Re: [PATCH v6 06/34] monitor: add 'chardev' property to Monitor base class

[Markus Armbruster <armbru@redhat.com>]

Daniel P. Berrangé <berrange@redhat.com> writes:

> This is associates both QMP and HMP monitors with a character
> device backend.
>
> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> Tested-by: Peter Krempa <pkrempa@redhat.com>
> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
> ---
>  chardev/char.c             |  3 ++-
>  gdbstub/system.c           |  4 ++--
>  include/monitor/monitor.h  |  4 ++--
>  monitor/hmp.c              | 10 ++++++--
>  monitor/monitor-internal.h |  2 ++
>  monitor/monitor.c          | 47 ++++++++++++++++++++++++++++++--------
>  monitor/qmp.c              | 18 ++++++++++-----
>  stubs/monitor-internal.c   |  2 +-
>  8 files changed, 67 insertions(+), 23 deletions(-)
>
> diff --git a/chardev/char.c b/chardev/char.c
> index c71ffa7963..22e5bae388 100644
> --- a/chardev/char.c
> +++ b/chardev/char.c
> @@ -804,8 +804,9 @@ static Chardev *qemu_chr_new_from_name(const char *label, const char *filename,
>      }
>  
>      if (qemu_opt_get_bool(opts, "mux", 0)) {
> +        const char *chardev_id = qemu_opts_id(opts);
>          assert(permit_mux_mon);
> -        monitor_new_hmp(NULL, chr, true, &err);
> +        monitor_new_hmp(NULL, chardev_id, true, &err);
>          if (err) {
>              error_report_err(err);
>              object_unparent(OBJECT(chr));
> diff --git a/gdbstub/system.c b/gdbstub/system.c
> index fda8ef9352..d681476f5e 100644
> --- a/gdbstub/system.c
> +++ b/gdbstub/system.c
> @@ -387,9 +387,9 @@ bool gdbserver_start(const char *device, Error **errp)
>          qemu_add_vm_change_state_handler(gdb_vm_state_change, NULL);
>  
>          /* Initialize a monitor terminal for gdb */
> -        mon_chr = qemu_chardev_new(NULL, TYPE_CHARDEV_GDB,
> +        mon_chr = qemu_chardev_new("gdbchrdev0", TYPE_CHARDEV_GDB,
>                                     NULL, NULL, &error_abort);

This chardev now gets ID "gdbchrdev0" instead of id_generate(ID_CHR).

If you truly need this, you have to mention it in the commit message.

> -        monitor_new_hmp(NULL, mon_chr, false, &error_abort);
> +        monitor_new_hmp(NULL, "gdbchrdev0", false, &error_abort);

If you don't, then passing mon_chr->label here should do the trick.

>      } else {
>          qemu_chr_fe_deinit(&gdbserver_system_state.chr, true);
>          mon_chr = gdbserver_system_state.mon_chr;
> diff --git a/include/monitor/monitor.h b/include/monitor/monitor.h
> index 83723f705a..bfbeedec9b 100644
> --- a/include/monitor/monitor.h
> +++ b/include/monitor/monitor.h
> @@ -29,9 +29,9 @@ bool monitor_cur_is_qmp(void);
>  void monitor_init_globals(void);
>  void monitor_init_globals_core(void);
>  char *monitor_compat_id(void);
> -void monitor_new_qmp(const char *id, Chardev *chr,
> +void monitor_new_qmp(const char *id, const char *chardev_id,
>                       bool pretty, Error **errp);
> -void monitor_new_hmp(const char *id, Chardev *chr,
> +void monitor_new_hmp(const char *id, const char *chardev_id,
>                       bool use_readline, Error **errp);
>  int monitor_new(MonitorOptions *opts, bool allow_hmp, Error **errp);
>  int monitor_new_opts(QemuOpts *opts, Error **errp);
> diff --git a/monitor/hmp.c b/monitor/hmp.c
> index da11e56854..be59b6ca65 100644
> --- a/monitor/hmp.c
> +++ b/monitor/hmp.c
> @@ -23,6 +23,7 @@
>   */
>  
>  #include "qemu/osdep.h"
> +#include "qapi/error.h"
>  #include <dirent.h>
>  #include "hw/core/qdev.h"
>  #include "hw/core/sysemu-cpu-ops.h"
> @@ -1538,22 +1539,27 @@ static void monitor_readline_flush(void *opaque)
>      monitor_flush(&mon->parent_obj);
>  }
>  
> -void monitor_new_hmp(const char *id, Chardev *chr,
> +void monitor_new_hmp(const char *id, const char *chardev_id,
>                       bool use_readline, Error **errp)

You now pass a chardev ID instead of a Chardev *.

I *guess* that's possible because chardev_new() asserts it's passed an
ID.

I should now check all callers of monitor_new_hmp() to make sure they
pass the correct ID.  On a Friday afternoon.  I hoist the white flag.

Such refactorings should be mentioned in the commit message.  Even
better would be a separate patch.

>  {
> +    ERRP_GUARD();
>      MonitorHMP *mon;
>      g_autofree char *autoid = id ? NULL : monitor_compat_id();
>      Object *obj = object_new_with_props(TYPE_MONITOR_HMP,
>                                          object_get_objects_root(),
>                                          id ? id : autoid,
>                                          errp,
> +                                        "chardev", chardev_id,
>                                          NULL);
> +
>      if (!obj) {
>          return;
>      }
> +
>      mon = MONITOR_HMP(obj);
>  
> -    if (!qemu_chr_fe_init(&mon->parent_obj.chr, chr, errp)) {
> +    monitor_complete(MONITOR(mon), errp);
> +    if (*errp) {
>          object_unparent(OBJECT(mon));
>          return;
>      }

[...]