Re: [PATCH v2 1/3] support: Add support_process_state_wait

[Florian Weimer <fweimer@redhat.com>]

* Adhemerval Zanella:

> +/* Wait for process PID to reach state STATE.  It can be a combination of
> +   multiple possible states ('process_state_running | process_state_sleeping')
> +   where the function return when any of these state are observed.
> +   The timeout POLL_TS might be used to wait before polling the PID status
> +   information from the kernel.  A NULL value will set to use a default
> +   value (10000000 ns).
> +   For an invalid state not represented by SUPPORT_PROCESS_STATE, it fallbacks
> +   to a 2 second sleep.  */
> +void support_process_state_wait (pid_t pid, enum support_process_state state,
> +				 struct timespec *poll_ts);

The comment should say what happens if the timeout expires.  And I
believe the argument should be a const struct timespec *.  But I don't
see how controlling the poll interval is useful.

> diff --git a/support/xgetline.c b/support/xgetline.c
> new file mode 100644
> index 0000000000..5b010bcf56
> --- /dev/null
> +++ b/support/xgetline.c
> @@ -0,0 +1,34 @@
> +/* fopen with error checking.
> +   Copyright (C) 2016-2019 Free Software Foundation, Inc.
> +   This file is part of the GNU C Library.
> +
> +   The GNU C Library is free software; you can redistribute it and/or
> +   modify it under the terms of the GNU Lesser General Public
> +   License as published by the Free Software Foundation; either
> +   version 2.1 of the License, or (at your option) any later version.
> +
> +   The GNU C Library is distributed in the hope that it will be useful,
> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> +   Lesser General Public License for more details.
> +
> +   You should have received a copy of the GNU Lesser General Public
> +   License along with the GNU C Library; if not, see
> +   <https://www.gnu.org/licenses/>.  */
> +
> +#include <support/xstdio.h>
> +
> +#include <support/check.h>
> +#include <errno.h>
> +
> +ssize_t
> +xgetline (char **lineptr, size_t *n, FILE *stream)
> +{
> +  int old_errno = errno;
> +  errno = 0;
> +  ssize_t ret = getline (lineptr, n, stream);
> +  if (ret == -1 && errno != 0)
> +    FAIL_EXIT1 ("getline failed: %m");
> +  errno = old_errno;
> +  return ret;
> +}

I think it would be clearer to use ferror to check for errors.

Rest of the patch looks okay.

Thanks,
Florian