linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Vivek Goyal <vgoyal@redhat.com>
To: linux-kernel@vger.kernel.org,
	linux-security-module@vger.kernel.org, kexec@lists.infradead.org
Cc: akpm@linux-foundation.org, zohar@linux.vnet.ibm.com,
	d.kasatkin@samsung.com, ebiederm@xmission.com, hpa@zytor.com,
	matthew.garrett@nebula.com, vgoyal@redhat.com
Subject: [PATCH 11/16] ptrace: Do not allow ptrace() from unsigned process to signed one
Date: Tue, 10 Sep 2013 17:44:26 -0400	[thread overview]
Message-ID: <1378849471-10521-12-git-send-email-vgoyal@redhat.com> (raw)
In-Reply-To: <1378849471-10521-1-git-send-email-vgoyal@redhat.com>

Do not allow unsigned processes to ptrace() signed ones otherwise they can
modify the address space of signed processes and whole purpose of signature
verification is defeated.

Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
---
 fs/binfmt_elf.c      | 49 +++++++++++++++++++++++++++++++++++++++++++++++--
 security/commoncap.c | 11 +++++++++++
 2 files changed, 58 insertions(+), 2 deletions(-)

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 22a8272..8f2286e 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -568,6 +568,43 @@ static unsigned long randomize_stack_top(unsigned long stack_top)
 #endif
 }
 
+#ifdef CONFIG_BINFMT_ELF_SIG
+/* check if current is being ptraced by tracer which is unsigned */
+static bool ptraced_by_unsafe_tracer(void)
+{
+	struct task_struct *child = current, *parent;
+	bool ret = false;
+	const struct cred *tcred;
+
+	/* Make sure parent does not change due to tracer ptrace detach */
+	read_lock(&tasklist_lock);
+
+	if (!child->ptrace) {
+		ret = false;
+		goto out;
+	}
+
+	parent = child->parent;
+	rcu_read_lock();
+	tcred = __task_cred(parent);
+	if (!tcred->proc_signed)
+		ret = true;
+	rcu_read_unlock();
+
+	/*
+	 * Make sure parent is memlocked too otherwise it might be signed
+	 * but still being swapped out and is open to address space
+	 * modifications.
+	 */
+	if (!test_bit(MMF_VM_LOCKED, &parent->mm->flags))
+		ret = true;
+
+out:
+	read_unlock(&tasklist_lock);
+	return ret;
+}
+#endif
+
 static int load_elf_binary(struct linux_binprm *bprm)
 {
 	struct file *interpreter = NULL; /* to shut gcc up */
@@ -951,8 +988,16 @@ static int load_elf_binary(struct linux_binprm *bprm)
 			send_sig(SIGKILL, current, 0);
 			goto out_free_dentry;
 		}
-		/* Signature verification successful */
-		bprm->cred->proc_signed = true;
+
+		/*
+		 * Signature verification successful. If this process is
+		 * is being ptraced at the time of exec() and tracer is
+		 * not signed, do not set proc_signed, otherwise unsigned
+		 * tracer could change signed tracee's address space,
+		 * effectively nullifying singature checking.
+		 */
+		if (!ptraced_by_unsafe_tracer())
+			bprm->cred->proc_signed = true;
 	}
 #endif
 
diff --git a/security/commoncap.c b/security/commoncap.c
index c44b6fe..4d7f90e 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -146,6 +146,12 @@ int cap_ptrace_access_check(struct task_struct *child, unsigned int mode)
 	rcu_read_lock();
 	cred = current_cred();
 	child_cred = __task_cred(child);
+
+	if (mode != PTRACE_MODE_READ && child_cred->proc_signed &&
+	    !cred->proc_signed) {
+		ret = -EPERM;
+		goto out;
+	}
 	if (cred->user_ns == child_cred->user_ns &&
 	    cap_issubset(child_cred->cap_permitted, cred->cap_permitted))
 		goto out;
@@ -178,6 +184,11 @@ int cap_ptrace_traceme(struct task_struct *parent)
 	rcu_read_lock();
 	cred = __task_cred(parent);
 	child_cred = current_cred();
+
+	if (child_cred->proc_signed && !cred->proc_signed) {
+		ret = -EPERM;
+		goto out;
+	}
 	if (cred->user_ns == child_cred->user_ns &&
 	    cap_issubset(child_cred->cap_permitted, cred->cap_permitted))
 		goto out;
-- 
1.8.3.1


  parent reply	other threads:[~2013-09-10 21:46 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-10 21:44 [PATCH 00/16] [RFC PATCH] Signed kexec support Vivek Goyal
2013-09-10 21:44 ` [PATCH 01/16] mm: vm_brk(), align the length to page boundary Vivek Goyal
2013-09-10 21:44 ` [PATCH 02/16] integrity: Add a function to determine digital signature length Vivek Goyal
2013-09-10 21:44 ` [PATCH 03/16] ima: Allow adding more memory locking metadata after digital signature v2 Vivek Goyal
2013-09-10 21:44 ` [PATCH 04/16] integrity: Allow digital signature verification with a given keyring ptr Vivek Goyal
2013-09-11 17:34   ` Mimi Zohar
2013-09-10 21:44 ` [PATCH 05/16] integrity: Export a function to retrieve hash algo used in digital signature Vivek Goyal
2013-09-10 21:44 ` [PATCH 06/16] ima: export new IMA functions for signature verification Vivek Goyal
2013-09-10 21:44 ` [PATCH 07/16] mm: Define a task flag MMF_VM_LOCKED for memlocked tasks and don't allow munlock Vivek Goyal
2013-09-10 21:44 ` [PATCH 08/16] binfmt_elf: Elf executable signature verification Vivek Goyal
2013-09-10 21:44 ` [PATCH 09/16] ima: define functions to appraise memory buffer contents Vivek Goyal
2013-09-10 21:44 ` [PATCH 10/16] keyctl: Introduce a new operation KEYCTL_VERIFY_SIGNATURE Vivek Goyal
2013-09-10 21:44 ` Vivek Goyal [this message]
2013-09-10 21:44 ` [PATCH 12/16] binfmt_elf: Do not mark process signed if binary has elf interpreter Vivek Goyal
2013-09-10 21:44 ` [PATCH 13/16] kexec: Allow only signed processes to call sys_kexec() in secureboot mode Vivek Goyal
2013-09-10 21:44 ` [PATCH 14/16] kexec: Export sysfs attributes for secureboot and secure modules to user space Vivek Goyal
2013-09-10 22:40   ` Greg KH
2013-09-11 13:44     ` Vivek Goyal
2013-09-10 22:57   ` Josh Boyer
2013-09-11 13:51     ` Vivek Goyal
2013-09-10 21:44 ` [PATCH 15/16] bootparam: Pass acpi_rsdp pointer in bootparam Vivek Goyal
2013-09-10 22:52   ` H. Peter Anvin
2013-09-11 11:44     ` Borislav Petkov
2013-09-11 13:45       ` Vivek Goyal
2013-09-11 14:32         ` Borislav Petkov
2013-09-12  7:34           ` Dave Young
2013-09-12 12:53             ` Borislav Petkov
     [not found]               ` <20130912131930.GC28500@redhat.com>
2013-09-12 14:25                 ` Borislav Petkov
2013-09-12 14:34               ` Matthew Garrett
2013-09-12 14:42                 ` Borislav Petkov
2013-09-13  7:12               ` Dave Young
2013-09-13 11:26                 ` Borislav Petkov
2013-09-10 21:44 ` [PATCH 16/16] mount: Add a flag to not follow symlink at the end of mount point Vivek Goyal
2013-09-12  3:40 ` [PATCH 00/16] [RFC PATCH] Signed kexec support Greg KH
2013-09-12 11:43   ` Vivek Goyal
2013-09-12 16:17     ` Greg KH
2013-09-12 18:24       ` Mimi Zohar
     [not found]         ` <20130916142852.GB20753@redhat.com>
2013-09-18 14:51           ` Andrea Adami
2013-09-23 17:15             ` Vivek Goyal
2013-09-16 14:24       ` Vivek Goyal

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=1378849471-10521-12-git-send-email-vgoyal@redhat.com \
    --to=vgoyal@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=d.kasatkin@samsung.com \
    --cc=ebiederm@xmission.com \
    --cc=hpa@zytor.com \
    --cc=kexec@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=matthew.garrett@nebula.com \
    --cc=zohar@linux.vnet.ibm.com \
    /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).