Boost regression notification (2005-10-06 [RC_1_33_0])
Boost regression test failures ------------------------------ Report time: 2005-10-06T15:24:21Z This report lists all regression test failures on release platforms. Detailed report: http://engineering.meta-comm.com/boost-regression/CVS-RC_1_33_0/developer/is... 228 failures in 7 libraries: foreach (1) graph (2) iostreams (13) multi_index (1) range (3) serialization (202) wave (6) |foreach| rvalue_const: vc-7_1 |graph| graphviz_test: cw-8_3 cw-9_4 |iostreams| gzip_test: cw-9_5-darwin gcc-2.95.3-stlport-4.6.2-linux gcc-3.2.3-linux gcc-3.3.6-linux gcc-3.4.4-linux gcc-3_3-darwin gcc-3_4_3-sunos gcc-4.0.1-linux gcc-4_0-darwin intel-8.1-linux intel-9.0-linux mingw-3_4_2 vc-7_1 |multi_index| test_serialization: cw-9_5-darwin |range| array: vc-6_5 vc-6_5-stlport vc-7_0 |serialization| test_binary_text_archive: vc-7_1 test_binary_text_archive_dll: vc-7_1 test_class_info_save_text_archive: vc-7_1 test_class_info_save_text_archive_dll: vc-7_1 test_delete_pointer_binary_archive: cw-9_5-darwin test_delete_pointer_binary_archive_dll: cw-9_5-darwin test_delete_pointer_text_archive: cw-9_5-darwin test_delete_pointer_text_warchive: cw-9_5-darwin test_delete_pointer_xml_archive: cw-9_5-darwin test_delete_pointer_xml_warchive: cw-9_5-darwin test_demo_polymorphic: cw-9_5-darwin test_demo_xml_load: gcc-3_3-darwin gcc-4_0-darwin intel-win32-8_1 test_demo_xml_load_dll: gcc-3_3-darwin gcc-4_0-darwin test_deque_binary_archive: cw-9_5-darwin test_deque_binary_archive_dll: cw-9_5-darwin test_deque_text_archive: cw-9_5-darwin test_deque_text_warchive: cw-9_5-darwin test_deque_xml_archive: cw-9_5-darwin test_deque_xml_warchive: cw-9_5-darwin test_derived_text_archive: vc-7_1 test_derived_text_archive_dll: vc-7_1 test_exported_binary_archive: gcc-3.2.3-linux gcc-4.0.1-linux intel-9.0-linux vc-6_5 test_exported_binary_archive_dll: gcc-4.0.1-linux vc-6_5 test_exported_text_archive: gcc-3.2.3-linux gcc-4.0.1-linux vc-6_5 vc-7_1 test_exported_text_archive_dll: gcc-3.2.3-linux gcc-4.0.1-linux vc-6_5 vc-7_1 test_exported_text_warchive: gcc-4.0.1-linux vc-6_5 test_exported_text_warchive_dll: gcc-4.0.1-linux vc-6_5 test_exported_xml_archive: gcc-4.0.1-linux intel-8.1-linux vc-6_5 test_exported_xml_archive_dll: gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux vc-6_5 test_exported_xml_warchive: gcc-3.2.3-linux gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux vc-6_5 test_exported_xml_warchive_dll: gcc-3.2.3-linux gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux vc-6_5 test_list_binary_archive: cw-9_5-darwin test_list_binary_archive_dll: cw-9_5-darwin test_list_ptrs_binary_archive: cw-9_5-darwin test_list_ptrs_binary_archive_dll: cw-9_5-darwin test_list_ptrs_text_archive: cw-9_5-darwin test_list_ptrs_text_warchive: cw-9_5-darwin test_list_ptrs_xml_archive: cw-9_5-darwin test_list_ptrs_xml_warchive: cw-9_5-darwin test_list_ptrs_xml_warchive_dll: gcc-3_4_3-sunos test_list_text_archive: cw-9_5-darwin test_list_text_warchive: cw-9_5-darwin test_list_xml_archive: cw-9_5-darwin test_list_xml_warchive: cw-9_5-darwin test_map_binary_archive: cw-9_5-darwin test_map_binary_archive_dll: cw-9_5-darwin test_map_text_archive: cw-9_5-darwin gcc-3_4_3-sunos test_map_text_warchive: cw-9_5-darwin test_map_xml_archive: cw-9_5-darwin test_map_xml_warchive: cw-9_5-darwin test_mi_text_archive: vc-7_1 test_mi_text_archive_dll: vc-7_1 test_no_rtti_binary_archive: vc-6_5 test_no_rtti_binary_archive_dll: vc-6_5 test_no_rtti_text_archive: vc-6_5 test_no_rtti_text_archive_dll: vc-6_5 test_no_rtti_text_warchive: vc-6_5 test_no_rtti_text_warchive_dll: vc-6_5 test_no_rtti_xml_archive: vc-6_5 test_no_rtti_xml_archive_dll: vc-6_5 test_no_rtti_xml_warchive: vc-6_5 test_no_rtti_xml_warchive_dll: vc-6_5 test_non_default_ctor2_text_archive: vc-7_1 test_non_default_ctor2_text_archive_dll: vc-7_1 test_non_default_ctor_text_archive: vc-7_1 test_non_default_ctor_text_archive_dll: vc-7_1 test_non_intrusive_text_archive: vc-7_1 test_non_intrusive_text_archive_dll: vc-7_1 test_null_ptr_text_archive: vc-7_1 test_null_ptr_text_archive_dll: vc-7_1 test_object_text_archive: vc-7_1 test_object_text_archive_dll: vc-7_1 test_polymorphic_binary_archive: cw-9_5-darwin test_polymorphic_binary_archive_dll: cw-9_5-darwin test_polymorphic_text_archive: cw-9_5-darwin test_polymorphic_text_warchive: cw-9_5-darwin test_polymorphic_xml_archive: cw-9_5-darwin test_polymorphic_xml_warchive: cw-9_5-darwin test_primitive_text_archive: vc-7_1 test_primitive_text_archive_dll: vc-7_1 test_private_ctor: cw-9_5-darwin test_registered_text_archive: vc-7_1 test_registered_text_archive_dll: vc-7_1 test_reset_object_address: cw-9_5-darwin test_set_binary_archive: cw-9_5-darwin test_set_binary_archive_dll: cw-9_5-darwin test_set_text_archive: cw-9_5-darwin test_set_text_warchive: cw-9_5-darwin test_set_xml_archive: cw-9_5-darwin test_set_xml_warchive: cw-9_5-darwin test_shared_ptr_132_binary_archive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_binary_archive_dll: vc-6_5 test_shared_ptr_132_text_archive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_text_archive_dll: gcc-3_4_3-sunos vc-6_5 test_shared_ptr_132_text_warchive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_text_warchive_dll: vc-6_5 test_shared_ptr_132_xml_archive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_xml_archive_dll: vc-6_5 test_shared_ptr_132_xml_warchive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_xml_warchive_dll: vc-6_5 test_shared_ptr_binary_archive: borland-5_6_4 vc-6_5 test_shared_ptr_binary_archive_dll: borland-5_6_4 vc-6_5 test_shared_ptr_text_archive: borland-5_6_4 vc-6_5 test_shared_ptr_text_archive_dll: borland-5_6_4 vc-6_5 test_shared_ptr_text_warchive: borland-5_6_4 vc-6_5 test_shared_ptr_text_warchive_dll: borland-5_6_4 vc-6_5 test_shared_ptr_xml_archive: borland-5_6_4 vc-6_5 test_shared_ptr_xml_archive_dll: borland-5_6_4 vc-6_5 test_shared_ptr_xml_warchive: borland-5_6_4 vc-6_5 test_shared_ptr_xml_warchive_dll: borland-5_6_4 vc-6_5 test_split_text_archive: vc-7_1 test_split_text_archive_dll: vc-7_1 test_split_text_warchive: gcc-3_4_3-sunos test_tracking_text_archive: vc-7_1 test_tracking_text_archive_dll: vc-7_1 test_unregistered_binary_archive: gcc-3.2.3-linux gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux test_unregistered_binary_archive_dll: gcc-3.2.3-linux gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux test_unregistered_text_archive: gcc-3.2.3-linux gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux vc-7_1 test_unregistered_text_archive_dll: gcc-3.2.3-linux gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux vc-7_1 test_unregistered_xml_archive: gcc-3.2.3-linux gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux test_unregistered_xml_archive_dll: gcc-3.2.3-linux gcc-4.0.1-linux intel-8.1-linux intel-9.0-linux test_unregistered_xml_warchive: gcc-4.0.1-linux test_unregistered_xml_warchive_dll: gcc-4.0.1-linux test_variant_binary_archive: cw-9_5-darwin test_variant_binary_archive_dll: cw-9_5-darwin test_variant_text_archive: cw-9_5-darwin test_variant_text_warchive: cw-9_5-darwin test_variant_xml_archive: cw-9_5-darwin test_variant_xml_warchive: cw-9_5-darwin test_vector_binary_archive: cw-9_5-darwin test_vector_binary_archive_dll: cw-9_5-darwin test_vector_text_archive: cw-9_5-darwin test_vector_text_warchive: cw-9_5-darwin test_vector_xml_archive: cw-9_5-darwin test_vector_xml_warchive: cw-9_5-darwin test_void_cast: gcc-3_4_3-sunos test_void_cast_dll: gcc-3_4_3-sunos |wave| test_re2c_lexer: cw-9_4 test_slex_lexer: cw-9_4 testwave: cw-9_4 gcc-3.4.4-linux gcc-3_4_3-sunos gcc-4.0.1-linux
This report lists all regression test failures on release platforms.
We really need to get 1.33.1 out the door, but this list of regressions seems to be getting worse not better. Is there anything that can be done to improve things? Any library authors stuck for ideas and need assistance? John.
John Maddock wrote:
This report lists all regression test failures on release platforms.
We really need to get 1.33.1 out the door, but this list of regressions seems to be getting worse not better. Is there anything that can be done to improve things? Any library authors stuck for ideas and need assistance?
I have not changed anything lately in the Wave library what could have caused the regressions.
|wave| test_re2c_lexer: cw-9_4 test_slex_lexer: cw-9_4 testwave: cw-9_4 gcc-3.4.4-linux gcc-3_4_3-sunos gcc-4.0.1-linux
The cw-9_4 errors are compiler problems, which sometimes seem to occur, sometimes not. I've already contacted Stefan Slapeta, but he wasn't able to figure out what's going on there. The gcc-3_4_3-sunos problem seems to be a test system problem on the target machine (look at the logs here: http://tinyurl.com/d9ees, there are bogus entries consisting out of lots of '?'...). The gcc-3.4.4-linux and gcc-4.0.1-linux errors are new to me and I have to investigate it further. Regards Hartmut
On 10/6/05, Hartmut Kaiser <hartmut.kaiser@gmail.com> wrote:
John Maddock wrote:
This report lists all regression test failures on release platforms.
We really need to get 1.33.1 out the door, but this list of regressions seems to be getting worse not better. Is there anything that can be done to improve things? Any library authors stuck for ideas and need assistance?
I have not changed anything lately in the Wave library what could have caused the regressions.
|wave| test_re2c_lexer: cw-9_4 test_slex_lexer: cw-9_4 testwave: cw-9_4 gcc-3.4.4-linux gcc-3_4_3-sunos gcc-4.0.1-linux
The cw-9_4 errors are compiler problems, which sometimes seem to occur, sometimes not. I've already contacted Stefan Slapeta, but he wasn't able to figure out what's going on there.
The gcc-3_4_3-sunos problem seems to be a test system problem on the target machine (look at the logs here: http://tinyurl.com/d9ees, there are bogus entries consisting out of lots of '?'...).
Yeah, and they look stale. I see a similar failure in the Serialization library as well. I think I'm running up against a full disk now that I look at things. I don't have the power to clean up other folks' messes here, so I may be unable to get a complete test run for some time. I'll do what I can to have my co-workers free up some space. -- Caleb Epstein caleb dot epstein at gmail dot com
I think I'm running up against a full disk now that I look at things. I don't have the power to clean up other folks' messes here, so I may be unable to get a complete test run for some time. I'll do what I can to have my co-workers free up some space.
How about running with -sBUILD="<debug-symbols>off" ? It make a *huge* difference in disk space requirements, especially for gcc. John.
John Maddock wrote:
I think I'm running up against a full disk now that I look at things. I don't have the power to clean up other folks' messes here, so I may be unable to get a complete test run for some time. I'll do what I can to have my co-workers free up some space.
How about running with -sBUILD="<debug-symbols>off" ?
It make a *huge* difference in disk space requirements, especially for gcc.
Now, that's some helpful information! Thanks for sharing it with us! Regards, m Send instant messages to your online friends http://au.messenger.yahoo.com
On 10/7/05, Martin Wille <mw8329@yahoo.com.au> wrote:
How about running with -sBUILD="<debug-symbols>off" ?
It make a *huge* difference in disk space requirements, especially for gcc.
Now, that's some helpful information! Thanks for sharing it with us!
Indeed. I never thought of doing this. Thanks for the heads-up. -- Caleb Epstein caleb dot epstein at gmail dot com
John Maddock wrote:
We really need to get 1.33.1 out the door, but this list of regressions seems to be getting worse not better. Is there anything that can be done to improve things? Any library authors stuck for ideas and need assistance?
(This message basically is a repost of a problem I already had reported on the testing list.) We have an additional problem that does not become visible in the reports Doug sends to this list. For gcc 4.0.2, there are a lot of additional regressions. They look like a compiler problem to me, but I didn't seriously investigate the problem. If someone would be able to track the problems down, be to a bug in Boost or in the compiler, then this would definitely help. (The gcc 4.0.1 toolset is relevant for the release. I think we should not ship Boost 1.33.1 without being able to say what exactly goes wrong with gcc 4.0.2.) The complete list of test results (including the toolsets not considered relevant for the release) can be found at http://tinyurl.com/9s72x [1]. Mainly affected by the new problems are: IOstreams, Iterator and Spirit. (Jonathan already tried to fix IOstreams, but without luck). Regards, m [1] http://engineering.meta-comm.com/boost-regression/CVS-RC_1_33_0/developer/su... Send instant messages to your online friends http://au.messenger.yahoo.com
Martin Wille wrote:
We have an additional problem that does not become visible in the reports Doug sends to this list. For gcc 4.0.2, there are a lot of additional regressions. They look like a compiler problem to me, but I didn't seriously investigate the problem.
Aren't some of those regressions only when _GLIBCXX_DEBUG is defined? It could be that the iterator-related tests are failing due to bugs in the debug mode, but I've not looked into them yet. Is GCC's debug mode a release-critical platform ? http://tinyurl.com/9eqr2 This crc-test is a compiler error, has anyone reported it? If not I'll add it to GCC's bugzilla. This is a reduced version that triggers the error: #include <boost/crc.hpp> // for boost::crc_basic, etc. // Run tests on CRCs below a byte in size (here, 3 bits) void small_crc_test1() { unsigned char const samples[4][4] = { { 0x3A, 0xC4, 0x08, 0x06 }, { 0x42, 0xC5, 0x0A, 0x41 }, { 0x4A, 0xC5, 0x08, 0x22 }, { 0x52, 0xC4, 0x08, 0x05 } }; boost::crc<3, 0x03, 0, 0, false, false>(samples[0], 4); } jon
On Oct 7, 2005, at 7:22 PM, Jonathan Wakely wrote:
Aren't some of those regressions only when _GLIBCXX_DEBUG is defined? It could be that the iterator-related tests are failing due to bugs in the debug mode, but I've not looked into them yet. Is GCC's debug mode a release-critical platform ?
No. Doug
Arkadiy Vertleyb wrote:
"Douglas Gregor" <doug.gregor@gmail.com> wrote
Is GCC's debug mode a release-critical platform ?
No.
How can this be? How useful is a feature that can be used in the release mode oinly? I think its usefullness is pretty close to zero...
I agree with your questioning, Arkadiy. People will want to use the debug mode. It's probably the best thing added to recent libstdc++ distributions. Severals bugs in Boost code were discovered using _GLIBCXX_DEBUG. I'd consider it a big loss not to be able to use it anymore. Regards, m Send instant messages to your online friends http://au.messenger.yahoo.com
Jonathan Wakely wrote:
http://tinyurl.com/9eqr2 This crc-test is a compiler error, has anyone reported it? If not I'll
This is now http://gcc.gnu.org/PR24277 jon
This is now http://gcc.gnu.org/PR24277
And I've posted the is_incrementable problems as http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24275 and a fix has already been scheduled for the next release. John.
John Maddock wrote:
This is now http://gcc.gnu.org/PR24277
And I've posted the is_incrementable problems as http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24275 and a fix has already been scheduled for the next release.
Thanks, John! I probably spent six hours trying to find a workaround, with no success. -- Jonathan Turkanis www.kangaroologic.com
Martin Wille wrote:
If someone would be able to track the problems down, be to a bug in Boost or in the compiler, then this would definitely help. (The gcc 4.0.1 toolset is relevant for the release. I think we should not ship Boost 1.33.1 without being able to say what exactly goes wrong with gcc 4.0.2.)
The complete list of test results (including the toolsets not considered relevant for the release) can be found at http://tinyurl.com/9s72x [1].
Mainly affected by the new problems are: IOstreams, Iterator and Spirit. (Jonathan already tried to fix IOstreams, but without luck).
I fixed the problems inside iostreams, but unfortunately the trick used to implement is_incrementable (in boost/detail) doesn't seem to work on 4.0.2; I'm not yet prepared to say it's a gcc bug, but it may be. I've checked in a change to is_incrementable in the RC_1_33_0 branch, restricted to gcc 4.0.2. We'll see tomorrow if it works.
Regards, m
-- Jonathan Turkanis www.kangaroologic.com
Jonathan Turkanis wrote:
Martin Wille wrote:
Mainly affected by the new problems are: IOstreams, Iterator and Spirit. (Jonathan already tried to fix IOstreams, but without luck).
I fixed the problems inside iostreams, but unfortunately the trick used to implement is_incrementable (in boost/detail) doesn't seem to work on 4.0.2; I'm not yet prepared to say it's a gcc bug, but it may be.
I've checked in a change to is_incrementable in the RC_1_33_0 branch, restricted to gcc 4.0.2. We'll see tomorrow if it works.
It looks like it helped in a handful of cases, but not everywhere. I've got one more trick I'd like to try. -- Jonathan Turkanis www.kangaroologic.com
Jonathan Turkanis wrote:
Jonathan Turkanis wrote:
Martin Wille wrote:
Mainly affected by the new problems are: IOstreams, Iterator and Spirit. (Jonathan already tried to fix IOstreams, but without luck).
I fixed the problems inside iostreams, but unfortunately the trick used to implement is_incrementable (in boost/detail) doesn't seem to work on 4.0.2; I'm not yet prepared to say it's a gcc bug, but it may be.
I've checked in a change to is_incrementable in the RC_1_33_0 branch, restricted to gcc 4.0.2. We'll see tomorrow if it works.
It looks like it helped in a handful of cases, but not everywhere. I've got one more trick I'd like to try.
Okay, I give up. I installed gcc 4.0.2 and tried about five different things, but nothing worked. I've put everything back the way it was. -- Jonathan Turkanis www.kangaroologic.com
participants (2)
-
Arkadiy Vertleyb -
Caleb Epstein -
Douglas Gregor -
Douglas Gregor -
Hartmut Kaiser -
John Maddock -
Jonathan Turkanis -
Jonathan Wakely -
Martin Wille