linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Eric W. Biederman" <ebiederm@xmission.com>
To: linux-kernel@vger.kernel.org
Cc: "Alexey Gladkov" <legion@kernel.org>,
	"Kees Cook" <keescook@chromium.org>,
	"Shuah Khan" <shuah@kernel.org>,
	"Christian Brauner" <brauner@kernel.org>,
	"Solar Designer" <solar@openwall.com>,
	"Ran Xiaokai" <ran.xiaokai@zte.com.cn>,
	containers@lists.linux-foundation.org,
	"Michal Koutný" <mkoutny@suse.com>,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	stable@vger.kernel.org
Subject: [PATCH 2/8] ucounts: Fix set_cred_ucounts
Date: Thu, 10 Feb 2022 20:13:18 -0600	[thread overview]
Message-ID: <20220211021324.4116773-2-ebiederm@xmission.com> (raw)
In-Reply-To: <87o83e2mbu.fsf@email.froward.int.ebiederm.org>

Michal Koutný <mkoutny@suse.com> wrote:
> Tasks are associated to multiple users at once. Historically and as per
> setrlimit(2) RLIMIT_NPROC is enforce based on real user ID.
>
> The commit 21d1c5e386bc ("Reimplement RLIMIT_NPROC on top of ucounts")
> made the accounting structure "indexed" by euid and hence potentially
> account tasks differently.
>
> The effective user ID may be different e.g. for setuid programs but
> those are exec'd into already existing task (i.e. below limit), so
> different accounting is moot.
>
> Some special setresuid(2) users may notice the difference, justifying
> this fix.

I looked at the cred->ucount is only used for rlimit operations that
were previously stored in cred->user.  Making the fact cred->ucount
can refer to a different user from cred->user a bug working will all
rlimits not just RLIMIT_NPROC.

So fix set_cred_ucounts to always use the real uid not the effective uid.

Further simplify set_cred_ucounts by noticing that set_cred_ucounts
somehow retained a draft version of the check to see if alloc_ucounts
was needed that checks the new->user and new->user_ns against the
current_real_cred(), when nothing matters for setting the ucounts
field of a struct cred except the other fields in that same struct
cred.

So delete the confusing and wrong check against the
current_real_cred(), and all of it's intermediate variables.

Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20220207121800.5079-4-mkoutny@suse.com
Reported-by: Michal Koutný <mkoutny@suse.com>
Fixes: 21d1c5e386bc ("Reimplement RLIMIT_NPROC on top of ucounts")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
---
 kernel/cred.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/kernel/cred.c b/kernel/cred.c
index 473d17c431f3..933155c96922 100644
--- a/kernel/cred.c
+++ b/kernel/cred.c
@@ -665,21 +665,16 @@ EXPORT_SYMBOL(cred_fscmp);
 
 int set_cred_ucounts(struct cred *new)
 {
-	struct task_struct *task = current;
-	const struct cred *old = task->real_cred;
 	struct ucounts *new_ucounts, *old_ucounts = new->ucounts;
 
-	if (new->user == old->user && new->user_ns == old->user_ns)
-		return 0;
-
 	/*
 	 * This optimization is needed because alloc_ucounts() uses locks
 	 * for table lookups.
 	 */
-	if (old_ucounts->ns == new->user_ns && uid_eq(old_ucounts->uid, new->euid))
+	if (old_ucounts->ns == new->user_ns && uid_eq(old_ucounts->uid, new->uid))
 		return 0;
 
-	if (!(new_ucounts = alloc_ucounts(new->user_ns, new->euid)))
+	if (!(new_ucounts = alloc_ucounts(new->user_ns, new->uid)))
 		return -EAGAIN;
 
 	new->ucounts = new_ucounts;
-- 
2.29.2


  parent reply	other threads:[~2022-02-11  2:14 UTC|newest]

Thread overview: 78+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-07 12:17 [RFC PATCH 0/6] RLIMIT_NPROC in ucounts fixups Michal Koutný
2022-02-07 12:17 ` [RFC PATCH 1/6] set_user: Perform RLIMIT_NPROC capability check against new user credentials Michal Koutný
2022-02-10  1:14   ` Solar Designer
2022-02-10  1:57     ` Eric W. Biederman
2022-02-11 20:32     ` Eric W. Biederman
2022-02-12 22:14       ` Solar Designer
2022-02-15 11:55     ` Michal Koutný
2022-02-07 12:17 ` [RFC PATCH 2/6] set*uid: Check RLIMIT_PROC against new credentials Michal Koutný
2022-02-07 12:17 ` [RFC PATCH 3/6] cred: Count tasks by their real uid into RLIMIT_NPROC Michal Koutný
2022-02-07 12:17 ` [RFC PATCH 4/6] ucounts: Allow root to override RLIMIT_NPROC Michal Koutný
2022-02-10  0:21   ` Eric W. Biederman
2022-02-07 12:17 ` [RFC PATCH 5/6] selftests: Challenge RLIMIT_NPROC in user namespaces Michal Koutný
2022-02-10  1:22   ` Shuah Khan
2022-02-15  9:45     ` Michal Koutný
2022-02-07 12:18 ` [RFC PATCH 6/6] selftests: Test RLIMIT_NPROC in clone-created " Michal Koutný
2022-02-10  1:25   ` Shuah Khan
2022-02-15  9:34     ` Michal Koutný
2022-02-08 13:54 ` [RFC PATCH 0/6] RLIMIT_NPROC in ucounts fixups Eric W. Biederman
2022-02-11  2:01 ` [PATCH 0/8] ucounts: RLIMIT_NPROC fixes Eric W. Biederman
2022-02-11  2:13   ` [PATCH 1/8] ucounts: Fix RLIMIT_NPROC regression Eric W. Biederman
2022-02-14 18:37     ` Michal Koutný
2022-02-16 15:22       ` Eric W. Biederman
2022-02-11  2:13   ` Eric W. Biederman [this message]
2022-02-15 11:10     ` [PATCH 2/8] ucounts: Fix set_cred_ucounts Michal Koutný
2022-02-11  2:13   ` [PATCH 3/8] ucounts: Fix and simplify RLIMIT_NPROC handling during setuid()+execve Eric W. Biederman
2022-02-12 23:17     ` Solar Designer
2022-02-14 15:10       ` Eric W. Biederman
2022-02-14 17:43         ` Eric W. Biederman
2022-02-15 10:25         ` Michal Koutný
2022-02-16 15:35           ` Eric W. Biederman
2022-02-11  2:13   ` [PATCH 4/8] ucounts: Only except the root user in init_user_ns from RLIMIT_NPROC Eric W. Biederman
2022-02-15 10:54     ` Michal Koutný
2022-02-16 15:41       ` Eric W. Biederman
2022-02-11  2:13   ` [PATCH 5/8] ucounts: Handle wrapping in is_ucounts_overlimit Eric W. Biederman
2022-02-12 22:36     ` Solar Designer
2022-02-14 15:23       ` Eric W. Biederman
2022-02-14 15:23         ` Eric W. Biederman
2022-02-15 11:25         ` Michal Koutný
2022-02-14 17:16       ` David Laight
2022-02-11  2:13   ` [PATCH 6/8] ucounts: Handle inc_rlimit_ucounts wrapping in fork Eric W. Biederman
2022-02-11 11:34     ` Alexey Gladkov
2022-02-11 17:50       ` Eric W. Biederman
2022-02-11 18:32         ` Shuah Khan
2022-02-11 18:40         ` Alexey Gladkov
2022-02-11 19:56           ` Eric W. Biederman
2022-02-11  2:13   ` [PATCH 7/8] rlimit: For RLIMIT_NPROC test the child not the parent for capabilites Eric W. Biederman
2022-02-11  2:13   ` [PATCH 8/8] ucounts: Use the same code to enforce RLIMIT_NPROC in fork and exec Eric W. Biederman
2022-02-11 18:22   ` [PATCH 0/8] ucounts: RLIMIT_NPROC fixes Shuah Khan
2022-02-11 19:23     ` Eric W. Biederman
2022-02-15 11:37     ` Michal Koutný
2022-02-16 15:56   ` [PATCH v2 0/5] " Eric W. Biederman
2022-02-16 15:58     ` [PATCH v2 1/5] rlimit: Fix RLIMIT_NPROC enforcement failure caused by capability calls in set_user Eric W. Biederman
2022-02-16 17:42       ` Solar Designer
2022-02-16 15:58     ` [PATCH v2 2/5] ucounts: Enforce RLIMIT_NPROC not RLIMIT_NPROC+1 Eric W. Biederman
2022-02-16 15:58     ` [PATCH v2 3/5] ucounts: Base set_cred_ucounts changes on the real user Eric W. Biederman
2022-02-16 15:58     ` [PATCH v2 4/5] ucounts: Move RLIMIT_NPROC handling after set_user Eric W. Biederman
2022-02-16 15:58     ` [PATCH v2 5/5] ucounts: Handle wrapping in is_ucounts_overlimit Eric W. Biederman
2022-02-16 17:28       ` Shuah Khan
2022-02-18 15:34     ` [GIT PULL] ucounts: RLIMIT_NPROC fixes for v5.17 Eric W. Biederman
2022-02-20 19:05       ` pr-tracker-bot
2022-03-03  0:12       ` [GIT PULL] ucounts: Regression fix " Eric W. Biederman
2022-03-03  0:30         ` pr-tracker-bot
2022-02-12 15:32 ` [RFC PATCH 0/6] RLIMIT_NPROC in ucounts fixups Etienne Dechamps
2022-02-15 10:11   ` Michal Koutný
2022-02-23  0:57     ` Eric W. Biederman
2022-02-23 18:00       ` How should rlimits, suid exec, and capabilities interact? Eric W. Biederman
2022-02-23 19:44         ` Andy Lutomirski
2022-02-23 21:28           ` Willy Tarreau
2022-02-23 19:50         ` Linus Torvalds
2022-02-24  1:24           ` Eric W. Biederman
2022-02-24  1:41             ` Linus Torvalds
2022-02-24  2:12               ` Eric W. Biederman
2022-02-24 15:41                 ` [PATCH] ucounts: Fix systemd LimigtNPROC with private users regression Eric W. Biederman
2022-02-24 16:28                   ` Kees Cook
2022-02-24 18:53                     ` Michal Koutný
2022-02-25  0:29                     ` Eric W. Biederman
2022-02-24  3:00               ` How should rlimits, suid exec, and capabilities interact? David Laight
2022-02-24  1:32           ` Eric W. Biederman

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=20220211021324.4116773-2-ebiederm@xmission.com \
    --to=ebiederm@xmission.com \
    --cc=brauner@kernel.org \
    --cc=containers@lists.linux-foundation.org \
    --cc=keescook@chromium.org \
    --cc=legion@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mkoutny@suse.com \
    --cc=ran.xiaokai@zte.com.cn \
    --cc=shuah@kernel.org \
    --cc=solar@openwall.com \
    --cc=stable@vger.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;
as well as URLs for NNTP newsgroup(s).