- Boost - lists.boost.org

Boost regression notification (2006-12-05 [RC_1_34_0])
by Douglas Gregor 05 Dec '06

05 Dec '06

Boost Regression test failures Report time: 2006-12-05T13:59:11Z This report lists all regression test failures on release platforms. Detailed report: http://engineering.meta-comm.com/boost-regression/CVS-RC_1_34_0/developer/i… 26 failures in 9 libraries concept_check (8) iostreams (1) math (5) serialization (3) signals (4) test (2) thread (1) utility (1) xpressive (1) |concept_check| class_concept_fail_expected: borland-5.6.4 borland-5.8.2 msvc-6.5 msvc-7.0 qcc-3.3.5_cpp qcc-3.3.5_gpp concept_check_fail_expected: qcc-3.3.5_cpp qcc-3.3.5_gpp |iostreams| finite_state_filter_test: cw-9.4 |math| common_factor_test: borland-5.8.2 msvc-6.5 msvc-7.0 qcc-3.3.5_cpp qcc-3.3.5_gpp |serialization| test_no_rtti_xml_archive_dll: qcc-3.3.5_gpp test_reset_object_address: msvc-7.0 test_reset_object_address_dll: msvc-7.0 |signals| deletion_test: msvc-6.5 ordering_test: msvc-6.5 signal_n_test: msvc-6.5 trackable_test: msvc-6.5 |test| prg_exec_fail2: darwin-4.0.1 prg_exec_fail3: cw-9.4 |thread| test_mutex_lib: gcc-3.3.6 |utility| operators_test: gcc-3.4.5_linux_x86_64 |xpressive| test3u: qcc-3.3.5_cpp

1 0

[serialization] [msvc-7.0] test_reset_object_address
by Vladimir Prus 05 Dec '06

05 Dec '06

The test fails with: test_reset_object_address.cpp ..\boost\archive\detail\oserializer.hpp(533) : fatal error C1204: compiler limit : internal structure overflow See: http://tinyurl.com/wrtu2 Does this sound to any MSVC guru as an easily workaroundable issue? Note that this is regression -- the test appears to have worked in previous version of the library. But given that 7.0 is some transitional version, there's no point spending *too* much cycles in this. Unless a workaround is found by the end of this week, I'll mark this failure as expected. Thanks, Volodya

2 1

[signals] msvc-6.5 failures
by Vladimir Prus 05 Dec '06

05 Dec '06

There are four failures of the signals library on msvc-6.5, like this http://tinyurl.com/tc4et All are access violations on runtime. Doug, do you think you'll have the time next week to look into this? Anybody else is willing to debug the problem? If not, I don't think it makes much sense to hold a release because of problems with many-years-old compiler, so I'll just mark the failures as expected. Thanks, Volodya

8 13

[thread] Some thoughts about "atomic"
by Roland Schwarz 05 Dec '06

05 Dec '06

After having tried to learn about atomic from reading through several articles, processor specs and mailing lists, I am more confused than before. I would be glad if we could (re)start a discussion about the topic. Perhaps I am not the only one to benefit from this. Following are some things I learned, but this might be wrong, and I would appreciate clarification. Also some questions: 1) atomicity (in this specialized context) is about optimizing the pattern: enter_critical_section; do_something; leave_critical_section; by making use of processor/platform specific means. In particular in presence of multiple processors. I.e. an atomic lib is primarily about performance. 2) atomicity better would be addressed by the compiler, given a suitable memory model, than as a library. 3) Despite 2) it would be possible to write a library, but it will be hard to get processor independent semantics. E.g. there is one concept of read/write/full memory barriers or another of acquire/release semantics for SMP. 4) Does there exist a canonical set of atomic primitives, from which others can be built? E.g. given atomic_test_and_set, atomic_load and atomic_store is it possible to emulate atomic_compare_and_swap? (without use of system mutices of course.) 5) Is it worth the effort to create a library with processor independent semantics, at the price of not being optimal? E.g. by doing away with the various kinds of barriers, instead simply requiring atomicity and full memory barrier semantics for the operation? Which operations, besides load and store would be essential? Sorry if this is not the perfect list to discuss the topic, but I think boost could possibly benefit from such a library, as previous discussions let me believe. Regards, Roland

5 13

[serialization] test_variant_xml_archive
by Vladimir Prus 05 Dec '06

05 Dec '06

Robert, for all appearances, this test failure: http://tinyurl.com/y5gv7s is the problem with the test itself, not with the library. The error message is: Error E2304 ..\libs\serialization\test\test_variant.cpp 171: Constant variable '_Stz' must be initialized The problem is that I can't find that many lines in the file in question. Do you have an idea what's this variable is and how to fix it? Thanks, Volodya

3 6

Geometry/Vec lib swizzle syntax prototype
by Michael Fawcett 05 Dec '06

05 Dec '06

I have uploaded a small prototype that demonstrates the syntax mentioned in the thread below. It is in the Vault under Math - Geometry as swizzle_demo.zip. http://article.gmane.org/gmane.comp.lib.boost.devel/149076/match=geometry+v… Here's a small example showing its use: float4 b(1, 2, 3, 4); float4 a = b.xxxx(); // a is now 1, 1, 1, 1 a = b.wwwz(); // a is now 4, 4, 4, 3 a = b.yzzw(); // a is now 2, 3, 3, 4 a = b.wzyx(); // a is now 4, 3, 2, 1 float4 c(10, 20, 30, 40); c.yz() = b.zy(); // c is now 10, 3, 2, 40 This mimics Cg, HLSL, and GLSL syntax closely. The swizzle syntax is implemented as function calls so that it doesn't increase the size of the vector struct (important for using those structs directly for graphics API calls). The prototype code does not implement arithmetic operators for simplicity's sake, but my local copy has it implemented. This isn't meant as a full library, only a demonstration showing that the swizzle syntax can be achieved rather simply using the Boost PP and TypeTraits libraries. One item that could be changed is the multiple template parameters to the classes. In the discussion that took place during the review of Andy Little's library the prospect of heterogeneous vectors took place. In my opinion heterogeneous vectors work with little expense to the library writer and transparently to the user. It would also "just work" with Andy's Quan library, or the other Units library being developed. This is pretty off-topic for the syntax that I am demonstrating, I just wanted to mention that I hadn't removed it from this demo. It is in no way necessary, but removal would require the macros to be changed a little bit. I have only compiled using VS 7.1 and 8.0. I am by no means a PP or template guru, so I would welcome your thoughts and critiques. --Michael Fawcett

3 6

Boost regression notification (2006-12-04 [RC_1_34_0])
by Douglas Gregor 05 Dec '06

05 Dec '06

Boost Regression test failures Report time: 2006-12-04T13:08:52Z This report lists all regression test failures on release platforms. Detailed report: http://engineering.meta-comm.com/boost-regression/CVS-RC_1_34_0/developer/i… 31 failures in 9 libraries concept_check (8) iostreams (1) math (6) random (1) serialization (7) signals (4) test (2) utility (1) xpressive (1) |concept_check| class_concept_fail_expected: borland-5.6.4 borland-5.8.2 msvc-6.5 msvc-7.0 qcc-3.3.5_cpp qcc-3.3.5_gpp concept_check_fail_expected: qcc-3.3.5_cpp qcc-3.3.5_gpp |iostreams| finite_state_filter_test: cw-9.4 |math| common_factor_test: borland-5.8.2 cw-9.4 msvc-6.5 msvc-7.0 qcc-3.3.5_cpp qcc-3.3.5_gpp |random| random_test: intel-linux-9.0 |serialization| test_no_rtti_xml_archive_dll: qcc-3.3.5_gpp test_reset_object_address: msvc-7.0 test_reset_object_address_dll: msvc-7.0 test_variant_xml_archive: borland-5.8.2 test_variant_xml_archive_dll: borland-5.8.2 test_variant_xml_warchive: borland-5.8.2 test_variant_xml_warchive_dll: borland-5.8.2 |signals| deletion_test: msvc-6.5 ordering_test: msvc-6.5 signal_n_test: msvc-6.5 trackable_test: msvc-6.5 |test| prg_exec_fail2: darwin-4.0.1 prg_exec_fail3: cw-9.4 |utility| operators_test: gcc-3.4.5_linux_x86_64 |xpressive| test3u: qcc-3.3.5_cpp

3 2

guid library
by Andy 04 Dec '06

04 Dec '06

I have uploaded a new version of the guid library to the boost vault (guid_v5.zip). There are only minor bug fixes compared with guid_v4.zip (which no longer exists). The biggest change is that it now compiles cleanly with gcc as well as msvc. There are now 2 guid files in the vault (guid_v3.zip and guid_v5.zip). What do most people do in this situation? Keep both around, or remove the older one? What would people like me to do? Thanks, Andy.

1 0

Re: [boost] Numerical Quadrature Library
by Sohail Somani 04 Dec '06

04 Dec '06

> -----Original Message----- > From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Franck Stauffer > > Hello, > > I have been using boost for quite a while, and I was wondering if > there is any interest (or ongoing development) in a numerical > quadrature library. I have seen some old discussions on the mailing > lists archive, but couldn't tell wether something was still in the > works. Hi, Even if its not an official boost review/library, I think posting this kind of code is useful because a lot of us solve these problems over and over again. Your library sounds interesting, albeit a lot like mine, so please post! (Can't post mine b/c its not... Mine... Weird eh) Sohail

1 0

Re: [boost] Yet another bignum type
by Oliver Kullmann 04 Dec '06

04 Dec '06

On Fri, Dec 01, 2006 at 10:16:11AM -0000, Martin Bonner wrote: > > One question I have though. > > There is no "the set of integers" etc. > > What is Z if it is not "the set of integers"? > > Given the precision of the rest of your reply I am pretty sure I have > misunderstood something here, but what? > Ask a mathematician to *define* Z: - Some (especially in these days) will not understand what you are talking about (I imagine). - Most will answer that *up to isomorphism* there is exactly one structure (Z, +, *, 0, 1) fulfilling some axioms --- this amounts to the situation in generic programming, where we have a concept of integers and associated operations, and we write generic algorithms using this concept (like generic graph algorithms using the Boost graph library concepts (however the Z-situation is a bit different since all models are isomorphic here, which is not so for graphs)). But in our situation this is not enough --- we need a model. - Finally, there might be some enlightened ones left, which are actually able to construct Z. The standard way (most appropriate for general mathematics in my opinion) is based on set theory, and would first construct natural numbers as the set 0 := empty set and n+1 := set of numbers 0, ..., n. Z then would be constructed by the ordinary quotient construction (so an element of Z is an equivalence class of pairs of natural numbers; this is the same construction as to get the rational numbers from the integers, only this time addition is used instead of multiplication). So as a generic concept there might be many implementations of Z, but this doesn't matter. This covers the situation regarding the structure of Z as a commutative ring. However, when going into the implementation, that is, really defining the *set* Z (and the operations) there are many choices. For a standard implementation there is a good chance for getting the above one. However there are other choices; an extreme case (which is useful under circumstances) starts with "the" real numbers and extracts the integers. Another approach, more down-to-earth this time, would avoid the equivalence classes, and would give a concrete representation. Finally there are (in special circles) alternative set theories and alternative foundations. The main argument about the big-number-library is (as far as I understand it) about the concepts involved. And these ambiguities (what are "really" the integers "for us"), which are much less critical for (ordinary) mathematics, are nevertheless similar to the "leak" of abstraction in (ordinary) mathematics: An integer is typically more an algebraic thing, and less ambiguous, but once you make a natural number out of it, then it suddenly is a cardinal numbers (for counting things) and an ordinal number (for indexing things), which involve both other (different) abstractions. Still on the algebraic side, some handwaving goes on with modular arithmetic. Hope that helped a bit (and did not create more confusion ;-)). Oliver

1 0