linux-kernel - Re: [PATCH 4/6] scsi: ufs: exynos: Enable PRDT pre-fetching with UFSHCD_CAP

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: <CADrjBPpju3MmZbNy1uaPzAWTWrmNHx0nT+03DmkM3p5qFEUHdA@mail.gmail.com>
Date: Fri, 7 Mar 2025 15:10:33 +0000
From: Peter Griffin <peter.griffin@...aro.org>
To: Eric Biggers <ebiggers@...nel.org>
Cc: alim.akhtar@...sung.com, James.Bottomley@...senpartnership.com, 
	martin.petersen@...cle.com, krzk@...nel.org, linux-scsi@...r.kernel.org, 
	linux-samsung-soc@...r.kernel.org, linux-arm-kernel@...ts.infradead.org, 
	linux-kernel@...r.kernel.org, willmcvicker@...gle.com, 
	tudor.ambarus@...aro.org, andre.draszik@...aro.org, bvanassche@....org, 
	kernel-team@...roid.com
Subject: Re: [PATCH 4/6] scsi: ufs: exynos: Enable PRDT pre-fetching with UFSHCD_CAP_CRYPTO

Hi Eric,

Thanks for your review feedback.

On Wed, 5 Mar 2025 at 02:40, Eric Biggers <ebiggers@...nel.org> wrote:
>
> On Wed, Feb 26, 2025 at 10:04:12PM +0000, Peter Griffin wrote:
> > PRDT_PREFETCH_ENABLE[31] bit should be set when desctype field of
> > fmpsecurity0 register is type2 (double file encryption) or type3
> > (file and disk excryption). Setting this bit enables PRDT
> > pre-fetching on both TXPRDT and RXPRDT.
> >
> > Signed-off-by: Peter Griffin <peter.griffin@...aro.org>
>
> I assume you mean that desctype 3 provides "support for file and disk
> encryption"?

Yes, the PRDT_PREFETCH_ENABLE description in the commit message I
copied from the datasheet. But I can re-word it like you suggest if
you think that it's clearer? I notice now there is also a typo with
the word 'encryption' which I can also fix.

This patch came about whilst comparing UFS SFR register dumps of
upstream and downstream drivers. I noticed that PRDT_PREFETCH_ENABLE
is enabled downstream but not upstream, and after checking the
datasheet description it seemed like we should set this if
exynos_ufs_fmp_init() completed and set CFG_DESCTYPE_3.

> The driver does use desctype 3, but it only uses the "file
> encryption".  So this confused me a bit.  (BTW, in FMP terminology, "file
> encryption" seems to mean "use the key provided in the I/O request", and "disk
> encryption" seems to mean "use some key the firmware provided somehow".  They
> can be cascaded, and the intended use cases are clearly file and disk encryption
> respectively, but they don't necessarily have to be used that way.)

Thanks for the additional context :)

Peter