- Devel - libvirt List Archives

[PATCH v2] Fix linkage to libutil and libkvm on FreeBSD 11
by Daniel P. Berrangé 10 Sep '20

10 Sep '20

We are currently adding -lutil and -lkvm to the linker using the add_project_link_arguments method. On FreeBSD 11.4, this results in build errors because the args appear too early in the command line. We need to pass the libraries as dependencies so that they get placed at the same point in the linker args as other dependencies. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- meson.build | 17 ++++++++--------- src/bhyve/meson.build | 2 ++ src/util/meson.build | 2 ++ 3 files changed, 12 insertions(+), 9 deletions(-) diff --git a/meson.build b/meson.build index 33eaa9ff56..bcb978292b 100644 --- a/meson.build +++ b/meson.build @@ -1086,7 +1086,8 @@ endif # Check for BSD kvm (kernel memory interface) if host_machine.system() == 'freebsd' kvm_dep = cc.find_library('kvm') - add_project_link_arguments('-lkvm', language: 'c') +else + kvm_dep = dependency('', required: false) endif libiscsi_version = '1.18.0' @@ -1203,11 +1204,10 @@ have_gnu_gettext_tools = false if not get_option('nls').disabled() have_gettext = cc.has_function('gettext') if not have_gettext - intl_lib = cc.find_library('intl', required: false) - have_gettext = intl_lib.found() - if have_gettext - add_project_link_arguments('-lintl', language: 'c') - endif + intl_dep = cc.find_library('intl', required: false) + have_gettext = intl_dep.found() + else + intl_dep = dependency('', required: false) endif if not have_gettext and get_option('nls').enabled() error('gettext() is required to build libvirt') @@ -1235,6 +1235,8 @@ if not get_option('nls').disabled() have_gnu_gettext_tools = true endif endif +else + intl_dep = dependency('', required: false) endif numactl_dep = cc.find_library('numa', required: get_option('numactl')) @@ -1402,9 +1404,6 @@ if udev_dep.found() endif util_dep = cc.find_library('util', required: false) -if util_dep.found() - add_project_link_arguments('-lutil', language: 'c') -endif if not get_option('virtualport').disabled() if cc.has_header_symbol('linux/if_link.h', 'IFLA_PORT_MAX') diff --git a/src/bhyve/meson.build b/src/bhyve/meson.build index 7d54718820..975f93a9c0 100644 --- a/src/bhyve/meson.build +++ b/src/bhyve/meson.build @@ -21,7 +21,9 @@ if conf.has('WITH_BHYVE') ], dependencies: [ access_dep, + kvm_dep, src_dep, + util_dep, ], include_directories: [ conf_inc_dir, diff --git a/src/util/meson.build b/src/util/meson.build index f7092cc3f1..bf556e7ae6 100644 --- a/src/util/meson.build +++ b/src/util/meson.build @@ -187,12 +187,14 @@ virt_util_lib = static_library( dbus_dep, devmapper_dep, gnutls_dep, + intl_dep, libnl_dep, m_dep, numactl_dep, secdriver_dep, src_dep, thread_dep, + util_dep, win32_dep, yajl_dep, ], -- 2.26.2

3 2

[libvirt PATCH] esx: generator: fix free of elements in lists
by Pino Toscano 10 Sep '20

10 Sep '20

When a list is freed, we iterate through all the items, invoking the free function for each; the actual free function called for each element is the function of the actual type of each element, and thus the @_next pointer in the element struct has the same type as the element itself. Currently, the free function gets the parent of the current element type, and invoke its free function to continue freeing the list. However, in case the hierarchy of the classes has more than 1 level (i.e. Class <- SubClass <- SubSubClass), the invoked free function is only the parent class' one, and not the actual base class of the hierarchy. To fix that, change the generator to get the ancestor of a class, and invoking that instead. Also, avoid to set the @_next back, as it is not needed. Fixes commits 5cff36e39ae691fbd7c40597df1732eecf294150 and f76c6dde2e33233566e886d96e76b5fe0c102d9a. Signed-off-by: Pino Toscano <ptoscano(a)redhat.com> --- scripts/esx_vi_generator.py | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/scripts/esx_vi_generator.py b/scripts/esx_vi_generator.py index 863c8af964..be9b3949e8 100755 --- a/scripts/esx_vi_generator.py +++ b/scripts/esx_vi_generator.py @@ -751,13 +751,13 @@ class Object(GenericObject): source += "{\n" if self.features & Object.FEATURE__LIST: - if self.extends is not None: + ancestor = get_ancestor(self) + if ancestor: # avoid "dereferencing type-punned pointer will break # strict-aliasing rules" warnings - source += " esxVI_%s *next = (esxVI_%s *)item->_next;\n\n" \ - % (self.extends, self.extends) - source += " esxVI_%s_Free(&next);\n" % self.extends - source += " item->_next = (esxVI_%s *)next;\n\n" % self.name + source += " esxVI_%s *baseNext = (esxVI_%s *)item->_next;\n" \ + % (ancestor, ancestor) + source += " esxVI_%s_Free(&baseNext);\n\n" % ancestor else: source += " esxVI_%s_Free(&item->_next);\n\n" % self.name @@ -1250,6 +1250,21 @@ def is_known_type(type): type in enums_by_name) +def get_ancestor(obj): + if not obj.extends: + return None + ancestor = None + try: + ancestor = ancestor_by_name[obj.extends] + except KeyError: + parent = objects_by_name[obj.extends] + ancestor = get_ancestor(parent) + if not ancestor: + ancestor = parent.name + ancestor_by_name[name] = ancestor + return ancestor + + def open_file(filename): return open(filename, "wt") @@ -1341,6 +1356,7 @@ managed_objects_by_name = {} enums_by_name = {} methods_by_name = {} block = None +ancestor_by_name = {} # parse input file -- 2.26.2

3 2

[libvirt PATCH 0/2] meson: Add tests_timeout build option
by Andrea Bolognani 10 Sep '20

10 Sep '20

See patch 2/2 for the rationale. Andrea Bolognani (2): meson: Rename tests-related build options meson: Add tests_timeout build option libvirt.spec.in | 2 +- meson.build | 8 +++++--- meson_options.txt | 5 +++-- tests/meson.build | 6 +++--- 4 files changed, 12 insertions(+), 9 deletions(-) -- 2.26.2

5 17

[PATCH] network: allow accept_ra == 0 when enabling ipv6 forwarding
by Ian Wienand 10 Sep '20

10 Sep '20

The checks modified here were added with 00d28a78b5d1f6eaf79f06ac59e31c568af9da37 to avoid losing routes on hosts. However, tools such as systemd-networking and NetworkManager manage RA's in userspace and thus IPv6 may be up and working on an interface even with accept_ra == 0. This modifies the check to only error if an interface's accept_ra is already set to "1"; as noted inline this seems to when it is likely that enabling forwarding may change the RA acceptance behaviour of the interface. I have noticed this because I am using the IPv6 NAT features enabled with 927acaedec7effbe67a154d8bfa0e67f7d08e6c7. I am using this on my laptop which switches between wired and wireless connections; both of which are configured in an unremarkable way by Fedora's NetworkManager and get configured for IPv6 via SLAAC and whatever NetworkManager magic it does. With this I can define and start a libvirt network with <nat ipv6="yes"> and <ip family='ipv6' address='fc00:abcd:ef12:34::' prefix='64'> and it seems to "just work" for guests. Signed-off-by: Ian Wienand <iwienand(a)redhat.com> --- src/util/virnetdevip.c | 41 +++++++++++++++++++++++++++-------------- 1 file changed, 27 insertions(+), 14 deletions(-) diff --git a/src/util/virnetdevip.c b/src/util/virnetdevip.c index 409f062c5c..de27cacfc9 100644 --- a/src/util/virnetdevip.c +++ b/src/util/virnetdevip.c @@ -496,7 +496,7 @@ virNetDevIPGetAcceptRA(const char *ifname) } struct virNetDevIPCheckIPv6ForwardingData { - bool hasRARoutes; + bool hasKernelRARoutes; /* Devices with conflicting accept_ra */ char **devices; @@ -552,15 +552,26 @@ virNetDevIPCheckIPv6ForwardingCallback(struct nlmsghdr *resp, if (!ifname) return -1; - accept_ra = virNetDevIPGetAcceptRA(ifname); - VIR_DEBUG("Checking route for device %s (%d), accept_ra: %d", ifname, ifindex, accept_ra); - if (accept_ra != 2 && virNetDevIPCheckIPv6ForwardingAddIF(data, &ifname) < 0) + accept_ra = virNetDevIPGetAcceptRA(ifname); + /* 0 = do no accept RA + * 1 = accept if forwarding disabled + * 2 = ovveride and accept RA when forwarding enabled + * + * When RA is managed by userspace (systemd-networkd or + * NetworkManager) accept_ra is unset and we don't need to + * worry about it. If it is 1, enabling forwarding might + * change the behaviour so the user needs to be warned. + */ + if (accept_ra == 0) + return 0; + + if (accept_ra == 1 && virNetDevIPCheckIPv6ForwardingAddIF(data, &ifname) < 0) return -1; - data->hasRARoutes = true; + data->hasKernelRARoutes = true; return 0; } @@ -590,11 +601,13 @@ virNetDevIPCheckIPv6ForwardingCallback(struct nlmsghdr *resp, VIR_DEBUG("Checking multipath route nexthop device %s (%d), accept_ra: %d", ifname, nh->rtnh_ifindex, accept_ra); - if (accept_ra != 2 && virNetDevIPCheckIPv6ForwardingAddIF(data, &ifname) < 0) - return -1; + if (accept_ra == 1) { + if (virNetDevIPCheckIPv6ForwardingAddIF(data, &ifname) < 0) + return -1; + data->hasKernelRARoutes = true; + } VIR_FREE(ifname); - data->hasRARoutes = true; len -= NLMSG_ALIGN(nh->rtnh_len); VIR_WARNINGS_NO_CAST_ALIGN @@ -613,7 +626,7 @@ virNetDevIPCheckIPv6Forwarding(void) struct rtgenmsg genmsg; size_t i; struct virNetDevIPCheckIPv6ForwardingData data = { - .hasRARoutes = false, + .hasKernelRARoutes = false, .devices = NULL, .ndevices = 0 }; @@ -644,11 +657,11 @@ virNetDevIPCheckIPv6Forwarding(void) goto cleanup; } - valid = !data.hasRARoutes || data.ndevices == 0; + valid = !data.hasKernelRARoutes || data.ndevices == 0; /* Check the global accept_ra if at least one isn't set on a per-device basis */ - if (!valid && data.hasRARoutes) { + if (!valid && data.hasKernelRARoutes) { int accept_ra = virNetDevIPGetAcceptRA(NULL); valid = accept_ra == 2; VIR_DEBUG("Checked global accept_ra: %d", accept_ra); @@ -663,9 +676,9 @@ virNetDevIPCheckIPv6Forwarding(void) } virReportError(VIR_ERR_INTERNAL_ERROR, - _("Check the host setup: enabling IPv6 forwarding with " - "RA routes without accept_ra set to 2 is likely to cause " - "routes loss. Interfaces to look at: %s"), + _("Check the host setup: interface has accept_ra set to 1 " + "and enabling forwarding without accept_ra set to 2 is " + "likely to cause routes loss. Interfaces to look at: %s"), virBufferCurrentContent(&buf)); } -- 2.26.2

3 8

[libvirt] [PATCH-for-4.2] hw/mips: Deprecate the r4k machine
by Philippe Mathieu-Daudé 10 Sep '20

10 Sep '20

The r4k machine was introduced in 2005 (6af0bf9c7) and its last logical change was in 2005 (9542611a6). After we can count 164 maintenance commits (QEMU API changes) with the exception of 1 fix in 2015 (memory leak, commit 3ad9fd5a). This machine was introduced as a proof of concept to run a MIPS CPU. 2 years later, the Malta machine was add (commit 5856de80) modeling a real platform. Note also this machine has no specification except 5 lines in the header of this file: * emulates a simple machine with ISA-like bus. * ISA IO space mapped to the 0x14000000 (PHYS) and * ISA memory at the 0x10000000 (PHYS, 16Mb in size). * All peripherial devices are attached to this "bus" with * the standard PC ISA addresses. It is time to deprecate this obsolete machine. Users are recommended to use the Malta board, which hardware is well documented. Signed-off-by: Philippe Mathieu-Daudé <philmd(a)redhat.com> --- qemu-deprecated.texi | 5 +++++ hw/mips/mips_r4k.c | 1 + MAINTAINERS | 2 +- 3 files changed, 7 insertions(+), 1 deletion(-) diff --git a/qemu-deprecated.texi b/qemu-deprecated.texi index 4b4b7425ac..05265b43c8 100644 --- a/qemu-deprecated.texi +++ b/qemu-deprecated.texi @@ -266,6 +266,11 @@ The 'scsi-disk' device is deprecated. Users should use 'scsi-hd' or @section System emulator machines +@subsection mips r4k platform (since 4.2) + +This machine type is very old and unmaintained. Users should use the 'malta' +machine type instead. + @subsection pc-0.12, pc-0.13, pc-0.14 and pc-0.15 (since 4.0) These machine types are very old and likely can not be used for live migration diff --git a/hw/mips/mips_r4k.c b/hw/mips/mips_r4k.c index 70024235ae..0b79ad26cb 100644 --- a/hw/mips/mips_r4k.c +++ b/hw/mips/mips_r4k.c @@ -294,6 +294,7 @@ void mips_r4k_init(MachineState *machine) static void mips_machine_init(MachineClass *mc) { + mc->deprecation_reason = "use malta machine type instead"; mc->desc = "mips r4k platform"; mc->init = mips_r4k_init; mc->block_default_type = IF_IDE; diff --git a/MAINTAINERS b/MAINTAINERS index 5e5e3e52d6..3b3a88e264 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -972,7 +972,7 @@ F: hw/net/mipsnet.c R4000 M: Aurelien Jarno <aurelien(a)aurel32.net> R: Aleksandar Rikalo <aleksandar.rikalo(a)rt-rk.com> -S: Maintained +S: Obsolete F: hw/mips/mips_r4k.c Fulong 2E -- 2.21.0

8 21

[PATCH 0/3] Several small documentation fixes
by Matt Coleman 10 Sep '20

10 Sep '20

This patchset fixes several spelling and punctuation errors in the documentation. Where British vs. American conventions differ, I used Google Trends to determine which form is most commonly used, globally. Matt Coleman (3): Fix a typo in compiling.html.in: that -> from Fix errors in contribute.html.in Fix a typo in hacking.rst: "contirbute" -> "contribute" docs/compiling.html.in | 2 +- docs/contribute.html.in | 6 +++--- docs/hacking.rst | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) -- 2.27.0

4 8

[PATCH] meson: src: Fix DESTDIR handling while creating dirs
by Jan Kiszka 09 Sep '20

09 Sep '20

From: Jan Kiszka <jan.kiszka(a)siemens.com> If the target path contains a link with an absolute path (e.g. /var/run -> /run), makedirs will target the wrong location. Resolve the path first, then append DESTDIR again if needed. Signed-off-by: Jan Kiszka <jan.kiszka(a)siemens.com> --- scripts/meson-install-dirs.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/scripts/meson-install-dirs.py b/scripts/meson-install-dirs.py index 14ec6b93e8..e762e44712 100644 --- a/scripts/meson-install-dirs.py +++ b/scripts/meson-install-dirs.py @@ -6,4 +6,7 @@ import sys destdir = os.environ.get('DESTDIR', os.sep) for dirname in sys.argv[1:]: - os.makedirs(os.path.join(destdir, dirname.strip(os.sep)), exist_ok=True) + realdir = os.path.realpath(os.path.join(destdir, dirname.strip(os.sep))) + if not realdir.startswith(destdir): + realdir = os.path.join(destdir, realdir.strip(os.sep)) + os.makedirs(realdir, exist_ok=True) -- 2.26.2

5 8

[PATCH] disk storage: fix allocation size for pool format dos
by Sebastian Mitterle 09 Sep '20

09 Sep '20

The changed condition was always false because the function was always called with boundary values 0. Use the free extent's start value to get its start offset from the cylinder boundary and determine if the needed size for allocation needs to be expanded too in case the offset doesn't fit within extra bytes for alignment. This fixes an issue where vol-create-from will call qemu-img convert to create a destination volume of same capacity as the source volume and qemu-img will error 'Cannot grow device files' due to the partition being too small for the source although both destination partition and source volume have the same capacity. Signed-off-by: Sebastian Mitterle <smitterl(a)redhat.com> --- src/storage/storage_backend_disk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/storage/storage_backend_disk.c b/src/storage/storage_backend_disk.c index a6d4e41220..ec0679d353 100644 --- a/src/storage/storage_backend_disk.c +++ b/src/storage/storage_backend_disk.c @@ -691,7 +691,7 @@ virStorageBackendDiskPartBoundaries(virStoragePoolObjPtr pool, if (def->source.format == VIR_STORAGE_POOL_DISK_DOS) { /* align to cylinder boundary */ neededSize += extraBytes; - if ((*start % cylinderSize) > extraBytes) { + if ((dev->freeExtents[i].start % cylinderSize) > extraBytes) { /* add an extra cylinder if the offset can't fit within the extra bytes we have */ neededSize += cylinderSize; -- 2.25.2

2 2

[PATCH] doc: add some examples for IPv6 NAT configuration
by Ian Wienand 09 Sep '20

09 Sep '20

Add some expanded examples for the nat ipv6 introduced with 927acaedec7effbe67a154d8bfa0e67f7d08e6c7. Unfortunately while for IPv4 it's well-known what addresses ranges are useful for NAT, with IPv6 unless you enjoy digging through RFC's going back-and-forth over unique local addresses and the meaning of the word "site" it's generally much less obvious. I've tried to add some details on choosing a range inline with RFC 4193 and then some pointers for when it maybe doesn't work in the guest as you first expect despite you doing what the RFC's say! Signed-off-by: Ian Wienand <iwienand(a)redhat.com> --- docs/formatnetwork.html.in | 47 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 47 insertions(+) diff --git a/docs/formatnetwork.html.in b/docs/formatnetwork.html.in index fb740111b1..94a4cab4d1 100644 --- a/docs/formatnetwork.html.in +++ b/docs/formatnetwork.html.in @@ -1209,6 +1209,53 @@ </ip> </network></pre> + <h3><a id="examplesNATv6">IPv6 NAT based network</a></h3> + + <p> + Below is a variation for also providing IPv6 NAT. This can be + especially useful when using multiple interfaces where some, + such as WiFi cards, can not be bridged (usually on a laptop), + making it difficult to provide end-to-end IPv6 routing. + </p> + + <pre> +<network> + <name>default6</name> + <bridge name="virbr0"/> + <forward mode="nat"> + <nat ipv6='yes'> + <port start='1024' end='65535'/> + </nat> + + <ip address="192.168.122.1" netmask="255.255.255.0"> + <dhcp> + <range start="192.168.122.2" end="192.168.122.254"/> + </dhcp> + </ip> + <ip family="ipv6" address="fdXX:XXXX:XXXX:NNNN:: prefix="64"/> + </ip> +</network></pre> + + <p>IPv6 NAT addressing has some caveats over the more straight + forward IPv4 case. + <a href="https://tools.ietf.org/html/rfc4193">RFC 4193</a> + defines the address range <tt>fd00::/8</tt> for <tt>/48</tt> IPv6 + private networks. It should be concatenated with a random 40-bit + string (i.e. 10 random hexadecimal digits replacing the <tt>X</tt> + values above, RFC 4193 provides + an <a href="https://tools.ietf.org/html/rfc4193#section-3.2.2">algorithm</a> + if you do not have a source of sufficient randomness). This + leaves <tt>0</tt> through <tt>ffff</tt> for subnets (<tt>N</tt> + above) which you can use at will.</p> + + <p>Many operating systems will not consider these addresses as + preferential to IPv4, due to some practial history of these + addresses being present but unroutable and causing networking + issues. On many Linux distributions, you may need to + override <tt>/etc/gai.conf</tt> with values + from <a href="https://www.ietf.org/rfc/rfc3484.txt">RFC 3484</a> + to have your IPv6 NAT network correctly preferenced over IPv4.</p> + <h3><a id="examplesRoute">Routed network config</a></h3> <p> -- 2.26.2

2 2

[PATCH] client: fix memory leak in client msg
by Hao Wang 09 Sep '20

09 Sep '20

>From 3ad3fae4f2562a11bef8dcdd25b6a7e0b00d4e2c Mon Sep 17 00:00:00 2001 From: Hao Wang <wanghao232(a)huawei.com> Date: Sat, 18 Jul 2020 15:43:30 +0800 Subject: [PATCH] client: fix memory leak in client msg When closing client->waitDispatch in virNetClientIOEventLoopRemoveAll or virNetClientIOEventLoopRemoveDone, VIR_FREE() is called to free call->msg directly, resulting in leak of the memory call->msg->buffer points to. Use virNetMessageFree(call->msg) instead of VIR_FREE(call->msg). Signed-off-by: Hao Wang <wanghao232(a)huawei.com> --- src/rpc/virnetclient.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c index 441f1502a6..f899493b64 100644 --- a/src/rpc/virnetclient.c +++ b/src/rpc/virnetclient.c @@ -1519,7 +1519,7 @@ static bool virNetClientIOEventLoopRemoveDone(virNetClientCallPtr call, if (call->expectReply) VIR_WARN("Got a call expecting a reply but without a waiting thread"); virCondDestroy(&call->cond); - VIR_FREE(call->msg); + virNetMessageFree(call->msg); VIR_FREE(call); } @@ -1546,7 +1546,7 @@ virNetClientIOEventLoopRemoveAll(virNetClientCallPtr call, VIR_DEBUG("Removing call %p", call); virCondDestroy(&call->cond); - VIR_FREE(call->msg); + virNetMessageFree(call->msg); VIR_FREE(call); return true; } -- 2.23.0

3 2