[PATCH] conf: Generate MAC address instead of keeping all zeroes
by Martin Kletzander
When we parse <mac address="00:00:00:00:00:00"/> we keep that in memory
and pass it down to the hypervisor. However, that MAC address is not
strictly valid as it is not marked as locally administered (bit 0x02)
but it is not even globally unique. It is also used for loopback device
on Linux, for example. And QEMU sees such MAC address just as "not
specified" and generates a new one that libvirt does not even know
about. So to make the overall experience better we now generate it if
the supplied one is all clear.
Resolves: https://issues.redhat.com/browse/RHEL-974
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
src/conf/domain_conf.c | 2 +-
src/util/virmacaddr.c | 5 +++++
src/util/virmacaddr.h | 1 +
.../network-interface-mac-clear.xml | 21 +++++++++++++++++++
.../network-interface-mac-clear.xml | 21 +++++++++++++++++++
tests/genericxml2xmltest.c | 4 +++-
6 files changed, 52 insertions(+), 2 deletions(-)
create mode 100644 tests/genericxml2xmlindata/network-interface-mac-clear.xml
create mode 100644 tests/genericxml2xmloutdata/network-interface-mac-clear.xml
diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
index bb4f1fdb948d..652bd09b21b8 100644
--- a/src/conf/domain_conf.c
+++ b/src/conf/domain_conf.c
@@ -9675,7 +9675,7 @@ virDomainNetDefParseXML(virDomainXMLOption *xmlopt,
return NULL;
}
- if (!macaddr) {
+ if (!macaddr || virMacAddrIsAllClear(&def->mac)) {
virDomainNetGenerateMAC(xmlopt, &def->mac);
def->mac_generated = true;
}
diff --git a/src/util/virmacaddr.c b/src/util/virmacaddr.c
index 073f298b5b66..e06bb200fc68 100644
--- a/src/util/virmacaddr.c
+++ b/src/util/virmacaddr.c
@@ -246,6 +246,11 @@ virMacAddrIsBroadcastRaw(const unsigned char s[VIR_MAC_BUFLEN])
return memcmp(virMacAddrBroadcastAddrRaw, s, sizeof(*s)) == 0;
}
+bool virMacAddrIsAllClear(const virMacAddr *addr)
+{
+ return !virMacAddrCmpRaw(addr, (const unsigned char[VIR_MAC_BUFLEN]){0});
+}
+
void
virMacAddrFree(virMacAddr *addr)
{
diff --git a/src/util/virmacaddr.h b/src/util/virmacaddr.h
index f32b58805a61..7b9eb7443bd1 100644
--- a/src/util/virmacaddr.h
+++ b/src/util/virmacaddr.h
@@ -58,6 +58,7 @@ int virMacAddrParseHex(const char* str,
bool virMacAddrIsUnicast(const virMacAddr *addr);
bool virMacAddrIsMulticast(const virMacAddr *addr);
bool virMacAddrIsBroadcastRaw(const unsigned char s[VIR_MAC_BUFLEN]);
+bool virMacAddrIsAllClear(const virMacAddr *addr);
void virMacAddrFree(virMacAddr *addr);
G_DEFINE_AUTOPTR_CLEANUP_FUNC(virMacAddr, virMacAddrFree);
diff --git a/tests/genericxml2xmlindata/network-interface-mac-clear.xml b/tests/genericxml2xmlindata/network-interface-mac-clear.xml
new file mode 100644
index 000000000000..41beda8a79bb
--- /dev/null
+++ b/tests/genericxml2xmlindata/network-interface-mac-clear.xml
@@ -0,0 +1,21 @@
+<domain type='qemu'>
+ <name>QEMUGuest1</name>
+ <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid>
+ <memory unit='KiB'>219136</memory>
+ <currentMemory unit='KiB'>219136</currentMemory>
+ <vcpu placement='static'>1</vcpu>
+ <os>
+ <type arch='x86_64' machine='pc'>hvm</type>
+ <boot dev='hd'/>
+ </os>
+ <clock offset='utc'/>
+ <on_poweroff>destroy</on_poweroff>
+ <on_reboot>restart</on_reboot>
+ <on_crash>destroy</on_crash>
+ <devices>
+ <interface type='bridge'>
+ <mac address='00:00:00:00:00:00'/>
+ <source bridge='br0'/>
+ </interface>
+ </devices>
+</domain>
diff --git a/tests/genericxml2xmloutdata/network-interface-mac-clear.xml b/tests/genericxml2xmloutdata/network-interface-mac-clear.xml
new file mode 100644
index 000000000000..a7935fa9f4de
--- /dev/null
+++ b/tests/genericxml2xmloutdata/network-interface-mac-clear.xml
@@ -0,0 +1,21 @@
+<domain type='qemu'>
+ <name>QEMUGuest1</name>
+ <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid>
+ <memory unit='KiB'>219136</memory>
+ <currentMemory unit='KiB'>219136</currentMemory>
+ <vcpu placement='static'>1</vcpu>
+ <os>
+ <type arch='x86_64' machine='pc'>hvm</type>
+ <boot dev='hd'/>
+ </os>
+ <clock offset='utc'/>
+ <on_poweroff>destroy</on_poweroff>
+ <on_reboot>restart</on_reboot>
+ <on_crash>destroy</on_crash>
+ <devices>
+ <interface type='bridge'>
+ <mac address='52:54:00:00:00:00'/>
+ <source bridge='br0'/>
+ </interface>
+ </devices>
+</domain>
diff --git a/tests/genericxml2xmltest.c b/tests/genericxml2xmltest.c
index 3501eadf5597..bf160b7e0bef 100644
--- a/tests/genericxml2xmltest.c
+++ b/tests/genericxml2xmltest.c
@@ -187,6 +187,7 @@ mymain(void)
DO_TEST("cpu-cache-disable");
DO_TEST("network-interface-mac-check");
+ DO_TEST_DIFFERENT("network-interface-mac-clear");
DO_TEST_DIFFERENT("chardev-tcp");
DO_TEST_FAIL_ACTIVE("chardev-tcp-missing-host");
@@ -255,4 +256,5 @@ mymain(void)
return ret == 0 ? EXIT_SUCCESS : EXIT_FAILURE;
}
-VIR_TEST_MAIN(mymain)
+VIR_TEST_MAIN_PRELOAD(mymain,
+ VIR_TEST_MOCK("virrandom"))
--
2.42.0
1 year, 7 months
[PATCH (pushed)] build: Fix logic bug determining whether running with optimization
by Peter Krempa
The conversion from ternary to a 'if' clause was wrong and thus didn't
properly increase the stack size where needed but only where not
actually needed.
Fixes: b68faa99d9f16c2f504b23737040d25d072ee85d
Signed-off-by: Peter Krempa <pkrempa(a)redhat.com>
---
meson.build | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meson.build b/meson.build
index e0ee4f2f21..3bf15d93a4 100644
--- a/meson.build
+++ b/meson.build
@@ -255,7 +255,7 @@ if cc.get_id() == 'clang' and get_option('optimization') == '0'
endif
# sanitizer instrumentation may enlarge stack frames
-if get_option('b_sanitize') == 'none'
+if get_option('b_sanitize') != 'none'
stack_frame_size = 32768
endif
--
2.41.0
1 year, 7 months
[PATCH (pushed)] build: Fix assignment into 'stack_frame_size' when sanitizer is enabled
by Peter Krempa
Instead of an assignment into the 'stack_frame_size' variable when
sanitizers are enabled I've accidentally compared the value against the
requested size.
Fix the typo.
Fixes: b68faa99d9f16c2f504b23737040d25d072ee85d
Signed-off-by: Peter Krempa <pkrempa(a)redhat.com>
---
meson.build | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meson.build b/meson.build
index d29761bc71..e0ee4f2f21 100644
--- a/meson.build
+++ b/meson.build
@@ -256,7 +256,7 @@ endif
# sanitizer instrumentation may enlarge stack frames
if get_option('b_sanitize') == 'none'
- stack_frame_size == 32768
+ stack_frame_size = 32768
endif
# array_bounds=2 check triggers false positive on some GCC
--
2.41.0
1 year, 7 months
[PATCH 0/2] build: Fix build with clang without optimization
by Peter Krempa
Add a note for users how optimization is supposed to be controled with
meson and fix the build by using 4k stack flame on clang without
optimization.
Peter Krempa (2):
docs: compiling: Add a note about use of CFLAGS for optimization
build: Work around clang's stack size calculation without optimization
docs/compiling.rst | 5 +++++
meson.build | 11 ++++++++++-
2 files changed, 15 insertions(+), 1 deletion(-)
--
2.41.0
1 year, 7 months
[PATCH] Revert "build: Decrease maximum stack frame size to 2048"
by Peter Krempa
Build fails with this patch with 'clang' when optimizations are
disabled in long functions which have many helper variables declared in
nested blocks such as for-loops.
As there is no clean solution for now, let's keep the stack frame size
at 4k.
This reverts commit 42bc76cdb8486ef502200f3bce9e3faebdd78103.
Signed-off-by: Peter Krempa <pkrempa(a)redhat.com>
---
meson.build | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meson.build b/meson.build
index 6b6f7ccb7c..dd7a4e8d34 100644
--- a/meson.build
+++ b/meson.build
@@ -248,7 +248,7 @@ alloc_max = run_command(
)
# sanitizer instrumentation may enlarge stack frames
-stack_frame_size = get_option('b_sanitize') == 'none' ? 2048 : 32768
+stack_frame_size = get_option('b_sanitize') == 'none' ? 4096 : 32768
# array_bounds=2 check triggers false positive on some GCC
# versions when using sanitizers. Seen on Fedora 34 with
--
2.41.0
1 year, 7 months
[PATCH] docs, passt: Clarify some niche passt usage
by Martin Kletzander
Change example logfile path and clarify how complicated all things passt
are. I chose not to create the non-existing directory because it could
open a whole new can of worms.
Also explain missing `dev` attribute of `<portForward/>`
Resolves: https://issues.redhat.com/browse/RHEL-1833
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
docs/formatdomain.rst | 21 +++++++++++++++------
1 file changed, 15 insertions(+), 6 deletions(-)
diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst
index 68f54ab3ed62..bc469e5f9f94 100644
--- a/docs/formatdomain.rst
+++ b/docs/formatdomain.rst
@@ -4902,14 +4902,23 @@ When the passt backend is used, the ``<backend>`` attribute
``logFile`` can be used to tell the passt process for this interface
where to write its message log, and the ``<source>`` attribute ``dev``
can tell it to use a particular host interface to derive the routes
-given to the guest for forwarding traffic upstream.
+given to the guest for forwarding traffic upstream. Due to the design
+decisions of passt, if using SELinux, the log file is recommended to
+reside in the runtime directory of a user under which the passt
+process will run, most probably ``/run/user/$UID`` where ``$UID`` is
+the UID of the user, e.g. ``qemu``. Beware that libvirt does not
+create this directory if it does not already exist to avoid possible,
+however unlikely, issues, especially since this logfile attribute is
+meant mostly for debugging.
Additionally, when passt is used, multiple ``<portForward>`` elements
can be added to forward incoming network traffic for the host to this
guest interface. Each ``<portForward>`` must have a ``proto``
-attribute (set to ``tcp`` or ``udp``) and optional original
-``address`` (if not specified, then all incoming sessions to any host
-IP for the given proto/port(s) will be forwarded to the guest).
+attribute (set to ``tcp`` or ``udp``), optional original ``address``
+(if not specified, then all incoming sessions to any host IP for the
+given proto/port(s) will be forwarded to the guest), and an optional
+``dev`` attribute to limit the forwarded traffic to a specific host
+interface.
The decision of which ports to forward is described with zero or more
``<range>`` subelements of ``<portForward>`` (if there is no
@@ -4934,7 +4943,7 @@ ports **with the exception of some subset**.
<devices>
...
<interface type='user'>
- <backend type='passt' logFile='/tmp/passt.log'/>
+ <backend type='passt' logFile='/run/user/$UID/passt-domain.log'/>
<mac address="00:11:22:33:44:55"/>
<source dev='eth0'/>
<ip family='ipv4' address='172.17.2.4' prefix='24'/>
@@ -4946,7 +4955,7 @@ ports **with the exception of some subset**.
<range start='5000' end='5020' to='6000'/>
<range start='5010' end='5015' exclude='yes'/>
</portForward>
- <portForward proto='tcp' address='2001:db8:ac10:fd01::1:10'>
+ <portForward proto='tcp' address='2001:db8:ac10:fd01::1:10' dev='eth0'>
<range start='80'/>
<range start='443' to='344'/>
</portForward>
--
2.41.0
1 year, 7 months
[PATCH] conf, schema: Switch iothread/poll values to unsignedLong
by Martin Kletzander
They represent nanoseconds, and we accept such values already. Not that
anyone would use such values in the wild, but even one person testing
QEMU could put in a bigger value and will be bothered with validation
errors after every `virsh edit`. Also add a test for it.
Resolves: https://issues.redhat.com/browse/RHEL-1717
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
src/conf/schemas/domaincommon.rng | 6 +++---
tests/genericxml2xmlindata/iothreadids.xml | 23 ++++++++++++++++++++++
tests/genericxml2xmltest.c | 2 ++
3 files changed, 28 insertions(+), 3 deletions(-)
create mode 100644 tests/genericxml2xmlindata/iothreadids.xml
diff --git a/src/conf/schemas/domaincommon.rng b/src/conf/schemas/domaincommon.rng
index de3bd1c35c55..2f9ba31c0aec 100644
--- a/src/conf/schemas/domaincommon.rng
+++ b/src/conf/schemas/domaincommon.rng
@@ -876,17 +876,17 @@
<element name="poll">
<optional>
<attribute name="max">
- <ref name="unsignedInt"/>
+ <ref name="unsignedLong"/>
</attribute>
</optional>
<optional>
<attribute name="grow">
- <ref name="unsignedInt"/>
+ <ref name="unsignedLong"/>
</attribute>
</optional>
<optional>
<attribute name="shrink">
- <ref name="unsignedInt"/>
+ <ref name="unsignedLong"/>
</attribute>
</optional>
</element>
diff --git a/tests/genericxml2xmlindata/iothreadids.xml b/tests/genericxml2xmlindata/iothreadids.xml
new file mode 100644
index 000000000000..671a4672958d
--- /dev/null
+++ b/tests/genericxml2xmlindata/iothreadids.xml
@@ -0,0 +1,23 @@
+<domain type='kvm'>
+ <name>foo</name>
+ <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid>
+ <memory unit='KiB'>219136</memory>
+ <currentMemory unit='KiB'>219136</currentMemory>
+ <vcpu placement='static'>1</vcpu>
+ <iothreads>1</iothreads>
+ <iothreadids>
+ <iothread id='8' thread_pool_min='2147483647' thread_pool_max='2147483647'>
+ <poll max='9223372036854775807' grow='456' shrink='789'/>
+ </iothread>
+ </iothreadids>
+ <os>
+ <type arch='i686' machine='pc'>hvm</type>
+ <boot dev='hd'/>
+ </os>
+ <clock offset='utc'/>
+ <on_poweroff>destroy</on_poweroff>
+ <on_reboot>restart</on_reboot>
+ <on_crash>destroy</on_crash>
+ <devices>
+ </devices>
+</domain>
diff --git a/tests/genericxml2xmltest.c b/tests/genericxml2xmltest.c
index 3501eadf5597..ce8073e85a30 100644
--- a/tests/genericxml2xmltest.c
+++ b/tests/genericxml2xmltest.c
@@ -249,6 +249,8 @@ mymain(void)
DO_TEST("cpu-phys-bits-emulate");
DO_TEST("cpu-phys-bits-passthrough");
+ DO_TEST("iothreadids");
+
virObjectUnref(caps);
virObjectUnref(xmlopt);
--
2.41.0
1 year, 7 months
[libvirt PATCH 00/33] ci: Unify the GitLab CI jobs with local executions && adopt lcitool container executions
by Erik Skultety
Technically a v2 of:
https://listman.redhat.com/archives/libvir-list/2023-February/237552.html
However, the approach here is slightly different and what that series said
about migration to lcitool container executions as a replacement for
ci/Makefile is actually done here. One of the core problems of the above
pointed out in review was that more Shell logic was introduced including CLI
parsing, conditional executions, etc. which we fought hard to get rid of in the
past. I reworked the Shell functions quite a bit and dropped whatever extra
Shell logic the original series added.
Obviously we can't get rid of Shell completely because of .gitlab-ci.yml and so
I merely extracted the recipes into functions which are then sourced as
ci/build.sh and executed. Now, that on its own would hide the actual commands
being run in the GitLab job log, so before any command is actually executed, it
is formatted with a color sequence so we don't miss that information as that
would be a regression to the status quo.
Lastly, this series then takes the effort inside the ci/build.sh script and
basically mirrors whatever GitLab would do to run a job inside a local
container which is executed by lcitool (yes, we already have that capability).
Please give this a try and I'm already looking forward to comments as I'd like
to expand this effort to local VM executions running the TCK integration tests,
so this series is quite important in that regard.
Erik Skultety (33):
ci: build.sh: Add variables from .gitlab-ci.yml
ci: build.sh: Add GIT_ROOT env helper variable
ci: build.sh: Don't mention that MESON_ARGS are available via CLI
ci: build.sh: Add a wrapper function over meson's setup
ci: build.sh: Add a wrapper function executing 'shell' commands
ci: build.sh: Add a wrapper function over the 'build' job
ci: build.sh: Add a helper function to create the dist tarball
ci: build.sh: Add a wrapper function over the 'test' job
ci: build.sh: Add a wrapper function over the 'codestyle' job
ci: build.sh: Add a wrapper function over the 'potfile' job
ci: build.sh: Add a wrapper function over the 'rpmbuild' job
ci: build.sh: Add a wrapper function over the 'website' job
ci: build.sh: Drop changing working directory to CI_CONT_DIR
ci: build.sh: Drop direct invocation of meson/ninja commands
ci: build.sh: Drop MESON_ARGS definition from global level
gitlab-ci.yml: Add 'after_script' stage to prep for artifact
collection
.gitlab-ci.yml: Convert the native build job to the build.sh usage
.gitlab-ci.yml: Convert the cross build job to the build.sh usage
.gitlab-ci.yml: Convert the website build job to the build.sh usage
.gitlab-ci.yml: Convert the codestyle job to the build.sh usage
.gitlab-ci.yml: Convert the potfile job to the build.sh usage
ci: helper: Drop _lcitool_get_targets method
ci: helper: Don't make ':' literal a static part of the image tag
ci: helper: Add --lcitool-path CLI option
ci: helper: Add a job argparse subparser
ci: helper: Add a required_deps higher order helper/decorator
ci: helper: Add Python code hangling git clones
ci: helper: Add a helper to create a local repo clone Pythonic way
ci: helper: Rework _lcitool_run method logic
ci: helper: Add an action to run the container workload via lcitool
ci: helper: Drop original actions
ci: helper: Drop the --meson-args/--ninja-args CLI options
ci: helper: Drop the _make_run method
.gitlab-ci.yml | 47 +++++++------
ci/build.sh | 105 +++++++++++++++++++++++++----
ci/helper | 176 ++++++++++++++++++++++++++++---------------------
3 files changed, 218 insertions(+), 110 deletions(-)
--
2.41.0
1 year, 7 months
[PATCH] tests: Remove unused symlink
by Martin Kletzander
The test does not use VIR_TEST_DIFFERENT anyway, so it's probably a
leftover.
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
Pushed as trivial
tests/genericxml2xmloutdata/device-backenddomain.xml | 1 -
1 file changed, 1 deletion(-)
delete mode 120000 tests/genericxml2xmloutdata/device-backenddomain.xml
diff --git a/tests/genericxml2xmloutdata/device-backenddomain.xml b/tests/genericxml2xmloutdata/device-backenddomain.xml
deleted file mode 120000
index f19471e3b50c..000000000000
--- a/tests/genericxml2xmloutdata/device-backenddomain.xml
+++ /dev/null
@@ -1 +0,0 @@
-../genericxml2xmlindata/device-backenddomain.xml
\ No newline at end of file
--
2.42.0
1 year, 7 months
