[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <253y1t5weqa.fsf@mtr-vdi-124.i-did-not-set--mail-host-address--so-tickle-me>
Date: Mon, 24 Oct 2022 16:09:01 +0300
From: Aurelien Aptel <aaptel@...dia.com>
To: Leon Romanovsky <leon@...nel.org>
Cc: netdev@...r.kernel.org, davem@...emloft.net, kuba@...nel.org,
edumazet@...gle.com, pabeni@...hat.com, saeedm@...dia.com,
tariqt@...dia.com, linux-nvme@...ts.infradead.org,
sagi@...mberg.me, hch@....de, kbusch@...nel.org, axboe@...com,
chaitanyak@...dia.com, smalin@...dia.com, ogerlitz@...dia.com,
yorayz@...dia.com, borisp@...dia.com, aurelien.aptel@...il.com,
malin1024@...il.com
Subject: Re: [PATCH v6 01/23] net: Introduce direct data placement tcp offload
Leon Romanovsky <leon@...nel.org> writes:
>> __u8 csum_not_inet:1;
>> +#ifdef CONFIG_ULP_DDP
>> + __u8 ulp_ddp:1;
>> + __u8 ulp_crc:1;
>> +#define IS_ULP_DDP(skb) ((skb)->ulp_ddp)
>> +#define IS_ULP_CRC(skb) ((skb)->ulp_crc)
>> +#else
>> +#define IS_ULP_DDP(skb) (0)
>> +#define IS_ULP_CRC(skb) (0)
>
> All users of this define are protected by ifdef CONFIG_ULP_DDP.
> It is easier to wrap user of IS_ULP_DDP() too and remove #else
> lag from here.
We have changed the macros to inline functions (as suggested by Jakub).
There are other users in later patches which are not wrapped. In fact
we could remove all the #ifdef wraps in this commit but I followed the
conventions of the file.
>> + * ulp_ddp.h
>> + * Author: Boris Pismenny <borisp@...dia.com>
>> + * Copyright (C) 2022 NVIDIA CORPORATION & AFFILIATES.
>
> The official format is:
> Copyright (C) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
> ^^^^ ^^^^^^^^^^^^^^^^^^
Sure, will update.
>> +config ULP_DDP
>> + bool "ULP direct data placement offload"
>> + default n
>
> No need to set "n" explicitly, it is already default.
Sure, will update.
Thanks
Powered by blists - more mailing lists