Hey Jean, On 5 October 2016 at 21:21, Jean Guyomarc'h <jean.guyoma...@gmail.com> wrote:
> I think you misunderstood my idea. By "blacklist", I meant what you > implemented there > https://git.enlightenment.org/core/efl.git/commit/?id=119e35 > b79127c7688d8782635fbe1a9642a851bc > : define a "void" EINA_PREFETCH_* when the compiler is known not handle it. > I am happy with your patch. I'll test with osx this evening to check. > Yes. No. I understood you and just didn't write proper English. The idea I hated was to leave the build broken as it is, blame the compiler and wait. Anyway, it's patched up now. :) > > Jean > > On Wed, Oct 5, 2016 at 12:32 PM, Jean-Philippe André <j...@videolan.org> > wrote: > > > Hi, > > > > On 5 October 2016 at 18:07, Jean Guyomarc'h <jean.guyoma...@gmail.com> > > wrote: > > > > > Hi raster, > > > > > > any update on this? Until this is fixed, osx (and clang) builds are > > doomed > > > :/ > > > Maybe we can blacklist compilers known to crash when using > > EINA_PREFETCH_*? > > > It does not feel very elegant, but it would unbreak the build. > > > > > > > I hate this idea, plus the fallback code does not even compile (because { > > list ?: } is not valid code). > > I'll add a #ifdef __clang__. But we still need to report the bug to LLVM > > maintainers. > > > > > > > > Jean > > > > > > On Sun, Oct 2, 2016 at 4:01 PM, Jean Guyomarc'h < > > jean.guyoma...@gmail.com> > > > wrote: > > > > > > > Ok, I haven't reported the bug yet (because it has to go through > > > > Apple's bugtracker), and it's boring... > > > > If you want your gold star, I can let you fill the bug report :-) You > > > > introduced this after all :p > > > > Jean > > > > > > > > > > > > On Sun, Oct 2, 2016 at 3:49 PM, Carsten Haitzler < > ras...@rasterman.com > > > > > > > wrote: > > > > > On Sun, 2 Oct 2016 10:21:01 +0200 "Jean Guyomarc'h" < > > > > jean.guyoma...@gmail.com> > > > > > said: > > > > > > > > > >> Congratulations, you found an llvm bug :D > > > > >> llvm 8.0.0 on osx now segfaults when compiling a file with > eina_list > > > > >> > > > > >> Llvm tells me to report the bug, so I will, but I wonder what we > > > > >> should do to handle this... > > > > >> Jean > > > > > > > > > > yup. same issue. > > > > > > > > > > 2 warnings generated. > > > > > 2 warnings generated. > > > > > 2 warnings generated. > > > > > 2 warnings generated. > > > > > #0 0x00007fc74e7707d8 llvm::sys::PrintStackTrace(llv > m::raw_ostream&) > > > > (/usr/bin/../lib/libLLVM-3.8.so+0x5ef7d8) > > > > > #1 0x00007fc74e76e3c6 llvm::sys::RunSignalHandlers() > > (/usr/bin/../lib/ > > > > libLLVM-3.8.so+0x5ed3c6) > > > > > #2 0x00007fc74e76e5aa (/usr/bin/../lib/libLLVM-3.8.so+0x5ed5aa) > > > > > #3 0x00007fc74d8770b0 __restore_rt (/usr/bin/../lib/libc.so.6+ > > 0x330b0) > > > > > #4 0x00007fc74ec7fba7 llvm::SelectionDAGBuilder::getControlRoot() > > > > (/usr/bin/../lib/libLLVM-3.8.so+0xafeba7) > > > > > #5 0x00007fc74ecc32e3 llvm::SelectionDAGISel:: > > > > SelectBasicBlock(llvm::ilist_iterator<llvm::Instruction const>, > > > > llvm::ilist_iterator<llvm::Instruction const>, bool&) > > (/usr/bin/../lib/ > > > > libLLVM-3.8.so+0xb422e3) > > > > > #6 0x00007fc74ecc6395 llvm::SelectionDAGISel:: > > > > SelectAllBasicBlocks(llvm::Function const&) > > > (/usr/bin/../lib/libLLVM-3.8. > > > > so+0xb45395) > > > > > #7 0x00007fc74ecc7d7c llvm::SelectionDAGISel:: > > > > runOnMachineFunction(llvm::MachineFunction&) > > > (/usr/bin/../lib/libLLVM-3.8. > > > > so+0xb46d7c) > > > > > #8 0x00007fc74fba8b04 (/usr/bin/../lib/libLLVM-3.8.so+0x1a27b04) > > > > > #9 0x00007fc74e8bd8c2 llvm::FPPassManager:: > > > runOnFunction(llvm::Function&) > > > > (/usr/bin/../lib/libLLVM-3.8.so+0x73c8c2) > > > > > #10 0x00007fc74e8bdc8b llvm::FPPassManager:: > > runOnModule(llvm::Module&) > > > > (/usr/bin/../lib/libLLVM-3.8.so+0x73cc8b) > > > > > #11 0x00007fc74e8bdfb4 llvm::legacy::PassManagerImpl: > > > :run(llvm::Module&) > > > > (/usr/bin/../lib/libLLVM-3.8.so+0x73cfb4) > > > > > #12 0x00000000006e9313 clang::EmitBackendOutput( > > > clang::DiagnosticsEngine&, > > > > clang::CodeGenOptions const&, clang::TargetOptions const&, > > > > clang::LangOptions const&, llvm::StringRef, llvm::Module*, > > > > clang::BackendAction, llvm::raw_pwrite_stream*) > > > > (/usr/bin/clang-3.8+0x6e9313) > > > > > #13 0x0000000000a42db1 (/usr/bin/clang-3.8+0xa42db1) > > > > > #14 0x0000000000b8aaca clang::ParseAST(clang::Sema&, bool, bool) > > > > (/usr/bin/clang-3.8+0xb8aaca) > > > > > #15 0x00000000009c2a2e clang::FrontendAction::Execute() > > > > (/usr/bin/clang-3.8+0x9c2a2e) > > > > > #16 0x00000000009968a6 clang::CompilerInstance:: > > ExecuteAction(clang:: > > > FrontendAction&) > > > > (/usr/bin/clang-3.8+0x9968a6) > > > > > #17 0x0000000000a41c33 clang::ExecuteCompilerInvocation( > > > clang::CompilerInstance*) > > > > (/usr/bin/clang-3.8+0xa41c33) > > > > > #18 0x00000000006a7918 cc1_main(llvm::ArrayRef<char const*>, char > > > > const*, void*) (/usr/bin/clang-3.8+0x6a7918) > > > > > #19 0x00000000006a3b38 main (/usr/bin/clang-3.8+0x6a3b38) > > > > > #20 0x00007fc74d864291 __libc_start_main > (/usr/bin/../lib/libc.so.6+ > > > > 0x20291) > > > > > #21 0x00000000006a5889 _start (/usr/bin/clang-3.8+0x6a5889) > > > > > Stack dump: > > > > > 0. Program arguments: /usr/bin/clang-3.8 -cc1 -triple > > > > x86_64-unknown-linux-gnu -emit-obj -mrelax-all -disable-free > > > > -disable-llvm-verifier -main-file-name ephysics_world.cpp > > > > -mrelocation-model pic -pic-level 2 -mthread-model posix > > > -mdisable-fp-elim > > > > -fmath-errno -masm-verbose -mconstructor-aliases -munwind-tables > > > > -fuse-init-array -target-cpu x86-64 -dwarf-column-info > > > -debugger-tuning=gdb > > > > -ffunction-sections -fdata-sections -coverage-file > > > > /home/raster/C/efl/src/lib/ephysics/.libs/lib_ephysics_ > > > > libephysics_la-ephysics_world.o -resource-dir > > > /usr/bin/../lib/clang/3.8.1 > > > > -dependency-file lib/ephysics/.deps/lib_ephysics_libephysics_la- > > > ephysics_world.Tpo > > > > -sys-header-deps -MP -MT lib/ephysics/lib_ephysics_ > > > > libephysics_la-ephysics_world.lo -D HAVE_CONFIG_H -I . -I .. -I > > > > ../src/lib/efl -I /usr/include/bullet -I ../src/lib/ephysics -I > > > > ../src/lib/ephysics -I ../src/bindings/cxx/ephysics -I > > > > ../src/bindings/ephysics -I ../src/lib/emile -I ../src/lib/emile -I > > > > ../src/lib/eo -I ../src/lib/eo -I ../src/lib > > > > > /ecore -I ../src/lib/ecore -I ../src/lib/efl -I ../src/lib/efl -I > > > > ../src/lib/evas -I ../src/lib/evas -I ../src/lib/eina -I > > ../src/lib/eina > > > -D > > > > EFL_EPHYSICS_BUILD=1 -D PIC -D PIC -internal-isystem > > > > /bin/../lib64/gcc/x86_64-pc-linux-gnu/6.1.1/../../../../ > > > include/c++/6.1.1 > > > > -internal-isystem /bin/../lib64/gcc/x86_64-pc- > > > linux-gnu/6.1.1/../../../../ > > > > include/c++/6.1.1/x86_64-pc-linux-gnu -internal-isystem > > > > /bin/../lib64/gcc/x86_64-pc-linux-gnu/6.1.1/../../../../ > > > include/c++/6.1.1/backward > > > > -internal-isystem /usr/local/include -internal-isystem > > > > /usr/bin/../lib/clang/3.8.1/include -internal-externc-isystem > /include > > > > -internal-externc-isystem /usr/include -Wall -Wextra -Wpointer-arith > > > > -Wno-missing-field-initializers -std=gnu++11 -fdeprecated-macro > > > > -fdebug-compilation-dir /home/raster/C/efl/src -ferror-limit 19 > > > > -fmessage-length 196 -fvisibility hidden -fobjc-runtime=gcc > > > > -fcxx-exceptions -fexceptions -fdiagnostics-show-option > > > -fcolor-diagnostics > > > > -o lib/ephysics/.libs/lib_ephysics_libephysics_la-ephysi > > > > > cs_world.o -x c++ lib/ephysics/ephysics_world.cpp > > > > > 1. <eof> parser at end of file > > > > > 2. Code generation > > > > > 3. Running pass 'Function Pass Manager' on module > > > > 'lib/ephysics/ephysics_world.cpp'. > > > > > 4. Running pass 'X86 DAG->DAG Instruction Selection' on > function > > > > '@_Z16_simulate_worldsPv' > > > > > clang-3.8: error: unable to execute command: Segmentation fault > (core > > > > dumped) > > > > > clang-3.8: error: clang frontend command failed due to signal (use > -v > > > to > > > > see invocation) > > > > > clang version 3.8.1 (tags/RELEASE_381/final) > > > > > Target: x86_64-unknown-linux-gnu > > > > > Thread model: posix > > > > > InstalledDir: /bin > > > > > clang-3.8: note: diagnostic msg: PLEASE submit a bug report to > > > > http://llvm.org/bugs/ and include the crash backtrace, preprocessed > > > > source, and associated run script. > > > > > clang-3.8: note: diagnostic msg: > > > > > ******************** > > > > > > > > > > PLEASE ATTACH THE FOLLOWING FILES TO THE BUG REPORT: > > > > > Preprocessed source(s) and associated run script(s) are located at: > > > > > clang-3.8: note: diagnostic msg: /tmp/ephysics_world-8afc31.cpp > > > > > clang-3.8: note: diagnostic msg: /tmp/ephysics_world-8afc31.sh > > > > > clang-3.8: note: diagnostic msg: > > > > > > > > > > ******************** > > > > > make[5]: *** [Makefile:46771: lib/ephysics/lib_ephysics_ > > > > libephysics_la-ephysics_world.lo] Error 1 > > > > > > > > > > > > > > >> On Sat, Oct 1, 2016 at 5:38 PM, Carsten Haitzler < > > > ras...@rasterman.com> > > > > wrote: > > > > >> > raster pushed a commit to branch master. > > > > >> > > > > > >> > http://git.enlightenment.org/core/efl.git/commit/?id= > > > > a2d507d3bfd12b84620eb7361b8fd8f3a5ba0dda > > > > >> > > > > > >> > commit a2d507d3bfd12b84620eb7361b8fd8f3a5ba0dda > > > > >> > Author: Carsten Haitzler (Rasterman) <ras...@rasterman.com> > > > > >> > Date: Sat Oct 1 17:51:25 2016 +0900 > > > > >> > > > > > >> > eina list - make use of prefetch for minor speedups > > > > >> > > > > > >> > i see a speedup of about 8% over a series of list walking > and > > > > freeing > > > > >> > functions given this change. it's a small speedup but still > > not > > > > too > > > > >> > shabby just for some prefetches thrown in. ymmv depending on > > > > memory > > > > >> > subsystem, memory speed itself, cpu and architecture. > > > > >> > > > > > >> > @optimize > > > > >> > --- > > > > >> > src/lib/eina/eina_list.h | 34 +++++++++++++++++++++++++----- > ---- > > > > >> > 1 file changed, 25 insertions(+), 9 deletions(-) > > > > >> > > > > > >> > diff --git a/src/lib/eina/eina_list.h b/src/lib/eina/eina_list.h > > > > >> > index e35943e..146601d 100644 > > > > >> > --- a/src/lib/eina/eina_list.h > > > > >> > +++ b/src/lib/eina/eina_list.h > > > > >> > @@ -1411,12 +1411,16 @@ EAPI int > > > > eina_list_data_idx(const > > > > >> > Eina_List *list, void *data) > > > > >> > * For destructive operations such as this, consider > > > > >> > * using EINA_LIST_FOREACH_SAFE(). > > > > >> > */ > > > > >> > -#define EINA_LIST_FOREACH(list, l, data) \ > > > > >> > +#define EINA_LIST_FOREACH(list, l, _data)\ > > > > >> > for (l = list, \ > > > > >> > - data = eina_list_data_get(l); \ > > > > >> > + _data = eina_list_data_get(l), \ > > > > >> > + l ? (EINA_PREFETCH(l->next), EINA_PREFETCH(_data)) : > > > > EINA_PREFETCH > > > > >> > (l); \ > > > > >> > + \ > > > > >> > l; \ > > > > >> > + \ > > > > >> > l = eina_list_next(l), \ > > > > >> > - data = eina_list_data_get(l)) > > > > >> > + _data = eina_list_data_get(l), \ > > > > >> > + l ? (EINA_PREFETCH(l->next), EINA_PREFETCH(_data)) : > > > > EINA_PREFETCH > > > > >> > (l)) > > > > >> > > > > > >> > /** > > > > >> > * @def EINA_LIST_REVERSE_FOREACH > > > > >> > @@ -1469,12 +1473,14 @@ EAPI int > > > > eina_list_data_idx(const > > > > >> > Eina_List *list, void *data) > > > > >> > * For destructive operations such as this, consider > > > > >> > * using EINA_LIST_REVERSE_FOREACH_SAFE(). > > > > >> > */ > > > > >> > -#define EINA_LIST_REVERSE_FOREACH(list, l, data) \ > > > > >> > +#define EINA_LIST_REVERSE_FOREACH(list, l, _data)\ > > > > >> > for (l = eina_list_last(list), \ > > > > >> > - data = eina_list_data_get(l); \ > > > > >> > + _data = eina_list_data_get(l), \ > > > > >> > + l ? (EINA_PREFETCH(l->prev), EINA_PREFETCH(_data)) : > > > > EINA_PREFETCH > > > > >> > (l); \ l; \ > > > > >> > l = eina_list_prev(l), \ > > > > >> > - data = eina_list_data_get(l)) > > > > >> > + _data = eina_list_data_get(l), \ > > > > >> > + l ? (EINA_PREFETCH(l->prev), EINA_PREFETCH(_data)) : > > > > EINA_PREFETCH > > > > >> > (l)) > > > > >> > > > > > >> > /** > > > > >> > * @def EINA_LIST_FOREACH_SAFE > > > > >> > @@ -1525,11 +1531,15 @@ EAPI int > > > > eina_list_data_idx(const > > > > >> > Eina_List *list, void *data) > > > > >> > #define EINA_LIST_FOREACH_SAFE(list, l, l_next, data) \ > > > > >> > for (l = list, \ > > > > >> > l_next = eina_list_next(l), \ > > > > >> > + EINA_PREFETCH(l_next), \ > > > > >> > data = eina_list_data_get(l); \ > > > > >> > + EINA_PREFETCH(data), \ > > > > >> > l; \ > > > > >> > l = l_next, \ > > > > >> > l_next = eina_list_next(l), \ > > > > >> > - data = eina_list_data_get(l)) > > > > >> > + EINA_PREFETCH(l_next), \ > > > > >> > + data = eina_list_data_get(l), \ > > > > >> > + EINA_PREFETCH(data)) > > > > >> > > > > > >> > /** > > > > >> > * @def EINA_LIST_REVERSE_FOREACH_SAFE > > > > >> > @@ -1582,11 +1592,15 @@ EAPI int > > > > eina_list_data_idx(const > > > > >> > Eina_List *list, void *data) > > > > >> > #define EINA_LIST_REVERSE_FOREACH_SAFE(list, l, l_prev, data) > \ > > > > >> > for (l = eina_list_last(list), \ > > > > >> > l_prev = eina_list_prev(l), \ > > > > >> > + EINA_PREFETCH(l_prev), \ > > > > >> > data = eina_list_data_get(l); \ > > > > >> > + EINA_PREFETCH(data), \ > > > > >> > l; \ > > > > >> > l = l_prev, \ > > > > >> > l_prev = eina_list_prev(l), \ > > > > >> > - data = eina_list_data_get(l)) > > > > >> > + EINA_PREFETCH(l_prev), \ > > > > >> > + data = eina_list_data_get(l), \ > > > > >> > + EINA_PREFETCH(data)) > > > > >> > > > > > >> > /** > > > > >> > * @def EINA_LIST_FREE > > > > >> > @@ -1624,9 +1638,11 @@ EAPI int > > > > eina_list_data_idx(const > > > > >> > Eina_List *list, void *data) > > > > >> > * @see eina_list_free() > > > > >> > */ > > > > >> > #define EINA_LIST_FREE(list, data) \ > > > > >> > - for (data = eina_list_data_get(list); \ > > > > >> > + for (data = eina_list_data_get(list), \ > > > > >> > + list ? EINA_PREFETCH((list)->next) : > EINA_PREFETCH(list); > > \ > > > > >> > list; \ > > > > >> > list = eina_list_remove_list(list, list), \ > > > > >> > + list ? EINA_PREFETCH((list)->next) : > EINA_PREFETCH(list), > > \ > > > > >> > data = eina_list_data_get(list)) > > > > >> > > > > > >> > #include "eina_inline_list.x" > > > > >> > > > > > >> > -- > > > > >> > > > > > >> > > > > > >> > > > > > > > > > > > > > > > -- > > > > > ------------- Codito, ergo sum - "I code, therefore I am" > > > -------------- > > > > > The Rasterman (Carsten Haitzler) ras...@rasterman.com > > > > > > > > > > > > ------------------------------------------------------------ > > > ------------------ > > > Check out the vibrant tech community on one of the world's most > > > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > > > _______________________________________________ > > > enlightenment-devel mailing list > > > enlightenment-devel@lists.sourceforge.net > > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > > > > > > > > > > -- > > Jean-Philippe André > > ------------------------------------------------------------ > > ------------------ > > Check out the vibrant tech community on one of the world's most > > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > > _______________________________________________ > > enlightenment-devel mailing list > > enlightenment-devel@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > _______________________________________________ > enlightenment-devel mailing list > enlightenment-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > -- Jean-Philippe André ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot _______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
