[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <401273a6-37f7-8a27-457b-dbd73a0bbc97@gmail.com>
Date: Fri, 3 Dec 2021 12:37:34 +0100
From: Heiner Kallweit <hkallweit1@...il.com>
To: Hayes Wang <hayeswang@...ltek.com>,
Jakub Kicinski <kuba@...nel.org>,
David Miller <davem@...emloft.net>
Cc: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
nic_swsd <nic_swsd@...ltek.com>
Subject: Re: [RFC PATCH 3/4] r8169: support CMAC
On 03.12.2021 08:57, Hayes Wang wrote:
> Heiner Kallweit <hkallweit1@...il.com>
>> Sent: Tuesday, November 30, 2021 4:47 AM
>> To: Hayes Wang <hayeswang@...ltek.com>; Jakub Kicinski <kuba@...nel.org>;
> [...]
>>> +struct rtl_dash {
>>> + struct rtl8169_private *tp;
>>> + struct pci_dev *pdev_cmac;
>>> + void __iomem *cmac_ioaddr;
>>> + struct cmac_desc *tx_desc, *rx_desc;
>>> + struct page *tx_buf, *rx_buf;
>>> + struct dash_tx_info tx_info[CMAC_DESC_NUM];
>>> + struct tasklet_struct tl;
>>
>> Please see the following in include/linux/interrupt.h:
>>
>> /* Tasklets --- multithreaded analogue of BHs.
>>
>> This API is deprecated. Please consider using threaded IRQs instead:
>
> How about replacing the tasklet with work?
> It seems that the bottom half of threaded IRQ
> is the work, too.
>
Right, it's running in process context. It's to a certain extent
comparable with the efforts to move NAPI processing from softirg
context to threads.
> Best Regards,
> Hayes
>
>
Powered by blists - more mailing lists