Kernel KVM virtualization development
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Tarun Sahu" <tarunsahu@google.com>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	joro@8bytes.org, "Mike Rapoport" <rppt@kernel.org>,
	"Pratyush Yadav" <pratyush@kernel.org>,
	maz@kernel.org, arnd@arndb.de,
	"David Matlack" <dmatlack@google.com>,
	jgg@nvidia.com, robin.murphy@arm.com, loganodell@google.com,
	oupton@kernel.org, kevin.tian@intel.com, alex@shazbot.org,
	dwmw2@infradead.org, pbonzini@redhat.com, seanjc@google.com,
	will@kernel.org, "Pasha Tatashin" <pasha.tatashin@soleen.com>,
	"Alexander Graf" <graf@amazon.com>,
	"Bjorn Helgaas" <bhelgaas@google.com>,
	baolu.lu@linux.intel.com
Cc: sagis@google.com, linux-mm@kvack.org, kvmarm@lists.linux.dev,
	linux-pci@vger.kernel.org, kexec@lists.infradead.org,
	skhawaja@google.com, kvm@vger.kernel.org
Subject: Re: [RFC PATCH v2 4/4] mm: memfd_luo: Add the 2-bit pair feature header
Date: Thu, 22 Oct 2026 15:26:21 +0000	[thread overview]
Message-ID: <sashiko-outbox-181698@kernel.org> (raw)
In-Reply-To: <20261021152527.375070-5-tarunsahu@google.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] mm: memfd_luo: missing feature header validation in memfd_luo_finish()
--

commit b960724786ac11078fdfa9ded6bcd2355170cf0b
Author: Tarun Sahu <tarunsahu@google.com>

mm: memfd_luo: Add the 2-bit pair feature header

This commit embeds a 128-byte struct luo_feature_hdr at the start of struct
memfd_luo_ser and defines enum memfd_luo_feature with MEMFD_LUO_FEAT_CORE as
a required feature. It initializes the feature header during
memfd_luo_preserve() and validates it during memfd_luo_retrieve().

> diff --git a/mm/memfd_luo.c b/mm/memfd_luo.c
> index 64f8b8cb75bdb..ea1f4c8cd2591 100644
> --- a/mm/memfd_luo.c
> +++ b/mm/memfd_luo.c

[ ... ]

> @@ -525,6 +531,14 @@ static int memfd_luo_retrieve(struct liveupdate_file_op_args *args)
>  	if (!ser)
>  		return -EINVAL;
>  
> +	err = luo_feature_hdr_validate(&ser->features, memfd_luo_features);
> +	if (err) {
> +		pr_err("memfd_luo_ser feature header validation failed (size=%llu, flags[0]=%#llx): %pe\n",
> +		       ser->features.size, ser->features.flags[0],
> +		       ERR_PTR(err));
> +		goto free_ser;
> +	}
> +
>  	/* Make sure the file only has seals supported by this version. */
>  	if (ser->seals & ~MEMFD_LUO_ALL_SEALS) {
>  		err = -EOPNOTSUPP;

[Severity: Medium]
Here in memfd_luo_retrieve(), ser->features is validated via
luo_feature_hdr_validate() before accessing any MEMFD_LUO_FEAT_CORE fields,
jumping directly to kho_restore_free(ser) on failure.

However, when a preserved memfd is not retrieved in the target kernel before
the session is finished or closed (for example via luo_session_release() ->
luo_session_finish_one() -> luo_file_finish() -> luo_file_finish_one() ->
memfd_luo_finish() with args->retrieve_status == 0), memfd_luo_finish()
reads ser->nr_folios and ser->folios without validating ser->features first:

mm/memfd_luo.c:memfd_luo_finish() {
	if (args->retrieve_status)
		return;

	ser = phys_to_virt(args->serialized_data);
	if (!ser)
		return;

	if (ser->nr_folios) {
		folios_ser = kho_restore_vmalloc(&ser->folios);
		if (!folios_ser)
			goto out;

		memfd_luo_discard_folios(folios_ser, ser->nr_folios);
		vfree(folios_ser);
	}

out:
	kho_restore_free(ser);
}

According to the ABI contract in include/linux/kho/abi/memfd.h, fields of a
LUO_FEAT_REQUIRED feature such as nr_folios and folios are only guaranteed
present once luo_feature_hdr_validate() succeeds.

Should memfd_luo_finish() also call luo_feature_hdr_validate(&ser->features,
memfd_luo_features) and skip to kho_restore_free(ser) if validation fails,
so it does not pass unvalidated MEMFD_LUO_FEAT_CORE fields to
kho_restore_vmalloc() and memfd_luo_discard_folios()?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261021152527.375070-1-tarunsahu@google.com?part=4

      parent reply	other threads:[~2026-10-22 15:26 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-21 15:25 [RFC PATCH v2 0/4] liveupdate: Feature header ABI and named handler Tarun Sahu
2026-10-21 15:25 ` [RFC PATCH v2 1/4] liveupdate: Rename handler compatible to name Tarun Sahu
2026-10-21 17:11   ` Pasha Tatashin
2026-10-21 18:16     ` tarunsahu
2026-10-21 19:11       ` Pasha Tatashin
2026-10-21 19:35         ` tarunsahu
2026-10-23  7:44           ` Mike Rapoport
2026-10-21 15:25 ` [RFC PATCH v2 2/4] liveupdate: Select file handler by name in the preserve ioctl Tarun Sahu
2026-10-21 17:17   ` Pasha Tatashin
2026-10-21 23:02     ` David Matlack
2026-10-22  1:14       ` Pasha Tatashin
2026-10-22 17:57         ` David Matlack
2026-10-22 12:50     ` tarunsahu
2026-10-22 13:42       ` Pasha Tatashin
2026-10-21 15:25 ` [RFC PATCH v2 3/4] liveupdate: Replace compatibility string with a 2-bit pair feature header Tarun Sahu
2026-10-21 17:27   ` Pasha Tatashin
2026-10-21 17:40     ` Jason Gunthorpe
2026-10-21 17:55       ` Pasha Tatashin
2026-10-21 18:40     ` tarunsahu
2026-10-21 18:46       ` David Matlack
2026-10-21 18:50         ` Pasha Tatashin
2026-10-21 23:28           ` David Matlack
2026-10-22  1:26             ` Pasha Tatashin
2026-10-22  2:57               ` David Matlack
2026-10-22  3:32                 ` Pasha Tatashin
2026-10-22  3:52                   ` David Matlack
2026-10-22  4:11                     ` Pasha Tatashin
2026-10-22 17:51                       ` David Matlack
2026-10-22 20:53                         ` Pasha Tatashin
2026-10-21 23:55   ` David Matlack
2026-10-22  0:04     ` David Matlack
2026-10-21 15:25 ` [RFC PATCH v2 4/4] mm: memfd_luo: Add the " Tarun Sahu
2026-10-21 17:38   ` Pasha Tatashin
2026-10-21 23:18     ` David Matlack
2026-10-22  0:41       ` Pasha Tatashin
2026-10-22  2:51         ` David Matlack
2026-10-22  3:50           ` Pasha Tatashin
2026-10-22 15:26   ` sashiko-bot [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=sashiko-outbox-181698@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=alex@shazbot.org \
    --cc=arnd@arndb.de \
    --cc=baolu.lu@linux.intel.com \
    --cc=bhelgaas@google.com \
    --cc=dmatlack@google.com \
    --cc=dwmw2@infradead.org \
    --cc=graf@amazon.com \
    --cc=jgg@nvidia.com \
    --cc=joro@8bytes.org \
    --cc=kevin.tian@intel.com \
    --cc=kexec@lists.infradead.org \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=linux-mm@kvack.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=loganodell@google.com \
    --cc=maz@kernel.org \
    --cc=oupton@kernel.org \
    --cc=pasha.tatashin@soleen.com \
    --cc=pbonzini@redhat.com \
    --cc=pratyush@kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=rppt@kernel.org \
    --cc=sagis@google.com \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=seanjc@google.com \
    --cc=skhawaja@google.com \
    --cc=tarunsahu@google.com \
    --cc=will@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