2021-12-27 09:09:34

by Hangyu Hua

Subject: [PATCH] m68k/kernel: array out of bound access in process_uboot_commandline

When the size of commandp >= size, array out of bound write occurs because
len == 0.

Signed-off-by: Hangyu Hua <[email protected]>
---
arch/m68k/kernel/uboot.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/arch/m68k/kernel/uboot.c b/arch/m68k/kernel/uboot.c
index 928dbd33fc4a..05eca6f653b5 100644
--- a/arch/m68k/kernel/uboot.c
+++ b/arch/m68k/kernel/uboot.c
@@ -101,5 +101,8 @@ __init void process_uboot_commandline(char *commandp, int size)
}

parse_uboot_commandline(commandp, len);
- commandp[len - 1] = 0;
+ if (len > 0)
+ commandp[len - 1] = 0;
+ else
+ commandp[0] = 0;
}
--
2.25.1

2021-12-27 09:28:04

by Andreas Schwab

Subject: Re: [PATCH] m68k/kernel: array out of bound access in process_uboot_commandline

On Dez 27 2021, Hangyu Hua wrote:

> diff --git a/arch/m68k/kernel/uboot.c b/arch/m68k/kernel/uboot.c
> index 928dbd33fc4a..05eca6f653b5 100644
> --- a/arch/m68k/kernel/uboot.c
> +++ b/arch/m68k/kernel/uboot.c
> @@ -101,5 +101,8 @@ __init void process_uboot_commandline(char *commandp, int size)
> }
>
> parse_uboot_commandline(commandp, len);
> - commandp[len - 1] = 0;
> + if (len > 0)
> + commandp[len - 1] = 0;
> + else
> + commandp[0] = 0;

If len == 0 then even commandp[0] is OOB.

--
Andreas Schwab, [email protected]
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510 2552 DF73 E780 A9DA AEC1
"And now for something completely different."

2021-12-27 11:53:14

by Hangyu Hua

Subject: Re: [PATCH] m68k/kernel: array out of bound access in process_uboot_commandline

But commandp[len -1] is used to add a zero-terminated. If we don't use
commandp[0]=0 in len == 0,
than commandp will not have a zero-terminated. I think strings may
make some errors beacause of this.

Thanks.


Andreas Schwab <[email protected]> 于2021年12月27日周一 17:19写道：
>
> On Dez 27 2021, Hangyu Hua wrote:
>
> > diff --git a/arch/m68k/kernel/uboot.c b/arch/m68k/kernel/uboot.c
> > index 928dbd33fc4a..05eca6f653b5 100644
> > --- a/arch/m68k/kernel/uboot.c
> > +++ b/arch/m68k/kernel/uboot.c
> > @@ -101,5 +101,8 @@ __init void process_uboot_commandline(char *commandp, int size)
> > }
> >
> > parse_uboot_commandline(commandp, len);
> > - commandp[len - 1] = 0;
> > + if (len > 0)
> > + commandp[len - 1] = 0;
> > + else
> > + commandp[0] = 0;
>
> If len == 0 then even commandp[0] is OOB.
>
> --
> Andreas Schwab, [email protected]
> GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510 2552 DF73 E780 A9DA AEC1
> "And now for something completely different."

2021-12-27 11:56:50

by Andreas Schwab

Subject: Re: [PATCH] m68k/kernel: array out of bound access in process_uboot_commandline

On Dez 27 2021, Hangyu Hua wrote:

> If we don't use
> commandp[0]=0 in len == 0,
> than commandp will not have a zero-terminated.

That doesn't make sense. There is no room for the zero.

--
Andreas Schwab, [email protected]
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510 2552 DF73 E780 A9DA AEC1
"And now for something completely different."

2021-12-27 12:50:17

by Hangyu Hua

Subject: Re: [PATCH] m68k/kernel: array out of bound access in process_uboot_commandline

You are right. I will resubmit the patch later.

Thanks.

Andreas Schwab <[email protected]> 于2021年12月27日周一 19:56写道：
>
> On Dez 27 2021, Hangyu Hua wrote:
>
> > If we don't use
> > commandp[0]=0 in len == 0,
> > than commandp will not have a zero-terminated.
>
> That doesn't make sense. There is no room for the zero.
>
> --
> Andreas Schwab, [email protected]
> GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510 2552 DF73 E780 A9DA AEC1
> "And now for something completely different."