linux-kernel - Re: [PATCH v3 0/1] libata: Add ATA_HORKAGE_NONCQ_ON

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] [day] [month] [year] [list]

Message-ID: <a76b2acd-edfa-55b2-907c-f1faa8ce4914@redhat.com>
Date:   Tue, 31 Aug 2021 09:38:04 +0200
From:   Hans de Goede <hdegoede@...hat.com>
To:     Damien Le Moal <Damien.LeMoal@....com>,
        Kate Hsuan <hpa@...hat.com>, Jens Axboe <axboe@...nel.dk>
Cc:     "linux-ide@...r.kernel.org" <linux-ide@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3 0/1] libata: Add ATA_HORKAGE_NONCQ_ON_AMD for Samsung
 860 and 870 SSD.

Hi,

On 8/30/21 11:45 PM, Damien Le Moal wrote:
> On 2021/08/30 23:43, Kate Hsuan wrote:
>> Many users reported the issue when running the system with Samsung 860,
>> 870 SSD, and AMD chipset. Therefore, completely disabling the NCQ can
>> avoid this issue.
>>
>> Entire disabling NCQ for Samsung 860/870 SSD will cause I/O performance
>> drop. In this case, a flag ATA_HORKAGE_NONCQ_ON_AMD is introduced to
>> used to perform an additional check for these SSDs. If it finds its parent
>> ATA controller is AMD, the NCQ will be disabled. Otherwise, the NCQ is kept
>> to enable.
> 
> For a single patch, generally, a cover letter is not needed. Especially so in
> this case since your cover letter message is the same as the patch commit message.
> 
>>
>> Changes since v3
>> * Modified the flag from ATA_HORKAGE_NONCQ_ON_ASMEDIA_AMD_MARVELL to
>>   ATA_HORKAGE_NONCQ_ON_AMD.
>> * Modified and fixed the code to completely disable NCQ on AMD controller.
> 
> Technically, this is a v2 right ? Also, by "completely", did you mean "always" ?
> (see patch comments).

Right, as mentioned in my reply to the v1 which was accidentally labelled v2
I suggested to just make this v3 to avoid confusion, otherwise we would have
2 v2-s.

Regards,

Hans