|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Xen Release 4.14.0: Couple of "all warnings being treated as errors" issues and ongoing docs/man issue in make world
Looking to build 4.14.0 on an LFS 10.0 system, so with GCC 10.2.0.
The "all warnings being treated as errors" I'm sure, will have been
picked up by now, but the issue with the man pages is something
I have been seeing for a while now.
The configure options are a little idiosyncratic but I can't see why they
would unmask the issues seen here, but, for completeness
./configure --prefix=/usr \
--disable-seabios \
--disable-qemu-traditional \
--disable-rombios \
--disable-stubdom
after which,
make PYTHON=/usr/bin/python3 world
first generates these four
libxlu_pci.c:32:18: error: 'func' may be used uninitialized in this
function [-Werror=maybe-uninitialized]
32 | pcidev->func = func;
| ~~~~~~~~~~~~~^~~~~~
libxlu_pci.c:51:29: note: 'func' was declared here
51 | unsigned dom, bus, dev, func, vslot = 0;
| ^~~~
libxlu_pci.c:31:17: error: 'dev' may be used uninitialized in this
function [-Werror=maybe-uninitialized]
31 | pcidev->dev = dev;
| ~~~~~~~~~~~~^~~~~
libxlu_pci.c:51:24: note: 'dev' was declared here
51 | unsigned dom, bus, dev, func, vslot = 0;
| ^~~
libxlu_pci.c:30:17: error: 'bus' may be used uninitialized in this
function [-Werror=maybe-uninitialized]
30 | pcidev->bus = bus;
| ~~~~~~~~~~~~^~~~~
libxlu_pci.c:51:19: note: 'bus' was declared here
51 | unsigned dom, bus, dev, func, vslot = 0;
| ^~~
libxlu_pci.c:29:20: error: 'dom' may be used uninitialized in this
function [-Werror=maybe-uninitialized]
29 | pcidev->domain = domain;
| ~~~~~~~~~~~~~~~^~~~~~~~
libxlu_pci.c:51:14: note: 'dom' was declared here
51 | unsigned dom, bus, dev, func, vslot = 0;
| ^~~
cc1: all warnings being treated as errors
make[5]: *** [/usr/src/xen/xen-4.14.0/tools/libxl/../../tools/Rules.mk:216:
libxlu_pci.o] Error 1
make[5]: Leaving directory '/usr/src/xen/xen-4.14.0/tools/libxl'
which I "fixed" by adding an
-Wno-maybe-uninitialized
to the CFLAGS in the tools/libxl Makefile
after which the make world then goes on to fail here
libxl_utils.c: In function 'libxl__prepare_sockaddr_un':
libxl_utils.c:1262:5: error: 'strncpy' specified bound 108 equals
destination size [-Werror=stringop-truncation]
1262 | strncpy(un->sun_path, path, sizeof(un->sun_path));
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
make[5]: *** [/usr/src/xen/xen-4.14.0/tools/libxl/../../tools/Rules.mk:216:
libxl_utils.o] Error 1
make[5]: Leaving directory '/usr/src/xen/xen-4.14.0/tools/libxl'
make[4]: *** [/usr/src/xen/xen-4.14.0/tools/../tools/Rules.mk:240:
subdir-install-libxl] Error 2
make[4]: Leaving directory '/usr/src/xen/xen-4.14.0/tools'
make[3]: *** [/usr/src/xen/xen-4.14.0/tools/../tools/Rules.mk:235:
subdirs-install] Error 2
make[3]: Leaving directory '/usr/src/xen/xen-4.14.0/tools'
make[2]: *** [Makefile:72: install] Error 2
make[2]: Leaving directory '/usr/src/xen/xen-4.14.0/tools'
make[1]: *** [Makefile:134: install-tools] Error 2
make[1]: Leaving directory '/usr/src/xen/xen-4.14.0'
make: *** [Makefile:170: world] Error 2
which I fixed by setting
-Wno-maybe-uninitialized -Wno-stringop-truncation
as the extra CFLAGS in the modified tools/libxl Makefile
After that point, the build gets as far as
make[2]: Leaving directory '/usr/src/xen/xen-4.14.0/tools'
make -C docs install
make[2]: Entering directory '/usr/src/xen/xen-4.14.0/docs'
/usr/bin/pod2man --release=4.14.0 --name=xenhypfs -s 1 -c "Xen"
man/xenhypfs.1.pod man1/xenhypfs.1
Can't write-open man1/xenhypfs.1: No such file or directory at
/usr/bin/pod2man line 69.
make[2]: *** [Makefile:176: man1/xenhypfs.1] Error 2
make[2]: Leaving directory '/usr/src/xen/xen-4.14.0/docs'
make[1]: *** [Makefile:153: install-docs] Error 2
make[1]: Leaving directory '/usr/src/xen/xen-4.14.0'
make: *** [Makefile:170: world] Error 2
and this is the interesting bit.
Firstly, nothing that the make ins being run from the top-level
docs directory, looking at
pkg xen:xen-4.14.0> ls docs/
INDEX configure.ac man/
Makefile designs/ misc/
README.colo features/ parse-support-md*
README.remus figs/ process/
README.source gen-html-index specs/
admin-guide/ glossary.rst support-matrix-generate*
conf.py guest-guide/ xen-headers*
config.status* hypervisor-guide/
configure* index.rst
shows that there isn't a man1 subdir within the source tree?
Furthermore, looking at
pkg xen:xen-4.14.0> ls docs/man/
xen-pci-device-reservations.7.pod xentop.1.pod
xen-pv-channel.7.pod xentrace.8.pod
xen-tscmode.7.pod xentrace_format.1.pod
xen-vbd-interface.7.pandoc xl-disk-configuration.5.pod
xen-vtpm.7.pod xl-network-configuration.5.pod
xen-vtpmmgr.7.pod xl-numa-placement.7.pod
xenhypfs.1.pod xl.1.pod
xenstore-chmod.1.pod xl.1.pod.in
xenstore-ls.1.pod xl.cfg.5.pod
xenstore-read.1.pod xl.cfg.5.pod.in
xenstore-write.1.pod xl.conf.5.pod
xenstore.1.pod xlcpupool.cfg.5.pod
suggests that all of the man pages POS files, that one might
expect, given the make output above, to be in man1, man5,
man7 and man8 subdirs, are all in the one "man" directory.
I have seen this "docs/man? failure" with 'make world' in a few past
builds and have found, in the past, that splitting out the 'make world'
into these two parts, with a creation of the seemingly expected dirs
in between, vis:
make clean
mkdir docs/man1 docs/man5 docs/man7 docs/man8
make dist
has then seen the build complete.
I am thinking though, that the 'make world' should just work,
out of the tarball ?
So is there something missing that should be re-arranging the
man page sources into seperate manN subirs, or should the
pod2man command be given different arguments?
Hoping that info is of some use to you,
Kevin
|
 |
Lists.xenproject.org is hosted with RackSpace, monitoring our |