Fix FTBFS on GCC 14 (boo#1222493) #516

This fixes type errors like this: In file included from /usr/include/lttng/tracepoint-event.h:69, from …-build/include/tracing/librados.h:4143, from …/src/tracing/librados.c:6 : …-build/include/tracing/librados.h: In function ‘lttng_ust__event_probe__librados___rados_mon_command_exit’: …-build/include/tracing/librados.h:477:9: error: initialization of ‘size_t’ {aka ‘long unsigned int’} from ‘size_t *’ {aka ‘long unsigned int *’} makes integer from pointer without a cast 477 | ceph_ctf_integerp(size_t, outslen, outslen) | ^~~~~~~~~~~~~~~~~ GCC 14 will likely treat these type mismatches as an error and fail the build. Signed-off-by: Florian Weimer <[email protected]> (cherry picked from commit f9aea91)

Several Ceph APIs use bool * types, which correspond to libcpp.bool * types in Cython. The bint type has an incorrect size 4 and cannot be used as a replacement. This prevents a compilation failure with future compilers: …-build/src/pybind/rbd/rbd.c: In function ‘__pyx_pf_3rbd_3RBD_104namespace_exists’: …-build/src/pybind/rbd/rbd.c:42165:76: error: passing argument 3 of ‘rbd_namespace_exists’ from incompatible pointer type 42165 | __pyx_v_ret = rbd_namespace_exists(__pyx_v__ioctx, __pyx_v__name, (&__pyx_v__exists)); | ~^~~~~~~~~~~~~~~~~ | | | int * In file included from …-build/src/pybind/rbd/rbd.c:1268: …/src/include/rbd/librbd.h:1496:45: note: expected ‘_Bool *’ but argument is of type ‘int *’ 1496 | bool *exists); | ^ Signed-off-by: Florian Weimer <[email protected]> (cherry picked from commit a49d154)

The following problem has been reported by Kaleb Keithley: ``` /builddir/build/BUILD/ceph-18.2.1/src/osd/osd_types.h: In lambda function: /builddir/build/BUILD/ceph-18.2.1/src/common/dout.h:184:73: error: call to non-‘constexpr’ function ‘virtual unsigned int DoutPrefixProvider::get_subsys() const’ 184 | dout_impl(pdpp->get_cct(), ceph::dout::need_dynamic(pdpp->get_subsys()), v) \ | ~~~~~~~~~~~~~~~~^~ /builddir/build/BUILD/ceph-18.2.1/src/common/dout.h:155:58: note: in definition of macro ‘dout_impl’ 155 | return (cctX->_conf->subsys.template should_gather<sub, v>()); \ | ^~~ /builddir/build/BUILD/ceph-18.2.1/src/osd/osd_types.h:3617:3: note: in expansion of macro ‘ldpp_dout’ 3617 | ldpp_dout(dpp, 10) << "build_prior all_probe " << all_probe << dendl; | ^~~~~~~~~ ``` For details of the problem and the idea behind the fix, please refer to the comment this commit brings to `dout.h`. The minimized replicator that the facilitated Goldbot-based investigation: ```cpp namespace ceph::dout { template<typename T> struct dynamic_marker_t { T value; // constexpr ctor isn't needed as it's an aggregate type constexpr operator T() const { return value; } }; template<typename T> constexpr dynamic_marker_t<T> need_dynamic(T&& t) { return dynamic_marker_t<T>{ std::forward<T>(t) }; } template<typename T> struct is_dynamic : public std::false_type {}; template<typename T> struct is_dynamic<dynamic_marker_t<T>> : public std::true_type {}; } // ceph::dout struct subsys_t { template <unsigned SubV, int LvlV> bool should_gather() const { return true; } bool should_gather(const unsigned sub, int level) const { return false; } }; static subsys_t subsys; do { \ const bool should_gather = [&](const auto cctX) { \ if constexpr (ceph::dout::is_dynamic<decltype(sub)>::value || \ ceph::dout::is_dynamic<decltype(v)>::value) { \ std::cout << "the dynamic path" << std::endl; \ return subsys.should_gather(sub, v); \ } else { \ /* The parentheses are **essential** because commas in angle \ * brackets are NOT ignored on macro expansion! A language's \ * limitation, sorry. */ \ std::cout << "the static path" << std::endl; \ /*return subsys.should_gather(sub, v);*/ \ return (subsys.template should_gather<sub, v>()); \ } \ }(cct); \ } while (0) if (decltype(auto) pdpp = (dpp); pdpp) /* workaround -Wnonnull-compare for 'this' */ \ dout_impl(42, sub, v) if (decltype(auto) pdpp = (dpp); pdpp) /* workaround -Wnonnull-compare for 'this' */ \ dout_impl(42, ceph::dout::need_dynamic(42), v) int main() { std::random_device dev; std::mt19937 rng(dev()); std::uniform_int_distribution<std::mt19937::result_type> dist6(1,6); // distribution in range [1, 6] int sub = dist6(rng); ldpp_dout("mocked out", sub); //ldpp_subdout("mocked out", 4, 3); } ``` Fixes: https://tracker.ceph.com/issues/64050 Signed-off-by: Radoslaw Zarzynski <[email protected]> (cherry picked from commit 0eace4e)

without this change, clang16 fails to compile, due to conversion from "bool" to "int". see error: https://paste.sh/QybPVNdh#OtffZeqJKcsqUVMupPYD11Kl Signed-off-by: yuval Lifshitz <[email protected]> (cherry picked from commit c67fb73)

Without this, the build fails on GCC 14: [ 3456s] /home/abuild/rpmbuild/BUILD/ceph-16.2.15-68-gbb20a17289a/src/rapidjson/include/rapidjson/document.h: In member function ‘rapidjson::GenericStringRef<CharType>& rapidjson::GenericStringRef<CharType>::operator=(const rapidjson::GenericStringRef<CharType>&)’: [ 3456s] /home/abuild/rpmbuild/BUILD/ceph-16.2.15-68-gbb20a17289a/src/rapidjson/include/rapidjson/document.h:319:82: error: assignment of read-only member ‘rapidjson::GenericStringRef<CharType>::length’ [ 3456s] 319 | GenericStringRef& operator=(const GenericStringRef& rhs) { s = rhs.s; length = rhs.length; } [ 3456s] | ~~~~~~~^~~~~~~~~~~~ [ 3456s] make[2]: *** [src/rgw/CMakeFiles/rgw_common.dir/build.make:1774: src/rgw/CMakeFiles/rgw_common.dir/rgw_crypt.cc.o] Error 1 [ 3456s] make[1]: *** [CMakeFiles/Makefile2:8011: src/rgw/CMakeFiles/rgw_common.dir/all] Error 2 The copy assignment operator was disabled in upstream rapidjson a very, very, very long time ago: Tencent/rapidjson@862c39b This method of patching a submodule is pretty heinous, even for me. The alternative would be to fork ceph's (very) old rapidjson to the SUSE org, apply this patch to that fork, and update ceph's .gitmodules file to point to that instead of the ceph one. Signed-off-by: Tim Serong <[email protected]>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix FTBFS on GCC 14 (boo#1222493) #516

Fix FTBFS on GCC 14 (boo#1222493) #516

Commits on Apr 9, 2024