Re: [Qemu-devel] [PATCH V5 3/3] linux-user: make host_to_target_cmsg support SO_TIMESTAMP cmsg_type

[Peter Maydell]

On 24 July 2012 23:33, Jing Huang <address@hidden> wrote:
>
> Signed-off-by: Jing Huang <address@hidden>
> Reviewed-by: Peter Maydell <address@hidden>

Please don't add a reviewed-by line for me unless I specifically
say it's OK to do so.

> ---
>  linux-user/syscall.c |   20 ++++++++++++++++----
>  1 files changed, 16 insertions(+), 4 deletions(-)
>
> diff --git a/linux-user/syscall.c b/linux-user/syscall.c
> index cc2a1c3..a5fa7a7 100644
> --- a/linux-user/syscall.c
> +++ b/linux-user/syscall.c
> @@ -1347,16 +1347,28 @@ static inline abi_long host_to_target_cmsg(struct 
> target_msghdr *target_msgh,
>          target_cmsg->cmsg_type = tswap32(cmsg->cmsg_type);
>          target_cmsg->cmsg_len = tswapal(TARGET_CMSG_LEN(len));
>
> -        if (cmsg->cmsg_level != TARGET_SOL_SOCKET || cmsg->cmsg_type != 
> SCM_RIGHTS) {
> -            gemu_log("Unsupported ancillary data: %d/%d\n", 
> cmsg->cmsg_level, cmsg->cmsg_type);
> -            memcpy(target_data, data, len);
> -        } else {
> +        if ((cmsg->cmsg_level == TARGET_SOL_SOCKET) &&
> +                                (cmsg->cmsg_type == SCM_RIGHTS)) {
>              int *fd = (int *)data;
>              int *target_fd = (int *)target_data;
>              int i, numfds = len / sizeof(int);
>
>              for (i = 0; i < numfds; i++)
>                  target_fd[i] = tswap32(fd[i]);
> +        } else if ((cmsg->cmsg_level == TARGET_SOL_SOCKET) &&
> +                                (cmsg->cmsg_type == SO_TIMESTAMP) &&
> +                                (len == sizeof(struct timeval))) {
> +            /* copy struct timeval to target */
> +            struct timeval *tv = (struct timeval *)data;
> +            struct target_timeval *target_tv =
> +                                 (struct target_timeval *)target_data;
> +
> +            target_tv->tv_sec = tswapl(tv->tv_sec);
> +            target_tv->tv_usec = tswapl(tv->tv_usec);

I hate to make you roll yet another version of these patches but I've
just noticed:
these should be tswapal(), not tswapl(), because the fields in
target_timeval are of type abi_ulong. (This makes a difference
only for a few targets which define TARGET_ABI32 to indicate that
the target's "long" type is 32 bits even though pointers are 64 bits.)

That really is the last thing, I promise.

-- PMM