Re: [PATCH v4 1/1] x86_64: fix KASan shadow region page tables

[Andrey Ryabinin <a.ryabinin@samsung.com>]

On 06/03/2015 10:44 AM, Ingo Molnar wrote:
> 
> * Alexander Popov <alpopov@ptsecurity.com> wrote:
> 
>>  #ifdef CONFIG_KASAN
>>  void __init kasan_map_early_shadow(pgd_t *pgd);
>> +void __init kasan_early_init(void);
>>  void __init kasan_init(void);
>>  #else
>>  static inline void kasan_map_early_shadow(pgd_t *pgd) { }
>> +static inline void kasan_early_init(void) { }
>>  static inline void kasan_init(void) { }
>>  #endif
>>  
>> diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
>> index 2b55ee6..e9a84a1 100644
>> --- a/arch/x86/kernel/head64.c
>> +++ b/arch/x86/kernel/head64.c
>> @@ -161,11 +161,15 @@ asmlinkage __visible void __init x86_64_start_kernel(char * real_mode_data)
>>  	/* Kill off the identity-map trampoline */
>>  	reset_early_page_tables();
>>  
>> -	kasan_map_early_shadow(early_level4_pgt);
>> -
>> -	/* clear bss before set_intr_gate with early_idt_handler */
>> +	/*
>> +	 * Clear bss before kasan_early_init and set_intr_gate
>> +	 * with early_idt_handler
>> +	 */
>>  	clear_bss();
>>  
>> +	kasan_early_init();
>> +	kasan_map_early_shadow(early_level4_pgt);
> 
> So why isn't kasan_map_early_shadow() called in kasan_early_init()?
> 
> High level x86 init code should not be polluted with too many low level details.
> 

Agreed. Eventually, with the patch bellow, we could get rid of the second
kasan_map_early_shadow(init_level4_pgt) call in x86_64_start_kernel().
Make it static, and call it from kasan_early_init() only.

------------------------------------------------------
From: Andrey Ryabinin <a.ryabinin@samsung.com>
Subject: [PATCH] x86_64: remove not needed clear_page for init_level4_page

Commit 8170e6bed465 ("x86, 64bit: Use a #PF handler to materialize
early mappings on demand") introduced clear_page(init_level4_pgt);
call in x86_64_start_kernel(). However, this clear_page is useless
because init_level4_page already filled with zeroes in head_64.S

Commit message in 8170e6bed465 says that this clear_page() was
dropped in v7, but it accidentally reappeared in later versions
of that patchset.

Signed-off-by: Andrey Ryabinin <a.ryabinin@samsung.com>
---
 arch/x86/kernel/head64.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 5a46681..6a6eefd 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -177,7 +177,6 @@ asmlinkage __visible void __init x86_64_start_kernel(char * real_mode_data)
 	 */
 	load_ucode_bsp();

-	clear_page(init_level4_pgt);
 	/* set init_level4_pgt kernel high mapping*/
 	init_level4_pgt[511] = early_level4_pgt[511];