[PATCH v2] um: add a UML specific futex implementation

[Richard Weinberger]

On Thu, Nov 12, 2020 at 8:57 PM <anton.ivanov at cambridgegreys.com> wrote:
>
> From: Anton Ivanov <anton.ivanov at cambridgegreys.com>
>
> The generic asm futex implementation emulates atomic access to
> memory by doing a get_user followed by put_user. These translate
> to two mapping operations on UML with paging enabled in the
> meantime. This, in turn may end up changing interrupts,
> invoking the signal loop, etc.
>
> This replaces the generic implementation by a mapping followed
> by an operation on the mapped segment. It is for now limited
> to 64 bit as the 32 bit may also need to take care of highmem,
> etc.
>
> Signed-off-by: Anton Ivanov <anton.ivanov at cambridgegreys.com>
> ---
>  arch/um/include/asm/futex.h   |  20 ++++++
>  arch/um/kernel/skas/uaccess.c | 119 ++++++++++++++++++++++++++++++++++
>  2 files changed, 139 insertions(+)
>  create mode 100644 arch/um/include/asm/futex.h
>
> diff --git a/arch/um/include/asm/futex.h b/arch/um/include/asm/futex.h
> new file mode 100644
> index 000000000000..9af35ab66b6e
> --- /dev/null
> +++ b/arch/um/include/asm/futex.h
> @@ -0,0 +1,20 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +#ifndef _ASM_GENERIC_FUTEX_H
> +#define _ASM_GENERIC_FUTEX_H
> +
> +#ifdef CONFIG_64BIT
> +
> +#include <linux/futex.h>
> +#include <linux/uaccess.h>
> +#include <asm/errno.h>
> +
> +
> +extern int arch_futex_atomic_op_inuser(int op, u32 oparg, int *oval, u32 __user *uaddr);
> +extern int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
> +                             u32 oldval, u32 newval);
> +
> +#else
> +#include <"asm-generic/futex.h">
> +#endif /* CONFIG_64BIT */
> +
> +#endif
> diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c
> index 2dec915abe6f..21a8f2b283bb 100644
> --- a/arch/um/kernel/skas/uaccess.c
> +++ b/arch/um/kernel/skas/uaccess.c
> @@ -11,6 +11,7 @@
>  #include <asm/current.h>
>  #include <asm/page.h>
>  #include <kern_util.h>
> +#include <asm/futex.h>
>  #include <os.h>
>
>  pte_t *virt_to_pte(struct mm_struct *mm, unsigned long addr)
> @@ -248,3 +249,121 @@ long __strnlen_user(const void __user *str, long len)
>         return 0;
>  }
>  EXPORT_SYMBOL(__strnlen_user);
> +
> +/**
> + * arch_futex_atomic_op_inuser() - Atomic arithmetic operation with constant
> + *                       argument and comparison of the previous
> + *                       futex value with another constant.
> + *
> + * @encoded_op:        encoded operation to execute
> + * @uaddr:     pointer to user space address
> + *
> + * Return:
> + * 0 - On success
> + * -EFAULT - User access resulted in a page fault
> + * -EAGAIN - Atomic operation was unable to complete due to contention
> + * -ENOSYS - Operation not supported
> + */
> +int arch_futex_atomic_op_inuser(int op, u32 oparg, int *oval, u32 __user *uaddr)

Does this even build in 32bits mode?
AFAICT the generic futex implementation has
arch_futex_atomic_op_inuser() as static inline...

-- 
Thanks,
//richard