mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Ahmad Fatoum <a.fatoum@pengutronix.de>
To: barebox@lists.infradead.org
Cc: Ahmad Fatoum <a.fatoum@pengutronix.de>
Subject: [PATCH] fs: efivarfs: allow creating variables under any GUID in loader mode
Date: Wed, 26 Aug 2026 11:37:13 +0200	[thread overview]
Message-ID: <20260826093720.2496298-1-a.fatoum@pengutronix.de> (raw)

efivars_create() refuses to create variables outside the barebox vendor
GUID. That dates from when barebox was only an EFI payload: on somebody
else's firmware, whose NVRAM holds the platform's own state, a shell that
can create Boot#### or BootOrder under the global GUID is a liability,
and the payload only ever needs barebox-env-<guid> anyway.

The EFI loader took the same efivarfs over unchanged, where the argument
does not hold. The store is barebox' own in-memory buffer backed by a
file it writes itself, and everything the loader starts - the UEFI
Shell, GRUB, the OS - can already create variables under any GUID. Only
the barebox shell was held back:

  barebox:/ echo -o /efivarfs/Timeout-8be4df61-93ca-11d2-aa0d-00e098032b8c 5
  open: Operation not permitted

Lift the restriction when running as the loader, so board scripts can
set up global variables without going through /env/data/init.efivars.
The payload keeps confining itself to its own GUID.

Assisted-by: Claude:opus-5
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
---
 fs/efivarfs.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/fs/efivarfs.c b/fs/efivarfs.c
index a4d8cc8d0268..1f53c06d66d3 100644
--- a/fs/efivarfs.c
+++ b/fs/efivarfs.c
@@ -66,12 +66,17 @@ static int efivars_create(struct device *dev, const char *pathname,
 	if (pathname[0] == '/')
 		pathname++;
 
-	/* deny creating files with other vendor GUID than our own */
 	ret = efivarfs_parse_filename(pathname, &vendor, &name);
 	if (ret)
 		return -ENOENT;
 
-	if (efi_guidcmp(vendor, EFI_BAREBOX_VENDOR_GUID))
+	/*
+	 * As a payload, barebox is a guest on somebody else's firmware and
+	 * confines itself to its own vendor GUID when creating variables.
+	 * As the loader, the variable store is barebox' own and anything it
+	 * boots may already create variables under any GUID.
+	 */
+	if (!efi_is_loader() && efi_guidcmp(vendor, EFI_BAREBOX_VENDOR_GUID))
 		return -EPERM;
 
 	inode = xzalloc(sizeof(*inode));
-- 
2.47.3




             reply	other threads:[~2026-08-26 10:23 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-26  9:37 Ahmad Fatoum [this message]
2026-08-28 12:52 ` Sascha Hauer

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=20260826093720.2496298-1-a.fatoum@pengutronix.de \
    --to=a.fatoum@pengutronix.de \
    --cc=barebox@lists.infradead.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