On Thu, Jun 13, 2019 at 12:13:33PM +0100, Cristian Marussi wrote:
> Added a simple fake_sigreturn testcase which builds a ucontext_t
> with a badly sized header and place it onto the stack.
> Expects a SIGSEGV on test PASS.
>
> Signed-off-by: Cristian Marussi <cristian.marussi@xxxxxxx>
> ---
> .../arm64/signal/testcases/.gitignore | 1 +
> .../testcases/fake_sigreturn_bad_size.c | 40 +++++++++++++++++++
> 2 files changed, 41 insertions(+)
> create mode 100644 tools/testing/selftests/arm64/signal/testcases/fake_sigreturn_bad_size.c
>
> diff --git a/tools/testing/selftests/arm64/signal/testcases/.gitignore b/tools/testing/selftests/arm64/signal/testcases/.gitignore
> index c353b7bd899d..9ad1735e0018 100644
> --- a/tools/testing/selftests/arm64/signal/testcases/.gitignore
> +++ b/tools/testing/selftests/arm64/signal/testcases/.gitignore
> @@ -8,3 +8,4 @@ mangle_pstate_invalid_mode_el3
> mangle_pstate_ssbs_regs
> fake_sigreturn_misaligned
> fake_sigreturn_bad_magic
> +fake_sigreturn_bad_size
> diff --git a/tools/testing/selftests/arm64/signal/testcases/fake_sigreturn_bad_size.c b/tools/testing/selftests/arm64/signal/testcases/fake_sigreturn_bad_size.c
> new file mode 100644
> index 000000000000..d54f79d63bf6
> --- /dev/null
> +++ b/tools/testing/selftests/arm64/signal/testcases/fake_sigreturn_bad_size.c
> @@ -0,0 +1,40 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/* Copyright (C) 2019 ARM Limited */
> +
> +#include <asm/sigcontext.h>
> +#include <ucontext.h>
> +
> +#include "test_signals_utils.h"
> +#include "testcases.h"
> +
> +struct a_sigframe sf;
> +
> +static int fake_sigreturn_bad_size_run(struct tdescr *td,
> + siginfo_t *si, ucontext_t *uc)
> +{
> + struct _aarch64_ctx *head = GET_SF_RESV_HEAD(sf);
> +
> + /* just to fill the ucontext_t with something real */
> + if (!get_current_context(td, &sf.uc))
> + return 1;
> +
> + /* Tampering size to an unplausible number should cause a SEGV */
> + head = get_terminator(head, GET_SF_RESV_SIZE(sf), NULL);
> + if (head) {
> + head->magic = ESR_MAGIC;
Add a comment to say why we're using this magic?
> + head->size = GET_SF_RESV_SIZE(sf) * 2;
If we have extra_context, this might not be an invalid size, because
extra_context can be much larger than sizeof(__reserved) ... but also
we will overrun anyway, because extra_context should already be only
just large enough to accommodate the original terminator.
As with the previous test, we should probably throw away the
extra_context record (if any) and replace it with this new record.
We should check the actual boundary condition here, rather than just
testing with a giant size: we want to make the kernel detects any
overrun, however small.
We should also write a valid terminator after the new record, so
that we know the kernel is detecting the size overrun instead of
rejecting the next record because it is garbage.
> +
> + ASSERT_BAD_CONTEXT(&sf.uc);
> + fake_sigreturn(&sf, sizeof(sf), 16);
> + }
> +
> + return 1;
> +}
[...]
Cheers
---Dave
