Commit Graph

59 Commits

Author SHA1 Message Date
Michal Privoznik
dd98a6edb9 Drop virAsprintf()
Now that function is no longer used, it can be dropped.

Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
2019-11-12 16:15:59 +01:00
Michal Privoznik
0ad79c1d01 m4: Don't suggest attribute malloc
With glib inclusion, some of its functions have
__attribute__((__malloc__)) which make compiler realize we want
to use the same attribute for some trivial functions of ours. For
instance qemuDomainManagedSavePath(). I don't see any real
benefit into using the attribute, so disable that suggestion.

In fact, wrong use of the attribute may lead to mysterious bugs:

  https://gitlab.gnome.org/GNOME/glib/issues/1465

Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
ACKed-by: Peter Krempa <pkrempa@redhat.com>
2019-10-18 10:55:36 +02:00
Daniel P. Berrangé
44e7f02915 util: rewrite auto cleanup macros to use glib's equivalent
To facilitate porting over to glib, this rewrites the auto cleanup
macros to use glib's equivalent.

As a result it is now possible to use g_autoptr/VIR_AUTOPTR, and
g_auto/VIR_AUTOCLEAN, g_autofree/VIR_AUTOFREE interchangably, regardless
of which macros were used to declare the cleanup types.

Within the scope of any single method, code must remain consistent
using either GLib or Libvirt macros, never mixing both. New code
must preferentially use the GLib macros, and old code will be
converted incrementally.

Reviewed-by: Ján Tomko <jtomko@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-10-14 10:54:42 +01:00
Daniel P. Berrangé
58e7c9bc05 build: probe for glib-2 library in configure
Prepare for linking with glib by probing for it at configure
time. Per supported platforms target, the min glib versions on
relevant distros are:

  RHEL-8: 2.56.1
  RHEL-7: 2.50.3
  Debian (Buster): 2.58.3
  Debian (Stretch): 2.50.3
  OpenBSD (Ports): 2.58.3
  FreeBSD (Ports): 2.56.3
  OpenSUSE Leap 15: 2.54.3
  SLE12-SP2: 2.48.2
  Ubuntu (Xenial): 2.48.0
  macOS (Homebrew): 2.56.0

This suggests that a minimum glib of 2.48 is a reasonable target.
This aligns with the minimum version required by qemu too.

We must disable the bad-function-cast warning as various GLib APIs
and macros will trigger this.

Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-10-14 10:54:42 +01:00
Daniel P. Berrangé
412cc0f403 build: stop clang complaining about redefined typedefs
Clang's gnu99 mode is not quite the same as GCC's. It will complain
about redefined typedefs being a C11 feature, while GCC does not
complain and allows them in gnu99 mode.

Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-10-08 17:00:35 +01:00
Daniel P. Berrangé
4d2b96655f m4: fix setting of warning flags
When adding the -std=gnu99 flag, we set $wantwarn instead
of appending to it. This meant all the compiler warnings
were accidentally discarded.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-10-08 13:41:19 +01:00
Daniel P. Berrangé
23605f58bf build: ask for -std=gnu99 explicitly
We previously got -std=gnu99 secretly enabled as a side-effect
of requesting the 'stdarg' gnulib module. We rely on some
extensions from c99/gnu99 and while RHEL-7 supports this, it
still defaults to gnu89.  RHEL-7 also supports some newer
standards but declares them experimental/incomplete, so sticking
with gnu99 is best bet for now & matches historical usage.

Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-10-07 15:50:57 +01:00
Daniel P. Berrangé
a067b781e9 build: bump min required gcc to 4.8
The previous bump to 4.4 was done in:

  commit 24241c236e
  Author: Daniel P. Berrange <berrange@redhat.com>
  Date:   Wed Jul 5 10:35:32 2017 +0100

    Require use of GCC 4.4 or CLang compilers

with 4.4 picked due to RHEL-6. Since we dropped RHEL-6, the
next oldest distro is RHEL-7 (4.8.5), and thus we pick 4.8
as the new min.

Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-08-07 14:26:12 +01:00
Daniel P. Berrangé
67966ad518 m4: enforce that all enum cases are listed in switch statements
As a general rule any time we switch() on something that is an enum, we
want to have a case for every enum constant. The -Wswitch warning will
report any switch where we've violated this rule, except if that switch
has a default case.

Unfortunately it is reasonable to want to list all enum constants *and*
also have a default case. To get a warning in that scenario requires
that we turn on -Wswitch-enum.

In a few cases where we explicitly don't want to list all enum cases, we
can discard the enum type checking by casting the value to a plain int.

Reviewed-by: John Ferlan <jferlan@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2018-02-21 16:59:48 +00:00
Daniel P. Berrangé
9148293018 m4: disable gcc8 -Wcast-function-type warnings from -Wextra
The -Wextra flag bundle gained a new warning -Wcast-function-type.
This complains if you cast between two function prototypes where
the number of parameters or their data types are not compatible.
Unfortunately we need such "bad" function casts for our event
callbacks. It is possible to silence the warning by first casting
to the generic "void (*)(void)" function prototype, but that is
rather ugly to add throughout libvirt code.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2018-02-16 14:37:04 +00:00
Ján Tomko
33dd664c50 Raise the frame limit for tests
After the latest CPU additions, the build fails with clang:
cputest.c:905:1: error: stack frame size of 26136 bytes
  in function 'mymain' [-Werror,-Wframe-larger-than=]

Raise the relaxed limit which is used for tests.
2018-01-18 10:16:41 +01:00
Andrea Bolognani
fbe315bdc8 m4: Disable -Wdisabled-optimization
After b4f7793ce2, qemuxml2xmltest has apparently become big enough
to trigger a compilation error when using --enable-test-coverage on
aarch64:

    CC       qemuxml2xmltest.o
  qemuxml2xmltest.c: In function 'mymain':
  qemuxml2xmltest.c:1216:1: error: const/copy propagation disabled: 4361 basic blocks and 99285 registers [-Werror=disabled-optimization]
   }
   ^
  qemuxml2xmltest.c:1216:1: error: PRE disabled: 4361 basic blocks and 99285 registers [-Werror=disabled-optimization]
  qemuxml2xmltest.c:1216:1: error: const/copy propagation disabled: 4361 basic blocks and 99285 registers [-Werror=disabled-optimization]
  qemuxml2xmltest.c:1216:1: error: const/copy propagation disabled: 4361 basic blocks and 99285 registers [-Werror=disabled-optimization]

However, as the GCC documentation states, this warning is not really
caused by issues in our code, so it makes sense to disable it.

Signed-off-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
2017-09-06 09:06:26 +02:00
Eric Blake
b887d7b0c7 maint: correct quoting for gl_WARN_ADD
3 out of 4 uses of gl_WARN_ADD() were incorrectly adding "" around
the argument, which in turn resulted in the argument being used
unquoted (configure had gl_positive=""-fstack-protector-all"",
rather than the intended gl_positive="-fstack-protector-all").

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
2017-08-28 10:15:05 -05:00
Daniel P. Berrange
1c21dd711c m4: workaround clang/glibc problem with isnan()
When building libvirt with clang we get bogus warnings about
'double' being promoted to 'long double' when calling isnan().

  https://bugzilla.redhat.com/show_bug.cgi?id=1472437

Detect this broken isnan() / compiler combination and disable
the -Wdouble-promotion flag.

Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2017-08-11 17:48:06 +01:00
Daniel P. Berrange
f1acc4130c Temporarily disable format truncation warnings
GCC 7.1 introduces a new -Wformat-truncation warning
flag that reports if it thinks the maximum possible
size of the formatted output will exceed the provided
fixed buffer. This is enabled automatically by the
-Wformat warning flag. There are quite a few places
hit by this in libvirt which need rewriting. This is
non-trivial work in some places, so temporarily
disable the new warning until those fixes can be
implemented.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2017-06-15 14:22:54 +01:00
Daniel P. Berrange
1ba693994a Disable the -Wduplicated-branches warning
Depending on the platform/architecture, a number of conditionals
in libvirt code expand the same on both branches. This is expected
behaviour and harmless, so disable the warning to avoid creating
unexpected build failures

Two examples, mingw32:

../../src/util/vircommand.c: In function 'virCommandWait':
../../src/util/vircommand.c:2562:51: error: this condition has identical branches [-Werror=duplicated-branches]
             *exitstatus = cmd->rawStatus ? status : WEXITSTATUS(status);
                                                   ^
and gcc7.1

In file included from util/virobject.c:28:0:
util/virobject.c: In function 'virClassNew':
util/viratomic.h:176:46: error: this condition has identical branches [-Werror=duplicated-branches]
            (void)(0 ? *(atomic) ^ *(atomic) : 0);                      \
                                             ^
util/virobject.c:144:20: note: in expansion of macro 'virAtomicIntInc'
    klass->magic = virAtomicIntInc(&magicCounter);
                   ^~~~~~~~~~~~~~~

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2017-06-15 13:26:51 +01:00
Ján Tomko
e73889b631 Split out -Wframe-larger-than warning from WARN_CLFAGS
Introduce STRICT_FRAME_LIMIT_CFLAGS that will be used for
production code and RELAXED_FRAME_LIMIT_CFLAGS for tests.

Raising the limit for tests allows building them with clang
with optimizations disabled.
2017-04-06 12:29:35 +02:00
Pavel Hrdina
ca01aeb57f configure: use LIBVIRT_ARG_ENABLE macro
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
2016-12-21 15:39:37 +01:00
Pavel Hrdina
d713a6b120 build: add GCC 6.0 -Wlogical-op workaround
fdstream.c: In function 'virFDStreamWrite':
fdstream.c:390:29: error: logical 'or' of equal expressions [-Werror=logical-op]
        if (errno == EAGAIN || errno == EWOULDBLOCK) {
                            ^~

Fedora rawhide now uses gcc 6.0 and there is a bug with -Wlogical-op
producing false warnings.

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69602

Use GCC pragma push/pop and ignore -Wlogical-op for GCC that supports
push/pop pragma and also has this bug.

Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
2016-04-11 12:09:14 +02:00
Pavel Hrdina
7fb81831fc build: cleanup GCC < 4.6 -Wlogical-op workaround
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
2016-04-11 12:09:11 +02:00
Eric Blake
3502f79198 maint: update to latest gnulib
Time to update to new gnulib before a release.

gcc 5.1 introduced a new -Wformat-signedness, and new gnulib now
turns it on by default.  However, it is still rather lame at the
moment, because it warns for enums, even though there is no way
to control the signeness of an enum which does not use any members
that are negative or larger than INT_MAX, and even though such an
enum would always print the same for both %d and %u:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66249

In file included from ../../src/util/virarch.c:26:0:
../../src/util/virarch.c: In function 'virArchFromHost':
../../src/util/virarch.c:180:15: error: format '%d' expects argument of type 'int', but argument 9 has type 'unsigned int' [-Werror=format=]
     VIR_DEBUG("Mapped %s to %d (%s)",

So this patch turns off the new warning as part of enabling all
other new gcc 5.1 warnings that gnulib now enables.

* .gnulib: Update to latest, in part for gcc 5.1 interaction.
* m4/virt-compile-warnings.m4: Ignore -Wformat-signedness, for now.

Signed-off-by: Eric Blake <eblake@redhat.com>
2015-05-26 10:53:12 -06:00
Daniel P. Berrange
87fa184910 Conditionalize use of -Wno-suggest-attribute=format pragma
Many GCC versions don't understand -Wno-suggest-attribute=format
so the pragma must only be used when supported
2015-01-16 17:12:27 +00:00
Ján Tomko
046a441f3d build: prefer -fstack-protector-strong to -all
Try -fstack-protector-strong first on Linux. If that fails,
fall back to -fstack-protector-all.
2014-06-12 08:16:03 +02:00
Ján Tomko
5567baa973 build: remove ssp-buffer-size
This option only makes sense for -fstack-protector.
With -fstack-protector-all or -fstack-protector-strong,
functions are protected regardless of buffer size.

https://bugzilla.redhat.com/show_bug.cgi?id=1105456
2014-06-12 08:16:03 +02:00
Ján Tomko
fa3519cbe5 build: remove duplicit warning suppression
These warnings have already been added to $dontwarn.
2014-06-12 08:15:54 +02:00
Richard W.M. Jones
fe881ae086 aarch64: Disable -fstack-protector.
I'm using gcc-4.8.2-7.fc21.aarch64 which does not appear to support
this option.
2014-01-01 14:21:21 +00:00
Daniel P. Berrange
a7a1244a73 Remove python binding
The python binding now lives in

  http://libvirt.org/git/?p=libvirt-python.git

that repo also provides an RPM which is upgrade compatible
with the old libvirt-python sub-RPM.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2013-11-25 16:08:00 +00:00
Eric Blake
21114ce9c2 maint: update to latest gnulib
Since we're about to freeze, it's time to pick up the latest
upstream gnulib.  Among other changes, gnulib now guarantees the
use of some -f flags that we were previously manually adding.

* .gnulib: Update to latest, in part for warning improvements.
* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Drop
flags that are now guaranteed by gnulib.
* bootstrap: Resync to gnulib.

Signed-off-by: Eric Blake <eblake@redhat.com>
2013-09-24 06:53:07 -06:00
Eric Blake
10ec64105b maint: slightly reduce configure size
Rather than inlining gl_WARN_ADD loads of time, we can shave about
17k size off of the configure script by delaying it to a cleanup
shell loop.

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Track a
list of things to check, rather than inlining multiple checks.

Signed-off-by: Eric Blake <eblake@redhat.com>
2013-08-16 13:31:05 -06:00
Eric Blake
ff38690be9 maint: the compiler is not always named gcc
https://bugzilla.redhat.com/show_bug.cgi?id=994589 complained that
even when using a cross-compiler not named 'gcc', the configure
output confusingly referred to gcc.

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Use a
more generic statement in configure output.

Signed-off-by: Eric Blake <eblake@redhat.com>
2013-08-07 09:44:52 -06:00
Ján Tomko
4b91dc24d1 Fix build with clang
Partially revert cdd703f's revert of c163410, as linking with clang
with --param=ssp-buffer-size=4 still fails with:
"argument unused during compilation".
2013-07-04 11:35:59 +02:00
Eric Blake
cdd703f475 maint: update to latest gnulib
The latest mingw headers on Fedora 19 fail to build with gnulib
without an update.

Meanwhile, now that upstream gnulib has better handling of -W
probing for clang, we can drop some of our own solutions in
favor of upstream; thus this reverts commit c1634100, "Correctly
detect warning flags with clang".

* .gnulib: Update to latest, for mingw and clang.

Signed-off-by: Eric Blake <eblake@redhat.com>
2013-07-03 11:02:33 -06:00
Roman Bogorodskiy
cc7cd6232e FreeBSD: disable buggy -fstack-protector-all
FreeBSD ships an old gcc 4.2.1 which generates
bogus code, e.g. getsockopt() call returns
struct xucred with bogus values, which doesn't even
allow to connect to libvirtd:

error: Failed to find group record for gid '1284660778': No error: 0

So roll back to just -fstack-protector on FreeBSD.
2013-05-15 15:20:52 -06:00
Daniel P. Berrange
71b54636f0 Don't duplicate compiler warning flags when linking
Automake already passes all CFLAGS to the linker too, so it
is not necessary to set WARN_LDFLAGS in addition to the
WARN_CFLAGS variable.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2013-05-14 15:02:37 +01:00
Daniel P. Berrange
c163410044 Correctly detect warning flags with clang
Clang will happily claim to support any warning flags
unless the -Werror and -Wunknown-warning-option flags
are set. Thus we need to make sure these are set when
testing for clags.

We must also set the clang specific warning flags
-Wno-unused-command-line-argument to avoid a warning
from the ssp-buffer-size flag when linking .o files.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2013-05-14 15:00:22 +01:00
Daniel P. Berrange
c43b685a1d Ensure consistent enablement of gcc 'diagnostic' pragma
The virt-compile-warnings.m4 file would do an explicit
check for whether the compile could use the 'diagnostic'
pragma push/pop feature. The src/internal.h file would
then only enable it for GCC >= 4.6

This breaks with clang which supports the pragma but
does not claim GCC 4.6 compat. Export a variable from
the m4 check to the header file so they are consistent.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2013-05-14 15:00:22 +01:00
Daniel P. Berrange
e57aaa6fcf Disable cast-align warnings in various places
There are a number of places which generate cast alignment
warnings, which are difficult or impossible to address. Use
pragmas to disable the warnings in these few places

conf/nwfilter_conf.c: In function 'virNWFilterRuleDetailsParse':
conf/nwfilter_conf.c:1806:16: warning: cast increases required alignment of target type [-Wcast-align]
         item = (nwItemDesc *)((char *)nwf + att[idx].dataIdx);
conf/nwfilter_conf.c: In function 'virNWFilterRuleDefDetailsFormat':
conf/nwfilter_conf.c:3238:16: warning: cast increases required alignment of target type [-Wcast-align]
         item = (nwItemDesc *)((char *)def + att[i].dataIdx);

storage/storage_backend_mpath.c: In function 'virStorageBackendCreateVols':
storage/storage_backend_mpath.c:247:17: warning: cast increases required alignment of target type [-Wcast-align]
         names = (struct dm_names *)(((char *)names) + next);

nwfilter/nwfilter_dhcpsnoop.c: In function 'virNWFilterSnoopDHCPDecode':
nwfilter/nwfilter_dhcpsnoop.c:994:15: warning: cast increases required alignment of target type [-Wcast-align]
         pip = (struct iphdr *) pep->eh_data;
nwfilter/nwfilter_dhcpsnoop.c:1004:11: warning: cast increases required alignment of target type [-Wcast-align]
     pup = (struct udphdr *) ((char *) pip + (pip->ihl << 2));

nwfilter/nwfilter_learnipaddr.c: In function 'procDHCPOpts':
nwfilter/nwfilter_learnipaddr.c:327:33: warning: cast increases required alignment of target type [-Wcast-align]
                 uint32_t *tmp = (uint32_t *)&dhcpopt->value;
nwfilter/nwfilter_learnipaddr.c: In function 'learnIPAddressThread':
nwfilter/nwfilter_learnipaddr.c:501:43: warning: cast increases required alignment of target type [-Wcast-align]
                     struct iphdr *iphdr = (struct iphdr*)(packet +
nwfilter/nwfilter_learnipaddr.c:538:43: warning: cast increases required alignment of target type [-Wcast-align]
                     struct iphdr *iphdr = (struct iphdr*)(packet +
nwfilter/nwfilter_learnipaddr.c:544:48: warning: cast increases required alignment of target type [-Wcast-align]
                         struct udphdr *udphdr= (struct udphdr *)

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2013-04-08 10:03:21 +01:00
Eric Blake
6e148567fd build: make broken -Wlogical-op test be gcc-only
Commit 8b8fcdea introduced a check for broken gcc -Wlogical-op,
but did not guard the check against non-gcc compilers, which might
lead to spurious failures when another compiler encounters an
unknown pragma.  Additionally, all of our compiler warning logic
should belong in a single file, and use cache variables to allow
overriding the decision at configure time if necessary.

* configure.ac (BROKEN_GCC_WLOGICALOP): Move...
* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): ...here,
and update to modern autoconf idioms.
2012-12-19 12:21:09 -07:00
Eric Blake
cf9bced084 build: improve FORTIFY_SOURCE usage
Based on this coreutils bug report, since coreutils copied what
libvirt had done:
https://lists.gnu.org/archive/html/bug-coreutils/2012-10/msg00121.html

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Allow
for user to pre-define _FORTIFY_SOURCE.
2012-10-23 14:00:32 -06:00
Eric Blake
814a8deaa1 build: avoid -Wno-format on new-enough gcc
Commit c579d6b added a sledgehammer to silence spurious warnings from
gcc 4.2, but in the process, it also silenced useful warnings from
gcc 4.3 through 4.5.  As a result, a bug slipped in to commit 0caccb58.

Tested with FreeBSD (gcc 4.2.1), RHEL 6.3 (gcc 4.4), and F17 (gcc 4.7.2),
where the former didn't trip on spurious warnings, and where the latter
two detected a revert of 2b804cf.

* m4/virt-compile-warnings.m4 (-Wno-format): Probe for the actual
spurious message, to once again allow gcc 4.4 to use -Wformat.
2012-10-02 09:48:23 -06:00
Eric Blake
892242519a build: fix build on older gcc
On RHEL 6.2, gcc 4.4.6 complains:
cc1: warning: command line option "-Wenum-compare" is valid for C++/ObjC++ but not for C
which in turn breaks a -Werror build.

Meanwhile, in Fedora 17, gcc 4.7.0, -Wenum-compare has been enhanced
to also work on C, but at the same time, it is documented that -Wall
now implicitly includes -Wenum-compare.

Therefore, it is sufficient to remove explicit checks for this option,
avoiding the warning from older gcc while still getting the
compile-time safety from newer gcc.

* m4/virt-compile-warnings.m4 (-Wenum-compare): Omit explicit check.
2012-09-07 14:15:22 -06:00
Eric Blake
c579d6b3d6 build: avoid warnings from gcc 4.2.1
OpenBSD ships with gcc 4.2.1, which annoyingly treats all format
strings as though they were also attribute((nonnull)).  The two
concepts are orthogonal, though, as evidenced by the number of
spurious warnings it generates on uses where we know that
virReportError specifically handles NULL instead of a format
string; worse, since we now force -Werror on git builds, it
prevents development builds on OpenBSD.

I hate to do this, as it disables ALL format checking on older
gcc, and therefore misses out on some useful checks (code that
happened to compile on Linux may still have type mismatches
when compiled on other platforms, as evidenced by the number
of times I have fixed formatting mismatches for uid_t as found
by warnings on Cygwin), but I don't see any other way to keep
-Werror alive and still compile on OpenBSD.

A more invasive change would be to make virReportError() mark
its format attribute as nonnull, and fix (a lot of) fallout;
we may end up doing that anyways as part of danpb's error
refactoring improvements, but not today.

* src/internal.h (ATTRIBUTE_FMT_PRINTF): Use preferred spellings.
* m4/virt-compile-warnings.m4 (-Wformat): Disable on older gcc.
2012-09-05 12:05:55 -06:00
Eric Blake
42af2167a4 build: avoid warnings on older gcc
A previous patch (c606671a) pulled in a newer version of
stat-time.h from gnulib, which causes some warnings in older gcc:

  CC     libvirt_driver_storage_la-storage_backend.lo
cc1: warnings being treated as errors
In file included from ../../src/storage/storage_backend.c:59:
../../gnulib/lib/stat-time.h:55: error: no previous prototype for 'get_stat_atime_ns' [-Wmissing-prototypes]

Upstream gnulib argues that these warnings are stupid (and I agree;
see <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54113>), and has
used a modern gcc feature (#pragma GCC diagnostic push) to avoid the
warning.  But we still aim to compile on RHEL 6.3, with gcc 4.4.6
(not to mention even older platforms like RHEL 5), and therefore
the warning trips up our default of development with -Werror.

It took me a while to figure out how to make our set of warnings
smaller on older gcc without losing the benefit of the warnings
when using newer gcc (such as the one on Fedora 17), but this
should do the trick.

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Avoid
warnings that gnulib can't silence on older gcc.
2012-08-15 13:30:21 -06:00
Hendrik Schwartke
7383c1d762 Added timestamps to storage volumes
The access, birth, modification and change times are added to
storage volumes and corresponding xml representations.  This
shows up in the XML in this format:

<timestamps>
  <atime>1341933637.027319099</atime>
  <mtime>1341933637.027319099</mtime>
</timestamps>

Signed-off-by: Eric Blake <eblake@redhat.com>
2012-08-02 17:14:17 -06:00
Eric Blake
1c2edf0fbd maint: make it easier to copy FORTIFY_SOURCE snippet
While libvirt intentionally avoids -Wundef (after all, C99
guarantees sane semantics of treating undefined macros as 0),
the glibc insanity of #warning on _FORTIFY_SOURCE coupled with
what some people feel is the black magic of autoconf means
that other projects are likely to copy our snippet verbatim.
We can be nicer to other projects by making it easier to
integrate into projects that use -Wundef.

Suggested by Christophe Fergeau.

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Be nice
to other projects using -Wundef.
2012-06-07 10:52:37 -06:00
Eric Blake
2af63b1c34 build: allow building with newer glibc-headers and -O0
glibc 2.15 (on Fedora 17) coupled with explicit disabling of
optimization during development dies a painful death:

In file included from /usr/include/limits.h:27:0,
                 from /usr/lib/gcc/x86_64-redhat-linux/4.7.0/include/limits.h:169,
                 from /usr/lib/gcc/x86_64-redhat-linux/4.7.0/include/syslimits.h:7,
                 from /usr/lib/gcc/x86_64-redhat-linux/4.7.0/include/limits.h:34,
                 from util/bitmap.c:26:
/usr/include/features.h:314:4: error: #warning _FORTIFY_SOURCE requires compiling with optimization (-O) [-Werror=cpp]
cc1: all warnings being treated as errors

Work around this by only conditionally defining _FORTIFY_SOURCE,
in the case where glibc can actually use it.  The trick is using
AH_VERBATIM instead of AC_DEFINE.

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Squelch
_FORTIFY_SOURCE when needed to avoid glibc #warnings.
2012-06-04 12:08:27 -06:00
Daniel P. Berrange
851117bd7a Enable all warnings permanently & default to -Werror for GIT builds
Given that we auto-detect whether each -Wxxxx flag is supported by
GCC, and we are warning-free and use automake silent rules, there
is no compelling reason to allow compile warnings to be disabled.

Replace the --enable-compile-warnings flag with a simpler
--enable-werror flag, which defaults to 'yes' if building
from GIT, or 'no' if building from tar.gz

This helps ensure that everyone writing patches for libvirt will
take care to fix their warning problems before submitting for
review

* autobuild.sh: Force -Werror
* configure.ac: Update for LIBVIRT_COMPILE_WARNINGS macro change
* m4/virt-compile-warnings.m4: Permanently enable all warnings,
  auto-enable Werror for GIT builds
2012-03-27 17:08:06 +01:00
Eric Blake
d940e3bdb9 build: silence some compiler warnings from gnulib
Gnulib claims that there are some classes of warnings that are
worth enabling during development, but where silencing those
warnings causes code bloat that is not necessary in an optimized
build.  The code bloat to silence the warnings is only enabled
by -Dlint.  Follow the lead of coreutils in setting up -Dlint
whenever full warnings are requested.

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Add
-Dlint, and move _FORTIFY_SOURCE to config.h instead of CFLAGS.
2012-01-19 13:14:10 -07:00
Eric Blake
f17e0e2182 build: reduce warnings from older gcc
Older gcc warns (on every file!) that -Wabi and -Wdeprecated only
make sense on C++ projects.  Newer gcc accepts these warnings for
C, but it is not clear that they can do anything useful, so it
is easier to just drop the warnings altogether.

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Silence
-Wabi and -Wdeprecated on older gcc.
Reported by Peter Krempa.
2011-12-05 10:14:55 -07:00
Eric Blake
75da320087 build: fix build at -O2 on rawhide
I had previously tested commit 059d746 with -O intentionally omitted
from my CFLAGS; but that means that I missed out on this warning
from gcc 4.6.2 when optimizations are enabled:

util/buf.c: In function 'virBufferGetIndent':
util/buf.c:86:1: error: function might be candidate for attribute 'pure' [-Werror=suggest-attribute=pure]

While it is probably a good idea to add the attributes and silence
this warning, it's also invasive; 'make -k' found more than 75 such
complaints.  And it doesn't help that gcc 4.6.2 is still buggy
(coreutils reported a case where gcc 4.6.2 incorrectly suggested
marking a function pure that incremented a global variable; fixed
in gcc 4.7).  So the best fix for now is to disable the warning.

It also doesn't help that I stumbled across another problem - gcc
documents that -Wsuggest-attribute=pure only warns if you use -O,
or if you use -fipa-pure-const.  But in practice, when I omitted -O
but added -fipa-pure-const, the warnings are fickle - I got warnings
for simple compilation that disappeared when I also added -fPIC.
And the way libtool compiles things is with -fPIC first, then without
-fPIC but with errors sent to /dev/null - which meant that without
disabling -Wsuggest-attribute=pure, I got a compile error with no
message.  :(  See http://debbugs.gnu.org/cgi/bugreport.cgi?bug=10197

* m4/virt-compile-warnings.m4 (LIBVIRT_COMPILE_WARNINGS): Silence
-Wsuggest-attribute warnings for now.
2011-12-02 13:00:57 -07:00