netdev - Re: [PATCH net-next v2] r8169: improve interrupt handling

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <83b2824d-187c-de76-f6e3-14084f216d59@gmail.com>
Date:   Mon, 26 Feb 2018 20:50:32 +0100
From:   Heiner Kallweit <hkallweit1@...il.com>
To:     David Miller <davem@...emloft.net>
Cc:     nic_swsd@...ltek.com, romieu@...zoreil.com, netdev@...r.kernel.org
Subject: Re: [PATCH net-next v2] r8169: improve interrupt handling

Am 26.02.2018 um 19:56 schrieb David Miller:
> From: Heiner Kallweit <hkallweit1@...il.com>
> Date: Sat, 24 Feb 2018 16:53:23 +0100
> 
>> @@ -736,8 +736,7 @@ struct ring_info {
>>  };
>>  
>>  enum features {
>> -	RTL_FEATURE_MSI		= (1 << 0),
>> -	RTL_FEATURE_GMII	= (1 << 1),
>> +	RTL_FEATURE_GMII	= (1 << 0),
>>  };
>>  
>>  struct rtl8169_counters {
>  ...
>> +	if (tp->mac_version <= RTL_GIGA_MAC_VER_06) {
> 
> Please, if you are going to keep the logic the same for the older
> chips, just keep the RTL_FEATURE_MSI flag around instead of adding
> new (and potentially regression causing) tests for this condition.
> 
I see your point. In the case here the condition is meant to be true
for chip versions:
- having the MSIEnable bit
- being PCI, not PCIe

Both is true for chip versions <= 06 only, as can be seen in different
places in the driver, e.g.
- where bit MSIEnable is defined comment says: /* 8169 only. Reserved in the 8168. */
- array rtl_chip_infos[] definition shows that only versions <= 06
  are named RTL8169xx and are marked as PCI

Last but not least condition "chip version <= 06" is used also in
other places in the driver when it's about the RTL8169xx PCI chips.

At least I'm convinced this gives enough confidence that we can get
rid of flag RTL_FEATURE_MSI.

> Thank you.
>