Re: [PATCH] tcp: fix tcp_fastopen unaligned access complaints on sparc
From: Eric Dumazet
Date: Thu Jan 12 2017 - 15:13:47 EST
On Thu, 2017-01-12 at 11:59 -0800, Shannon Nelson wrote:
> Fix up a data alignment issue on sparc by swapping the order
> of the cookie byte array field with the length field in
> struct tcp_fastopen_cookie
>
> This addresses log complaints like these:
> log_unaligned: 113 callbacks suppressed
> Kernel unaligned access at TPC[976490] tcp_try_fastopen+0x2d0/0x360
> Kernel unaligned access at TPC[9764ac] tcp_try_fastopen+0x2ec/0x360
> Kernel unaligned access at TPC[9764c8] tcp_try_fastopen+0x308/0x360
> Kernel unaligned access at TPC[9764e4] tcp_try_fastopen+0x324/0x360
> Kernel unaligned access at TPC[976490] tcp_try_fastopen+0x2d0/0x360
>
> Signed-off-by: Shannon Nelson <shannon.nelson@xxxxxxxxxx>
> ---
> include/linux/tcp.h | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/include/linux/tcp.h b/include/linux/tcp.h
> index fc5848d..95cda75 100644
> --- a/include/linux/tcp.h
> +++ b/include/linux/tcp.h
> @@ -62,8 +62,8 @@ static inline unsigned int tcp_optlen(const struct sk_buff *skb)
>
> /* TCP Fast Open Cookie as stored in memory */
> struct tcp_fastopen_cookie {
> - s8 len;
> u8 val[TCP_FASTOPEN_COOKIE_MAX];
> + s8 len;
> bool exp; /* In RFC6994 experimental option format */
> };
>
Strange... Do you have an explanation of why this patch would be
needed ? A compiler issue ?
s8 and u8 are bytes after all.