2023-12-21 10:18:18

by Yuntao Wang

Subject: [PATCH] x86/kexec: fix incorrect end address passed to kernel_ident_mapping_init()

kernel_ident_mapping_init() takes an exclusive memory range [pstart, pend)
where pend is not included in the range, while res represents an inclusive
memory range [start, end] where end is considered part of the range.

Therefore, passing res->end directly to kernel_ident_mapping_init() is
incorrect, the correct end address should be `res->end + 1`.

Signed-off-by: Yuntao Wang <[email protected]>
---
arch/x86/kernel/machine_kexec_64.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c
index 6f8df998890e..bc0a5348b4a6 100644
--- a/arch/x86/kernel/machine_kexec_64.c
+++ b/arch/x86/kernel/machine_kexec_64.c
@@ -44,7 +44,7 @@ static int mem_region_callback(struct resource *res, void *arg)
struct init_pgtable_data *data = arg;

return kernel_ident_mapping_init(data->info, data->level4p,
- res->start, res->end);
+ res->start, res->end + 1);
}

static int
--
2.43.0

2023-12-21 22:03:21

by Andrew Morton

Subject: Re: [PATCH] x86/kexec: fix incorrect end address passed to kernel_ident_mapping_init()

On Thu, 21 Dec 2023 18:17:02 +0800 Yuntao Wang <[email protected]> wrote:

> kernel_ident_mapping_init() takes an exclusive memory range [pstart, pend)
> where pend is not included in the range, while res represents an inclusive
> memory range [start, end] where end is considered part of the range.
>
> Therefore, passing res->end directly to kernel_ident_mapping_init() is
> incorrect, the correct end address should be `res->end + 1`.
>
> ...
>
> --- a/arch/x86/kernel/machine_kexec_64.c
> +++ b/arch/x86/kernel/machine_kexec_64.c
> @@ -44,7 +44,7 @@ static int mem_region_callback(struct resource *res, void *arg)
> struct init_pgtable_data *data = arg;
>
> return kernel_ident_mapping_init(data->info, data->level4p,
> - res->start, res->end);
> + res->start, res->end + 1);
> }

Thanks. When fixing a bug, please always full describe the
userspace-visible effects of that bug.

In this case I'm assuming "none", due to kernel_ident_mapping_init()'s
upward rounding of the address?

2023-12-22 15:03:29

by Yuntao Wang

Subject: Re: [PATCH] x86/kexec: fix incorrect end address passed to kernel_ident_mapping_init()

On Thu, 21 Dec 2023 14:03:00 -0800, Andrew Morton <[email protected]> wrote:

> On Thu, 21 Dec 2023 18:17:02 +0800 Yuntao Wang <[email protected]> wrote:
>
> > kernel_ident_mapping_init() takes an exclusive memory range [pstart, pend)
> > where pend is not included in the range, while res represents an inclusive
> > memory range [start, end] where end is considered part of the range.
> >
> > Therefore, passing res->end directly to kernel_ident_mapping_init() is
> > incorrect, the correct end address should be `res->end + 1`.
> >
> > ...
> >
> > --- a/arch/x86/kernel/machine_kexec_64.c
> > +++ b/arch/x86/kernel/machine_kexec_64.c
> > @@ -44,7 +44,7 @@ static int mem_region_callback(struct resource *res, void *arg)
> > struct init_pgtable_data *data = arg;
> >
> > return kernel_ident_mapping_init(data->info, data->level4p,
> > - res->start, res->end);
> > + res->start, res->end + 1);
> > }
>
> Thanks. When fixing a bug, please always full describe the
> userspace-visible effects of that bug.
>
> In this case I'm assuming "none", due to kernel_ident_mapping_init()'s
> upward rounding of the address?

Passing [start, end] rather than [start, end+1) to kernel_ident_mapping_init()
may result in the identity mapping for the end address not being set up.

For example, when res->start is equal to res->end, kernel_ident_mapping_init()
will not establish any identity mapping. Similarly, when the value of res->end
is a multiple of 2M and the page table maps 2M pages, kernel_ident_mapping_init()
will also not set up identity mapping for res->end.