Bug 1133240 - scan gets stuck indefinitely in tune_to_next_transponder()
Summary: scan gets stuck indefinitely in tune_to_next_transponder()
Keywords:
Status: CLOSED EOL
Alias: None
Product: Fedora
Classification: Fedora
Component: w_scan
Version: 19
Hardware: x86_64
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Felix Kaechele
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2014-08-23 17:27 UTC by Oliver Henshaw
Modified: 2015-02-18 11:03 UTC (History)
1 user (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2015-02-18 11:03:24 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
" w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan-20140118-1.fc19.x86_64 (11.05 KB, text/plain)
2014-08-23 17:29 UTC, Oliver Henshaw
no flags Details
" w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan-20140727-1.fc19.x86_64 (8.09 KB, text/plain)
2014-08-23 17:30 UTC, Oliver Henshaw
no flags Details
w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan 20140712 (10.00 KB, text/plain)
2014-09-05 13:40 UTC, Oliver Henshaw
no flags Details
"w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan 20140727 + fix 1 (11.80 KB, text/plain)
2014-09-05 13:42 UTC, Oliver Henshaw
no flags Details
"w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan 20140727 + fix 2 (9.31 KB, text/plain)
2014-09-05 13:44 UTC, Oliver Henshaw
no flags Details
"w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan 20140727 + fix 3 (10.05 KB, text/plain)
2014-09-05 13:45 UTC, Oliver Henshaw
no flags Details

Description Oliver Henshaw 2014-08-23 17:27:59 UTC
Description of problem:

The new version of w_scan currently in updates-testing fails to successfully scan and churns cpu until killed. Scanning with initial data from the Crystal Palace transmitter (UK) succeeds with the old w_scan-20140118 but fails with the new version.


Version-Release number of selected component (if applicable):

w_scan-20140727-1.fc19.x86_64
dtv-scan-tables-0-5.20140309git177b522.fc19.noarch


How reproducible:

Scan with
$ w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace

The output with the old and new w_scan is attached. The new version somehow mis-reads the muliplex frequencies from the NIT data, fails to tune to 0 Hz and then spins cpu without making progress part way through scanning the multiplexes for services.

$ pstack `pidof w_scan`
#0  0x0000000000416835 in tune_to_next_transponder (frontend_fd=frontend_fd@entry=3) at scan.c:2306
#1  0x0000000000402d3f in network_scan (tuning_data=<optimized out>, frontend_fd=3) at scan.c:2923
#2  main (argc=<optimized out>, argv=<optimized out>) at scan.c:4101

is typical.

I guess there's two problems here: whatever goes wrong with reading transmitter info from the dvb stream and the tune_to_next_transponder failing to return.


Additional info:

Note the initial scan file is slightly out of date compared to the upstream version, it's missing a new local DVB-T and a new DVB-T2 multiplex, but all other details are correct I think (you can see this in the old w_scan tuning output).

Comment 1 Oliver Henshaw 2014-08-23 17:29:46 UTC
Created attachment 929965 [details]
" w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan-20140118-1.fc19.x86_64

Comment 2 Oliver Henshaw 2014-08-23 17:30:48 UTC
Created attachment 929966 [details]
" w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan-20140727-1.fc19.x86_64

Comment 3 Felix Kaechele 2014-08-30 20:07:19 UTC
Please test this scratch build:
http://koji.fedoraproject.org/koji/taskinfo?taskID=7490933

Comment 4 Oliver Henshaw 2014-09-05 13:40:21 UTC
Created attachment 934791 [details]
w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan 20140712

This fixes the parsing of the DVB-T2 entry, but doesn't solve any of the other problems. I found w_scan-20140712.tar.bz2 on the w_scan site, which works slightly better than 20140727 but not as well as 20140118. I couldn't find any other intermediate tarballs corresponding to the (dev) entries in the ChangeLog, though.


20140712 regressions from 20140118:
- updates transponders with 0 kHz frequencies
- the initial tune skips the 506MHz and 529.833MHz frequencies
- the service finding skips the 506MHz and 529.833MHz frequencies
  -> finds 52 services (rather than 90 for 20140118)


20140727:
- doesn't finish, as described above

Comment 5 Oliver Henshaw 2014-09-05 13:42:31 UTC
Created attachment 934792 [details]
"w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan 20140727 + fix 1

I've identified the problem, but found a couple more regressions from 20140712 to 20140727.

Fix 1 - Don't short-circuit, and let the new_transponders list empty.

diff --git a/scan.c b/scan.c
index 52f40be..53b30fa 100644
--- a/scan.c
+++ b/scan.c
@@ -2308,7 +2308,7 @@ static int tune_to_next_transponder(int frontend_fd) {
      i = 0;
 
    //retry:                                                        /* NOTE: 
jump mark here. */
-     if (t->frequency && (tune_to_transponder(frontend_fd, t) == 0))
+     if (tune_to_transponder(frontend_fd, t) == 0)
         return 0;
 
    //other_freq: 

changes vs. 20140712:
+ finds services from 506MHz frequency - tries 0MHz then succeeds after "retrying with center_frequency = 506000000"
  -> finds 68 services
- outputs a 0MHz QPSK entry to stdout
- misses all but 506MHz, 529.833MHz and 537.833MHz QPSK entries from stdout

Comment 6 Oliver Henshaw 2014-09-05 13:44:06 UTC
Created attachment 934793 [details]
"w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan 20140727 + fix 2

Fix 2 - Don't set type in parse_nit

diff --git a/scan.c b/scan.c
index 53b30fa..dc42397 100644
--- a/scan.c
+++ b/scan.c
@@ -1274,7 +1274,7 @@ em_static void parse_nit(const unsigned char * buf, uint16_t section_length, uin
         }
 
      memset(&tn, 0, sizeof(tn));
-     tn.type                = current_tp->type;
+     tn.type = -1;
      tn.network_PID         = current_tp->network_PID;
      tn.network_id          = network_id;
      tn.original_network_id = original_network_id;

changes:
+ 0MHz QPSK entry is gone
- so are services from 506MHz frequency
  -> back to 52 services

Comment 7 Oliver Henshaw 2014-09-05 13:45:36 UTC
Created attachment 934794 [details]
"w_scan -c GB -x -I /usr/share/dvb/dvb-t/uk-CrystalPalace" output for w_scan 20140727 + fix 3

Fix 3 - auto_params fallout

diff --git a/scan.c b/scan.c
index dc42397..3b6714d 100644
--- a/scan.c
+++ b/scan.c
@@ -1304,7 +1304,7 @@ em_static void parse_nit(const unsigned char * buf, uint16_t section_length, uin
             tn.frequency = t->frequency;
         if (table_id == TABLE_NIT_ACT) {
            // only nit_actual should update transponders, too much garbage in satellite nit_other.
-           if (is_different_transponder_deep_scan(t, &tn, 0) && ((! t->locks_with_params) || is_auto_params(t))) { // || t->source != tn.source) {
+           if ( (is_different_transponder_deep_scan(t, &tn, 0) && ((! t->locks_with_params) || is_auto_params(t))) || t->source != tn.source) {
               /* some of the informations is still set to AUTO */
               print_transponder(buffer, t);
               info("        updating transponder:\n           (%s) 0x%.4X\n", buffer, t->source);

Changes:
+ the missing entres from stdout are back
= now back to the 20140712 status quo, with all its regressions.

Comment 8 Oliver Henshaw 2014-09-05 13:59:52 UTC
I don't necessarily think all, or even any, of these are correct fixes -there's probably better ways to stop zero frequencies wreaking havoc across the code. And parsing the correct frequencies might help too.


Another small issue I noticed, although it doesn't seem to affect me. locks_with_params is set to false and straight back to true if you don't find a lock:

diff --git a/scan.c b/scan.c
index 3b6714d..b5eafae 100644
--- a/scan.c
+++ b/scan.c
@@ -2246,17 +2246,16 @@ static int __tune_to_transponder(int frontend_fd, struct transponder * t, int v)
   t->locks_with_params = false;
  
   if (v > 0)
      info("----------no signal----------\n");
   else 
      info("\n");
   
   t->last_tuning_failed = 1;
-  t->locks_with_params = 1;
 
   /* tuning didnt work, retry with auto. */
   if (t->delsys != SYS_DVBS2) t->modulation = QAM_AUTO;
   t->pilot = PILOT_AUTO;
   t->coderate = FEC_AUTO;
   t->guard = GUARD_INTERVAL_AUTO;
   t->rolloff = ROLLOFF_AUTO;
   t->transmission = TRANSMISSION_MODE_AUTO;

Comment 9 Fedora End Of Life 2015-01-09 22:25:23 UTC
This message is a notice that Fedora 19 is now at end of life. Fedora 
has stopped maintaining and issuing updates for Fedora 19. It is 
Fedora's policy to close all bug reports from releases that are no 
longer maintained. Approximately 4 (four) weeks from now this bug will
be closed as EOL if it remains open with a Fedora 'version' of '19'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora 19 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora, you are encouraged  change the 'version' to a later Fedora 
version prior this bug is closed as described in the policy above.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events. Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

Comment 10 Fedora End Of Life 2015-02-18 11:03:24 UTC
Fedora 19 changed to end-of-life (EOL) status on 2015-01-06. Fedora 19 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
bug.

Thank you for reporting this bug and we are sorry it could not be fixed.


Note You need to log in before you can comment on or make changes to this bug.