diff options
| author | Ali Ahmet Memis <ali@iusegentoo.com> | 2026-08-21 01:45:27 +0000 |
|---|---|---|
| committer | Stafford Horne <shorne@gmail.com> | 2026-08-29 07:32:26 +0100 |
| commit | 78004e9a87f240df03e2f73120d291763c32e0a7 (patch) | |
| tree | 137c22dd795222931b9d2bf2501d37cd766741aa /security/selinux/ss/context.c | |
| download | linux-stable-78004e9a87f240df03e2f73120d291763c32e0a7.tar.gz linux-stable-78004e9a87f240df03e2f73120d291763c32e0a7.zip | |
openrisc: fix arbitrary kernel memory access via or1k_atomic syscallgrafted
sys_or1k_atomic() (syscall 244 in the "or1k" ABI) takes two user
pointers, v1 and v2, and swaps the words they point to in hand-written
assembly.
l.lwz r29,0(r4)
l.lwz r27,0(r5)
l.sw 0(r4),r27
l.sw 0(r5),r29
The pointers are not checked with access_ok(). The four memory
accesses also have no exception table entries.
A caller passes a kernel address as either pointer, and the syscall
reads from and writes to it directly.
This gives an unprivileged process a kernel read/write primitive. It
overwrites kernel data such as the sys_call_table, gaining code
execution in kernel context.
Check both pointers before entering the critical section. Add fixups
for the four memory accesses so faults on valid but unmapped user
addresses return -EFAULT.
[shorne@gmail.com: fix comment style]
Fixes: 9d02a4283e9c ("OpenRISC: Boot code")
Cc: stable@vger.kernel.org
Signed-off-by: Ali Ahmet Memis <ali@iusegentoo.com>
Signed-off-by: Stafford Horne <shorne@gmail.com>
Diffstat (limited to 'security/selinux/ss/context.c')
| -rw-r--r-- | security/selinux/ss/context.c | 32 |
1 files changed, 32 insertions, 0 deletions
diff --git a/security/selinux/ss/context.c b/security/selinux/ss/context.c new file mode 100644 index 000000000..a528b7f76 --- /dev/null +++ b/security/selinux/ss/context.c @@ -0,0 +1,32 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Implementations of the security context functions. + * + * Author: Ondrej Mosnacek <omosnacek@gmail.com> + * Copyright (C) 2020 Red Hat, Inc. + */ + +#include <linux/jhash.h> + +#include "context.h" +#include "mls.h" + +u32 context_compute_hash(const struct context *c) +{ + u32 hash = 0; + + /* + * If a context is invalid, it will always be represented by a + * context struct with only the len & str set (and vice versa) + * under a given policy. Since context structs from different + * policies should never meet, it is safe to hash valid and + * invalid contexts differently. The context_equal() function + * already operates under the same assumption. + */ + if (c->len) + return full_name_hash(NULL, c->str, c->len); + + hash = jhash_3words(c->user, c->role, c->type, hash); + hash = mls_range_hash(&c->range, hash); + return hash; +} |
