")
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Link: https://lkml.kernel.org/r/20180328032233.153055-1-edumazet@google.com

---
 arch/x86/lib/msr-smp.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/arch/x86/lib/msr-smp.c b/arch/x86/lib/msr-smp.c
index 761ba062afda..fee8b9c0520c 100644
--- a/arch/x86/lib/msr-smp.c
+++ b/arch/x86/lib/msr-smp.c
@@ -225,16 +225,13 @@ EXPORT_SYMBOL(wrmsrl_safe_on_cpu);
 
 int rdmsrl_safe_on_cpu(unsigned int cpu, u32 msr_no, u64 *q)
 {
+	u32 low, high;
 	int err;
-	struct msr_info rv;
 
-	memset(&rv, 0, sizeof(rv));
+	err = rdmsr_safe_on_cpu(cpu, msr_no, &low, &high);
+	*q = (u64)high << 32 | low;
 
-	rv.msr_no = msr_no;
-	err = smp_call_function_single(cpu, __rdmsr_safe_on_cpu, &rv, 1);
-	*q = rv.reg.q;
-
-	return err ? err : rv.err;
+	return err;
 }
 EXPORT_SYMBOL(rdmsrl_safe_on_cpu);
 

Re: [PATCH] x86, msr: fix rdmsrl_safe_on_cpu()

[Borislav Petkov]

On Tue, Mar 27, 2018 at 08:22:33PM -0700, Eric Dumazet wrote:
> When changing rdmsr_safe_on_cpu() to schedule, I missed that
> __rdmsr_safe_on_cpu() was also used by rdmsrl_safe_on_cpu()
> 
> Lets make rdmsrl_safe_on_cpu() a wrapper instead of copy/pasting
> the code I added for the completion handling.
> 
> Fixes: 07cde313b2d2 ("x86/msr: Allow rdmsr_safe_on_cpu() to schedule")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Reported-by: kbuild test robot <fengguang.wu@intel.com>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Borislav Petkov <bp@alien8.de>
> ---
>  arch/x86/lib/msr-smp.c | 11 ++++-------
>  1 file changed, 4 insertions(+), 7 deletions(-)
> 
> diff --git a/arch/x86/lib/msr-smp.c b/arch/x86/lib/msr-smp.c
> index 761ba062afdaf7f7d0603ed94ed6cc3e46b37f76..fee8b9c0520c9954e3cf9093332e1a9b53009b8b 100644
> --- a/arch/x86/lib/msr-smp.c
> +++ b/arch/x86/lib/msr-smp.c
> @@ -225,16 +225,13 @@ EXPORT_SYMBOL(wrmsrl_safe_on_cpu);
>  
>  int rdmsrl_safe_on_cpu(unsigned int cpu, u32 msr_no, u64 *q)
>  {
> +	u32 low, high;
>  	int err;
> -	struct msr_info rv;
>  
> -	memset(&rv, 0, sizeof(rv));
> +	err = rdmsr_safe_on_cpu(cpu, msr_no, &low, &high);
> +	*q = (u64)high << 32 | low;
>  
> -	rv.msr_no = msr_no;
> -	err = smp_call_function_single(cpu, __rdmsr_safe_on_cpu, &rv, 1);
> -	*q = rv.reg.q;
> -
> -	return err ? err : rv.err;
> +	return err;
>  }
>  EXPORT_SYMBOL(rdmsrl_safe_on_cpu);
>  
> -- 

I guess now that the rdmsr* side does this, you probably should convert
the wrmsr* side as well.

Thx.

-- 
Regards/Gruss,
    Boris.

Good mailing practices for 400: avoid top-posting and trim the reply.

Re: [PATCH] x86, msr: fix rdmsrl_safe_on_cpu()

[Eric Dumazet]

On 03/28/2018 03:08 AM, Borislav Petkov wrote:

> I guess now that the rdmsr* side does this, you probably should convert
> the wrmsr* side as well.

Yes indeed, thanks for the reminder.