On Fri, 2019-11-08 at 21:34 +0100, Arnd Bergmann wrote:
> The layout of struct timeval is different on sparc64 from
> anything else, and the patch I did long ago failed to take
> this into account.
>
> Change it now to handle sparc64 user space correctly again.
>
> Quite likely nobody cares about parallel ports on sparc64,
> but there is no reason not to fix it.
>
> Cc: stable@xxxxxxxxxxxxxxx
> Fixes: 9a450484089d ("lp: support 64-bit time_t user space")
> Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
> ---
> drivers/char/lp.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/drivers/char/lp.c b/drivers/char/lp.c
> index 7c9269e3477a..bd95aba1f9fe 100644
> --- a/drivers/char/lp.c
> +++ b/drivers/char/lp.c
> @@ -713,6 +713,10 @@ static int lp_set_timeout64(unsigned int minor, void __user *arg)
> if (copy_from_user(karg, arg, sizeof(karg)))
> return -EFAULT;
>
> + /* sparc64 suseconds_t is 32-bit only */
> + if (IS_ENABLED(CONFIG_SPARC64) && !in_compat_syscall())
> + karg[1] >>= 32;
> +
> return lp_set_timeout(minor, karg[0], karg[1]);
> }
>
It seems like it would make way more sense to use __kernel_old_timeval.
Then you don't have to explicitly handle the sparc64 oddity.
As it is, this still over-reads from user-space which might result in a
spurious -EFAULT.
Ben.
--
Ben Hutchings, Software Developer Codethink Ltd
https://www.codethink.co.uk/ Dale House, 35 Dale Street
Manchester, M1 2HF, United Kingdom
