Move networkMacMgrFileName into src/util/virmacmap.c and rename to
virMacMapFileName. We're about to move some more MacMgr processing
files into virnetworkobj and it doesn't make sense to have this helper
in the driver or in virnetworkobj.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Rather than assuming that what's passed to virObject{Ref|Unref}
would be a virObjectPtr as long as it's not NULL, let's do the
similar checks virObjectIsClass in order to prevent a possible
increment or decrement to some field at the obj->u.s.refs offset.
Signed-off-by: John Ferlan <jferlan@redhat.com>
The virObjectIsClass API has only ever checked object validity
based on if the @obj is not NULL and it was derived from some class.
While this has worked well in general, there is one additional
check that could be made prior to calling virClassIsDerivedFrom
which loops through the classes checking the magic number against
the klass expected magic number.
If by chance a non virObject is passed, rather than assuming the
void * @obj is a _virObject and thus offsetting to obj->klass,
obj->magic, and obj->parent, let's check that the void * @obj
has at least the "base part" of the magic number in the right
place and generate a more specific VIR_WARN message if not.
There are many consumers to virObjectIsClass, include the locking
primitives virObject{Lock|Unlock}, virObjectRWLock{Read|Write},
and virObjectRWUnlock. For those callers, the locking call will
not fail, but it also will not attempt a virMutex* call which
will "most likely" fail since the &obj->lock is used.
In order to avoid some possible future wrap on the 0xCAFExxxx
value, add a check during initialization that some new class
won't cause the wrap. Should be good for a few years at least!
It is still left up to the caller to handle the failed API calls
just as it would be if it passed a NULL opaque pointer anyobj.
If virObjectIsClass fails "internally" to virobject.c, create a
macro to generate the VIR_WARN describing what the problem is.
Also improve the checks and message a bit to indicate which was
the failure - whether the obj was NULL or just not the right class
Signed-off-by: John Ferlan <jferlan@redhat.com>
Rather than overload virObjectUnlock as commit id '77f4593b' has
done, create a separate virObjectRWUnlock API that will force the
consumers to make the proper decision regarding unlocking the
RWLock's. Similar to the RWLockRead and RWLockWrite, use the
virObjectGetRWLockableObj helper. This restores the virObjectUnlock
code to using the virObjectGetLockableObj.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Introduce a helper to handle the error path more cleanly. The same
as virObjectGetLockableObj in order to essentially follow the original
logic of commit 'b545f65d' to ensure that the input argument at least
has some validity before using.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Now that virObjectRWLockWrite exists to handle the virObjectRWLockable
objects, let's restore virObjectLock to only handle virObjectLockable
class locks. There still exists the possibility that the input @anyobj
isn't a valid object and the resource isn't truly locked, but that
also exists before commit id '77f4593b'.
This also restores some logic that commit id '77f4593b' removed
with respect to a common code path that commit id '10c2bb2b' had
introduced as virObjectGetLockableObj. This code path merely does
the same checks as the original virObjectLock commit 'b545f65d',
but in callable/reusable helper to ensure the @obj at least has
some validity before using.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Instead of making virObjectLock be the entry point for two
different types of locks, let's create a virObjectRWLockWrite API
which will only handle the virObjectRWLockableClass objects.
Use the new virObjectRWLockWrite for the virdomainobjlist code
in order to handle the Add, Remove, Rename, and Load operations
that need to be very synchronous.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Since the class it represents is based on virObjectRWLockableClass
and in order to make sure we differentiate just in case anyone somehow
believes they could use virObjectLockRead for a virObjectLockableClass,
let's rename the API to use the RW in the name. Besides the RW locks
refer to pthread_rwlock_{init|rdlock|wrlock|unlock|destroy} while the
other locks refer to pthread_mutex_{init|lock|unlock|destroy}.
Signed-off-by: John Ferlan <jferlan@redhat.com>
This way later patches can add another structures with virResctrl
prefix without the meaning being even more confusing than it needs to
be.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
That means that returning negative values means error and non-negative
values differ in meaning, but are all successful.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
It doesn't access anything from conf/ and ti will be needed to use
from other util/ places. This split makes the separation clearer.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Commit 81fb440b further qualified an if statement by adding the
boolean saveVlan to the condition. Coverity pointed out that this
change in the logic eliminated the need to check saveVlan in an
argument to virAsprintf().
Commit 9a94af6d restructured virHostdevReadNetConfig() so that it
would manually set ret = 0 after successfully reading the device's
config, but Coverity pointed out that "ret = 0" was erroneously placed
outside of an "else" clause, meaning that the the value of ret set in
the "if" clause was unnecessarily and incorrectly overwritten.
This patch moves ret = 0 into the else clause, which should silence
Coverity.
When using a VF from an SRIOV-capable network card in a guest (either
in macvtap passthrough mode, or via VFIO PCI device assignment), The
associated PF netdev must be online in order for the VF to be usable
by the guest. The guest, however, is not able to change the state of
the PF. And libvirt *could* set the PF online as needed, but that
could lead to the host receiving unexpected IPv6 traffic (since the
default for an unconfigured interface is to participate in IPv6
autoconf). For this reason, before assigning a VF to a guest, libvirt
verifies that the related PF netdev is online - if it isn't, then we
log an error and don't allow the guest startup to continue.
Until now, this check was done during virNetDevSetNetConfig(). This
works nicely because the same function is called both for macvtap
passthrough and for VFIO device assignment. But in the case of VFIO,
the VF has already been unbound from its netdev driver by the time we
get to virNetDevSetNetConfig(), and in the case of dual port Mellanox
NICs that have their VFs setup in single port mode, the *only* way to
determine the proper PF netdev to query for online status is via the
"phys_port_id" file that is in the VF netdev's sysfs directory. *BUT*
if we've unbound the VF from the netdev driver, then it doesn't *have*
a netdev sysfs directory.
So, in order to check the correct PF netdev for online status, this
patch moved the check earlier in the setup, into
virNetDevSaveNetConfig(), which is called *before* unbinding the VF
from its netdev driver.
(Note that this implies that if you are using VFIO device assignment
for the VFs of a Mellanox NIC that has the VFs programmed in single
port mode, you must let the VFs be bound to their net driver and use
"managed='yes'" in the device definition. To be more specific, this is
only true if the VFs in single port mode are using port *2* of the PF
- if the VFs are using only port 1, then the correct PF netdev will be
arrived at by default/chance))
This resolves: https://bugzilla.redhat.com/267191
virHostdevRestoreNetConfig() calls virNetDevReadNetConfig() to try and
read the "original config" of a netdev, and if that fails, it tries
again with a different directory/netdev name. This achieves the
desired effect (we end up finding the config wherever it may be), but
for each failure, virNetDevReadNetConfig() places a nice error message
in the system logs. Experience has shown that false-positive error
logs like this lead to erroneous bug reports, and can often mislead
those searching for *real* bugs.
This patch changes virNetDevReadNetConfig() to explicitly check if the
file exists before calling virFileReadAll(); if it doesn't exist,
virNetDevReadNetConfig() returns a success, but leaves all the
variables holding the results as NULL. (This makes sense if you define
the purpose of the function as "read a netdev's config from its config
file *if that file exists*).
To take advantage of that change, the caller,
virHostdevRestoreNetConfig() is modified to fail immediately if
virNetDevReadNetConfig() returns an error, and otherwise to try the
different directory/netdev name if adminMAC & vlan & MAC are all NULL
after the preceding attempt.
Mellanox ConnectX-3 dual port SRIOV NICs present a bit of a challenge
when assigning one of their VFs to a guest using VFIO device
assignment.
These NICs have only a single PCI PF device, and that single PF has
two netdevs sharing the single PCI address - one for port 1 and one
for port 2. When a VF is created it can also have 2 netdevs, or it can
be setup in "single port" mode, where the VF has only a single netdev,
and that netdev is connected either to port 1 or to port 2.
When the VF is created in dual port mode, you get/set the MAC
address/vlan tag for the port 1 VF by sending a netlink message to the
PF's port1 netdev, and you get/set the MAC address/vlan tag for the
port 2 VF by sending a netlink message to the PF's port 2 netdev. (Of
course libvirt doesn't have any way to describe MAC/vlan info for 2
ports in a single hostdev interface, so that's a bit of a moot point)
When the VF is created in single port mode, you can *set* the MAC/vlan
info by sending a netlink message to *either* PF netdev - the driver
is smart enough to understand that there's only a single netdev, and
set the MAC/vlan for that netdev. When you want to *get* it, however,
the driver is more accurate - it will return 00:00:00:00:00:00 for the
MAC if you request it from the port 1 PF netdev when the VF was
configured to be single port on port 2, or if you request if from the
port 2 PF netdev when the VF was configured to be single port on port
1.
Based on this information, when *getting* the MAC/vlan info (to save
the original setting prior to assignment), we determine the correct PF
netdev by matching phys_port_id between VF and PF.
(IMPORTANT NOTE: this implies that to do PCI device assignment of the
VFs on dual port Mellanox cards using <interface type='hostdev'>
(i.e. if you want the MAC address/vlan tag to be set), not only must
the VFs be configured in single port mode, but also the VFs *must* be
bound to the host VF net driver, and libvirt must use managed='yes')
By the time libvirt is ready to set the new MAC/vlan tag, the VF has
already been unbound from the host net driver and bound to
vfio-pci. This isn't problematic though because, as stated earlier,
when a VF is created in single port mode, commands to configure it can
be sent to either the port 1 PF netdev or the port 2 PF netdev.
When it is time to restore the original MAC/vlan tag, again the VF
will *not* be bound to a host net driver, so it won't be possible to
learn from sysfs whether to use the port 1 or port 2 PF netdev for the
netlink commands. And again, it doesn't matter which netdev you
use. However, we must keep in mind that we saved the original settings
to a file called "${PF}_${VFNUM}". To solve this problem, we just
check for the existence of ${PF1}_${VFNUM} and ${PF2}_${VFNUM}, and
use whichever one we find (since we know that only one can be there)
This patch updates functions in netdev.c to pay attention to
phys_port_id. It uses the new function virNetDevGetPhysPortID() to
learn the phys_port_id of a VF or PF, then sends that info to
virPCIGetNetName(), which has newly been modified to take an optional
phys_port_id.
A single PCI device may have multiple netdevs associated with it. Each
of those netdevs will have a different phys_port_id entry in
sysfs. This patch modifies virPCIGetNetName() to allow selecting one
of the potential many netdevs in two different ways:
1) by setting the "idx" argument, the caller can select the 1st (0),
2nd (1), etc. netdev from the PCI device's net subdirectory.
2) If the physPortID arg is set (to a null-terminated string) then
virPCIGetNetName() returns the netdev that has that phys_port_id in
the sysfs file of the same name in the netdev's directory.
On Linux each network device *can* (but not necessarily *does*) have
an attribute called phys_port_id which can be read from the file of
that name in the netdev's sysfs directory. The examples I've seen have
been a many-digit hexadecimal number (as an ASCII string).
This value can be useful when a single PCI device is associated with
multiple netdevs (e.g a dual port Mellanox SR-IOV NIC - this card has
a single PCI Physical Function (PF), and that PF has two netdevs
associated with it (the "net" subdirectory of the PF in sysfs has two
links rather than the usual single link to a netdev directory). Each
of the PF netdevs has a different phys_port_id. The Virtual Functions
(VF) are similar - the PF (a PCI device) has "n" VFs (also each of
these is a PCI device), each VF has two netdevs, and each of the VF
netdevs points back to the VF PCI device (with the "device" entry in
its sysfs directory) as well as having a phys_port_id matching the PF
netdev it is associated with.
virNetDevGetPhysPortID() simply attempts to read the phys_port_id for
the given netdev and return it to the caller. If this particular
netdev driver doesn't support phys_port_id, it returns NULL (*not* a
NULL-terminated string, but a NULL pointer) but still counts it as a
success.
We are given a string in @machinename, we never allocate it, just
merely use it for reading. We should not free it otherwise it
leads to double free:
==32191== Thread 17:
==32191== Invalid free() / delete / delete[] / realloc()
==32191== at 0x4C2D1A0: free (vg_replace_malloc.c:530)
==32191== by 0x54BBB84: virFree (viralloc.c:582)
==32191== by 0x2BC04499: qemuProcessStop (qemu_process.c:6313)
==32191== by 0x2BC500FF: processMonitorEOFEvent (qemu_driver.c:4724)
==32191== by 0x2BC502FC: qemuProcessEventHandler (qemu_driver.c:4769)
==32191== by 0x5550640: virThreadPoolWorker (virthreadpool.c:167)
==32191== by 0x554FBCF: virThreadHelper (virthread.c:206)
==32191== by 0x8F913D3: start_thread (in /lib64/libpthread-2.23.so)
==32191== by 0x928DE3C: clone (in /lib64/libc-2.23.so)
==32191== Address 0x31893d70 is 0 bytes inside a block of size 1,100 free'd
==32191== at 0x4C2D1A0: free (vg_replace_malloc.c:530)
==32191== by 0x54BBB84: virFree (viralloc.c:582)
==32191== by 0x54C1936: virCgroupValidateMachineGroup (vircgroup.c:343)
==32191== by 0x54C4B29: virCgroupNewDetectMachine (vircgroup.c:1550)
==32191== by 0x2BBDDA29: qemuConnectCgroup (qemu_cgroup.c:972)
==32191== by 0x2BC05DA7: qemuProcessReconnect (qemu_process.c:6822)
==32191== by 0x554FBCF: virThreadHelper (virthread.c:206)
==32191== by 0x8F913D3: start_thread (in /lib64/libpthread-2.23.so)
==32191== by 0x928DE3C: clone (in /lib64/libc-2.23.so)
==32191== Block was alloc'd at
==32191== at 0x4C2BE80: malloc (vg_replace_malloc.c:298)
==32191== by 0x4C2E35F: realloc (vg_replace_malloc.c:785)
==32191== by 0x54BB492: virReallocN (viralloc.c:245)
==32191== by 0x54BEDF2: virBufferGrow (virbuffer.c:150)
==32191== by 0x54BF3B9: virBufferVasprintf (virbuffer.c:408)
==32191== by 0x54BF324: virBufferAsprintf (virbuffer.c:381)
==32191== by 0x55BB271: virDomainGenerateMachineName (domain_conf.c:27078)
==32191== by 0x2BBD5B8F: qemuDomainGetMachineName (qemu_domain.c:9595)
==32191== by 0x2BBDD9B4: qemuConnectCgroup (qemu_cgroup.c:966)
==32191== by 0x2BC05DA7: qemuProcessReconnect (qemu_process.c:6822)
==32191== by 0x554FBCF: virThreadHelper (virthread.c:206)
==32191== by 0x8F913D3: start_thread (in /lib64/libpthread-2.23.so)
Moreover, make the @machinename 'const char *' to mark it
explicitly that we are not changing the passed string.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
The new virFileCache will nicely handle the caching logic for any data
that we would like to cache. For each type of data we will just need
to implement few handlers that will take care of creating, validating,
loading and saving the cached data.
The cached data must be an instance of virObject.
Currently we cache QEMU capabilities which will start using
virFileCache.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
It is more related to a domain as we might use it even when there is
no systemd and it does not use any dbus/systemd functions. In order
not to use code from conf/ in util/ pass machineName in cgroups code
as a parameter. That also fixes a leak of machineName in the lxc
driver and cleans up and de-duplicates some code.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
This reverts commit 328bd24443.
As it turns out, this is not portable and very Linux & glibc
specific. Worse, this may lead to not starving writers on Linux
but everywhere else. Revert this and if the starvation occurs
resolve it.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Up until now we only had virObjectLockable which uses mutexes for
mutually excluding each other in critical section. Well, this is
not enough. Future work will require RW locks so we might as well
have virObjectRWLockable which is introduced here.
Moreover, polymorphism is introduced to our code for the first
time. Yay! More specifically, virObjectLock will grab a write
lock, virObjectLockRead will grab a read lock then (what a
surprise right?). This has great advantage that an object can be
made derived from virObjectRWLockable in a single line and still
continue functioning properly (mutexes can be viewed as grabbing
write locks only). Then just those critical sections that can
grab a read lock need fixing. Therefore the resulting change is
going to be way smaller.
In order to avoid writer starvation, the object initializes RW
lock that prefers writers.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
We already have virRWLockInit. But this uses pthread defaults
which prefer reader to initialize the RW lock. This may lead to
writer starvation. Therefore we need to have the counterpart that
prefers writers. Now, according to the
pthread_rwlockattr_setkind_np() man page setting
PTHREAD_RWLOCK_PREFER_WRITER_NP attribute is no-op. Therefore we
need to use PTHREAD_RWLOCK_PREFER_WRITER_NONRECURSIVE_NP
attribute. So much for good enum value names.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
Currently, @port is type of string. Well, that's overkill and
waste of memory. Port is always an integer. Use it as such.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
A new function virNetDevOpenvswitchUpdateVlan has been created to instruct
OVS of the changes. qemuDomainChangeNet has been modified to handle the
update of the VLAN configuration for a running guest and rely on
virNetDevOpenvswitchUpdateVlan to do the actual update if needed.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
While searching for an element using a function it may be
desirable to know the element key for future operation.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
The purpose of this function is to tell if the current position
in given FD is in data section or a hole and how much bytes there
is remaining until the end of the section. This is achieved by
couple of lseeks(). The most important part is that we reposition
the FD back, so that the position is unchanged from the caller
POV. And until now the final lseek() back to the original
position was done with no check for errors. And I was convinced
that that's okay since nothing can go wrong. However, review
feedback from a related series persuaded me, that it's better to
be safe than sorry. Therefore, lets check if the final lseek()
succeeded and if it doesn't report an error.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Fill them in right away rather than having to figure out at runtime
whether they are necessary or not.
virStorageSourceNetworkDefaultPort does not need to be exported any
more.
This reverts commit e4b980c853.
When a binary links against a .a archive (as opposed to a shared library),
any symbols which are marked as 'weak' get silently dropped. As a result
when the binary later runs, those 'weak' functions have an address of
0x0 and thus crash when run.
This happened with virtlogd and virtlockd because they don't link to
libvirt.so, but instead just libvirt_util.a and libvirt_rpc.a. The
virRandomBits symbols was weak and so left out of the virtlogd &
virtlockd binaries, despite being required by virHashTable functions.
Various other binaries like libvirt_lxc, libvirt_iohelper, etc also
link directly to .a files instead of libvirt.so, so are potentially
at risk of dropping symbols leading to a later runtime crash.
This is normal linker behaviour because a weak symbol is not treated
as undefined, so nothing forces it to be pulled in from the .a You
have to force the linker to pull in weak symbols using -u$SYMNAME
which is not a practical approach.
This risk is silent bad linkage that affects runtime behaviour is
not acceptable for a fix that was merely trying to fix the test
suite. So stop using __weak__ again.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Currently the scan of the /proc/mounts file used to find cgroup mount
points doesn't take into account that mount points may hidden by other
mount points. For, example in certain Kubernetes environments the
/proc/mounts contains the following lines:
cgroup /sys/fs/cgroup/net_prio,net_cls cgroup ...
tmpfs /sys/fs/cgroup tmpfs ...
cgroup /sys/fs/cgroup/net_cls,net_prio cgroup ...
In this particular environment the first mount point is hidden by the
second one. The correct mount point is the third one, but libvirt will
never process it because it only checks the first mount point for each
controller (net_cls in this case). So libvirt will try to use the first
mount point, which doesn't actually exist, and the complete detection
process will fail.
To avoid that issue this patch changes the virCgroupDetectMountsFromFile
function so that when there are duplicates it takes the information from
the last line in /proc/mounts. This requires removing the previous
explicit condition to skip duplicates, and adding code to free the
memory used by the processing of duplicated lines.
Related-To: https://bugzilla.redhat.com/1468214
Related-To: https://github.com/kubevirt/libvirt/issues/4
Signed-off-by: Juan Hernandez <jhernand@redhat.com>
Currently all mockable functions are annotated with the 'noinline'
attribute. This is insufficient to guarantee that a function can
be reliably mocked with an LD_PRELOAD. The C language spec allows
the compiler to assume there is only a single implementation of
each function. It can thus do things like propagating constant
return values into the caller at compile time, or creating
multiple specialized copies of the function body each optimized
for a different caller. To prevent these optimizations we must
also set the 'noclone' and 'weak' attributes.
This fixes the test suite when libvirt.so is built with CLang
with optimization enabled.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
