Skip to contentSkip to conversation

babeltrace test failures on some architectures

DISCUSSION Jérémie Galarneau 4 participants 9 messages Jun 9, 2014 – Jun 20, 2014 6 messages not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): 20140604184126.GA26877@helmut

#1 lore raw

Hi Jon,

Would it be possible to run the tests on these machines and provide the TAP output?

Thanks,
Jérémie

Quoted 21 lines · Jon Bernard

I noticed babeltrace is having test failures on armel, hurd-i386, kfreebsd-{amd64,i386}, and sparc. These were passing in version 1.1 and the failures appeared in 1.2. Here are links to the build logs:

    armel:          https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=armel&ver=1.2.1-2&stamp=1395846913
    hurd-i386:      https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=hurd-i386&ver=1.2.1-2&stamp=1398158394
    kfreebsd-amd64: https://buildd.debian.org/status/architecture.php?a=kfreebsd-amd64&suite=sid
    kfreebsd-i386:  https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=kfreebsd-i386&ver=1.2.1-2&stamp=1395842518
    sparc:          https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=sparc&ver=1.2.1-2&stamp=1395859546

    buildd page:    https://buildd.debian.org/status/package.php?p=babeltrace

Cheers,

--
Jon

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

Signature · 8 lines
Jérémie Galarneau
EfficiOS Inc.
http://www.efficios.com

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

A message that is not in the archive (sent off-list or not yet received): CA+jJMxtn8XiLjT5NVURv5K8kECPPoPs1WF-0zTakJmRW=CO3mg@mail.gmail.com

#3 lore raw

Jérémie, the end of the build.log seem to contains the tests output.

Quoted 44 lines · Jérémie Galarneau

Hi Jon,

Would it be possible to run the tests on these machines and provide the TAP output?

Thanks,
Jérémie

Quoted 5 lines · Jon Bernard

I noticed babeltrace is having test failures on armel, hurd-i386, kfreebsd-{amd64,i386}, and sparc. These were passing in version 1.1 and the failures appeared in 1.2. Here are links to the build logs:

armel:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=armel&ver=1.2.1-2&stamp=1395846913

Quoted 1 line
    hurd-i386:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=hurd-i386&ver=1.2.1-2&stamp=1398158394

Quoted 1 line
    kfreebsd-amd64:

https://buildd.debian.org/status/architecture.php?a=kfreebsd-amd64&suite=sid

Quoted 1 line
    kfreebsd-i386:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=kfreebsd-i386&ver=1.2.1-2&stamp=1395842518

Quoted 1 line

sparc:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=sparc&ver=1.2.1-2&stamp=1395859546

Quoted 2 lines

buildd page:

https://buildd.debian.org/status/package.php?p=babeltrace

Quoted 10 lines

Cheers,

--
Jon

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

--
Jérémie Galarneau
EfficiOS Inc.
http://www.efficios.com

Quoted 5 lines

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

A message that is not in the archive (sent off-list or not yet received): CAEzWAeLEvv8FbHfwq8=sMeDW49DjO_gKoZ99z1Ff9s+dR6GuJQ@mail.gmail.com

#5 lore raw
Quoted 1 line · Brosseau, Yannick

Jérémie, the end of the build.log seem to contains the tests output.

If I can provide additional info that isn't in the logs, or run specialized tests, etc, just let me know.

Signature · 1 line
Jon
Quoted 52 lines
Quoted 49 lines · Jérémie Galarneau

Hi Jon,

Would it be possible to run the tests on these machines and provide the TAP output?

Thanks,
Jérémie

Quoted 5 lines · Jon Bernard

I noticed babeltrace is having test failures on armel, hurd-i386, kfreebsd-{amd64,i386}, and sparc. These were passing in version 1.1 and the failures appeared in 1.2. Here are links to the build logs:

armel:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=armel&ver=1.2.1-2&stamp=1395846913

Quoted 1 line
    hurd-i386:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=hurd-i386&ver=1.2.1-2&stamp=1398158394

Quoted 1 line
    kfreebsd-amd64:

https://buildd.debian.org/status/architecture.php?a=kfreebsd-amd64&suite=sid

Quoted 1 line
    kfreebsd-i386:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=kfreebsd-i386&ver=1.2.1-2&stamp=1395842518

Quoted 1 line

sparc:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=sparc&ver=1.2.1-2&stamp=1395859546

Quoted 2 lines

buildd page:

https://buildd.debian.org/status/package.php?p=babeltrace

Quoted 10 lines

Cheers,

--
Jon

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

--
Jérémie Galarneau
EfficiOS Inc.
http://www.efficios.com

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

A message that is not in the archive (sent off-list or not yet received): 20140609145201.GB25766@helmut

#7 lore raw

The build logs only seem to contain the "prove" output (invoked by run.sh).

Could you run tests/bin/test_trace_read and tests/lib/test_seek_empty_packet on armel and sparc? The output of both of these tests and the coredumps (if any) should give me something to work with.

Thanks,
Jérémie

Quoted 61 lines · Jon Bernard
Quoted 1 line · Brosseau, Yannick

Jérémie, the end of the build.log seem to contains the tests output.

If I can provide additional info that isn't in the logs, or run specialized tests, etc, just let me know.

--
Jon

Quoted 52 lines
Quoted 49 lines · Jérémie Galarneau

Hi Jon,

Would it be possible to run the tests on these machines and provide the TAP output?

Thanks,
Jérémie

Quoted 5 lines · Jon Bernard

I noticed babeltrace is having test failures on armel, hurd-i386, kfreebsd-{amd64,i386}, and sparc. These were passing in version 1.1 and the failures appeared in 1.2. Here are links to the build logs:

armel:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=armel&ver=1.2.1-2&stamp=1395846913

Quoted 1 line
    hurd-i386:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=hurd-i386&ver=1.2.1-2&stamp=1398158394

Quoted 1 line
    kfreebsd-amd64:

https://buildd.debian.org/status/architecture.php?a=kfreebsd-amd64&suite=sid

Quoted 1 line
    kfreebsd-i386:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=kfreebsd-i386&ver=1.2.1-2&stamp=1395842518

Quoted 1 line

sparc:

https://buildd.debian.org/status/fetch.php?pkg=babeltrace&arch=sparc&ver=1.2.1-2&stamp=1395859546

Quoted 2 lines

buildd page:

https://buildd.debian.org/status/package.php?p=babeltrace

Quoted 10 lines

Cheers,

--
Jon

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

--
Jérémie Galarneau
EfficiOS Inc.
http://www.efficios.com

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

Signature · 8 lines
Jérémie Galarneau
EfficiOS Inc.
http://www.efficios.com

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

A message that is not in the archive (sent off-list or not yet received): CA+jJMxsuEtk3QdDz1_a0MOBq0To7wB4+QnOW3Rkp0uJZ1eM1gw@mail.gmail.com

#9 lore raw
Quoted 6 lines · Jérémie Galarneau

The build logs only seem to contain the "prove" output (invoked by run.sh).

Could you run tests/bin/test_trace_read and tests/lib/test_seek_empty_packet on armel and sparc? The output of both of these tests and the coredumps (if any) should give me something to work with.

Here are the sparc results:

    $ ./tests/bin/test_trace_read
    1..17
    ./tests/bin/test_trace_read: line 34: 31489 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 1 - Run babeltrace with trace env-warning
    #   Failed test 'Run babeltrace with trace env-warning'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31508 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 2 - Run babeltrace with trace lttng-modules-2.0-pre5
    #   Failed test 'Run babeltrace with trace lttng-modules-2.0-pre5'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31527 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 3 - Run babeltrace with trace sequence
    #   Failed test 'Run babeltrace with trace sequence'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31546 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 4 - Run babeltrace with trace smalltrace
    #   Failed test 'Run babeltrace with trace smalltrace'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31565 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 5 - Run babeltrace with trace succeed1
    #   Failed test 'Run babeltrace with trace succeed1'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31584 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 6 - Run babeltrace with trace succeed2
    #   Failed test 'Run babeltrace with trace succeed2'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31603 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 7 - Run babeltrace with trace succeed3
    #   Failed test 'Run babeltrace with trace succeed3'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31622 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 8 - Run babeltrace with trace succeed4
    #   Failed test 'Run babeltrace with trace succeed4'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31641 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 9 - Run babeltrace with trace test
    #   Failed test 'Run babeltrace with trace test'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31660 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 10 - Run babeltrace with trace warnings
    #   Failed test 'Run babeltrace with trace warnings'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 34: 31679 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    not ok 11 - Run babeltrace with trace wk-heartbeat-u
    #   Failed test 'Run babeltrace with trace wk-heartbeat-u'
    #   in ./tests/bin/test_trace_read:main() at line 37.
    ./tests/bin/test_trace_read: line 40: 31698 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    ok 12 - Run babeltrace with invalid trace fail1
    ./tests/bin/test_trace_read: line 40: 31713 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    ok 13 - Run babeltrace with invalid trace fail2
    ./tests/bin/test_trace_read: line 40: 31728 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    ok 14 - Run babeltrace with invalid trace integer-range
    ok 15 - Run babeltrace with invalid trace lttng-modules-2.0-pre1
    ok 16 - Run babeltrace with invalid trace packet-based-metadata
    ./tests/bin/test_trace_read: line 40: 31773 Bus error               $BABELTRACE_BIN ${path} > /dev/null 2>&1
    ok 17 - Run babeltrace with invalid trace smalltrace
    # Looks like you failed 11 tests of 17.
    $ ./tests/lib/test_seek_empty_packet
    1..29
    Bus error

I'll gather the arm results tomorrow.

Signature · 1 line
Jon
#10 lore raw
In reply to a message not in the archive
Quoted 6 lines · Jérémie Galarneau

The build logs only seem to contain the "prove" output (invoked by run.sh).

Could you run tests/bin/test_trace_read and tests/lib/test_seek_empty_packet on armel and sparc? The output of both of these tests and the coredumps (if any) should give me something to work with.

With the attached patch applied for kfreebsd, I see these failures:

make[3]: Entering directory '/home/jbernard/babeltrace/tests' ./run.sh tests
bin/test_trace_read ........... ok lib/test_bitfield ............. ok lib/test_seek_empty_packet .... ok lib/test_seek_big_trace ....... ok lib/test_ctf_writer_complete ..
Dubious, test returned 139 (wstat 35584, 0x8b00) All 77 subtests passed

Test Summary Report
-------------------
lib/test_ctf_writer_complete (Wstat: 35584 Tests: 77 Failed: 0) Non-zero exit status: 139
Parse errors: No plan found in TAP output
Files=5, Tests=178, 26 wallclock secs ( 0.00 usr + 0.00 sys = 0.00 CPU)
Result: FAIL
Makefile:662: recipe for target 'check-am' failed

I'll submit a proper patch for HOST_NAME_MAX to the list.

Cheers,

Signature · 1 line
Jon
configure.ac +11 -0
index 80eba33..248044e 100644
@@ -50,6 +50,17 @@ AC_C_INLINE
 AC_TYPE_PID_T
 AC_TYPE_SIZE_T
 
+# FreeBSD does not define HOST_NAME_MAX
+AC_CHECK_DECL(HOST_NAME_MAX,,
+	[
+	 AC_CHECK_DECL(_POSIX_HOST_NAME_MAX,
+		       AC_DEFINE(HOST_NAME_MAX, _POSIX_HOST_NAME_MAX, []),
+		       AC_DEFINE(HOST_NAME_MAX, 256, [Define to 256 if neither have HOST_NAME_MAX nor _POSIX_HOST_NAME_MAX]),
+		       [[#include <limits.h>]])
+	],
+	[[#include <limits.h>]]
+)
+
 # Checks for library functions.
 AC_FUNC_MALLOC
 AC_FUNC_MMAP

_______________________________________________
lttng-dev mailing list
[email protected]
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

A message that is not in the archive (sent off-list or not yet received): 20140613035117.GA8741@helmut

#12 lore raw
Quoted 11 lines · Jon Bernard
Quoted 6 lines · Jérémie Galarneau

The build logs only seem to contain the "prove" output (invoked by run.sh).

Could you run tests/bin/test_trace_read and tests/lib/test_seek_empty_packet on armel and sparc? The output of both of these tests and the coredumps (if any) should give me something to work with.

[...]

I'll gather the arm results tomorrow.

Well, the armel box to which I have access shows all tests passing :) I'll have to query the buildd guys for more information and I'll let you know what I find.

Signature · 1 line
Jon
#13 lore raw
In reply to a message not in the archive
Quoted 6 lines · Jérémie Galarneau

The build logs only seem to contain the "prove" output (invoked by run.sh).

Could you run tests/bin/test_trace_read and tests/lib/test_seek_empty_packet on armel and sparc? The output of both of these tests and the coredumps (if any) should give me something to work with.

And finally, the hurd build errors (which are clearly the most important :)) are due to missing PATH_MAX. I'll send a patch for that to the list as well.

Signature · 1 line
Jon
#14 lore raw
In reply to a message not in the archive
Quoted 6 lines · Jérémie Galarneau

The build logs only seem to contain the "prove" output (invoked by run.sh).

Could you run tests/bin/test_trace_read and tests/lib/test_seek_empty_packet on armel and sparc? The output of both of these tests and the coredumps (if any) should give me something to work with.

I had a short moment to run gdb on one of the offending tests on sparc, here is the output:

    (gdb) run tests/ctf-traces/succeed/sequence
    Starting program: /home/jbernard/babeltrace/converter/.libs/lt-babeltrace tests/ctf-traces/succeed/sequence
    [Thread debugging using libthread_db enabled]
    Using host libthread_db library "/lib/sparc-linux-gnu/libthread_db.so.1".

    Program received signal SIGBUS, Bus error.
    0xf7f78af4 in make_node (scanner=0x7001cfe0, type=NODE_UNARY_EXPRESSION) at ctf-parser.y:373
    373             node->lineno = yyget_lineno(scanner->scanner);
    (gdb) bt
    #0  0xf7f78af4 in make_node (scanner=0x7001cfe0, type=NODE_UNARY_EXPRESSION) at ctf-parser.y:373
    #1  0xf7f7a708 in yyparse (scanner=0x7001cfe0, yyscanner=0x7001c9c8) at ctf-parser.y:1202
    #2  0xf7f6a6a4 in ctf_trace_metadata_read (td=0x7001d600, metadata_fp=<optimized out>, scanner=0x7001cfe0, append=<optimized out>) at ctf.c:1281
    #3  0xf7f6f470 in ctf_open_trace_read (metadata_fp=0x0, packet_seek=0xf7f6cc60 <ctf_packet_seek>, flags=0, path=0x70026630 "tests/ctf-traces/succeed/sequence",
        td=0x7001d600) at ctf.c:2111
    #4  ctf_open_trace (path=0x70026630 "tests/ctf-traces/succeed/sequence", flags=0, packet_seek=0xf7f6cc60 <ctf_packet_seek>, metadata_fp=<optimized out>)
        at ctf.c:2206
    #5  0xf7fafba0 in bt_context_add_trace (ctx=0x7001b950, path=0x70026630 "tests/ctf-traces/succeed/sequence", format_name=0x7001b990 "ctf", packet_seek=0x0,
        stream_list=0x0, metadata=0x0) at context.c:95
    #6  0x70003468 in bt_context_add_traces_recursive (ctx=0x7001b950, path=0x7001b968 "tests/ctf-traces/succeed/sequence", format_str=0x7001b990 "ctf",
        packet_seek=0x0) at babeltrace.c:529
    #7  0x70001ae8 in main (argc=<optimized out>, argv=<optimized out>) at babeltrace.c:738

The scanner pointer and ctf_scanner struct appear to be okay, the bus error appears to occur on scanner->scanner, which is of type yyscan_t.

I don't have time to look further at the moment, but I'll try next chance I get. Hope this helps.

Cheers,

Signature · 1 line
Jon