2023-03-24 05:28:34

by 电子

Subject: [PATCH] memblock: Correct calculation method for overflowing range @size

When memblock users to specify range where @base + @size overflows
and automatically cap it at maximum, The new size should be
PHYS_ADDR_MAX - @base + 1.

Assuming that base is 0, PHYS_ADDR_MAX is 0xff, which is 255 in decimal,
then @size should be 256 instead of 255

Signed-off-by: Hongbin Ji <[email protected]>
---
mm/memblock.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/memblock.c b/mm/memblock.c
index 25fd0626a9e7..f1683d1dae65 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -169,7 +169,7 @@ static enum memblock_flags __init_memblock choose_memblock_flags(void)
/* adjust *@size so that (@base + *@size) doesn't overflow, return new size */
static inline phys_addr_t memblock_cap_size(phys_addr_t base, phys_addr_t *size)
{
- return *size = min(*size, PHYS_ADDR_MAX - base);
+ return *size = min(*size, PHYS_ADDR_MAX - base + 1);
}

/*
--
2.34.1

2023-03-24 06:16:01

by 电子

Subject: Re:[PATCH] memblock: Correct calculation method for overflowing range @size

Sorry, the size here may be calculated according to this formula, cancel this patch




At 2023-03-24 13:23:51, "Hongbin Ji" <[email protected]> wrote:
>When memblock users to specify range where @base + @size overflows
>and automatically cap it at maximum, The new size should be
>PHYS_ADDR_MAX - @base + 1.
>
>Assuming that base is 0, PHYS_ADDR_MAX is 0xff, which is 255 in decimal,
>then @size should be 256 instead of 255
>
>Signed-off-by: Hongbin Ji <[email protected]>
>---
> mm/memblock.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
>diff --git a/mm/memblock.c b/mm/memblock.c
>index 25fd0626a9e7..f1683d1dae65 100644
>--- a/mm/memblock.c
>+++ b/mm/memblock.c
>@@ -169,7 +169,7 @@ static enum memblock_flags __init_memblock choose_memblock_flags(void)
> /* adjust *@size so that (@base + *@size) doesn't overflow, return new size */
> static inline phys_addr_t memblock_cap_size(phys_addr_t base, phys_addr_t *size)
> {
>- return *size = min(*size, PHYS_ADDR_MAX - base);
>+ return *size = min(*size, PHYS_ADDR_MAX - base + 1);
> }
>
> /*
>--
>2.34.1