On 08/10/2019 14.41, Daniel P. Berrangé wrote: > On Tue, Oct 08, 2019 at 02:35:20PM +0200, Thomas Huth wrote: >> On 08/10/2019 14.18, Aleksandar Markovic wrote: >>> >>> >>> On Tuesday, October 8, 2019, Thomas Huth <th...@redhat.com >>> <mailto:th...@redhat.com>> wrote: >>> >>> When running "make distclean" we currently leave a lot of generated >>> files in the build directory. These should be completely removed. >>> Some of the generated files are removed in the "clean" target (which >>> is a prerequisite for the "distclean" target), since binary files >>> should be removed in this step already. >>> >>> Signed-off-by: Thomas Huth <th...@redhat.com <mailto:th...@redhat.com>> >>> --- >>> v2: >>> - Remove tests/qemu-iotests/common.env in "distclean", not in "clean" >>> - Improved patch description >>> >>> Makefile | 6 +++--- >>> tests/Makefile.include | 12 +++++++++++- >>> 2 files changed, 14 insertions(+), 4 deletions(-) >>> >>> diff --git a/Makefile b/Makefile >>> index 30f0abfb42..767b1ffb25 100644 >>> --- a/Makefile >>> +++ b/Makefile >>> @@ -696,14 +696,14 @@ clean: recurse-clean >>> -exec rm {} + >>> rm -f $(edk2-decompressed) >>> rm -f $(filter-out %.tlb,$(TOOLS)) $(HELPERS-y) TAGS >>> cscope.* *.pod *~ */*~ >>> - rm -f fsdev/*.pod scsi/*.pod >>> + rm -f fsdev/*.pod scsi/*.pod docs/*.pod docs/*/*.pod >>> docs/*/.buildinfo >>> rm -f qemu-img-cmds.h >>> rm -f ui/shader/*-vert.h ui/shader/*-frag.h >>> @# May not be present in generated-files-y >>> rm -f trace/generated-tracers-dtrace.dtrace* >>> rm -f trace/generated-tracers-dtrace.h* >>> rm -f $(foreach f,$(generated-files-y),$(f) $(f)-timestamp) >>> - rm -f qapi-gen-timestamp >>> + rm -f qapi-gen-timestamp vhost-user-input >>> rm -rf qga/qapi-generated >>> rm -f config-all-devices.mak >>> >>> @@ -724,7 +724,7 @@ distclean: clean >>> rm -f tests/tcg/config-*.mak >>> rm -f config-all-devices.mak config-all-disas.mak config.status >>> rm -f $(SUBDIR_DEVICES_MAK) >>> - rm -f po/*.mo tests/qemu-iotests/common.env >>> + rm -f po/*.mo >>> rm -f roms/seabios/config.mak roms/vgabios/config.mak >>> rm -f qemu-doc.info <http://qemu-doc.info> qemu-doc.aux >>> qemu-doc.cp qemu-doc.cps >>> rm -f qemu-doc.fn qemu-doc.fns qemu-doc.info >>> <http://qemu-doc.info> qemu-doc.ky <http://qemu-doc.ky> qemu-doc.kys >>> diff --git a/tests/Makefile.include b/tests/Makefile.include >>> index 3543451ed3..694f193fb6 100644 >>> --- a/tests/Makefile.include >>> +++ b/tests/Makefile.include >>> @@ -1176,11 +1176,21 @@ check: check-block check-qapi-schema >>> check-unit check-softfloat check-qtest chec >>> check-clean: >>> rm -rf $(check-unit-y) tests/*.o $(QEMU_IOTESTS_HELPERS-y) >>> rm -rf $(sort $(foreach target,$(SYSEMU_TARGET_LIST), >>> $(check-qtest-$(target)-y)) $(check-qtest-generic-y)) >>> - rm -f tests/test-qapi-gen-timestamp >>> rm -rf $(TESTS_VENV_DIR) $(TESTS_RESULTS_DIR) >>> + rm -f tests/test-qapi-gen-timestamp tests/qht-bench$(EXESUF) \ >>> + tests/fp/fp-test tests/fp/*.out >>> tests/qapi-schema/*.test.* >>> >>> clean: check-clean >>> >>> +check-distclean: >>> + rm -f tests/qemu-iotests/common.env tests/qemu-iotests/check.* >>> + rm -f tests/test-qapi-types*.c tests/test-qapi-visit*.c \ >>> + tests/test-qapi-commands*.c tests/test-qapi-events*.c \ >>> + tests/test-qapi-emit-events.[ch] >>> tests/test-qapi-introspect.c \ >>> + tests/include/test-qapi-*.c >>> + >>> +distclean: check-distclean >>> + >>> >>> >>> Sorry for bad email format, I have some difficultues with mail accounts >>> and mail clients. >>> >>> Thomas, >>> >>> If I remember well, QAPI-related c files are generated while doing >>> 'make'. If that is true, these files should be deleted by 'make clean', >>> shouldn't they? >> >> I think that's a philosophical question. Should "make clean" delete all >> files that have been generated by a Makefile (i.e. not by "configure"), >> or rather mainly the binary files? >> >> If you have a look at our main Makefile in the topmost directory, you >> can see that we remove lots of files that are obviously generated by >> "make" in the "distclean" target, so it seems we have no clear rule what >> should be deleted in each step... > > Given that we're aiming to convert to meson, how much effort do we really > want to put into fixing this inconsistency ?
I don't think that we'll convert to meson within the next release, I rather assume that this will take quite a bit of additional time, so fixing this now seems reasonable to me. > I'd argue the best way to fix "clean" and "distclean" is to delete them > entirely. Instead enforce that builddir != srcdir when running configure, > so cleaning up simply means deleting the builddir sub-tree entirely. I could agree with "distclean", but as long as nobody cares to get such a "builddir != srcdir" merged, we should not remove it. And I don't agree for removing the "clean" target. Sometimes you want to recompile without destroying your "config.status", and for these cases it's quite useful. Thomas
