public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Eric Biggers <ebiggers@kernel.org>
To: "zheng.gong" <zheng.gong@samsung.com>
Cc: linux-scsi@vger.kernel.org, avri.altman@wdc.com,
	bvanassche@acm.org, quic_cang@quicinc.com,
	alim.akhtar@samsung.com, martin.petersen@oracle.com,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 0/1] scsi: ufs: Add crypto_keyslot_remap variant op
Date: Thu, 27 Nov 2025 19:47:36 -0800	[thread overview]
Message-ID: <20251128034736.GA18644@sol> (raw)
In-Reply-To: <20251128033709.1342579-1-zheng.gong@samsung.com>

On Fri, Nov 28, 2025 at 11:37:08AM +0800, zheng.gong wrote:
> Thank you very much for your feedback, Eric. I truly appreciate your review and the time you've taken to point out that a real user is required.
> You're right. Adding a new variant op without a clear use case would not be acceptable. Let me clarify the context behind this patch.

To clarify, there has to be an in-tree user.

> This hook is not theoretical. It is designed to replace an existing out-of-tree variant op (`crypto_keyslot_cfg`) used in Samsung's ExynosAuto UFS driver for multi-VM inline encryption.
> In production, each VM has its own keyslot range per hardware allocation, and the keyslot is remapped at request time:
> 
>     lrbp->crypto_key_slot += vm_id * UFS_KEYSLOTS;
> 
> This was already in use on automotive platforms.
> 
> But the reason this usage isn't visible in mainline is due to ExynosAuto's kernel architecture:
> 
> Starting from kernel 6.1, we adopt the dual-repository model (similar to Android Common Kernel):
> - `kernel.git`: Mainline-based, minimal patches
> - `exynosauto-modules.git`: Hosts platform-specific drivers (as .ko or built-in)
> 
> Our UFS driver, including FMP and IOV support, resides in `exynosauto-modules/drivers/ufs/*`. It couldnot be upstreamed due to:
> - Hardware-specific SMC calls
> - Security-specific key management
> - Non-public register interfaces

The upstream driver is drivers/ufs/host/ufs-exynos.c, so you'll need to
focus on adding this functionality to there, if it's actually needed.
What's happening downstream is irrelevant.

- Eric

  parent reply	other threads:[~2025-11-28  3:49 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20251112030559epcas5p13358e7b05ca6b39688530b9c8178527e@epcas5p1.samsung.com>
     [not found] ` <20251112030524.3545394-1-zheng.gong@samsung.com>
2025-11-12  3:05   ` [PATCH 1/1] scsi: ufs: crypto: Add ufs_hba_variant_ops::crypto_keyslot_remap zheng.gong
2025-11-12  3:10     ` Eric Biggers
2025-11-27  7:06       ` [PATCH v2 0/2] scsi: ufs: Add crypto_keyslot_remap support zheng.gong
2025-11-27  7:06         ` [PATCH v2 1/2] scsi: ufs: crypto: Add ufs_hba_variant_ops::crypto_keyslot_remap zheng.gong
2025-11-27  7:06         ` [PATCH v2 2/2] scsi: ufs: Add crypto keyslot remapping test module zheng.gong
2025-11-27 22:24         ` [PATCH v2 0/2] scsi: ufs: Add crypto_keyslot_remap support Eric Biggers
2025-11-28  3:37       ` [PATCH v3 0/1] scsi: ufs: Add crypto_keyslot_remap variant op zheng.gong
2025-11-28  3:37         ` [PATCH v3 1/1] scsi: ufs: crypto: Add ufs_hba_variant_ops::crypto_keyslot_remap zheng.gong
2025-11-28  3:47         ` Eric Biggers [this message]
2026-01-29  3:10       ` [PATCH v4 0/3] scsi: ufs: Add crypto_keyslot_remap support zheng.gong
2026-01-29  3:10         ` [PATCH v4 1/3] scsi: ufs: crypto: Add ufs_hba_variant_ops::crypto_keyslot_remap zheng.gong
2026-01-29 16:43           ` Bart Van Assche
2026-01-29  3:10         ` [PATCH v4 2/3] scsi: ufs: exynos: Support crypto keyslot remapping via DT zheng.gong
2026-01-29  3:10         ` [PATCH v4 3/3] dt-bindings: ufs: Add binding for ufs-keyslot-offset zheng.gong
2026-01-29  3:31         ` [PATCH v4 0/3] scsi: ufs: Add crypto_keyslot_remap support Eric Biggers
2026-01-29  6:17       ` zheng.gong
2026-01-29  7:42         ` Eric Biggers

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20251128034736.GA18644@sol \
    --to=ebiggers@kernel.org \
    --cc=alim.akhtar@samsung.com \
    --cc=avri.altman@wdc.com \
    --cc=bvanassche@acm.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=quic_cang@quicinc.com \
    --cc=zheng.gong@samsung.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox