Thank you, Eric!
Nice to know the recent change to wrap-at-100. Will this be reflected
somewhere, like, in Documentation/process/coding-style.rst?
Best,
-Ke
On 10/22/20, 12:49 AM, "Eric Dumazet" <eduma...@google.com> wrote:
On Thu, Oct 22, 2020 at 8:42 AM Ke Li <k...@akamai.com> wrote:
>
> In setsockopt(SO_MAX_PACING_RATE) on 64bit systems, sk_max_pacing_rate,
> after extended from 'u32' to 'unsigned long', takes unintentionally
> hiked value whenever assigned from an 'int' value with MSB=1, due to
> binary sign extension in promoting s32 to u64, e.g. 0x80000000 becomes
> 0xFFFFFFFF80000000.
>
> Thus inflated sk_max_pacing_rate causes subsequent getsockopt to return
> ~0U unexpectedly. It may also result in increased pacing rate.
>
> Fix by explicitly casting the 'int' value to 'unsigned int' before
> assigning it to sk_max_pacing_rate, for zero extension to happen.
>
> Fixes: 76a9ebe811fb ("net: extend sk_pacing_rate to unsigned long")
> Signed-off-by: Ji Li <j...@akamai.com>
> Signed-off-by: Ke Li <k...@akamai.com>
> Cc: Eric Dumazet <eduma...@google.com>
> ---
> v2: wrap the line in net/core/filter.c to less than 80 chars.
SGTM (the other version was also fine, the 80 chars rule has been
relaxed/changed to 100 recently)
Reviewed-by: Eric Dumazet <eduma...@google.com>
