lkml.org 
[lkml]   [2013]   [Nov]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 04/22] tools lib traceevent: Add jbd2 plugin
On Fri, 22 Nov 2013 23:27:57 +0900
Namhyung Kim <namhyung@kernel.org> wrote:

> [SNIP]
> > +#define MINORBITS 20
> > +#define MINORMASK ((1U << MINORBITS) - 1)
> > +
> > +#define MAJOR(dev) ((unsigned int) ((dev) >> MINORBITS))
> > +#define MINOR(dev) ((unsigned int) ((dev) & MINORMASK))
> > +
> > +unsigned long long process_jbd2_dev_to_name(struct trace_seq *s,
> > + unsigned long long *args)
>
> Isn't it better to make these functions static?

Yeah, but that was my mistake. Lets keep the fix separate from the
backport though. Thanks!

>
>
> > +{
> > + unsigned int dev = args[0];
> > +
> > + trace_seq_printf(s, "%d:%d", MAJOR(dev), MINOR(dev));
> > + return 0;
> > +}
> > +
> > +unsigned long long process_jiffies_to_msecs(struct trace_seq *s,
> > + unsigned long long *args)
> > +{
> > + unsigned long long jiffies = args[0];
> > +
> > + trace_seq_printf(s, "%lld", jiffies);
> > + return jiffies;
> > +}
> > +
> > +int PEVENT_PLUGIN_LOADER(struct pevent *pevent)
> > +{
> > + pevent_register_print_function(pevent,
> > + process_jbd2_dev_to_name,
> > + PEVENT_FUNC_ARG_STRING,
>
> Actually the function returns long long not string. But it seems the
> current code doesn't care about the return type.

Actually it's not representing what process_jbd2_dev_to_name() returns
(which will always return unsigned long long), but what
"jbd2_dev_to_name()" returns that is (was) defined in the kernel. That
was:

const char *jbd2_dev_to_name(dev_t device)

When registering a function to handle, you need to express the
prototype of that function (not the handler). The third argument is the
ret_type of that function.

But this is interesting, the ret_type doesn't seem to be used in
event_parse.c. The return value of the callback is only done in
eval_num_arg() where we could put a warning if the ret_type is not a
number.

Funny thing is, this patch is irrelevant with todays kernel due to this
commit:

commit 4862fd6047ed02e2726667c54d35f538eecc56aa
Author: Theodore Ts'o <tytso@mit.edu>
Date: Sun Jul 10 22:05:08 2011 -0400

jbd2: remove jbd2_dev_to_name() from jbd2 tracepoints

Using function calls in TP_printk causes perf heartburn, so print the
MAJOR/MINOR device numbers instead.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>



>
>
> > + "jbd2_dev_to_name",
> > + PEVENT_FUNC_ARG_INT,
> > + PEVENT_FUNC_ARG_VOID);
> > +
> > + pevent_register_print_function(pevent,
> > + process_jiffies_to_msecs,
> > + PEVENT_FUNC_ARG_LONG,
> > + "jiffies_to_msecs",
> > + PEVENT_FUNC_ARG_LONG,
> > + PEVENT_FUNC_ARG_VOID);
> > + return 0;
> > +}
>
> Shouldn't it unregister these functions when unloaded?

Probably ;-)

-- Steve

>
> Thanks,
> Namhyung
>



\
 
 \ /
  Last update: 2013-11-23 10:21    [W:0.371 / U:0.216 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site