Re: [PATCH] x86: use invd instead of wbinvd in real mode start code

2020-02-02 Thread Bin Meng
On Mon, Feb 3, 2020 at 12:41 PM Bin Meng  wrote:
>
> On Wed, Jan 8, 2020 at 7:09 PM Masahiro Yamada  wrote:
> >
> > I do not know why the boot code immediately after the system reset
> > should write-back the cache content. I think the cache invalidation
> > should be enough.
> >
> > I tested this commit with qemu-x86_defconfig, and it worked for me.
> >
> > Signed-off-by: Masahiro Yamada 
> > ---
> >
> >  arch/x86/cpu/start.S   | 2 +-
> >  arch/x86/cpu/start16.S | 2 +-
> >  2 files changed, 2 insertions(+), 2 deletions(-)
> >
>
> Reviewed-by: Bin Meng 
> Tested-by: Bin Meng 

applied to u-boot-x86, thanks!


Re: [PATCH] x86: use invd instead of wbinvd in real mode start code

2020-02-02 Thread Bin Meng
On Wed, Jan 8, 2020 at 7:09 PM Masahiro Yamada  wrote:
>
> I do not know why the boot code immediately after the system reset
> should write-back the cache content. I think the cache invalidation
> should be enough.
>
> I tested this commit with qemu-x86_defconfig, and it worked for me.
>
> Signed-off-by: Masahiro Yamada 
> ---
>
>  arch/x86/cpu/start.S   | 2 +-
>  arch/x86/cpu/start16.S | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
>

Reviewed-by: Bin Meng 
Tested-by: Bin Meng 


Re: [PATCH] x86: use invd instead of wbinvd in real mode start code

2020-01-29 Thread Simon Glass
On Wed, 8 Jan 2020 at 04:09, Masahiro Yamada  wrote:
>
> I do not know why the boot code immediately after the system reset
> should write-back the cache content. I think the cache invalidation
> should be enough.
>
> I tested this commit with qemu-x86_defconfig, and it worked for me.
>
> Signed-off-by: Masahiro Yamada 
> ---
>
>  arch/x86/cpu/start.S   | 2 +-
>  arch/x86/cpu/start16.S | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)

Reviewed-by: Simon Glass 


[PATCH] x86: use invd instead of wbinvd in real mode start code

2020-01-08 Thread Masahiro Yamada
I do not know why the boot code immediately after the system reset
should write-back the cache content. I think the cache invalidation
should be enough.

I tested this commit with qemu-x86_defconfig, and it worked for me.

Signed-off-by: Masahiro Yamada 
---

 arch/x86/cpu/start.S   | 2 +-
 arch/x86/cpu/start16.S | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/cpu/start.S b/arch/x86/cpu/start.S
index 01524635e9c8..26cf995db2d0 100644
--- a/arch/x86/cpu/start.S
+++ b/arch/x86/cpu/start.S
@@ -50,7 +50,7 @@ _x86boot_start:
movl%cr0, %eax
orl $(X86_CR0_NW | X86_CR0_CD), %eax
movl%eax, %cr0
-   wbinvd
+   invd
 
/*
 * Zero the BIST (Built-In Self Test) value since we don't have it.
diff --git a/arch/x86/cpu/start16.S b/arch/x86/cpu/start16.S
index 54f4ff6662a8..292e75050814 100644
--- a/arch/x86/cpu/start16.S
+++ b/arch/x86/cpu/start16.S
@@ -28,7 +28,7 @@ start16:
movl%cr0, %eax
orl $(X86_CR0_NW | X86_CR0_CD), %eax
movl%eax, %cr0
-   wbinvd
+   invd
 
/* load the temporary Global Descriptor Table */
 data32 cs  lidtidt_ptr
-- 
2.17.1