1S390 Debug Feature
2==================
3
4files: arch/s390/kernel/debug.c
5       include/asm-s390/debug.h
6
7Description:
8------------
9The goal of this feature is to provide a kernel debug logging API 
10where log records can be stored efficiently in memory, where each component 
11(e.g. device drivers) can have one separate debug log.
12One purpose of this is to inspect the debug logs after a production system crash
13in order to analyze the reason for the crash.
14If the system still runs but only a subcomponent which uses dbf fails,
15it is possible to look at the debug logs on a live system via the Linux
16debugfs filesystem.
17The debug feature may also very useful for kernel and driver development.
18
19Design:
20-------
21Kernel components (e.g. device drivers) can register themselves at the debug 
22feature with the function call debug_register(). This function initializes a 
23debug log for the caller. For each debug log exists a number of debug areas 
24where exactly one is active at one time.  Each debug area consists of contiguous
25pages in memory. In the debug areas there are stored debug entries (log records)
26which are written by event- and exception-calls. 
27
28An event-call writes the specified debug entry to the active debug
29area and updates the log pointer for the active area. If the end 
30of the active debug area is reached, a wrap around is done (ring buffer) 
31and the next debug entry will be written at the beginning of the active 
32debug area.
33
34An exception-call writes the specified debug entry to the log and
35switches to the next debug area. This is done in order to be sure
36that the records which describe the origin of the exception are not
37overwritten when a wrap around for the current area occurs.
38
39The debug areas themselves are also ordered in form of a ring buffer.
40When an exception is thrown in the last debug area, the following debug 
41entries are then written again in the very first area.
42
43There are three versions for the event- and exception-calls: One for
44logging raw data, one for text and one for numbers.
45
46Each debug entry contains the following data:
47
48- Timestamp
49- Cpu-Number of calling task
50- Level of debug entry (0...6)
51- Return Address to caller
52- Flag, if entry is an exception or not
53
54The debug logs can be inspected in a live system through entries in
55the debugfs-filesystem. Under the toplevel directory "s390dbf" there is
56a directory for each registered component, which is named like the
57corresponding component. The debugfs normally should be mounted to
58/sys/kernel/debug therefore the debug feature can be accessed under
59/sys/kernel/debug/s390dbf.
60
61The content of the directories are files which represent different views
62to the debug log. Each component can decide which views should be
63used through registering them with the function debug_register_view().
64Predefined views for hex/ascii, sprintf and raw binary data are provided.
65It is also possible to define other views. The content of
66a view can be inspected simply by reading the corresponding debugfs file.
67
68All debug logs have an actual debug level (range from 0 to 6).
69The default level is 3. Event and Exception functions have a 'level'
70parameter. Only debug entries with a level that is lower or equal
71than the actual level are written to the log. This means, when
72writing events, high priority log entries should have a low level
73value whereas low priority entries should have a high one.
74The actual debug level can be changed with the help of the debugfs-filesystem
75through writing a number string "x" to the 'level' debugfs file which is
76provided for every debug log. Debugging can be switched off completely
77by using "-" on the 'level' debugfs file.
78
79Example:
80
81> echo "-" > /sys/kernel/debug/s390dbf/dasd/level
82
83It is also possible to deactivate the debug feature globally for every
84debug log. You can change the behavior using  2 sysctl parameters in
85/proc/sys/s390dbf:
86There are currently 2 possible triggers, which stop the debug feature
87globally. The first possibility is to use the "debug_active" sysctl. If
88set to 1 the debug feature is running. If "debug_active" is set to 0 the
89debug feature is turned off.
90The second trigger which stops the debug feature is a kernel oops.
91That prevents the debug feature from overwriting debug information that
92happened before the oops. After an oops you can reactivate the debug feature
93by piping 1 to /proc/sys/s390dbf/debug_active. Nevertheless, its not
94suggested to use an oopsed kernel in a production environment.
95If you want to disallow the deactivation of the debug feature, you can use
96the "debug_stoppable" sysctl. If you set "debug_stoppable" to 0 the debug
97feature cannot be stopped. If the debug feature is already stopped, it
98will stay deactivated.
99
100Kernel Interfaces:
101------------------
102
103----------------------------------------------------------------------------
104debug_info_t *debug_register(char *name, int pages, int nr_areas,
105                             int buf_size);
106
107Parameter:    name:        Name of debug log (e.g. used for debugfs entry)
108              pages:       number of pages, which will be allocated per area
109              nr_areas:    number of debug areas
110              buf_size:    size of data area in each debug entry
111
112Return Value: Handle for generated debug area   
113              NULL if register failed 
114
115Description:  Allocates memory for a debug log     
116              Must not be called within an interrupt handler 
117
118---------------------------------------------------------------------------
119void debug_unregister (debug_info_t * id);
120
121Parameter:     id:   handle for debug log  
122
123Return Value:  none 
124
125Description:   frees memory for a debug log     
126               Must not be called within an interrupt handler 
127
128---------------------------------------------------------------------------
129void debug_set_level (debug_info_t * id, int new_level);
130
131Parameter:     id:        handle for debug log  
132               new_level: new debug level 
133
134Return Value:  none 
135
136Description:   Sets new actual debug level if new_level is valid. 
137
138---------------------------------------------------------------------------
139void debug_stop_all(void);
140
141Parameter:     none
142
143Return Value:  none
144
145Description:   stops the debug feature if stopping is allowed. Currently
146               used in case of a kernel oops.
147
148---------------------------------------------------------------------------
149debug_entry_t* debug_event (debug_info_t* id, int level, void* data, 
150                            int length);
151
152Parameter:     id:     handle for debug log  
153               level:  debug level           
154               data:   pointer to data for debug entry  
155               length: length of data in bytes       
156
157Return Value:  Address of written debug entry 
158
159Description:   writes debug entry to active debug area (if level <= actual 
160               debug level)    
161
162---------------------------------------------------------------------------
163debug_entry_t* debug_int_event (debug_info_t * id, int level, 
164                                unsigned int data);
165debug_entry_t* debug_long_event(debug_info_t * id, int level,
166                                unsigned long data);
167
168Parameter:     id:     handle for debug log  
169               level:  debug level           
170               data:   integer value for debug entry           
171
172Return Value:  Address of written debug entry 
173
174Description:   writes debug entry to active debug area (if level <= actual 
175               debug level)    
176
177---------------------------------------------------------------------------
178debug_entry_t* debug_text_event (debug_info_t * id, int level, 
179                                 const char* data);
180
181Parameter:     id:     handle for debug log  
182               level:  debug level           
183               data:   string for debug entry  
184
185Return Value:  Address of written debug entry 
186
187Description:   writes debug entry in ascii format to active debug area 
188               (if level <= actual debug level)     
189
190---------------------------------------------------------------------------
191debug_entry_t* debug_sprintf_event (debug_info_t * id, int level, 
192                                    char* string,...);
193
194Parameter:     id:    handle for debug log 
195               level: debug level
196               string: format string for debug entry 
197               ...: varargs used as in sprintf()
198
199Return Value:  Address of written debug entry
200
201Description:   writes debug entry with format string and varargs (longs) to 
202               active debug area (if level $<=$ actual debug level). 
203               floats and long long datatypes cannot be used as varargs.
204
205---------------------------------------------------------------------------
206
207debug_entry_t* debug_exception (debug_info_t* id, int level, void* data, 
208                                int length);
209
210Parameter:     id:     handle for debug log  
211               level:  debug level           
212               data:   pointer to data for debug entry  
213               length: length of data in bytes       
214
215Return Value:  Address of written debug entry 
216
217Description:   writes debug entry to active debug area (if level <= actual 
218               debug level) and switches to next debug area  
219
220---------------------------------------------------------------------------
221debug_entry_t* debug_int_exception (debug_info_t * id, int level, 
222                                    unsigned int data);
223debug_entry_t* debug_long_exception(debug_info_t * id, int level,
224                                    unsigned long data);
225
226Parameter:     id:     handle for debug log  
227               level:  debug level           
228               data:   integer value for debug entry           
229
230Return Value:  Address of written debug entry 
231
232Description:   writes debug entry to active debug area (if level <= actual 
233               debug level) and switches to next debug area  
234
235---------------------------------------------------------------------------
236debug_entry_t* debug_text_exception (debug_info_t * id, int level, 
237                                     const char* data);
238
239Parameter:     id:     handle for debug log  
240               level:  debug level           
241               data:   string for debug entry  
242
243Return Value:  Address of written debug entry 
244
245Description:   writes debug entry in ascii format to active debug area 
246               (if level <= actual debug level) and switches to next debug 
247               area  
248
249---------------------------------------------------------------------------
250debug_entry_t* debug_sprintf_exception (debug_info_t * id, int level,
251                                        char* string,...);
252
253Parameter:     id:    handle for debug log  
254               level: debug level  
255               string: format string for debug entry  
256               ...: varargs used as in sprintf()
257
258Return Value:  Address of written debug entry 
259
260Description:   writes debug entry with format string and varargs (longs) to 
261               active debug area (if level $<=$ actual debug level) and
262               switches to next debug area. 
263               floats and long long datatypes cannot be used as varargs.
264
265---------------------------------------------------------------------------
266
267int debug_register_view (debug_info_t * id, struct debug_view *view);
268
269Parameter:     id:    handle for debug log  
270               view:  pointer to debug view struct 
271
272Return Value:  0  : ok 
273               < 0: Error 
274
275Description:   registers new debug view and creates debugfs dir entry
276
277---------------------------------------------------------------------------
278int debug_unregister_view (debug_info_t * id, struct debug_view *view); 
279
280Parameter:     id:    handle for debug log  
281               view:  pointer to debug view struct 
282
283Return Value:  0  : ok 
284               < 0: Error 
285
286Description:   unregisters debug view and removes debugfs dir entry
287
288
289
290Predefined views:
291-----------------
292
293extern struct debug_view debug_hex_ascii_view;
294extern struct debug_view debug_raw_view;
295extern struct debug_view debug_sprintf_view;
296
297Examples
298--------
299
300/*
301 * hex_ascii- + raw-view Example
302 */
303
304#include <linux/init.h>
305#include <asm/debug.h>
306
307static debug_info_t* debug_info;
308
309static int init(void)
310{
311    /* register 4 debug areas with one page each and 4 byte data field */
312
313    debug_info = debug_register ("test", 1, 4, 4 );
314    debug_register_view(debug_info,&debug_hex_ascii_view);
315    debug_register_view(debug_info,&debug_raw_view);
316
317    debug_text_event(debug_info, 4 , "one ");
318    debug_int_exception(debug_info, 4, 4711);
319    debug_event(debug_info, 3, &debug_info, 4);
320
321    return 0;
322}
323
324static void cleanup(void)
325{
326    debug_unregister (debug_info);
327}
328
329module_init(init);
330module_exit(cleanup);
331
332---------------------------------------------------------------------------
333
334/*
335 * sprintf-view Example
336 */
337
338#include <linux/init.h>
339#include <asm/debug.h>
340
341static debug_info_t* debug_info;
342
343static int init(void)
344{
345    /* register 4 debug areas with one page each and data field for */
346    /* format string pointer + 2 varargs (= 3 * sizeof(long))       */
347
348    debug_info = debug_register ("test", 1, 4, sizeof(long) * 3);
349    debug_register_view(debug_info,&debug_sprintf_view);
350
351    debug_sprintf_event(debug_info, 2 , "first event in %s:%i\n",__FILE__,__LINE__);
352    debug_sprintf_exception(debug_info, 1, "pointer to debug info: %p\n",&debug_info);
353
354    return 0;
355}
356
357static void cleanup(void)
358{
359    debug_unregister (debug_info);
360}
361
362module_init(init);
363module_exit(cleanup);
364
365
366
367Debugfs Interface
368----------------
369Views to the debug logs can be investigated through reading the corresponding 
370debugfs-files:
371
372Example:
373
374> ls /sys/kernel/debug/s390dbf/dasd
375flush  hex_ascii  level pages raw
376> cat /sys/kernel/debug/s390dbf/dasd/hex_ascii | sort +1
37700 00974733272:680099 2 - 02 0006ad7e  07 ea 4a 90 | ....
37800 00974733272:682210 2 - 02 0006ade6  46 52 45 45 | FREE
37900 00974733272:682213 2 - 02 0006adf6  07 ea 4a 90 | ....
38000 00974733272:682281 1 * 02 0006ab08  41 4c 4c 43 | EXCP 
38101 00974733272:682284 2 - 02 0006ab16  45 43 4b 44 | ECKD
38201 00974733272:682287 2 - 02 0006ab28  00 00 00 04 | ....
38301 00974733272:682289 2 - 02 0006ab3e  00 00 00 20 | ... 
38401 00974733272:682297 2 - 02 0006ad7e  07 ea 4a 90 | ....
38501 00974733272:684384 2 - 00 0006ade6  46 52 45 45 | FREE
38601 00974733272:684388 2 - 00 0006adf6  07 ea 4a 90 | ....
387
388See section about predefined views for explanation of the above output!
389
390Changing the debug level
391------------------------
392
393Example:
394
395
396> cat /sys/kernel/debug/s390dbf/dasd/level
3973
398> echo "5" > /sys/kernel/debug/s390dbf/dasd/level
399> cat /sys/kernel/debug/s390dbf/dasd/level
4005
401
402Flushing debug areas
403--------------------
404Debug areas can be flushed with piping the number of the desired
405area (0...n) to the debugfs file "flush". When using "-" all debug areas
406are flushed.
407
408Examples:
409
4101. Flush debug area 0:
411> echo "0" > /sys/kernel/debug/s390dbf/dasd/flush
412
4132. Flush all debug areas:
414> echo "-" > /sys/kernel/debug/s390dbf/dasd/flush
415
416Changing the size of debug areas
417------------------------------------
418It is possible the change the size of debug areas through piping
419the number of pages to the debugfs file "pages". The resize request will
420also flush the debug areas.
421
422Example:
423
424Define 4 pages for the debug areas of debug feature "dasd":
425> echo "4" > /sys/kernel/debug/s390dbf/dasd/pages
426
427Stooping the debug feature
428--------------------------
429Example:
430
4311. Check if stopping is allowed
432> cat /proc/sys/s390dbf/debug_stoppable
4332. Stop debug feature
434> echo 0 > /proc/sys/s390dbf/debug_active
435
436lcrash Interface
437----------------
438It is planned that the dump analysis tool lcrash gets an additional command
439's390dbf' to display all the debug logs. With this tool it will be possible 
440to investigate the debug logs on a live system and with a memory dump after 
441a system crash.
442
443Investigating raw memory
444------------------------
445One last possibility to investigate the debug logs at a live
446system and after a system crash is to look at the raw memory
447under VM or at the Service Element.
448It is possible to find the anker of the debug-logs through
449the 'debug_area_first' symbol in the System map. Then one has
450to follow the correct pointers of the data-structures defined
451in debug.h and find the debug-areas in memory.
452Normally modules which use the debug feature will also have
453a global variable with the pointer to the debug-logs. Following
454this pointer it will also be possible to find the debug logs in
455memory.
456
457For this method it is recommended to use '16 * x + 4' byte (x = 0..n)
458for the length of the data field in debug_register() in
459order to see the debug entries well formatted.
460
461
462Predefined Views
463----------------
464
465There are three predefined views: hex_ascii, raw and sprintf. 
466The hex_ascii view shows the data field in hex and ascii representation 
467(e.g. '45 43 4b 44 | ECKD'). 
468The raw view returns a bytestream as the debug areas are stored in memory.
469
470The sprintf view formats the debug entries in the same way as the sprintf
471function would do. The sprintf event/exception functions write to the
472debug entry a pointer to the format string (size = sizeof(long)) 
473and for each vararg a long value. So e.g. for a debug entry with a format 
474string plus two varargs one would need to allocate a (3 * sizeof(long)) 
475byte data area in the debug_register() function.
476
477
478NOTE: If using the sprintf view do NOT use other event/exception functions
479than the sprintf-event and -exception functions.
480
481The format of the hex_ascii and sprintf view is as follows:
482- Number of area
483- Timestamp (formatted as seconds and microseconds since 00:00:00 Coordinated 
484  Universal Time (UTC), January 1, 1970)
485- level of debug entry
486- Exception flag (* = Exception)
487- Cpu-Number of calling task
488- Return Address to caller
489- data field
490
491The format of the raw view is:
492- Header as described in debug.h
493- datafield 
494
495A typical line of the hex_ascii view will look like the following (first line 
496is only for explanation and will not be displayed when 'cating' the view):
497
498area  time           level exception cpu caller    data (hex + ascii)
499--------------------------------------------------------------------------
50000    00964419409:440690 1 -         00  88023fe   
501
502
503Defining views
504--------------
505
506Views are specified with the 'debug_view' structure. There are defined
507callback functions which are used for reading and writing the debugfs files:
508
509struct debug_view {
510        char name[DEBUG_MAX_PROCF_LEN];  
511        debug_prolog_proc_t* prolog_proc; 
512        debug_header_proc_t* header_proc;
513        debug_format_proc_t* format_proc;
514        debug_input_proc_t*  input_proc;
515	void*                private_data;
516};
517
518where
519
520typedef int (debug_header_proc_t) (debug_info_t* id,
521                                   struct debug_view* view,
522                                   int area,
523                                   debug_entry_t* entry,
524                                   char* out_buf);
525
526typedef int (debug_format_proc_t) (debug_info_t* id,
527                                   struct debug_view* view, char* out_buf,
528                                   const char* in_buf);
529typedef int (debug_prolog_proc_t) (debug_info_t* id,
530                                   struct debug_view* view,
531                                   char* out_buf);
532typedef int (debug_input_proc_t) (debug_info_t* id,
533                                  struct debug_view* view,
534                                  struct file* file, const char* user_buf,
535                                  size_t in_buf_size, loff_t* offset);
536
537
538The "private_data" member can be used as pointer to view specific data.
539It is not used by the debug feature itself.
540
541The output when reading a debugfs file is structured like this:
542
543"prolog_proc output"
544
545"header_proc output 1"  "format_proc output 1"
546"header_proc output 2"  "format_proc output 2"
547"header_proc output 3"  "format_proc output 3"
548...
549
550When a view is read from the debugfs, the Debug Feature calls the
551'prolog_proc' once for writing the prolog.
552Then 'header_proc' and 'format_proc' are called for each 
553existing debug entry.
554
555The input_proc can be used to implement functionality when it is written to 
556the view (e.g. like with 'echo "0" > /sys/kernel/debug/s390dbf/dasd/level).
557
558For header_proc there can be used the default function
559debug_dflt_header_fn() which is defined in debug.h.
560and which produces the same header output as the predefined views.
561E.g:
56200 00964419409:440761 2 - 00 88023ec
563
564In order to see how to use the callback functions check the implementation
565of the default views!
566
567Example
568
569#include <asm/debug.h>
570
571#define UNKNOWNSTR "data: %08x"
572
573const char* messages[] =
574{"This error...........\n",
575 "That error...........\n",
576 "Problem..............\n",
577 "Something went wrong.\n",
578 "Everything ok........\n",
579 NULL
580};
581
582static int debug_test_format_fn(
583   debug_info_t * id, struct debug_view *view, 
584   char *out_buf, const char *in_buf
585)
586{
587  int i, rc = 0;
588
589  if(id->buf_size >= 4) {
590     int msg_nr = *((int*)in_buf);
591     if(msg_nr < sizeof(messages)/sizeof(char*) - 1)
592        rc += sprintf(out_buf, "%s", messages[msg_nr]);	
593     else
594        rc += sprintf(out_buf, UNKNOWNSTR, msg_nr);
595  }
596 out:
597   return rc;
598}
599
600struct debug_view debug_test_view = {
601  "myview",                 /* name of view */
602  NULL,                     /* no prolog */
603  &debug_dflt_header_fn,    /* default header for each entry */
604  &debug_test_format_fn,    /* our own format function */
605  NULL,                     /* no input function */
606  NULL                      /* no private data */
607};
608
609=====
610test:
611=====
612debug_info_t *debug_info;
613...
614debug_info = debug_register ("test", 0, 4, 4 ));
615debug_register_view(debug_info, &debug_test_view);
616for(i = 0; i < 10; i ++) debug_int_event(debug_info, 1, i);
617
618> cat /sys/kernel/debug/s390dbf/test/myview
61900 00964419734:611402 1 - 00 88042ca   This error...........
62000 00964419734:611405 1 - 00 88042ca   That error...........
62100 00964419734:611408 1 - 00 88042ca   Problem..............
62200 00964419734:611411 1 - 00 88042ca   Something went wrong.
62300 00964419734:611414 1 - 00 88042ca   Everything ok........
62400 00964419734:611417 1 - 00 88042ca   data: 00000005
62500 00964419734:611419 1 - 00 88042ca   data: 00000006
62600 00964419734:611422 1 - 00 88042ca   data: 00000007
62700 00964419734:611425 1 - 00 88042ca   data: 00000008
62800 00964419734:611428 1 - 00 88042ca   data: 00000009
629