linux-kernel - Re: [PATCH] mmc: core: add auto bkops support

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]

Date:	Mon, 27 Jun 2016 11:30:48 +0000
From:	Alex Lemberg <Alex.Lemberg@...disk.com>
To:	Ulf Hansson <ulf.hansson@...aro.org>,
	Adrian Hunter <adrian.hunter@...el.com>
CC:	Shawn Lin <shawn.lin@...k-chips.com>,
	Jaehoon Chung <jh80.chung@...sung.com>,
	"linux-mmc@...r.kernel.org" <linux-mmc@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Doug Anderson <dianders@...omium.org>,
	"linux-rockchip@...ts.infradead.org" 
	<linux-rockchip@...ts.infradead.org>
Subject: Re: [PATCH] mmc: core: add auto bkops support

Hi Ulf and Adrian,


On 6/27/16, 12:08 PM, "Ulf Hansson" <ulf.hansson@...aro.org> wrote:

>[...]
>
>>>> If I am not wrong, in current implementation of runtime suspend,
>>>> the driver stops BKOPS (send HPI) just before sending sleep command,
>>>> see _mmc_suspend(), depends on “MMC_CAP_AGGRESSIVE_PM” flag.
>>>> In this case, the eMMC device will not have enough time to perform internal
>>>> BKOPS in both – Manual and Auto BKOPS configurations.
>>>>
>>>
>>> ye, so it seems a pre-exiting issue before introducing auto bkops?
>>> I think we can push another patch to improve it but not handling
>>> it for this $SUBJECT, does it sound ok to you?
>>
>> Runtime suspend for eMMC has a default auto-suspend delay of 3 seconds
>> (refer mmc_blk_probe()).  Isn't that when auto bkops would happen?
>
>That's correct.
>
>So perhaps we should extend the default auto-suspend delay?

Due to the differences in flash management and technologies,
I suggest to make this value configurable, and not hard coded.
I believe it should be configured per vendor BKOPS needs.
The default value still can be set to 3 seconds, as done today…

>
>The SD case actually have the same issue, as I some background
>operations exists there as well.
>
>[...]
>
>Kind regards
>Uffe