| Message ID | 20200417203914.26358-1-sudipm.mukherjee@gmail.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 0ea3f5c42ffad95ef55c8b269649a1135e5d02f9 |
| Headers | show |
| Series | trace-cmd: Fix build with gcc-10 | expand |
On Fri, 17 Apr 2020 21:39:14 +0100 Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > The build fails while building with gcc-10 with the error: > multiple definition of `common_type_field' I'd say this is a bug in gcc. Both variables are static to the files. Why is it complaining? If its a new "feature" of gcc, I would look to see if there's a way to quiet the warning, and adding that to the Makefile instead. -- Steve > > Rename the field in trace-hist.c and trace-mem.c to satisfy gcc-10. >
On Mon, 20 Apr 2020 15:43:25 -0400 Steven Rostedt <rostedt@goodmis.org> wrote: > On Fri, 17 Apr 2020 21:39:14 +0100 > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > The build fails while building with gcc-10 with the error: > > multiple definition of `common_type_field' > > I'd say this is a bug in gcc. > > Both variables are static to the files. Why is it complaining? > > If its a new "feature" of gcc, I would look to see if there's a way to > quiet the warning, and adding that to the Makefile instead. > And when doing a search of "gcc 10" "multiple definitions" the first hit was this: https://patchwork.kernel.org/patch/11372767/ Which is Tzvetomir's patch (which I accepted, but didn't think too much about it). I was thinking of just adding -fcommon, but then took a look at the definition of this: "-fcommon In C code, this option controls the placement of global variables defined without an initializer, known as tentative definitions in the C standard. Tentative definitions are distinct from declarations of a variable with the extern keyword, which do not allocate storage. The default is -fno-common, which specifies that the compiler places uninitialized global variables in the BSS section of the object file. This inhibits the merging of tentative definitions by the linker so you get a multiple-definition error if the same variable is accidentally defined in more than one compilation unit. The -fcommon places uninitialized global variables in a common block. This allows the linker to resolve all tentative definitions of the same variable in different compilation units to the same object, or to a non-tentative definition. This behavior is inconsistent with C++, and on many targets implies a speed and code size penalty on global variable references. It is mainly useful to enable legacy code to link without errors. " OK, so this has a warning due to linking optimizations, and when there's common names, it causes linking to do some more tricks, and can cause issues when linking with C++ programs (like KernelShark). OK, I'll take this patch, and may even enable a -fno-common to find other cases of this without having to wait till my distro adds gcc 10 by default. -- Steve
On Mon, Apr 20, 2020 at 8:53 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > On Mon, 20 Apr 2020 15:43:25 -0400 > Steven Rostedt <rostedt@goodmis.org> wrote: > > > On Fri, 17 Apr 2020 21:39:14 +0100 > > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > > > The build fails while building with gcc-10 with the error: > > > multiple definition of `common_type_field' > > <snip> > > OK, I'll take this patch, and may even enable a -fno-common to find other > cases of this without having to wait till my distro adds gcc 10 by default. Thanks Steve. We have gcc-10 in the experimental release of Debian and I am using that to build. Looks like this was the only one case which gcc found.
On Mon, Apr 20, 2020 at 11:37 PM Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > On Mon, Apr 20, 2020 at 8:53 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > > > On Mon, 20 Apr 2020 15:43:25 -0400 > > Steven Rostedt <rostedt@goodmis.org> wrote: > > > > > On Fri, 17 Apr 2020 21:39:14 +0100 > > > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > > > > > The build fails while building with gcc-10 with the error: > > > > multiple definition of `common_type_field' > > > > <snip> > > > > OK, I'll take this patch, and may even enable a -fno-common to find other > > cases of this without having to wait till my distro adds gcc 10 by default. A gentle ping.
On Fri, 8 May 2020 13:46:19 +0100 Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > On Mon, Apr 20, 2020 at 11:37 PM Sudip Mukherjee > <sudipm.mukherjee@gmail.com> wrote: > > > > On Mon, Apr 20, 2020 at 8:53 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > > > > > On Mon, 20 Apr 2020 15:43:25 -0400 > > > Steven Rostedt <rostedt@goodmis.org> wrote: > > > > > > > On Fri, 17 Apr 2020 21:39:14 +0100 > > > > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > > > > > > > The build fails while building with gcc-10 with the error: > > > > > multiple definition of `common_type_field' > > > > > > <snip> > > > > > > OK, I'll take this patch, and may even enable a -fno-common to find other > > > cases of this without having to wait till my distro adds gcc 10 by default. > > A gentle ping. > Thanks for the reminder, I haven't forgotten about it, just had other things added on top of my "todo" list. I'll go apply this now. Cheers! -- Steve
On Fri, May 8, 2020 at 3:11 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > On Fri, 8 May 2020 13:46:19 +0100 > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > On Mon, Apr 20, 2020 at 11:37 PM Sudip Mukherjee > > <sudipm.mukherjee@gmail.com> wrote: > > > > > > On Mon, Apr 20, 2020 at 8:53 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > > > > > > > On Mon, 20 Apr 2020 15:43:25 -0400 > > > > Steven Rostedt <rostedt@goodmis.org> wrote: > > > > > > > > > On Fri, 17 Apr 2020 21:39:14 +0100 > > > > > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > > > > > > > > > The build fails while building with gcc-10 with the error: > > > > > > multiple definition of `common_type_field' > > > > > > > > <snip> > > > > > > > > OK, I'll take this patch, and may even enable a -fno-common to find other > > > > cases of this without having to wait till my distro adds gcc 10 by default. > > > > A gentle ping. > > > > Thanks for the reminder, I haven't forgotten about it, just had other > things added on top of my "todo" list. > > I'll go apply this now. Thanks. And now kernelshark fails. :( /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsGLWidget.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsGLWidget.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsDualMarker.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsDualMarker.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsTraceGraph.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsTraceGraph.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsTraceViewer.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsTraceViewer.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsMainWindow.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsMainWindow.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsQuickContextMenu.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/moc_KsQuickContextMenu.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsUtils.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsUtils.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsModels.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsModels.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsSession.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsSession.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsGLWidget.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsGLWidget.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsSearchFSM.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsSearchFSM.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsDualMarker.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsDualMarker.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsWidgetsLib.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsWidgetsLib.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsTraceGraph.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsTraceGraph.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsTraceViewer.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsTraceViewer.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsMainWindow.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsMainWindow.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsCaptureDialog.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsCaptureDialog.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsQuickContextMenu.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsQuickContextMenu.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsAdvFilteringDialog.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:43: multiple definition of `stringWidth'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:43: first defined here /usr/bin/ld: CMakeFiles/kshark-gui.dir/KsAdvFilteringDialog.cpp.o:/build/trace-cmd/kernel-shark/src/KsUtils.hpp:35: multiple definition of `fontHeight'; CMakeFiles/kshark-gui.dir/moc_KsUtils.cpp.o:/build/trace-cmd/kernel-shark/build/src/../../src/KsUtils.hpp:35: first defined here
On Sat, 9 May 2020 19:30:38 +0100 Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > Thanks for the reminder, I haven't forgotten about it, just had other > > things added on top of my "todo" list. > > > > I'll go apply this now. > > Thanks. And now kernelshark fails. :( > Have you tried doing a make clean and building again? Sometimes the updates leave objects in place that need to be recompiled, but the dependencies may miss them. -- Steve
On Sat, May 9, 2020 at 7:48 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > On Sat, 9 May 2020 19:30:38 +0100 > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > > > Thanks for the reminder, I haven't forgotten about it, just had other > > > things added on top of my "todo" list. > > > > > > I'll go apply this now. > > > > Thanks. And now kernelshark fails. :( > > > > Have you tried doing a make clean and building again? Sometimes the > updates leave objects in place that need to be recompiled, but the > dependencies may miss them. The build will be done in a new chroot every time so there is no chance of objects from a previous build. Just to make sure I did 'git clone' in a new folder and tried to build again. And it’s the same failure. Also did 'make clean' and built again with the same failure.
On Sat, 9 May 2020 20:11:09 +0100 Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > On Sat, May 9, 2020 at 7:48 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > > > On Sat, 9 May 2020 19:30:38 +0100 > > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > > > > > Thanks for the reminder, I haven't forgotten about it, just had other > > > > things added on top of my "todo" list. > > > > > > > > I'll go apply this now. > > > > > > Thanks. And now kernelshark fails. :( > > > > > > > Have you tried doing a make clean and building again? Sometimes the > > updates leave objects in place that need to be recompiled, but the > > dependencies may miss them. > > The build will be done in a new chroot every time so there is no > chance of objects from a previous build. Just to make sure I did 'git > clone' in a new folder and tried to build again. And it’s the same > failure. Also did 'make clean' and built again with the same failure. > > Hmm, I wonder if something like this fixes it? -- Steve diff --git a/kernel-shark/src/KsUtils.hpp b/kernel-shark/src/KsUtils.hpp index f44139bb..272a27d1 100644 --- a/kernel-shark/src/KsUtils.hpp +++ b/kernel-shark/src/KsUtils.hpp @@ -32,7 +32,7 @@ //! @cond Doxygen_Suppress -auto fontHeight = []() +static auto fontHeight = []() { QFont font; QFontMetrics fm(font); @@ -40,7 +40,7 @@ auto fontHeight = []() return fm.height(); }; -auto stringWidth = [](QString s) +static auto stringWidth = [](QString s) { QFont font; QFontMetrics fm(font);
On Sat, May 9, 2020 at 10:12 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > On Sat, 9 May 2020 20:11:09 +0100 > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > On Sat, May 9, 2020 at 7:48 PM Steven Rostedt <rostedt@goodmis.org> wrote: > > > > > > On Sat, 9 May 2020 19:30:38 +0100 > > > Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote: > > > > > > > > > Thanks for the reminder, I haven't forgotten about it, just had other > > > > > things added on top of my "todo" list. > > > > > > > > > > I'll go apply this now. > > > > > > > > Thanks. And now kernelshark fails. :( > > > > > > > > > > Have you tried doing a make clean and building again? Sometimes the > > > updates leave objects in place that need to be recompiled, but the > > > dependencies may miss them. > > > > The build will be done in a new chroot every time so there is no > > chance of objects from a previous build. Just to make sure I did 'git > > clone' in a new folder and tried to build again. And it’s the same > > failure. Also did 'make clean' and built again with the same failure. > > > > > > Hmm, I wonder if something like this fixes it? Yes, it does. :)
diff --git a/tracecmd/trace-hist.c b/tracecmd/trace-hist.c index c458c25..c4d8b40 100644 --- a/tracecmd/trace-hist.c +++ b/tracecmd/trace-hist.c @@ -27,7 +27,7 @@ static int kernel_stack_type; static int long_size; -static struct tep_format_field *common_type_field; +static struct tep_format_field *common_type_hist; static struct tep_format_field *common_pid_field; static struct tep_format_field *sched_wakeup_comm_field; static struct tep_format_field *sched_wakeup_new_comm_field; @@ -568,7 +568,7 @@ process_record(struct tep_handle *pevent, struct tep_record *record) unsigned long long val; int type; - tep_read_number_field(common_type_field, record->data, &val); + tep_read_number_field(common_type_hist, record->data, &val); type = val; if (type == function_type) @@ -956,8 +956,8 @@ static void do_trace_hist(struct tracecmd_input *handle) long_size = tracecmd_long_size(handle); - common_type_field = tep_find_common_field(event, "common_type"); - if (!common_type_field) + common_type_hist = tep_find_common_field(event, "common_type"); + if (!common_type_hist) die("Can't find a 'type' field?"); common_pid_field = tep_find_common_field(event, "common_pid"); diff --git a/tracecmd/trace-mem.c b/tracecmd/trace-mem.c index 465b182..76f38fa 100644 --- a/tracecmd/trace-mem.c +++ b/tracecmd/trace-mem.c @@ -30,7 +30,7 @@ static int kmem_cache_alloc_type; static int kmem_cache_alloc_node_type; static int kmem_cache_free_type; -static struct tep_format_field *common_type_field; +static struct tep_format_field *common_type_mem; static struct tep_format_field *kmalloc_callsite_field; static struct tep_format_field *kmalloc_bytes_req_field; @@ -369,7 +369,7 @@ process_record(struct tep_handle *pevent, struct tep_record *record) unsigned long long val; int type; - tep_read_number_field(common_type_field, record->data, &val); + tep_read_number_field(common_type_mem, record->data, &val); type = val; if (type == kmalloc_type) @@ -490,8 +490,8 @@ static void do_trace_mem(struct tracecmd_input *handle) ret = tep_data_type(pevent, record); event = tep_find_event(pevent, ret); - common_type_field = tep_find_common_field(event, "common_type"); - if (!common_type_field) + common_type_mem = tep_find_common_field(event, "common_type"); + if (!common_type_mem) die("Can't find a 'type' field?"); update_kmalloc(pevent);
The build fails while building with gcc-10 with the error: multiple definition of `common_type_field' Rename the field in trace-hist.c and trace-mem.c to satisfy gcc-10. Bug-Link: https://bugs.debian.org/957879 Signed-off-by: Sudip Mukherjee <sudipm.mukherjee@gmail.com> --- tracecmd/trace-hist.c | 8 ++++---- tracecmd/trace-mem.c | 8 ++++---- 2 files changed, 8 insertions(+), 8 deletions(-)