Re: [PATCH] tcp: Use BIT() for OPTION_* constants

[Eric Dumazet]

On 11/4/21 2:17 AM, David Laight wrote:
> From: Eric Dumazet
>> Sent: 03 November 2021 22:50
>>
>> On 11/3/21 3:17 PM, Leonard Crestez wrote:
>>> Extending these flags using the existing (1 << x) pattern triggers
>>> complaints from checkpatch. Instead of ignoring checkpatch modify the
>>> existing values to use BIT(x) style in a separate commit.
>>>
>>> Signed-off-by: Leonard Crestez <cdleonard@xxxxxxxxx>
>>>
>>
>> Yes, I guess checkpatch does not know that we currently use at most 16 bits :)
>>
>> u16 options = opts->options;
>>
>> Anyway, this seems fine.
> 
> Doesn't BIT() have a nasty habit of generating 64bit constants
> that just cause a different set of issues when inverted?
> It may be safe here - but who knows.

BIT() does not use/force 64bit constants, plain "unsigned long" ones.

Really this patch looks a nop to me.