> On 19 Jun 2019, at 16:09, Kevin 'ldir' Darbyshire-Bryant <ldir@xxxxxxxxxxxxxxxxxxxxxxx> wrote:
>
>
>
>> On 19 Jun 2019, at 15:13, David Miller <davem@xxxxxxxxxxxxx> wrote:
>>
>>
>> I've fixed this as follows, thanks:
>>
>> ====================
>> From 23cdf8752b26d4edbd60a6293bca492d83192d4d Mon Sep 17 00:00:00 2001
>> From: "David S. Miller" <davem@xxxxxxxxxxxxx>
>> Date: Wed, 19 Jun 2019 10:12:58 -0400
>> Subject: [PATCH] act_ctinfo: Don't use BIT() in UAPI headers.
>>
>> Use _BITUL() instead.
>>
>> Reported-by: Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx>
>> Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
>> ---
>> include/uapi/linux/tc_act/tc_ctinfo.h | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/include/uapi/linux/tc_act/tc_ctinfo.h b/include/uapi/linux/tc_act/tc_ctinfo.h
>> index da803e05a89b..32337304fbe5 100644
>> --- a/include/uapi/linux/tc_act/tc_ctinfo.h
>> +++ b/include/uapi/linux/tc_act/tc_ctinfo.h
>> @@ -27,8 +27,8 @@ enum {
>> #define TCA_CTINFO_MAX (__TCA_CTINFO_MAX - 1)
>>
>> enum {
>> - CTINFO_MODE_DSCP = BIT(0),
>> - CTINFO_MODE_CPMARK = BIT(1)
>> + CTINFO_MODE_DSCP = _BITUL(0),
>> + CTINFO_MODE_CPMARK = _BITUL(1)
>> };
>>
>> #endif
>> --
>> 2.20.1
>>
>
> Hi David,
>
> Thanks for that. Owe you a beer!
>
> Thinking out loud, doesn’t this also require:
>
> #include <linux/const.h>
>
>
> Or at least iproute2 would need to know about _BITUL as it doesn’t at present.
> Which also means iproute2’s Linux uapi shadow would also have to import
> linux/const.h. Or have I got wrong end of stick?
>
> Cheers,
>
> Kevin
>
Whilst out walking the dog I realised I’m a moron.
The CTINFO_MODE_FOO is only used within module, it shouldn’t even be exposed
to user space.
I’ll send a patch shortly. Sorry, I’m pretty embarrassed at how something so
apparently simple on the surface has had so many issues.
Cheers,
Kevin D-B
gpg: 012C ACB2 28C6 C53E 9775 9123 B3A2 389B 9DE2 334A
Attachment:
signature.asc
Description: Message signed with OpenPGP
