[PATCH] powerpc/32s: Fix shift-out-of-bounds in KASAN init

[Erhard F.]

On Tue, 30 Nov 2021 09:42:37 +0100
Christophe Leroy <christophe.leroy at csgroup.eu> wrote:

> ================================================================================
> UBSAN: shift-out-of-bounds in arch/powerpc/mm/kasan/book3s_32.c:22:23
> shift exponent -1 is negative
> CPU: 0 PID: 0 Comm: swapper Not tainted 5.15.5-gentoo-PowerMacG4 #9
> Call Trace:
> [c214be60] [c0ba0048] dump_stack_lvl+0x80/0xb0 (unreliable)
> [c214be80] [c0b99288] ubsan_epilogue+0x10/0x5c
> [c214be90] [c0b98fe0] __ubsan_handle_shift_out_of_bounds+0x94/0x138
> [c214bf00] [c1c0f010] kasan_init_region+0xd8/0x26c
> [c214bf30] [c1c0ed84] kasan_init+0xc0/0x198
> [c214bf70] [c1c08024] setup_arch+0x18/0x54c
> [c214bfc0] [c1c037f0] start_kernel+0x90/0x33c
> [c214bff0] [00003610] 0x3610
> ================================================================================
> 
> This happens when the directly mapped memory is a power of 2.
> 
> Fix it by checking the shift and set the result to 0 when shift is -1
> 
> Reported-by: Erhard Furtner <erhard_f at mailbox.org>
> Link: https://bugzilla.kernel.org/show_bug.cgi?id=215169
> Fixes: 7974c4732642 ("powerpc/32s: Implement dedicated kasan_init_region()")
> Signed-off-by: Christophe Leroy <christophe.leroy at csgroup.eu>
> ---
>  arch/powerpc/mm/kasan/book3s_32.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/powerpc/mm/kasan/book3s_32.c b/arch/powerpc/mm/kasan/book3s_32.c
> index 202bd260a009..35b287b0a8da 100644
> --- a/arch/powerpc/mm/kasan/book3s_32.c
> +++ b/arch/powerpc/mm/kasan/book3s_32.c
> @@ -19,7 +19,8 @@ int __init kasan_init_region(void *start, size_t size)
>  	block = memblock_alloc(k_size, k_size_base);
>  
>  	if (block && k_size_base >= SZ_128K && k_start == ALIGN(k_start, k_size_base)) {
> -		int k_size_more = 1 << (ffs(k_size - k_size_base) - 1);
> +		int shift = ffs(k_size - k_size_base);
> +		int k_size_more = shift ? 1 << (shift - 1) : 0;
>  
>  		setbat(-1, k_start, __pa(block), k_size_base, PAGE_KERNEL);
>  		if (k_size_more >= SZ_128K)
> -- 
> 2.33.1
> 

Tested Christophes patch applied on 5.15.5 which works fine. Thanks!

I'll close the bugzilla report once it gets into stable.

Regards
Erhard