On Tue, Sep 03, 2019 at 04:04:20PM +0200, Petr Mladek wrote: > On Mon 2019-09-02 19:01:39, Andy Shevchenko wrote: > > On Mon, Sep 02, 2019 at 04:39:35PM +0200, Petr Mladek wrote: > > > On Mon 2019-09-02 11:32:36, Sakari Ailus wrote: > > > > %pS and %ps are now the preferred conversion specifiers to print > > > > function > > > > names. The functionality is equivalent; remove the old, deprecated %pF > > > > and %pf support. > > > > > > Hmm, I see the following in master: > > > > > > $> git grep %pF > > > tools/lib/traceevent/Documentation/libtraceevent-func_apis.txt:or events > > > have "%pF" or "%pS" parameter in its format string. It is common to > > > > > > $> git grep %pf > > > tools/lib/traceevent/event-parse.c: if (asprintf(&format, > > > "%%pf: (NO FORMAT FOUND at %llx)\n", addr) < 0) > > > tools/lib/traceevent/event-parse.c: if (asprintf(&format, "%s: %s", > > > "%pf", printk->printk) < 0) > > > > > > I wonder how this is related to printk(). In each case, it seems > > > > It's going thru binary printf() I suppose. The fist stage just saves the > > format > > string and argument addresses or so and prints in later on when user is > > looking > > for human-readable output. > > It seems that vbin_printf() still thinks that %pf and %pF > handle function pointers. If I get it correctly, it just > stores the binary data and the formating is done when > tracing log is read. The idea is the function pointers > will stay the same. > > We need to fix/obsolete this path as well.
Agreed. I'll include a patch to do that in v6. -- Sakari Ailus [email protected]
