2021-12-10 03:22:35

by Yang Li

[permalink] [raw]
Subject: [PATCH -next 1/2] x86: Use rdmsrl instead of rdmsr

From: Li Yang <[email protected]>

In the current scenario, variable 'hi' set but not used,
rdmsrl() does not change the function, but it is more concise.

Eliminate the following clang warnings:
arch/x86/kernel/process.c:889:10: warning: variable 'hi' set but not
used [-Wunused-but-set-variable]

Reported-by: Abaci Robot <[email protected]>
Signed-off-by: Li Yang <[email protected]>
---
arch/x86/kernel/process.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 5d481038fe0b..b415f30278d1 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -886,7 +886,7 @@ void amd_e400_c1e_apic_setup(void)

void __init arch_post_acpi_subsys_init(void)
{
- u32 lo, hi;
+ u32 lo;

if (!boot_cpu_has_bug(X86_BUG_AMD_E400))
return;
@@ -896,7 +896,7 @@ void __init arch_post_acpi_subsys_init(void)
* the machine is affected K8_INTP_C1E_ACTIVE_MASK bits are set in
* MSR_K8_INT_PENDING_MSG.
*/
- rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi);
+ rdmsrl(MSR_K8_INT_PENDING_MSG, lo);
if (!(lo & K8_INTP_C1E_ACTIVE_MASK))
return;

--
2.34.0



2021-12-10 03:22:50

by Yang Li

[permalink] [raw]
Subject: [PATCH -next 2/2] x86: Fix kernel-doc

From: Li Yang <[email protected]>

Fix function name in process.c kernel-doc comment
to remove a warning found by running scripts/kernel-doc,
which is caused by using 'make W=1'.

arch/x86/kernel/process.c:411: warning: expecting prototype for
tss_update_io_bitmap(). Prototype was for native_tss_update_io_bitmap()
instead

Reported-by: Abaci Robot <[email protected]>
Signed-off-by: Li Yang <[email protected]>
---
arch/x86/kernel/process.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index b415f30278d1..290c5bf610f1 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -405,7 +405,7 @@ static void tss_copy_io_bitmap(struct tss_struct *tss, struct io_bitmap *iobm)
}

/**
- * tss_update_io_bitmap - Update I/O bitmap before exiting to usermode
+ * native_tss_update_io_bitmap - Update I/O bitmap before exiting to usermode
*/
void native_tss_update_io_bitmap(void)
{
--
2.34.0


2021-12-28 21:07:13

by Borislav Petkov

[permalink] [raw]
Subject: Re: [PATCH -next 1/2] x86: Use rdmsrl instead of rdmsr

On Fri, Dec 10, 2021 at 11:22:07AM +0800, Yang Li wrote:
> From: Li Yang <[email protected]>
>
> In the current scenario, variable 'hi' set but not used,
> rdmsrl() does not change the function, but it is more concise.
>
> Eliminate the following clang warnings:
> arch/x86/kernel/process.c:889:10: warning: variable 'hi' set but not
> used [-Wunused-but-set-variable]
>
> Reported-by: Abaci Robot <[email protected]>
> Signed-off-by: Li Yang <[email protected]>
> ---
> arch/x86/kernel/process.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
> index 5d481038fe0b..b415f30278d1 100644
> --- a/arch/x86/kernel/process.c
> +++ b/arch/x86/kernel/process.c
> @@ -886,7 +886,7 @@ void amd_e400_c1e_apic_setup(void)
>
> void __init arch_post_acpi_subsys_init(void)
> {
> - u32 lo, hi;
> + u32 lo;
>
> if (!boot_cpu_has_bug(X86_BUG_AMD_E400))
> return;
> @@ -896,7 +896,7 @@ void __init arch_post_acpi_subsys_init(void)
> * the machine is affected K8_INTP_C1E_ACTIVE_MASK bits are set in
> * MSR_K8_INT_PENDING_MSG.
> */
> - rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi);
> + rdmsrl(MSR_K8_INT_PENDING_MSG, lo);
> if (!(lo & K8_INTP_C1E_ACTIVE_MASK))
> return;
>
> --

I believe you folks have been warned a bunch of times to think first
before acting upon a warning your silly bot catches.

Because your "fix" shows that you didn't even look at rdmsrl()'s
definition and what the second variable width is.

gcc explains it this way:

arch/x86/kernel/process.c: In function ‘arch_post_acpi_subsys_init’:
./arch/x86/include/asm/msr.h:277:11: warning: conversion from ‘long long unsigned int’ to ‘u32’ {aka ‘unsigned int’} may change value [-Wconversion]
277 | ((val) = native_read_msr((msr)))
| ^~~~~~~~~~~~~~~
arch/x86/kernel/process.c:899:2: note: in expansion of macro ‘rdmsrl’
899 | rdmsrl(MSR_K8_INT_PENDING_MSG, lo);
| ^~~~~~

so I'm ignoring all submissions from you from now on.

--
Regards/Gruss,
Boris.

https://people.kernel.org/tglx/notes-about-netiquette