linux-kernel - Re: [PATCH 07/26] firmware: avoid pointless reference count bump

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] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAOQ4uxhUX9v4RvfXJUNODOpua+G97eDYwera664n0KRsexOmWw@mail.gmail.com>
Date: Sun, 24 Nov 2024 15:56:16 +0100
From: Amir Goldstein <amir73il@...il.com>
To: Christian Brauner <brauner@...nel.org>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>, Miklos Szeredi <miklos@...redi.hu>, 
	linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org
Subject: Re: [PATCH 07/26] firmware: avoid pointless reference count bump

On Sun, Nov 24, 2024 at 2:44 PM Christian Brauner <brauner@...nel.org> wrote:
>
> The creds are allocated via prepare_kernel_cred() which has already
> taken a reference.
>
> Signed-off-by: Christian Brauner <brauner@...nel.org>
> ---
>  drivers/base/firmware_loader/main.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/base/firmware_loader/main.c b/drivers/base/firmware_loader/main.c
> index 96a2c3011ca82148b4ba547764a1f92e252dbf5f..740ef6223a62ca37e776d1558f840f09c7c46c95 100644
> --- a/drivers/base/firmware_loader/main.c
> +++ b/drivers/base/firmware_loader/main.c
> @@ -912,7 +912,7 @@ _request_firmware(const struct firmware **firmware_p, const char *name,
>                 ret = -ENOMEM;
>                 goto out;
>         }
> -       old_cred = override_creds(get_new_cred(kern_cred));
> +       old_cred = override_creds(kern_cred);
>
>         ret = fw_get_filesystem_firmware(device, fw->priv, "", NULL);
>
> @@ -945,7 +945,6 @@ _request_firmware(const struct firmware **firmware_p, const char *name,
>                 ret = assign_fw(fw, device);
>
>         put_cred(revert_creds(old_cred));
> -       put_cred(kern_cred);

This may seem like nit picking, but I think that:
        revert_creds(old_cred));
        put_cred(kern_cred);

Is nicer. It is more balanced and it is more consistent with the majority of
patches in this series which in a balanced manner remove both the
get_new_cred from the override_creds line and the put_cred from the
revert_creds line.

If someone wanted to, both old_cred and kern_cred could be converted
to use scoped cleanup handlers. This is more apparent when the cleanup
is explicit on the local kern_cred var.

The same comment applies to a few other patches in this series.

Thanks,
Amir.