Re: [Qemu-devel] [PATCH v2 2/3] Add 'debug-threads' suboption to --name

[Dr. David Alan Gilbert]

* Laszlo Ersek (address@hidden) wrote:
> a few irrelevant comments below:
> 
> On 01/30/14 11:20, Dr. David Alan Gilbert (git) wrote:
> > From: "Dr. David Alan Gilbert" <address@hidden>
> > 
> > Add flag storage to qemu-thread-* to store the namethreads flag
> > 
> > Signed-off-by: Dr. David Alan Gilbert <address@hidden>
> > ---
> >  include/qemu/thread.h    | 1 +
> >  qemu-options.hx          | 7 +++++--
> >  util/qemu-thread-posix.c | 7 +++++++
> >  util/qemu-thread-win32.c | 8 ++++++++
> >  vl.c                     | 9 +++++++++
> >  5 files changed, 30 insertions(+), 2 deletions(-)
> > 
> > diff --git a/include/qemu/thread.h b/include/qemu/thread.h
> > index 3e32c65..bf1e110 100644
> > --- a/include/qemu/thread.h
> > +++ b/include/qemu/thread.h
> > @@ -59,5 +59,6 @@ void *qemu_thread_join(QemuThread *thread);
> >  void qemu_thread_get_self(QemuThread *thread);
> >  bool qemu_thread_is_self(QemuThread *thread);
> >  void qemu_thread_exit(void *retval);
> > +void qemu_thread_naming(bool enable);
> >  
> >  #endif
> > diff --git a/qemu-options.hx b/qemu-options.hx
> > index 56e5fdf..068da2d 100644
> > --- a/qemu-options.hx
> > +++ b/qemu-options.hx
> > @@ -328,9 +328,11 @@ possible drivers and properties, use @code{-device 
> > help} and
> >  ETEXI
> >  
> >  DEF("name", HAS_ARG, QEMU_OPTION_name,
> > -    "-name string1[,process=string2]\n"
> > +    "-name string1[,process=string2][,debug-threads=on|off]\n"
> >      "                set the name of the guest\n"
> > -    "                string1 sets the window title and string2 the process 
> > name (on Linux)\n",
> > +    "                string1 sets the window title and string2 the process 
> > name (on Linux)\n"
> > +    "                When debug-threads is enabled, individual threads are 
> > given a separate name (on Linux)\n"
> > +    "                NOTE: The thread names are for debugging and not a 
> > stable API.\n",
> >      QEMU_ARCH_ALL)
> >  STEXI
> >  @item -name @var{name}
> > @@ -339,6 +341,7 @@ Sets the @var{name} of the guest.
> >  This name will be displayed in the SDL window caption.
> >  The @var{name} will also be used for the VNC server.
> >  Also optionally set the top visible process name in Linux.
> > +Naming of individual threads can also be enabled on Linux to aid debugging.
> >  ETEXI
> >  
> >  DEF("uuid", HAS_ARG, QEMU_OPTION_uuid,
> > diff --git a/util/qemu-thread-posix.c b/util/qemu-thread-posix.c
> > index 37dd298..0fa6c81 100644
> > --- a/util/qemu-thread-posix.c
> > +++ b/util/qemu-thread-posix.c
> > @@ -27,6 +27,13 @@
> >  #include "qemu/thread.h"
> >  #include "qemu/atomic.h"
> >  
> > +static bool name_threads;
> > +
> > +void qemu_thread_naming(bool enable)
> > +{
> > +    name_threads = enable;
> > +}
> > +
> >  static void error_exit(int err, const char *msg)
> >  {
> >      fprintf(stderr, "qemu: %s: %s\n", msg, strerror(err));
> > diff --git a/util/qemu-thread-win32.c b/util/qemu-thread-win32.c
> > index 27a5217..e42cb77 100644
> > --- a/util/qemu-thread-win32.c
> > +++ b/util/qemu-thread-win32.c
> > @@ -16,6 +16,14 @@
> >  #include <assert.h>
> >  #include <limits.h>
> >  
> > +static bool name_threads;
> > +
> > +void qemu_thread_naming(bool enable)
> > +{
> > +    /* But note we don't actually name them on Windows yet */
> > +    name_threads = enable;
> > +}
> > +
> >  static void error_exit(int err, const char *msg)
> >  {
> >      char *pstr;
> > diff --git a/vl.c b/vl.c
> > index 5f993e4..77d6d9e 100644
> > --- a/vl.c
> > +++ b/vl.c
> > @@ -547,6 +547,12 @@ static QemuOptsList qemu_name_opts = {
> >              .name = "process",
> >              .type = QEMU_OPT_STRING,
> >              .help = "Sets the name of the QEMU process, as shown in top 
> > etc",
> > +        }, {
> > +            .name = "debug-threads",
> > +            .type = QEMU_OPT_BOOL,
> > +            .help = "When enabled, name the individual threads; defaults 
> > off.\n"
> 
> (1) same question about newlines applies

I'd assumed this landed in -help output and that looked OK.

> (2) the default setting is usually advertised in qemu-options.hx, not here

OK

> (3) the meaning of "default" is relative to the case when the option is
> specified and the option argument is not, *not* relative to when the
> option is missing. Cf.
> 
> (a)  -name debug-threads
> (b)  -name debug-threads=on
> (c)  [nothing]
> 
> The default as explained by the help text (in qemu-options.hx),
> independently of option argument type, is usually the difference between
> (a) and (b), not (a) and (c), nor between (b) and (c).
> 
> Here, due to the way boolean options are parsed in parse_option_bool(),
> (a) is identical to (b), so the default is "on". (See -msg timestamp in
> qemu-options.hx, it's similar.)
> 
> I *think*. But I could never really form a decisive understanding of the
> tradition here, so feel free to ignore it.

If 'default' is (a) then what is the word for (c) ? What I was really
trying to express was that if you don't pass the option at all it's
not going to name the threads.

> > +                    "NOTE: The thread names are for debugging and not a\n"
> > +                    "stable API.",
> >          },
> >          { /* End of list */ }
> >      },
> > @@ -1006,6 +1012,9 @@ static void parse_name(QemuOpts *opts)
> >  {
> >      const char *proc_name;
> >  
> > +    if (qemu_opt_get(opts, "debug-threads")) {
> > +        qemu_thread_naming(qemu_opt_get_bool(opts, "debug-threads", 
> > false));
> > +    }
> 
> (4) I'm not sure you need the surrounding "if". qemu_opt_get() and
> qemu_opt_get_bool() both start with qemu_opt_find().
> 
> With the "if", if the option is not passed, then you leave
> "name_threads" at its default value (== false).
> 
> Without the "if", if the option is not passed, then you overwrite
> "name_threads" with the option default (== false). I think the code
> would be simpler if you dropped the surrounding "if".

Yes, I think you're right, I was trying to guard against what happened
when you passed -name multiple times, but as you say the code already
does the right thing.

> >      qemu_name = qemu_opt_get(opts, "guest");
> >  
> >      proc_name = qemu_opt_get(opts, "process");
> > 
> 
> Anyway I shouldn't obsess about this (also because if I do and you
> respin, I get to review v3 too :))
> 
> Reviewed-by: Laszlo Ersek <address@hidden>

Thanks

Dave

> Thanks
> Laszlo
> 
--
Dr. David Alan Gilbert / address@hidden / Manchester, UK