On Tue, 2022-08-09 at 13:16 +0530, Immad Mir wrote:
> This patch fixes the ICE caused by valid_to_unchecked_state,
> at analyzer/sm-fd.cc by handling the m_start state in
> check_for_dup.
>
> Tested lightly on x86_64.
>
> gcc/analyzer/ChangeLog:
> PR analyzer/106551
> * sm-fd.cc (check_for_dup): handle the m_start
> state when transitioning the state of LHS
> of dup, dup2 and dup3 call.
>
> Signed-off-by: Immad Mir <mirim...@outlook.com>
> ---
> gcc/analyzer/sm-fd.cc | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/gcc/analyzer/sm-fd.cc b/gcc/analyzer/sm-fd.cc
> index 8bb76d72b05..c8b9930a7b6 100644
> --- a/gcc/analyzer/sm-fd.cc
> +++ b/gcc/analyzer/sm-fd.cc
> @@ -983,7 +983,7 @@ fd_state_machine::check_for_dup (sm_context
> *sm_ctxt, const supernode *node,
> case DUP_1:
> if (lhs)
> {
> - if (is_constant_fd_p (state_arg_1))
> + if (is_constant_fd_p (state_arg_1) || state_arg_1 ==
> m_start)
> sm_ctxt->set_next_state (stmt, lhs,
> m_unchecked_read_write);
> else
> sm_ctxt->set_next_state (stmt, lhs,
> @@ -1011,7 +1011,7 @@ fd_state_machine::check_for_dup (sm_context
> *sm_ctxt, const supernode *node,
> file descriptor i.e the first argument. */
> if (lhs)
> {
> - if (is_constant_fd_p (state_arg_1))
> + if (is_constant_fd_p (state_arg_1) || state_arg_1 ==
> m_start)
> sm_ctxt->set_next_state (stmt, lhs,
> m_unchecked_read_write);
> else
> sm_ctxt->set_next_state (stmt, lhs,
Thanks. The fix looks reasonable, but please can the patch also add a
reproducer to the test suite, covering each of the three dup/dup2/dup3
entrypoints - presumably the one from the bug can be used/adapted.
Dave
