linux-kernel - [PATCH v2] staging: android: ashmem: Disallow ashmem memory from being remapped

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

Date:   Mon, 27 Jan 2020 15:56:16 -0800
From:   Todd Kjos <tkjos@...gle.com>
To:     tkjos@...gle.com, surenb@...gle.com, gregkh@...uxfoundation.org,
        arve@...roid.com, devel@...verdev.osuosl.org,
        linux-kernel@...r.kernel.org, maco@...gle.com
Cc:     joel@...lfernandes.org, kernel-team@...roid.com,
        Jann Horn <jannh@...gle.com>, stable <stable@...r.kernel.org>
Subject: [PATCH v2] staging: android: ashmem: Disallow ashmem memory from
 being remapped

From: Suren Baghdasaryan <surenb@...gle.com>

When ashmem file is mmapped, the resulting vma->vm_file points to the
backing shmem file with the generic fops that do not check ashmem
permissions like fops of ashmem do. If an mremap is done on the ashmem
region, then the permission checks will be skipped. Fix that by disallowing
mapping operation on the backing shmem file.

Reported-by: Jann Horn <jannh@...gle.com>
Signed-off-by: Suren Baghdasaryan <surenb@...gle.com>
Cc: stable <stable@...r.kernel.org> # 4.4,4.9,4.14,4.18,5.4
Signed-off-by: Todd Kjos <tkjos@...gle.com>
---
 drivers/staging/android/ashmem.c | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

v2: update commit message as suggested by joelaf@...gle.com.

diff --git a/drivers/staging/android/ashmem.c b/drivers/staging/android/ashmem.c
index 74d497d39c5a..c6695354b123 100644
--- a/drivers/staging/android/ashmem.c
+++ b/drivers/staging/android/ashmem.c
@@ -351,8 +351,23 @@ static inline vm_flags_t calc_vm_may_flags(unsigned long prot)
 	       _calc_vm_trans(prot, PROT_EXEC,  VM_MAYEXEC);
 }
 
+static int ashmem_vmfile_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	/* do not allow to mmap ashmem backing shmem file directly */
+	return -EPERM;
+}
+
+static unsigned long
+ashmem_vmfile_get_unmapped_area(struct file *file, unsigned long addr,
+				unsigned long len, unsigned long pgoff,
+				unsigned long flags)
+{
+	return current->mm->get_unmapped_area(file, addr, len, pgoff, flags);
+}
+
 static int ashmem_mmap(struct file *file, struct vm_area_struct *vma)
 {
+	static struct file_operations vmfile_fops;
 	struct ashmem_area *asma = file->private_data;
 	int ret = 0;
 
@@ -393,6 +408,19 @@ static int ashmem_mmap(struct file *file, struct vm_area_struct *vma)
 		}
 		vmfile->f_mode |= FMODE_LSEEK;
 		asma->file = vmfile;
+		/*
+		 * override mmap operation of the vmfile so that it can't be
+		 * remapped which would lead to creation of a new vma with no
+		 * asma permission checks. Have to override get_unmapped_area
+		 * as well to prevent VM_BUG_ON check for f_ops modification.
+		 */
+		if (!vmfile_fops.mmap) {
+			vmfile_fops = *vmfile->f_op;
+			vmfile_fops.mmap = ashmem_vmfile_mmap;
+			vmfile_fops.get_unmapped_area =
+					ashmem_vmfile_get_unmapped_area;
+		}
+		vmfile->f_op = &vmfile_fops;
 	}
 	get_file(asma->file);
 
-- 
2.25.0.341.g760bfbb309-goog