From: Jay Wang <wanjay@amazon.com>
To: <ebiggers@kernel.org>
Cc: <bp@alien8.de>, <catalin.marinas@arm.com>, <davem@davemloft.net>,
<herbert@gondor.apana.org.au>, <ilya.okomin@oracle.com>,
<jay.wang.upstream@gmail.com>,
<linux-arm-kernel@lists.infradead.org>,
<linux-crypto@vger.kernel.org>, <linux-kbuild@vger.kernel.org>,
<linux-modules@vger.kernel.org>, <mcgrof@kernel.org>,
<mingo@redhat.com>, <nathan@kernel.org>, <nsc@kernel.org>,
<nstange@suse.de>, <petr.pavlu@suse.com>, <tglx@kernel.org>,
<vegard.nossum@oracle.com>, <wanjay@amazon.com>,
<will@kernel.org>, <x86@kernel.org>
Subject: RE: [PATCH 17/17] Add standalone crypto kernel module technical documentation
Date: Wed, 25 Feb 2026 17:35:45 +0000 [thread overview]
Message-ID: <20260225173551.8704-1-wanjay@amazon.com> (raw)
In-Reply-To: <20260225015517.GA162634@quark>
Thanks to everyone who has taken the time to review and comment.
Based on the discussion so far, it seems the core question is whether this feature is appropriate for upstream at all, or whether it should remain entirely downstream.
We’ve discussed this with a few folks, and to help clarify the discussion, let me outline what is actually required to use this approach and what the community would gain from upstreaming it:
1. Maintaining a crypto snapshot (for example, a source code snapshot used for streamlined FIPS re-validation processes such as NSRL).
2. Maintaining the loadable crypto build infrastructure itself (i.e., this patch set).
For (1), since this requires maintaining a specific snapshot, we acknowledge that this is inherently a downstream responsibility. We are not expecting upstream to provide or guarantee a stable in-kernel API (to clarify, a stable ABI is not strictly required here, since the crypto module can be rebuilt against newer kernels and still benefit from FIPS through the shorter NSRL re-validation process of roughly 3 months, compared to the full 12–18 month certification cycle). The upstream crypto subsystem should continue evolving independently.
For (2), since this feature mainly serves as infrastructure and is of interest to multiple distributions, upstreaming it could help reduce the effort each distribution would otherwise spend maintaining similar infrastructure patches independently.
We’d love to hear more thoughts on this. If the general consensus is that the downsides outweigh the benefits of merging this into mainline, we are happy to maintain it in a separate repository tracking the latest mainline and stable releases in order to keep the work publicly available.
Thanks again for taking the time to read and discuss this.
Jay
prev parent reply other threads:[~2026-02-25 17:36 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-12 2:42 [PATCH v1 00/17] crypto: Standalone crypto module (Series 1/4): Core implementation Jay Wang
2026-02-12 2:42 ` [PATCH 01/17] crypto: add Kconfig options for standalone crypto module Jay Wang
2026-02-12 2:42 ` [PATCH 02/17] crypto: add module entry for standalone crypto kernel module Jay Wang
2026-02-12 2:42 ` [PATCH 03/17] build: special compilation rule for building the standalone crypto module Jay Wang
2026-02-12 2:42 ` [PATCH 04/17] build: Add ELF marker for crypto-objs-m modules Jay Wang
2026-02-12 2:42 ` [PATCH 05/17] module: allow kernel module loading directly from memory Jay Wang
2026-02-12 2:42 ` [PATCH 06/17] crypto: add pluggable interface for builtin crypto modules Jay Wang
2026-02-12 2:42 ` [PATCH 07/17] crypto: dedicated ELF sections for collected crypto initcalls Jay Wang
2026-02-12 2:42 ` [PATCH 08/17] crypto: fips140: add crypto module loader Jay Wang
2026-02-12 2:42 ` [PATCH 09/17] build: embed the standalone crypto module into vmlinux Jay Wang
2026-02-12 2:42 ` [PATCH 10/17] build: add CONFIG_DEBUG_INFO_BTF_MODULES support for the standalone crypto kernel module Jay Wang
2026-02-12 2:42 ` [PATCH 11/17] Allow selective crypto module loading at boot based on FIPS mode Jay Wang
2026-02-12 2:42 ` [PATCH 12/17] Execute crypto initcalls during module initialization Jay Wang
2026-02-12 2:42 ` [PATCH 13/17] crypto/algapi.c: skip crypto_check_module_sig() for the standalone crypto module Jay Wang
2026-02-12 2:42 ` [PATCH 14/17] crypto: fips140: add module integrity self-check Jay Wang
2026-02-12 2:42 ` [PATCH 15/17] x86: crypto: to convert exported crypto symbols into pluggable interface for x86 cryptos Jay Wang
2026-02-12 2:42 ` [PATCH 16/17] arm64: crypto: to convert exported crypto symbols into pluggable interface for arm64 cryptos Jay Wang
2026-02-12 2:42 ` [PATCH 17/17] Add standalone crypto kernel module technical documentation Jay Wang
2026-02-25 1:55 ` Eric Biggers
2026-02-25 14:08 ` Christoph Hellwig
2026-02-25 17:35 ` Jay Wang [this message]
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=20260225173551.8704-1-wanjay@amazon.com \
--to=wanjay@amazon.com \
--cc=bp@alien8.de \
--cc=catalin.marinas@arm.com \
--cc=davem@davemloft.net \
--cc=ebiggers@kernel.org \
--cc=herbert@gondor.apana.org.au \
--cc=ilya.okomin@oracle.com \
--cc=jay.wang.upstream@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-modules@vger.kernel.org \
--cc=mcgrof@kernel.org \
--cc=mingo@redhat.com \
--cc=nathan@kernel.org \
--cc=nsc@kernel.org \
--cc=nstange@suse.de \
--cc=petr.pavlu@suse.com \
--cc=tglx@kernel.org \
--cc=vegard.nossum@oracle.com \
--cc=will@kernel.org \
--cc=x86@kernel.org \
/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