linux-kernel - Re: [PATCH] nvme-tcp: fix selinux denied when calling sock

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>] [day] [month] [year] [list]

Message-ID: <20250321050538.GA2038@lst.de>
Date: Fri, 21 Mar 2025 06:05:38 +0100
From: Christoph Hellwig <hch@....de>
To: 邵培杰 <shaopeijie@...tc.cn>
Cc: Christoph Hellwig <hch@....de>, kbusch <kbusch@...nel.org>,
	axboe <axboe@...nel.dk>, sagi <sagi@...mberg.me>,
	linux-nvme <linux-nvme@...ts.infradead.org>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	"zhang.guanghui " <zhang.guanghui@...tc.cn>,
	gechangzhong <gechangzhong@...tc.cn>
Subject: Re: [PATCH] nvme-tcp: fix selinux denied when calling sock_sendmsg

[note: your mail was mangled html, I'm trying to extract the answer]

On Fri, Mar 21, 2025 at 10:49:34AM +0800, 邵培杰 wrote:
> Sure. But on the target side, it's a bit different.
> All socket operations are done in the workqueue with domain kernel_t,
> the kernel_t can access other labels, so I think there's no issue.

Indeed.

> By
> the way, switching to sock_create_kern() could skip selinux avc
> permission check and might give a very small performance boost.

It also helps to avoid copy and paste bugs..

> I'll send an optimization patch later.

Thanks!