- Devel - libvirt List Archives

[PATCH 0/2] Add Granite Rapids cpu model
by Tim Wiederhake 22 Jul '24

22 Jul '24

Also, update the sync tool with new features introduced in qemu. Tim Wiederhake (2): sync_qemu_models_i386.py: Add missing features cpu_map: Add GraniteRapids CPU model src/cpu_map/index.xml | 1 + src/cpu_map/meson.build | 1 + src/cpu_map/sync_qemu_models_i386.py | 5 + src/cpu_map/x86_GraniteRapids.xml | 200 ++++++++++++++++++ .../domaincapsdata/qemu_8.1.0-q35.x86_64.xml | 1 + .../domaincapsdata/qemu_8.1.0-tcg.x86_64.xml | 1 + tests/domaincapsdata/qemu_8.1.0.x86_64.xml | 1 + .../domaincapsdata/qemu_8.2.0-q35.x86_64.xml | 1 + .../domaincapsdata/qemu_8.2.0-tcg.x86_64.xml | 1 + tests/domaincapsdata/qemu_8.2.0.x86_64.xml | 1 + .../domaincapsdata/qemu_9.0.0-q35.x86_64.xml | 1 + .../domaincapsdata/qemu_9.0.0-tcg.x86_64.xml | 1 + tests/domaincapsdata/qemu_9.0.0.x86_64.xml | 1 + .../domaincapsdata/qemu_9.1.0-q35.x86_64.xml | 1 + .../domaincapsdata/qemu_9.1.0-tcg.x86_64.xml | 1 + tests/domaincapsdata/qemu_9.1.0.x86_64.xml | 1 + 16 files changed, 219 insertions(+) create mode 100644 src/cpu_map/x86_GraniteRapids.xml -- 2.43.0

2 3

[PATCH] cpu_map: Add libcpuinfo as optional data source
by Tim Wiederhake 22 Jul '24

22 Jul '24

This adds an option to use libcpuinfo [1] as data source for libvirt's list of x86 cpu features. This is purely optional and does not change the script's behavior if libcpuinfo is not installed. libcpuinfo is a cross-vendor, cross-architecture source for CPU related information that has the capability to replace libvirt's dependence on qemu's cpu feature list. [1] https://gitlab.com/twiederh/libcpuinfo Signed-off-by: Tim Wiederhake <twiederh(a)redhat.com> --- src/cpu_map/libcpuinfo_aliases.xml | 75 +++++++++++++++++++++++++ src/cpu_map/sync_qemu_features_i386.py | 77 +++++++++++++++++++++++--- 2 files changed, 145 insertions(+), 7 deletions(-) create mode 100644 src/cpu_map/libcpuinfo_aliases.xml diff --git a/src/cpu_map/libcpuinfo_aliases.xml b/src/cpu_map/libcpuinfo_aliases.xml new file mode 100644 index 0000000000..75d243fead --- /dev/null +++ b/src/cpu_map/libcpuinfo_aliases.xml @@ -0,0 +1,75 @@ + + +<external_aliases> + <external_alias> + <type>feature</type> + <canonical>pclmulqdq</canonical> + <name>pclmuldq</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>ds-cpl</canonical> + <name>ds_cpl</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>sse4-1</canonical> + <name>sse4.1</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>sse4-2</canonical> + <name>sse4.2</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>tsc-adjust</canonical> + <name>tsc_adjust</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>lahf-lm</canonical> + <name>lahf_lm</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>cmp-legacy</canonical> + <name>cmp_legacy</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>nodeid-msr</canonical> + <name>nodeid_msr</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>perfctr-core</canonical> + <name>perfctr_core</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>perfctr-nb</canonical> + <name>perfctr_nb</name> + <domain>libvirt</domain> + </external_alias> + <external_alias> + <type>feature</type> + <canonical>fxsr-opt</canonical> + <name>fxsr_opt</name> + <domain>libvirt</domain> + </external_alias> +</external_aliases> diff --git a/src/cpu_map/sync_qemu_features_i386.py b/src/cpu_map/sync_qemu_features_i386.py index e4b1f7275a..3b3ad5a643 100755 --- a/src/cpu_map/sync_qemu_features_i386.py +++ b/src/cpu_map/sync_qemu_features_i386.py @@ -4,6 +4,11 @@ import argparse import os import re +try: + import pycpuinfo +except ImportError: + pycpuinfo = None + # features in qemu that we do not want in libvirt FEATURES_IGNORE = ( @@ -22,6 +27,7 @@ FEATURES_IGNORE = ( "kvm-steal-time", "kvmclock", "kvmclock-stable-bit", + "kvmclock2", "xstore", "xstore-en", @@ -295,6 +301,53 @@ def add_feature_qemu(query, data): add_feature_cpuid(eax, ecx, reg, bit, name) +def add_features_cpuinfo(): + def decode_bit(value): + for i in range(0, 64): + if value == (1 << i): + return i + + def decode_cpuid(v): + if v[0] != 0 and v[1] == 0 and v[2] == 0 and v[3] == 0: + reg, val = "eax", v[0] + if v[0] == 0 and v[1] != 0 and v[2] == 0 and v[3] == 0: + reg, val = "ebx", v[1] + if v[0] == 0 and v[1] == 0 and v[2] != 0 and v[3] == 0: + reg, val = "ecx", v[2] + if v[0] == 0 and v[1] == 0 and v[2] == 0 and v[3] != 0: + reg, val = "edx", v[3] + + return reg, decode_bit(val) + + x86 = pycpuinfo.Family.find("x86", "") + + for feature in pycpuinfo.features(): + if feature.family() != x86: + continue + + if list(feature.features()): + continue + + name = feature.name("libvirt") + if name in FEATURES_IGNORE: + continue + + cpuid = feature.extra_x86_cpuid() + if cpuid: + eax = cpuid[0] + ecx = cpuid[1] + if ecx == pycpuinfo.x86.CPUINFO_X86_CPUID_ECX_NONE: + ecx = None + reg, bit = decode_cpuid(cpuid[2:]) + add_feature_cpuid(eax, ecx, reg, bit, name) + + msr = feature.extra_x86_msr() + if msr: + index = msr[0] + bit = decode_bit(msr[1] | (msr[2] << 32)) + add_feature_msr(index, bit, name) + + # read the `feature_word_info` struct from qemu's cpu.c into a list of strings def read_cpu_c(path): pattern_comment = re.compile("/\\*.*?\\*/") @@ -450,6 +503,12 @@ def main(): nargs="?", type=os.path.realpath, ) + if pycpuinfo: + parser.add_argument( + "--libcpuinfo", + help="Use libcpuinfo as data source instead", + action="store_true", + ) parser.add_argument( "--output", "-o", @@ -459,14 +518,18 @@ def main(): ) args = parser.parse_args() - if not os.path.isdir(args.qemu): - parser.print_help() - exit("qemu source directory not found") + if pycpuinfo and args.libcpuinfo: + add_features_cpuinfo() + else: + if not os.path.isdir(args.qemu): + parser.print_help() + exit("qemu source directory not found") + + read_headers(args.qemu) + lines = read_cpu_c(args.qemu) + parse_feature_words(lines) + add_extra_features() - read_headers(args.qemu) - lines = read_cpu_c(args.qemu) - parse_feature_words(lines) - add_extra_features() write_output(args.output) print( -- 2.43.0

2 2

[PATCH] formatstorage: Document qcow2 default version change
by Peter Krempa 22 Jul '24

22 Jul '24

Based on discussion after commit f432114d9c was pushed it was pointed out that the documentation still mentions the older version. Fix the documentation to state the new version and introduce ambiguity for future updates. Fixes: f432114d9cf507a4047aa9dc1344b1c13356db08 Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- Posting this documentation update to document what happened rather than introduce (almost pointless) complication in adding a config file which is unlikely to be ever used. docs/formatstorage.rst | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/docs/formatstorage.rst b/docs/formatstorage.rst index 86e167d9cb..9d9a4143eb 100644 --- a/docs/formatstorage.rst +++ b/docs/formatstorage.rst @@ -700,10 +700,15 @@ host filesystem. It can contain the following child elements: Encryption <formatstorageencryption.html>`__ page for more information. ``compat`` Specify compatibility level. So far, this is only used for ``type='qcow2'`` - volumes. Valid values are ``0.10`` and ``1.1`` so far, specifying QEMU - version the images should be compatible with. If the ``feature`` element is - present, 1.1 is used. :since:`Since 1.1.0` If omitted, 0.10 is used. - :since:`Since 1.1.2` + volumes. Valid values are ``0.10`` (QCOW2 v2) and ``1.1`` (QCOW2 v3) so far. + The values were meant to specify QEMU version the images should be compatible + with. + + The default, if the ``feature`` element is present is ``1.1``. :since:`Since 1.1.0` + If ``feature`` is not present, ``0.10`` was used :since:`Since 1.1.2` and + :since:`Since 10.2.0` ``1.1`` is used as it's the default of ``qemu-img``. + + Any tool depending on a specific version should specify this field explicitly. ``nocow`` Turn off COW of the newly created volume. So far, this is only valid for a file image in btrfs file system. It will improve performance when the file -- 2.44.0

2 1

Plans for 10.6.0 release (freeze on Monday 29 Jul)
by Jiri Denemark 22 Jul '24

22 Jul '24

We are getting close to 10.6.0 release of libvirt. Since I'll be away from computers next week, I suggest moving the "last week in a month" freeze a bit and aim for the release on Monday 05 Aug after entering the freeze on Monday 29 Jul. There will be no RC2 this time. I hope this works for everyone. Jirka

1 0

[PATCH v2] qemu: add a monitor to /proc/$pid when killing times out
by Boris Fiuczynski 19 Jul '24

19 Jul '24

In cases when a QEMU process takes longer than the time sigterm and sigkill are issued to kill the process do not simply fail and leave the VM in state VIR_DOMAIN_SHUTDOWN until the daemon stops. Instead set up an fd on /proc/$pid and get notified when the QEMU process finally has terminated to cleanup the VM state. Resolves: https://issues.redhat.com/browse/RHEL-28819 Signed-off-by: Boris Fiuczynski <fiuczy(a)linux.ibm.com> --- src/qemu/qemu_domain.c | 8 +++ src/qemu/qemu_domain.h | 2 + src/qemu/qemu_driver.c | 18 +++++++ src/qemu/qemu_process.c | 115 ++++++++++++++++++++++++++++++++++++++-- src/qemu/qemu_process.h | 1 + 5 files changed, 139 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 2134b11038..8147ff02fd 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -1889,6 +1889,11 @@ qemuDomainObjPrivateFree(void *data) virChrdevFree(priv->devs); + if (priv->pidMonitored >= 0) { + virEventRemoveHandle(priv->pidMonitored); + priv->pidMonitored = -1; + } + /* This should never be non-NULL if we get here, but just in case... */ if (priv->mon) { VIR_ERROR(_("Unexpected QEMU monitor still active during domain deletion")); @@ -1934,6 +1939,8 @@ qemuDomainObjPrivateAlloc(void *opaque) priv->blockjobs = virHashNew(virObjectUnref); priv->fds = virHashNew(g_object_unref); + priv->pidMonitored = -1; + /* agent commands block by default, user can choose different behavior */ priv->agentTimeout = VIR_DOMAIN_AGENT_RESPONSE_TIMEOUT_BLOCK; priv->migMaxBandwidth = QEMU_DOMAIN_MIG_BANDWIDTH_MAX; @@ -11680,6 +11687,7 @@ qemuProcessEventFree(struct qemuProcessEvent *event) case QEMU_PROCESS_EVENT_RESET: case QEMU_PROCESS_EVENT_NBDKIT_EXITED: case QEMU_PROCESS_EVENT_MONITOR_EOF: + case QEMU_PROCESS_EVENT_SHUTDOWN_COMPLETED: case QEMU_PROCESS_EVENT_LAST: break; } diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index d777559119..a5092dd7f0 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -119,6 +119,7 @@ struct _qemuDomainObjPrivate { bool beingDestroyed; char *pidfile; + int pidMonitored; virDomainPCIAddressSet *pciaddrs; virDomainUSBAddressSet *usbaddrs; @@ -469,6 +470,7 @@ typedef enum { QEMU_PROCESS_EVENT_UNATTENDED_MIGRATION, QEMU_PROCESS_EVENT_RESET, QEMU_PROCESS_EVENT_NBDKIT_EXITED, + QEMU_PROCESS_EVENT_SHUTDOWN_COMPLETED, QEMU_PROCESS_EVENT_LAST } qemuProcessEventType; diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 9f3013e231..6b1e4084f6 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -4041,6 +4041,21 @@ processNbdkitExitedEvent(virDomainObj *vm, } +static void +processShutdownCompletedEvent(virQEMUDriver *driver, + virDomainObj *vm) +{ + if (virDomainObjBeginJob(vm, VIR_JOB_MODIFY) < 0) + return; + + if (virDomainObjIsActive(vm)) + qemuProcessStop(driver, vm, VIR_DOMAIN_SHUTOFF_UNKNOWN, + VIR_ASYNC_JOB_NONE, 0); + + virDomainObjEndJob(vm); +} + + static void qemuProcessEventHandler(void *data, void *opaque) { struct qemuProcessEvent *processEvent = data; @@ -4101,6 +4116,9 @@ static void qemuProcessEventHandler(void *data, void *opaque) case QEMU_PROCESS_EVENT_NBDKIT_EXITED: processNbdkitExitedEvent(vm, processEvent->data); break; + case QEMU_PROCESS_EVENT_SHUTDOWN_COMPLETED: + processShutdownCompletedEvent(driver, vm); + break; case QEMU_PROCESS_EVENT_LAST: break; } diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 25dfd04272..86ca495a20 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -25,6 +25,7 @@ #include <unistd.h> #include <signal.h> #include <sys/stat.h> +#include <sys/syscall.h> #if defined(__linux__) # include <linux/capability.h> #elif defined(__FreeBSD__) @@ -8387,9 +8388,107 @@ qemuProcessCreatePretendCmdBuild(virDomainObj *vm, } +typedef struct { + virDomainObj *vm; + int pidfd; +} qemuProcessInShutdownEventData; + + +static qemuProcessInShutdownEventData* +qemuProcessInShutdownEventDataNew(virDomainObj *vm, int pidfd) +{ + qemuProcessInShutdownEventData *d = g_new(qemuProcessInShutdownEventData, 1); + d->vm = virObjectRef(vm); + d->pidfd = pidfd; + return d; +} + + +static void +qemuProcessInShutdownEventDataFree(qemuProcessInShutdownEventData *d) +{ + virObjectUnref(d->vm); + VIR_FORCE_CLOSE(d->pidfd); + g_free(d); +} + + +static void +qemuProcessInShutdownPidfdCb(int watch, + int fd, + int events G_GNUC_UNUSED, + void *opaque) +{ + qemuProcessInShutdownEventData *data = opaque; + virDomainObj *vm = data->vm; + + VIR_DEBUG("vm=%p name=%s pid=%lld fd=%d", + vm, vm->def->name, (long long)vm->pid, fd); + + virEventRemoveHandle(watch); + + virObjectLock(vm); + + VIR_INFO("QEMU process %lld finally completed termination", + (long long)vm->pid); + + QEMU_DOMAIN_PRIVATE(vm)->pidMonitored = -1; + qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_SHUTDOWN_COMPLETED, + 0, 0, NULL); + + virObjectUnlock(vm); +} + + +static int +qemuProcessInShutdownStartMonitor(virDomainObj *vm) +{ + qemuDomainObjPrivate *priv = vm->privateData; + qemuProcessInShutdownEventData *data; + int pidfd; + int ret = -1; + + VIR_DEBUG("vm=%p name=%s pid=%lld pidMonitored=%d", + vm, vm->def->name, (long long)vm->pid, + priv->pidMonitored); + + if (priv->pidMonitored >= 0) { + VIR_DEBUG("Monitoring qemu in-shutdown process %i already set up", vm->pid); + goto cleanup; + } + + pidfd = syscall(SYS_pidfd_open, vm->pid, 0); + if (pidfd < 0) { + if (errno == ESRCH) /* process has already terminated */ + ret = 1; + goto cleanup; + } + + data = qemuProcessInShutdownEventDataNew(vm, pidfd); + if ((priv->pidMonitored = virEventAddHandle(pidfd, + VIR_EVENT_HANDLE_READABLE, + qemuProcessInShutdownPidfdCb, + data, + (virFreeCallback)qemuProcessInShutdownEventDataFree)) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("failed to monitor qemu in-shutdown process %1$i"), + vm->pid); + qemuProcessInShutdownEventDataFree(data); + goto cleanup; + } + VIR_DEBUG("Monitoring qemu in-shutdown process %i for termination", vm->pid); + ret = 0; + + cleanup: + return ret; +} + + int qemuProcessKill(virDomainObj *vm, unsigned int flags) { + int ret = -1; + VIR_DEBUG("vm=%p name=%s pid=%lld flags=0x%x", vm, vm->def->name, (long long)vm->pid, flags); @@ -8410,10 +8509,16 @@ qemuProcessKill(virDomainObj *vm, unsigned int flags) /* Request an extra delay of two seconds per current nhostdevs * to be safe against stalls by the kernel freeing up the resources */ - return virProcessKillPainfullyDelay(vm->pid, - !!(flags & VIR_QEMU_PROCESS_KILL_FORCE), - vm->def->nhostdevs * 2, - false); + ret = virProcessKillPainfullyDelay(vm->pid, + !!(flags & VIR_QEMU_PROCESS_KILL_FORCE), + vm->def->nhostdevs * 2, + false); + + if (ret < 0 && (flags & VIR_QEMU_PROCESS_KILL_MONITOR_ON_ERROR)) + if (qemuProcessInShutdownStartMonitor(vm) == 1) + ret = 0; /* process termination detected */ + + return ret; } @@ -8438,7 +8543,7 @@ qemuProcessBeginStopJob(virDomainObj *vm, * cleared inside qemuProcessStop */ priv->beingDestroyed = true; - if (qemuProcessKill(vm, killFlags) < 0) + if (qemuProcessKill(vm, killFlags|VIR_QEMU_PROCESS_KILL_MONITOR_ON_ERROR) < 0) goto error; /* Wake up anything waiting on domain condition */ diff --git a/src/qemu/qemu_process.h b/src/qemu/qemu_process.h index cb67bfcd2d..2324aeb7bd 100644 --- a/src/qemu/qemu_process.h +++ b/src/qemu/qemu_process.h @@ -180,6 +180,7 @@ typedef enum { VIR_QEMU_PROCESS_KILL_FORCE = 1 << 0, VIR_QEMU_PROCESS_KILL_NOWAIT = 1 << 1, VIR_QEMU_PROCESS_KILL_NOCHECK = 1 << 2, /* bypass the running vm check */ + VIR_QEMU_PROCESS_KILL_MONITOR_ON_ERROR = 1 << 3, /* on error enable process monitor */ } virQemuProcessKillMode; int qemuProcessKill(virDomainObj *vm, unsigned int flags); -- 2.45.0

2 2

[PATCH] security: AppArmor allow write when os loader readonly=no
by Miroslav Los 19 Jul '24

19 Jul '24

Since libvirt commit 3ef9b51b10e52886e8fe8d75e36d0714957616b7, the pflash storage for the os loader file follows its read-only flag, and qemu tries to open the file for writing if set so. This patches virt-aa-helper to generate the VM's AppArmor rules that allow this, using the same domain definition flag and default. Signed-off-by: Miroslav Los <mirlos(a)cisco.com> --- src/security/virt-aa-helper.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index 0374581f07..2f57664a4c 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -1001,9 +1001,14 @@ get_files(vahControl * ctl) if (vah_add_file(&buf, ctl->def->os.slic_table, "r") != 0) goto cleanup; - if (ctl->def->os.loader && ctl->def->os.loader->path) - if (vah_add_file(&buf, ctl->def->os.loader->path, "rk") != 0) + if (ctl->def->os.loader && ctl->def->os.loader->path) { + bool readonly = false; + virTristateBoolToBool(ctl->def->os.loader->readonly, &readonly); + if (vah_add_file(&buf, + ctl->def->os.loader->path, + readonly ? "rk" : "rwk") != 0) goto cleanup; + } if (ctl->def->os.loader && ctl->def->os.loader->nvram) { if (storage_source_add_files(ctl->def->os.loader->nvram, &buf, 0) < 0) -- 2.25.1

4 8

[PATCH 0/2] Warn on pthread errors
by Tim Wiederhake 18 Jul '24

18 Jul '24

libvirt currently exhibits undefined behavior due to pthread mutex misuse, e.g. destroying a locked mutex or attempting to unlock an already unlocked mutex. Add a warning if such a case is detected, so we can start on fixing the issues. Tim Wiederhake (2): virMutex: Warn on error DO NOT MERGE: virMutex: Fail loudly src/util/virthread.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) -- 2.43.0

3 8

[PATCH] qemu: add a monitor to /proc/$pid when killing times out
by Boris Fiuczynski 17 Jul '24

17 Jul '24

In cases when a QEMU process takes longer than the time sigterm and sigkill are issued to kill the process do not simply fail and leave the VM in state VIR_DOMAIN_SHUTDOWN until the daemon stops. Instead set up an fd on /proc/$pid and get notified when the QEMU process finally has terminated to cleanup the VM state. Resolves: https://issues.redhat.com/browse/RHEL-28819 Signed-off-by: Boris Fiuczynski <fiuczy(a)linux.ibm.com> --- src/qemu/qemu_domain.c | 8 +++ src/qemu/qemu_domain.h | 2 + src/qemu/qemu_driver.c | 18 ++++++ src/qemu/qemu_process.c | 127 ++++++++++++++++++++++++++++++++++++++-- src/qemu/qemu_process.h | 1 + 5 files changed, 151 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 2134b11038..96f4e41a11 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -1889,6 +1889,11 @@ qemuDomainObjPrivateFree(void *data) virChrdevFree(priv->devs); + if (priv->watchPid >= 0) { + virEventRemoveHandle(priv->watchPid); + priv->watchPid = -1; + } + /* This should never be non-NULL if we get here, but just in case... */ if (priv->mon) { VIR_ERROR(_("Unexpected QEMU monitor still active during domain deletion")); @@ -1934,6 +1939,8 @@ qemuDomainObjPrivateAlloc(void *opaque) priv->blockjobs = virHashNew(virObjectUnref); priv->fds = virHashNew(g_object_unref); + priv->watchPid = -1; + /* agent commands block by default, user can choose different behavior */ priv->agentTimeout = VIR_DOMAIN_AGENT_RESPONSE_TIMEOUT_BLOCK; priv->migMaxBandwidth = QEMU_DOMAIN_MIG_BANDWIDTH_MAX; @@ -11680,6 +11687,7 @@ qemuProcessEventFree(struct qemuProcessEvent *event) case QEMU_PROCESS_EVENT_RESET: case QEMU_PROCESS_EVENT_NBDKIT_EXITED: case QEMU_PROCESS_EVENT_MONITOR_EOF: + case QEMU_PROCESS_EVENT_SHUTDOWN_COMPLETED: case QEMU_PROCESS_EVENT_LAST: break; } diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index d777559119..e5366c6e8c 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -119,6 +119,7 @@ struct _qemuDomainObjPrivate { bool beingDestroyed; char *pidfile; + int watchPid; virDomainPCIAddressSet *pciaddrs; virDomainUSBAddressSet *usbaddrs; @@ -469,6 +470,7 @@ typedef enum { QEMU_PROCESS_EVENT_UNATTENDED_MIGRATION, QEMU_PROCESS_EVENT_RESET, QEMU_PROCESS_EVENT_NBDKIT_EXITED, + QEMU_PROCESS_EVENT_SHUTDOWN_COMPLETED, QEMU_PROCESS_EVENT_LAST } qemuProcessEventType; diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 9f3013e231..6b1e4084f6 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -4041,6 +4041,21 @@ processNbdkitExitedEvent(virDomainObj *vm, } +static void +processShutdownCompletedEvent(virQEMUDriver *driver, + virDomainObj *vm) +{ + if (virDomainObjBeginJob(vm, VIR_JOB_MODIFY) < 0) + return; + + if (virDomainObjIsActive(vm)) + qemuProcessStop(driver, vm, VIR_DOMAIN_SHUTOFF_UNKNOWN, + VIR_ASYNC_JOB_NONE, 0); + + virDomainObjEndJob(vm); +} + + static void qemuProcessEventHandler(void *data, void *opaque) { struct qemuProcessEvent *processEvent = data; @@ -4101,6 +4116,9 @@ static void qemuProcessEventHandler(void *data, void *opaque) case QEMU_PROCESS_EVENT_NBDKIT_EXITED: processNbdkitExitedEvent(vm, processEvent->data); break; + case QEMU_PROCESS_EVENT_SHUTDOWN_COMPLETED: + processShutdownCompletedEvent(driver, vm); + break; case QEMU_PROCESS_EVENT_LAST: break; } diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 25dfd04272..d6dbd7ba53 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -25,6 +25,7 @@ #include <unistd.h> #include <signal.h> #include <sys/stat.h> +#include <sys/syscall.h> #if defined(__linux__) # include <linux/capability.h> #elif defined(__FreeBSD__) @@ -8387,9 +8388,119 @@ qemuProcessCreatePretendCmdBuild(virDomainObj *vm, } +typedef struct { + virDomainObj *vm; + int pidfd; +} qemuProcessInShutdownEventData; + + +static qemuProcessInShutdownEventData* +qemuProcessInShutdownEventDataNew(virDomainObj *vm, int pidfd) +{ + qemuProcessInShutdownEventData *d = g_new(qemuProcessInShutdownEventData, 1); + d->vm = virObjectRef(vm); + d->pidfd = pidfd; + return d; +} + + +static void +qemuProcessInShutdownEventDataFree(qemuProcessInShutdownEventData *d) +{ + virObjectUnref(d->vm); + VIR_FORCE_CLOSE(d->pidfd); + g_free(d); +} + + +static void +qemuProcessInShutdownStopMonitor(virDomainObj *vm) +{ + qemuDomainObjPrivate *priv = vm->privateData; + + VIR_DEBUG("vm=%p name=%s pid=%lld watchPid=%d", + vm, vm->def->name, (long long)vm->pid, + priv->watchPid); + + virObjectLock(vm); + if (priv->watchPid >= 0) { + virEventRemoveHandle(priv->watchPid); + priv->watchPid = -1; + } + virObjectUnlock(vm); +} + + +static void +qemuProcessInShutdownPidfdCb(int watch G_GNUC_UNUSED, + int fd, + int events G_GNUC_UNUSED, + void *opaque) +{ + qemuProcessInShutdownEventData *data = opaque; + virDomainObj *vm = data->vm; + + VIR_DEBUG("vm=%p name=%s pid=%lld fd=%d", + vm, vm->def->name, (long long)vm->pid, fd); + + VIR_DEBUG("QEMU process %lld finally completed termination", + (long long)vm->pid); + qemuProcessInShutdownStopMonitor(vm); + + qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_SHUTDOWN_COMPLETED, + 0, 0, NULL); +} + + +static int +qemuProcessInShutdownStartMonitor(virDomainObj *vm) +{ + qemuDomainObjPrivate *priv = vm->privateData; + qemuProcessInShutdownEventData *data; + int pidfd; + int ret = -1; + + VIR_DEBUG("vm=%p name=%s pid=%lld watchPid=%d", + vm, vm->def->name, (long long)vm->pid, + priv->watchPid); + + if (priv->watchPid >= 0) { + VIR_DEBUG("Monitoring qemu in-shutdown process %i already set up", vm->pid); + goto cleanup; + } + + pidfd = syscall(SYS_pidfd_open, vm->pid, 0); + if (pidfd < 0) { + if (errno == ESRCH) /* process has already terminated */ + ret = 1; + goto cleanup; + } + + data = qemuProcessInShutdownEventDataNew(vm, pidfd); + if ((priv->watchPid = virEventAddHandle(pidfd, + VIR_EVENT_HANDLE_READABLE, + qemuProcessInShutdownPidfdCb, + data, + (virFreeCallback)qemuProcessInShutdownEventDataFree)) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("failed to monitor qemu in-shutdown process %1$i"), + vm->pid); + qemuProcessInShutdownEventDataFree(data); + goto cleanup; + } + VIR_DEBUG("Monitoring qemu in-shutdown process %i for termination", vm->pid); + ret = 0; + + cleanup: + return ret; +} + + int qemuProcessKill(virDomainObj *vm, unsigned int flags) { + int ret = -1; + VIR_DEBUG("vm=%p name=%s pid=%lld flags=0x%x", vm, vm->def->name, (long long)vm->pid, flags); @@ -8410,10 +8521,16 @@ qemuProcessKill(virDomainObj *vm, unsigned int flags) /* Request an extra delay of two seconds per current nhostdevs * to be safe against stalls by the kernel freeing up the resources */ - return virProcessKillPainfullyDelay(vm->pid, - !!(flags & VIR_QEMU_PROCESS_KILL_FORCE), - vm->def->nhostdevs * 2, - false); + ret = virProcessKillPainfullyDelay(vm->pid, + !!(flags & VIR_QEMU_PROCESS_KILL_FORCE), + vm->def->nhostdevs * 2, + false); + + if (ret < 0 && (flags & VIR_QEMU_PROCESS_KILL_MONITOR_ON_ERROR)) + if (qemuProcessInShutdownStartMonitor(vm) == 1) + ret = 0; /* process termination detected */ + + return ret; } @@ -8438,7 +8555,7 @@ qemuProcessBeginStopJob(virDomainObj *vm, * cleared inside qemuProcessStop */ priv->beingDestroyed = true; - if (qemuProcessKill(vm, killFlags) < 0) + if (qemuProcessKill(vm, killFlags|VIR_QEMU_PROCESS_KILL_MONITOR_ON_ERROR) < 0) goto error; /* Wake up anything waiting on domain condition */ diff --git a/src/qemu/qemu_process.h b/src/qemu/qemu_process.h index cb67bfcd2d..2324aeb7bd 100644 --- a/src/qemu/qemu_process.h +++ b/src/qemu/qemu_process.h @@ -180,6 +180,7 @@ typedef enum { VIR_QEMU_PROCESS_KILL_FORCE = 1 << 0, VIR_QEMU_PROCESS_KILL_NOWAIT = 1 << 1, VIR_QEMU_PROCESS_KILL_NOCHECK = 1 << 2, /* bypass the running vm check */ + VIR_QEMU_PROCESS_KILL_MONITOR_ON_ERROR = 1 << 3, /* on error enable process monitor */ } virQemuProcessKillMode; int qemuProcessKill(virDomainObj *vm, unsigned int flags); -- 2.45.0

1 0

[PATCH] qemu: virtiofs: format --rlimit-nofile
by Adam Julis 17 Jul '24

17 Jul '24

Resolves: https://gitlab.com/libvirt/libvirt/-/issues/485 Signed-off-by: Adam Julis <ajulis(a)redhat.com> --- src/qemu/qemu_virtiofs.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/qemu/qemu_virtiofs.c b/src/qemu/qemu_virtiofs.c index 0e3c7dbb58..703f1226a2 100644 --- a/src/qemu/qemu_virtiofs.c +++ b/src/qemu/qemu_virtiofs.c @@ -194,6 +194,9 @@ qemuVirtioFSBuildCommandLine(virQEMUDriverConfig *cfg, if (fs->thread_pool_size >= 0) virCommandAddArgFormat(cmd, "--thread-pool-size=%i", fs->thread_pool_size); + if (fs->rlimit_nofile > 0) + virCommandAddArgFormat(cmd, "--rlimit-nofile=%llu", fs->rlimit_nofile); + if (cfg->virtiofsdDebug) { if (virBitmapIsBitSet(fs->caps, QEMU_VHOST_USER_FS_FEATURE_SEPARATE_OPTIONS)) virCommandAddArgList(cmd, "--log-level", "debug", NULL); -- 2.45.2

2 1

[PATCH] vmx: Be even more lax when trying to comprehend serial ports
by Martin Kletzander 16 Jul '24

16 Jul '24

So much can happen in the fileName field of the VMX that the easiest thing is to silently report a serial type="null". This effectively reverts commits de81bdb8d4cd and 62c53db0421a, but keeps the test files to show the fix is still in place. There is one instance where an error gets reset, but since that is a rare case on its own and on top of that does not happen in any of our long-running daemons with a logfile that might get monitored it should be fine to leave it there. Resolves: https://issues.redhat.com/browse/RHEL-32182 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- src/vmx/vmx.c | 30 +++++++++++++----------------- 1 file changed, 13 insertions(+), 17 deletions(-) diff --git a/src/vmx/vmx.c b/src/vmx/vmx.c index e5bc2d793c66..227744d06258 100644 --- a/src/vmx/vmx.c +++ b/src/vmx/vmx.c @@ -2975,9 +2975,6 @@ virVMXParseSerial(virVMXContext *ctx, virConf *conf, int port, char fileName_name[48] = ""; g_autofree char *fileName = NULL; - char vspc_name[48] = ""; - g_autofree char *vspc = NULL; - char network_endPoint_name[48] = ""; g_autofree char *network_endPoint = NULL; @@ -3000,7 +2997,6 @@ virVMXParseSerial(virVMXContext *ctx, virConf *conf, int port, VMX_BUILD_NAME(startConnected); VMX_BUILD_NAME(fileType); VMX_BUILD_NAME(fileName); - VMX_BUILD_NAME(vspc); VMX_BUILD_NAME_EXTRA(network_endPoint, "network.endPoint"); /* vmx:present */ @@ -3030,10 +3026,6 @@ virVMXParseSerial(virVMXContext *ctx, virConf *conf, int port, if (virVMXGetConfigString(conf, fileName_name, &fileName, true) < 0) goto cleanup; - /* vmx:fileName -> def:data.file.path */ - if (virVMXGetConfigString(conf, vspc_name, &vspc, true) < 0) - goto cleanup; - /* vmx:network.endPoint -> def:data.tcp.listen */ if (virVMXGetConfigString(conf, network_endPoint_name, &network_endPoint, true) < 0) { @@ -3065,21 +3057,25 @@ virVMXParseSerial(virVMXContext *ctx, virConf *conf, int port, (*def)->target.port = port; (*def)->source->type = VIR_DOMAIN_CHR_TYPE_PIPE; (*def)->source->data.file.path = g_steal_pointer(&fileName); - } else if (STRCASEEQ(fileType, "network") && (vspc || !fileName || STREQ(fileName, ""))) { - (*def)->target.port = port; - (*def)->source->type = VIR_DOMAIN_CHR_TYPE_NULL; } else if (STRCASEEQ(fileType, "network")) { (*def)->target.port = port; (*def)->source->type = VIR_DOMAIN_CHR_TYPE_TCP; - if (!(parsedUri = virURIParse(fileName))) - goto cleanup; + if (!(parsedUri = virURIParse(fileName))) { + /* + * Ignore anything we cannot parse since there are many variations + * that could lead to unusable or non-representable serial ports + * which are very commonly seen and the main consumer of this driver + * (virt-v2v) ignores them anyway, so let's at least not error out. + */ + virResetLastError(); + (*def)->source->type = VIR_DOMAIN_CHR_TYPE_NULL; + return 0; + } if (parsedUri->port == 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, - _("VMX entry '%1$s' doesn't contain a port part"), - fileName_name); - goto cleanup; + (*def)->source->type = VIR_DOMAIN_CHR_TYPE_NULL; + return 0; } (*def)->source->data.tcp.host = g_strdup(parsedUri->server); -- 2.45.1

2 1

[PATCH] ci: Refresh generated files
by Michal Privoznik 15 Jul '24

15 Jul '24

This is supposed to unstuck FreeBSD as it switched to Python-3.11. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- ci/cirrus/freebsd-13.vars | 2 +- ci/cirrus/freebsd-14.vars | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/ci/cirrus/freebsd-13.vars b/ci/cirrus/freebsd-13.vars index 278b5834f6..f24b921300 100644 --- a/ci/cirrus/freebsd-13.vars +++ b/ci/cirrus/freebsd-13.vars @@ -11,6 +11,6 @@ MAKE='/usr/local/bin/gmake' NINJA='/usr/local/bin/ninja' PACKAGING_COMMAND='pkg' PIP3='/usr/local/bin/pip-3.8' -PKGS='augeas bash-completion ca_root_nss ccache codespell cppi curl cyrus-sasl diffutils fusefs-libs gettext git glib gmake gnugrep gnutls gsed libpcap libpciaccess libssh libssh2 libxml2 libxslt meson ninja perl5 pkgconf polkit py39-black py39-docutils py39-flake8 py39-pytest python3 qemu readline yajl' +PKGS='augeas bash-completion ca_root_nss ccache codespell cppi curl cyrus-sasl diffutils fusefs-libs gettext git glib gmake gnugrep gnutls gsed libpcap libpciaccess libssh libssh2 libxml2 libxslt meson ninja perl5 pkgconf polkit py311-black py311-docutils py311-flake8 py311-pytest python3 qemu readline yajl' PYPI_PKGS='' PYTHON='/usr/local/bin/python3' diff --git a/ci/cirrus/freebsd-14.vars b/ci/cirrus/freebsd-14.vars index 278b5834f6..f24b921300 100644 --- a/ci/cirrus/freebsd-14.vars +++ b/ci/cirrus/freebsd-14.vars @@ -11,6 +11,6 @@ MAKE='/usr/local/bin/gmake' NINJA='/usr/local/bin/ninja' PACKAGING_COMMAND='pkg' PIP3='/usr/local/bin/pip-3.8' -PKGS='augeas bash-completion ca_root_nss ccache codespell cppi curl cyrus-sasl diffutils fusefs-libs gettext git glib gmake gnugrep gnutls gsed libpcap libpciaccess libssh libssh2 libxml2 libxslt meson ninja perl5 pkgconf polkit py39-black py39-docutils py39-flake8 py39-pytest python3 qemu readline yajl' +PKGS='augeas bash-completion ca_root_nss ccache codespell cppi curl cyrus-sasl diffutils fusefs-libs gettext git glib gmake gnugrep gnutls gsed libpcap libpciaccess libssh libssh2 libxml2 libxslt meson ninja perl5 pkgconf polkit py311-black py311-docutils py311-flake8 py311-pytest python3 qemu readline yajl' PYPI_PKGS='' PYTHON='/usr/local/bin/python3' -- 2.44.2

2 1

[PATCH 0/2] qemu: Add support for pauth Arm CPU feature
by Andrea Bolognani 15 Jul '24

15 Jul '24

Andrea Bolognani (2): cpu_map: Add pauth Arm CPU feature tests: Add coverage for pauth Arm CPU feature src/cpu_map/arm_features.xml | 3 ++ ...aarch64-features-pauth.aarch64-latest.args | 31 +++++++++++++++++++ .../aarch64-features-pauth.aarch64-latest.xml | 28 +++++++++++++++++ .../aarch64-features-pauth.xml | 17 ++++++++++ tests/qemuxmlconftest.c | 1 + 5 files changed, 80 insertions(+) create mode 100644 tests/qemuxmlconfdata/aarch64-features-pauth.aarch64-latest.args create mode 100644 tests/qemuxmlconfdata/aarch64-features-pauth.aarch64-latest.xml create mode 100644 tests/qemuxmlconfdata/aarch64-features-pauth.xml -- 2.45.2

2 7

[PATCH] docs: use real examples for QEMU cli passthrough
by Daniel P. Berrangé 15 Jul '24

15 Jul '24

User feedback has shown that the examples are not clear enough to illustrate the cli passthrough concept in action. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- docs/drvqemu.rst | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/docs/drvqemu.rst b/docs/drvqemu.rst index 1ee4b1e366..e52d9ffd47 100644 --- a/docs/drvqemu.rst +++ b/docs/drvqemu.rst @@ -595,8 +595,11 @@ Example: <emulator>/usr/bin/qemu-system-x86_64</emulator> </devices> <qemu:commandline> - <qemu:arg value='-newarg'/> - <qemu:env name='QEMU_ENV' value='VAL'/> + <qemu:arg value='-device'/> + <qemu:arg value='ipmi-bmc-sim,id=bmc0'/> + <qemu:arg value='-device'/> + <qemu:arg value='smbus-impi,bmc=bmc0'/> + <qemu:env name='QEMU_MODULE_DIR' value='/usr/lib64/qemu'/> </qemu:commandline> </domain> -- 2.45.1

2 1

[PATCH v3 00/17] hw/sd/sdcard: Accumulation of cleanups and fixes
by Philippe Mathieu-Daudé 11 Jul '24

11 Jul '24

Since v2: - Tested-by from Cédric recorded - more patches added :S Since v1: - various patches merged, few more added Various SD card cleanups and fixes accumulated over the years. Various have been useful to help integrating eMMC support (which will come later). Full series for testing: https://gitlab.com/philmd/qemu/-/tags/emmc-v4 Cédric Le Goater (1): hw/sd/sdcard: Introduce definitions for EXT_CSD register Philippe Mathieu-Daudé (16): hw/sd/sdcard: Deprecate support for spec v1.10 hw/sd/sdcard: Use spec v3.01 by default hw/sd/sdcard: Track last command used to help logging hw/sd/sdcard: Trace block offset in READ/WRITE data accesses hw/sd/sdcard: Trace requested address computed by sd_req_get_address() hw/sd/sdcard: Do not store vendor data on block drive (CMD56) hw/sd/sdcard: Send WRITE_PROT bits MSB first (CMD30) hw/sd/sdcard: Send NUM_WR_BLOCKS bits MSB first (ACMD22) hw/sd/sdcard: Use READY_FOR_DATA definition instead of magic value hw/sd/sdcard: Assign SDCardStates enum values hw/sd/sdcard: Simplify sd_inactive_state handling hw/sd/sdcard: Restrict SWITCH_FUNCTION to sd_transfer_state (CMD6) hw/sd/sdcard: Add direct reference to SDProto in SDState hw/sd/sdcard: Extract sd_blk_len() helper tests/qtest: Disable npcm7xx_sdhci tests using hardcoded RCA hw/sd/sdcard: Generate random RCA value docs/about/deprecated.rst | 6 ++ hw/sd/sdmmc-internal.h | 97 +++++++++++++++++++++ hw/sd/sd.c | 145 ++++++++++++++++++------------- tests/qtest/npcm7xx_sdhci-test.c | 7 ++ hw/sd/trace-events | 6 +- 5 files changed, 199 insertions(+), 62 deletions(-) -- 2.41.0

4 34

[PATCH] qemu: Don't leave beingDestroyed=true on inactive domain
by Jiri Denemark 11 Jul '24

11 Jul '24

Recent commit v10.4.0-87-gd9935a5c4f made a reasonable change to only reset beingDestroyed back to false when vm->def->id is reset to make sure other code can detect a domain is (about to become) inactive. It even added a comment saying any caller of qemuProcessBeginStopJob is supposed to call qemuProcessStop to clear beingDestroyed. But not every caller really does so because they first call qemuProcessBeginStopJob and then check whether a domain is still running. If not the qemuProcessStop call is skipped leaving beingDestroyed=true. In case of a persistent domain this may block incoming migrations of such domain as the migration code would think the domain died unexpectedly (even though it's still running). The qemuProcessBeginStopJob function is a wrapper around virDomainObjBeginJob, but virDomainObjEndJob was used directly for cleanup. This patch introduces a new qemuProcessEndStopJob wrapper around virDomainObjEndJob to properly undo everything qemuProcessBeginStopJob did. https://issues.redhat.com/browse/RHEL-43309 Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/qemu/qemu_driver.c | 4 ++-- src/qemu/qemu_process.c | 20 ++++++++++++++++---- src/qemu/qemu_process.h | 1 + 3 files changed, 19 insertions(+), 6 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index f898d85667..9f3013e231 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -2102,7 +2102,7 @@ qemuDomainDestroyFlags(virDomainPtr dom, endjob: if (ret == 0) qemuDomainRemoveInactive(driver, vm, 0, false); - virDomainObjEndJob(vm); + qemuProcessEndStopJob(vm); cleanup: virDomainObjEndAPI(&vm); @@ -3888,7 +3888,7 @@ processMonitorEOFEvent(virQEMUDriver *driver, endjob: qemuDomainRemoveInactive(driver, vm, 0, false); - virDomainObjEndJob(vm); + qemuProcessEndStopJob(vm); } diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 7cbe521a6e..25dfd04272 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -8421,7 +8421,8 @@ qemuProcessKill(virDomainObj *vm, unsigned int flags) * qemuProcessBeginStopJob: * * Stop all current jobs by killing the domain and start a new one for - * qemuProcessStop. + * qemuProcessStop. The caller has to make sure qemuProcessEndStopJob is + * called to properly cleanup the job. */ int qemuProcessBeginStopJob(virDomainObj *vm, @@ -8448,8 +8449,9 @@ qemuProcessBeginStopJob(virDomainObj *vm, goto error; /* priv->beingDestroyed is deliberately left set to 'true' here. Caller - * is supposed to call qemuProcessStop, which will reset it after - * 'vm->def->id' is set to -1 */ + * is supposed to call qemuProcessStop (which will reset it after + * 'vm->def->id' is set to -1) and/or qemuProcessEndStopJob to do proper + * cleanup. */ return 0; error: @@ -8458,6 +8460,16 @@ qemuProcessBeginStopJob(virDomainObj *vm, } +void +qemuProcessEndStopJob(virDomainObj *vm) +{ + if (!virDomainObjIsActive(vm)) + QEMU_DOMAIN_PRIVATE(vm)->beingDestroyed = false; + + virDomainObjEndJob(vm); +} + + void qemuProcessStop(virQEMUDriver *driver, virDomainObj *vm, virDomainShutoffReason reason, @@ -8800,7 +8812,7 @@ qemuProcessAutoDestroy(virDomainObj *dom, qemuDomainRemoveInactive(driver, dom, 0, false); - virDomainObjEndJob(dom); + qemuProcessEndStopJob(dom); virObjectEventStateQueue(driver->domainEventState, event); } diff --git a/src/qemu/qemu_process.h b/src/qemu/qemu_process.h index c1ea949215..cb67bfcd2d 100644 --- a/src/qemu/qemu_process.h +++ b/src/qemu/qemu_process.h @@ -169,6 +169,7 @@ typedef enum { int qemuProcessBeginStopJob(virDomainObj *vm, virDomainJob job, bool forceKill); +void qemuProcessEndStopJob(virDomainObj *vm); void qemuProcessStop(virQEMUDriver *driver, virDomainObj *vm, virDomainShutoffReason reason, -- 2.45.2

2 1

[PATCH] virt-host-validate: Rework calling of driver validation
by Michal Privoznik 11 Jul '24

11 Jul '24

It all started with me looking at the --help output which also printed "bhyve" as supported hypervisor type. Well, it's not on my Linux machine. To resolve this, I'm just creating a static array of { "$driver", callback() } pairs and iterating over it. The array is then initialized at compile time with supported drivers. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tools/virt-host-validate.c | 80 +++++++++++++++++++++----------------- 1 file changed, 45 insertions(+), 35 deletions(-) diff --git a/tools/virt-host-validate.c b/tools/virt-host-validate.c index 426648a5d3..365b8acd92 100644 --- a/tools/virt-host-validate.c +++ b/tools/virt-host-validate.c @@ -29,6 +29,7 @@ #include "internal.h" #include "virerror.h" #include "virgettext.h" +#include "virglibutil.h" #include "virt-host-validate-common.h" #if WITH_QEMU @@ -44,26 +45,58 @@ # include "virt-host-validate-ch.h" #endif +typedef struct _virValidateCallbacks virValidateCallbacks; +struct _virValidateCallbacks { + const char *name; + int (*callback)(void); +}; + +static virValidateCallbacks validateCallbacks[] = { +#if WITH_QEMU + { "qemu", virHostValidateQEMU }, +#endif +#if WITH_LXC + { "lxc", virHostValidateLXC }, +#endif +#if WITH_BHYVE + { "bhyve", virHostValidateBhyve }, +#endif +#if WITH_CH + { "ch", virHostValidateCh }, +#endif +}; + static void show_help(FILE *out, const char *argv0) { + g_autofree char *hvs = NULL; + char *hvs_list[G_N_ELEMENTS(validateCallbacks) + 1] = { }; + size_t i; + + for (i = 0; i < G_N_ELEMENTS(validateCallbacks); i++) { + hvs_list[i] = g_strdup_printf(" - %1$s", validateCallbacks[i].name); + } + + hvs = g_strjoinv("\n", hvs_list); + + for (i = 0; i < G_N_ELEMENTS(validateCallbacks); i++) { + g_free(hvs_list[i]); + } + fprintf(out, _("\n" "syntax: %1$s [OPTIONS] [HVTYPE]\n" "\n" " Hypervisor types:\n" "\n" - " - qemu\n" - " - lxc\n" - " - bhyve\n" - " - ch\n" + "%2$s\n" "\n" " Options:\n" " -h, --help Display command line help\n" " -v, --version Display command version\n" " -q, --quiet Don't display progress information\n" "\n"), - argv0); + argv0, hvs); } static void @@ -87,6 +120,7 @@ main(int argc, char **argv) int ret = EXIT_SUCCESS; bool quiet = false; bool usedHvname = false; + size_t i; if (virGettextInitialize() < 0 || virErrorInitialize() < 0) { @@ -126,37 +160,13 @@ main(int argc, char **argv) virValidateSetQuiet(quiet); -#if WITH_QEMU - if (!hvname || STREQ(hvname, "qemu")) { - usedHvname = true; - if (virHostValidateQEMU() < 0) - ret = EXIT_FAILURE; + for (i = 0; i < G_N_ELEMENTS(validateCallbacks); i++) { + if (!hvname || STREQ(hvname, validateCallbacks[i].name)) { + usedHvname = true; + if (validateCallbacks[i].callback() < 0) + ret = EXIT_FAILURE; + } } -#endif - -#if WITH_LXC - if (!hvname || STREQ(hvname, "lxc")) { - usedHvname = true; - if (virHostValidateLXC() < 0) - ret = EXIT_FAILURE; - } -#endif - -#if WITH_BHYVE - if (!hvname || STREQ(hvname, "bhyve")) { - usedHvname = true; - if (virHostValidateBhyve() < 0) - ret = EXIT_FAILURE; - } -#endif - -#if WITH_CH - if (!hvname || STREQ(hvname, "ch")) { - usedHvname = true; - if (virHostValidateCh() < 0) - ret = EXIT_FAILURE; - } -#endif if (hvname && !usedHvname) { fprintf(stderr, _("%1$s: unsupported hypervisor name %2$s\n"), -- 2.44.2

2 1

[PATCH] virt-host-validate: Drop extra "PASS"
by Michal Privoznik 11 Jul '24

11 Jul '24

If virt-host-validate is ran on a SEV-SNP capable machine, an extra "PASS" is printed out. This is because virHostValidateAMDSev() prints "PASS" and then returns 1 (indicating success) which in turn makes the caller (virHostValidateSecureGuests()) print "PASS" again. Just drop the extra printing in the caller and let virHostValidateAMDSev() do all the printing. Fixes: 1a8f646f291775d2423ce4e4df62ad69f06ab827 Resolves: https://issues.redhat.com/browse/RHEL-46868 Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tools/virt-host-validate-common.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/tools/virt-host-validate-common.c b/tools/virt-host-validate-common.c index a29a5b6d5f..86db4815c3 100644 --- a/tools/virt-host-validate-common.c +++ b/tools/virt-host-validate-common.c @@ -488,11 +488,7 @@ int virHostValidateSecureGuests(const char *hvname, return VIR_VALIDATE_FAILURE(level); } } else if (hasAMDSev) { - int rc = virHostValidateAMDSev(hvname, level); - - if (rc > 0) - virValidatePass(); - return rc; + return virHostValidateAMDSev(hvname, level); } virValidateFail(level, -- 2.44.2

2 1

[PATCH v1 0/1] Add support for Vfio Stats in Libvirt
by Kshitij Jha 10 Jul '24

10 Jul '24

Hi, This is a solution to add support for data transferred by Vfio devices during the Live Migration in Libvirt. Currently libvirt only supports a few essential stats like memory transferred, dirty rate and number of iteration etc. returned as part of virDomainGetJobStats for each iteration. However it does not return stats for Vfio devices like data transferred by the device during the migration With this change virDomainGetJobStats will be able to support Vfio stats. Regards Kshitij Jha Kshitij Jha (1): Include support for Vfio stats during Migration include/libvirt/libvirt-domain.h | 9 +++++++++ src/qemu/qemu_domainjob.c | 6 ++++++ src/qemu/qemu_monitor.h | 1 + src/qemu/qemu_monitor_json.c | 12 ++++++++++++ 4 files changed, 28 insertions(+) -- 2.22.3

2 2

[PATCH] network: allow "modify" option for DNS-Txt records
by Adam Julis 10 Jul '24

10 Jul '24

The "modify" command allows to replace an existing record (its text value). The primary key is the name of the record. If duplicity or missing record detected, throw error. Tests in networkxml2xmlupdatetest.c contain replacements of an existing DNS-text record and failure due to non-existing record. Resolves: https://gitlab.com/libvirt/libvirt/-/issues/639 Signed-off-by: Adam Julis <ajulis(a)redhat.com> --- src/conf/network_conf.c | 18 ++++++++----- .../dns-txt-record-modify-fail.xml | 1 + .../dns-txt-record-modify-success.xml | 1 + .../nat-network-dns-txt-modify-ok.xml | 26 +++++++++++++++++++ tests/networkxml2xmlupdatetest.c | 9 +++++++ 5 files changed, 49 insertions(+), 6 deletions(-) create mode 100644 tests/networkxml2xmlupdatein/dns-txt-record-modify-fail.xml create mode 100644 tests/networkxml2xmlupdatein/dns-txt-record-modify-success.xml create mode 100644 tests/networkxml2xmlupdateout/nat-network-dns-txt-modify-ok.xml diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c index fc387f9566..dd362b6ab2 100644 --- a/src/conf/network_conf.c +++ b/src/conf/network_conf.c @@ -3351,12 +3351,6 @@ virNetworkDefUpdateDNSTxt(virNetworkDef *def, bool isAdd = (command == VIR_NETWORK_UPDATE_COMMAND_ADD_FIRST || command == VIR_NETWORK_UPDATE_COMMAND_ADD_LAST); - if (command == VIR_NETWORK_UPDATE_COMMAND_MODIFY) { - virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", - _("DNS TXT records cannot be modified, only added or deleted")); - goto cleanup; - } - if (virNetworkDefUpdateCheckElementName(def, ctxt->node, "txt") < 0) goto cleanup; @@ -3395,6 +3389,18 @@ virNetworkDefUpdateDNSTxt(virNetworkDef *def, virNetworkDNSTxtDefClear(&dns->txts[foundIdx]); VIR_DELETE_ELEMENT(dns->txts, foundIdx, dns->ntxts); + } else if (command == VIR_NETWORK_UPDATE_COMMAND_MODIFY) { + + if (foundIdx == dns->ntxts) { + virReportError(VIR_ERR_OPERATION_INVALID, + _("couldn't locate a matching DNS TXT record in network %1$s"), + def->name); + goto cleanup; + } + + VIR_FREE(dns->txts[foundIdx].value); + dns->txts[foundIdx].value = g_strdup(txt.value); + } else { virNetworkDefUpdateUnknownCommand(command); goto cleanup; diff --git a/tests/networkxml2xmlupdatein/dns-txt-record-modify-fail.xml b/tests/networkxml2xmlupdatein/dns-txt-record-modify-fail.xml new file mode 100644 index 0000000000..75ed475fe1 --- /dev/null +++ b/tests/networkxml2xmlupdatein/dns-txt-record-modify-fail.xml @@ -0,0 +1 @@ +<txt name='notexisted' value='modified example'/> diff --git a/tests/networkxml2xmlupdatein/dns-txt-record-modify-success.xml b/tests/networkxml2xmlupdatein/dns-txt-record-modify-success.xml new file mode 100644 index 0000000000..e16c352253 --- /dev/null +++ b/tests/networkxml2xmlupdatein/dns-txt-record-modify-success.xml @@ -0,0 +1 @@ +<txt name='example' value='modified example'/> diff --git a/tests/networkxml2xmlupdateout/nat-network-dns-txt-modify-ok.xml b/tests/networkxml2xmlupdateout/nat-network-dns-txt-modify-ok.xml new file mode 100644 index 0000000000..4b4dda094a --- /dev/null +++ b/tests/networkxml2xmlupdateout/nat-network-dns-txt-modify-ok.xml @@ -0,0 +1,26 @@ +<network> + <name>default</name> + <uuid>81ff0d90-c91e-6742-64da-4a736edb9a9b</uuid> + <forward dev='eth1' mode='nat'> + <interface dev='eth1'/> + </forward> + <bridge name='virbr0' stp='on' delay='0'/> + <dns> + <txt name='example' value='modified example'/> + </dns> + <ip address='192.168.122.1' netmask='255.255.255.0'> + <dhcp> + <range start='192.168.122.2' end='192.168.122.254'/> + <host mac='00:16:3e:77:e2:ed' name='a.example.com' ip='192.168.122.10'/> + <host mac='00:16:3e:3e:a9:1a' name='b.example.com' ip='192.168.122.11'/> + </dhcp> + </ip> + <ip family='ipv4' address='192.168.123.1' netmask='255.255.255.0'> + </ip> + <ip family='ipv6' address='2001:db8:ac10:fe01::1' prefix='64'> + </ip> + <ip family='ipv6' address='2001:db8:ac10:fd01::1' prefix='64'> + </ip> + <ip family='ipv4' address='10.24.10.1'> + </ip> +</network> diff --git a/tests/networkxml2xmlupdatetest.c b/tests/networkxml2xmlupdatetest.c index 59e6ce98e5..875cede035 100644 --- a/tests/networkxml2xmlupdatetest.c +++ b/tests/networkxml2xmlupdatetest.c @@ -306,6 +306,15 @@ mymain(void) "dns-txt-record-snowman", "nat-network-dns-txt-record", VIR_NETWORK_UPDATE_COMMAND_DELETE); + DO_TEST("modify-dns-txt-record", + "dns-txt-record-modify-success", + "nat-network-dns-txt-record", + "nat-network-dns-txt-modify-ok", + VIR_NETWORK_UPDATE_COMMAND_MODIFY); + DO_TEST_FAIL("modify-missing-dns-txt-record", + "dns-txt-record-modify-fail", + "nat-network-dns-txt-record", + VIR_NETWORK_UPDATE_COMMAND_MODIFY); section = VIR_NETWORK_SECTION_DNS_SRV; -- 2.45.2

2 1

[PATCH] network: allow "modify" option for DNS hostname
by Adam Julis 10 Jul '24

10 Jul '24

The "modify" command allows you to replace an existing record (its hostname, sub-elements). IP address acts as the primary key. If it is not found, the attempt ends with an error message. If the XML contains a duplicate address, it will select the last one. Tests in networkxml2xmlupdatetest.c contain replacements of an existing DNS-Host record and failure due to non-existing record. Resolves: https://gitlab.com/libvirt/libvirt/-/issues/639 Signed-off-by: Adam Julis <ajulis(a)redhat.com> --- src/conf/network_conf.c | 28 ++++++++++++++----- .../dns-host-modify-not-existing.xml | 4 +++ .../dns-host-modify.xml | 5 ++++ .../nat-network-dns-hosts-modified.xml | 28 +++++++++++++++++++ tests/networkxml2xmlupdatetest.c | 9 ++++++ 5 files changed, 67 insertions(+), 7 deletions(-) create mode 100644 tests/networkxml2xmlupdatein/dns-host-modify-not-existing.xml create mode 100644 tests/networkxml2xmlupdatein/dns-host-modify.xml create mode 100644 tests/networkxml2xmlupdateout/nat-network-dns-hosts-modified.xml diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c index f5ccf4bd12..2a541cd5b0 100644 --- a/src/conf/network_conf.c +++ b/src/conf/network_conf.c @@ -3138,19 +3138,13 @@ virNetworkDefUpdateDNSHost(virNetworkDef *def, unsigned int fflags G_GNUC_UNUSED) { size_t i, j, k; - int foundIdx = -1, ret = -1; + int foundIdx = -1, ret = -1, foundIdxModify = -1; virNetworkDNSDef *dns = &def->dns; virNetworkDNSHostDef host = { 0 }; bool isAdd = (command == VIR_NETWORK_UPDATE_COMMAND_ADD_FIRST || command == VIR_NETWORK_UPDATE_COMMAND_ADD_LAST); int foundCt = 0; - if (command == VIR_NETWORK_UPDATE_COMMAND_MODIFY) { - virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", - _("DNS HOST records cannot be modified, only added or deleted")); - goto cleanup; - } - if (virNetworkDefUpdateCheckElementName(def, ctxt->node, "host") < 0) goto cleanup; @@ -3163,6 +3157,12 @@ virNetworkDefUpdateDNSHost(virNetworkDef *def, if (virSocketAddrEqual(&host.ip, &dns->hosts[i].ip)) foundThisTime = true; + /* modify option required index of matching ip-address, the loop under + * this comment could affect results of found index foundThisTime, + * so the foundIdxModify is there used instead */ + if (foundThisTime) + foundIdxModify = i; + /* when adding we want to only check duplicates of address since having * multiple addresses with the same hostname is a legitimate configuration */ if (!isAdd) { @@ -3213,6 +3213,20 @@ virNetworkDefUpdateDNSHost(virNetworkDef *def, virNetworkDNSHostDefClear(&dns->hosts[foundIdx]); VIR_DELETE_ELEMENT(dns->hosts, foundIdx, dns->nhosts); + } else if (command == VIR_NETWORK_UPDATE_COMMAND_MODIFY) { + + if (foundCt == 0) { + virReportError(VIR_ERR_OPERATION_INVALID, + _("couldn't locate a matching DNS HOST record in network %1$s"), + def->name); + goto cleanup; + } + + virNetworkDNSHostDefClear(&dns->hosts[foundIdxModify]); + + memcpy(&dns->hosts[foundIdxModify], &host, sizeof(virNetworkDNSHostDef)); + memset(&host, 0, sizeof(virNetworkDNSHostDef)); + } else { virNetworkDefUpdateUnknownCommand(command); goto cleanup; diff --git a/tests/networkxml2xmlupdatein/dns-host-modify-not-existing.xml b/tests/networkxml2xmlupdatein/dns-host-modify-not-existing.xml new file mode 100644 index 0000000000..357fccd110 --- /dev/null +++ b/tests/networkxml2xmlupdatein/dns-host-modify-not-existing.xml @@ -0,0 +1,4 @@ +<host ip='192.168.122.333'> + <hostname>shared</hostname> + <hostname>names</hostname> +</host> diff --git a/tests/networkxml2xmlupdatein/dns-host-modify.xml b/tests/networkxml2xmlupdatein/dns-host-modify.xml new file mode 100644 index 0000000000..78b9fd88a6 --- /dev/null +++ b/tests/networkxml2xmlupdatein/dns-host-modify.xml @@ -0,0 +1,5 @@ +<host ip='192.168.122.2'> + <hostname>Another</hostname> + <hostname>decent</hostname> + <hostname>names</hostname> +</host> diff --git a/tests/networkxml2xmlupdateout/nat-network-dns-hosts-modified.xml b/tests/networkxml2xmlupdateout/nat-network-dns-hosts-modified.xml new file mode 100644 index 0000000000..8fcaad15d1 --- /dev/null +++ b/tests/networkxml2xmlupdateout/nat-network-dns-hosts-modified.xml @@ -0,0 +1,28 @@ +<network> + <name>default</name> + <uuid>81ff0d90-c91e-6742-64da-4a736edb9a9c</uuid> + <forward dev='eth0' mode='nat'> + <interface dev='eth0'/> + </forward> + <bridge name='virbr0' stp='on' delay='0'/> + <dns forwardPlainNames='no'> + <host ip='192.168.122.122'> + <hostname>pudding</hostname> + </host> + <host ip='192.168.122.1'> + <hostname>host</hostname> + <hostname>gateway</hostname> + </host> + <host ip='192.168.122.2'> + <hostname>Another</hostname> + <hostname>decent</hostname> + <hostname>names</hostname> + </host> + <host ip='fd8f:1391:3a82:150::c0a8:9603'> + <hostname>shared</hostname> + <hostname>names</hostname> + </host> + </dns> + <ip address='192.168.122.1' netmask='255.255.255.0'> + </ip> +</network> diff --git a/tests/networkxml2xmlupdatetest.c b/tests/networkxml2xmlupdatetest.c index afe2b1f574..383cbf85ce 100644 --- a/tests/networkxml2xmlupdatetest.c +++ b/tests/networkxml2xmlupdatetest.c @@ -276,6 +276,15 @@ mymain(void) "nat-network-dns-hosts", "nat-network-no-hosts", VIR_NETWORK_UPDATE_COMMAND_DELETE); + DO_TEST("modify-dns-host", + "dns-host-modify", + "nat-network-dns-hosts", + "nat-network-dns-hosts-modified", + VIR_NETWORK_UPDATE_COMMAND_MODIFY); + DO_TEST_FAIL("modify-dns-host-not-existing", + "dns-host-modify-not-existing", + "nat-network-dns-hosts", + VIR_NETWORK_UPDATE_COMMAND_MODIFY); section = VIR_NETWORK_SECTION_DNS_TXT; -- 2.45.2

2 1

[PATCH] domain_conf: comment not match the code below
by Adam Julis 10 Jul '24

10 Jul '24

The outdated comment refers to a non-existent member in the virDomainObj structure. Signed-off-by: Adam Julis <ajulis(a)redhat.com> --- src/conf/domain_conf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 7033b4e9fe..115862bbde 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -4449,7 +4449,7 @@ virDomainObjGetDefs(virDomainObj *vm, * @live: set to true if live config was returned (may be omitted) * * Helper function to resolve @flags and return the correct domain pointer - * object. This function returns one of @vm->def or @vm->persistentDef + * object. This function returns one of @vm->def or @vm->newDef * according to @flags. @live is set to true if the live vm config will be * returned. This helper should be used only in APIs that guarantee * that @flags contains exactly one of VIR_DOMAIN_AFFECT_LIVE or @@ -4489,7 +4489,7 @@ virDomainObjGetOneDefState(virDomainObj *vm, * @flags: for virDomainModificationImpact * * Helper function to resolve @flags and return the correct domain pointer - * object. This function returns one of @vm->def or @vm->persistentDef + * object. This function returns one of @vm->def or @vm->newDef * according to @flags. This helper should be used only in APIs that guarantee * that @flags contains exactly one of VIR_DOMAIN_AFFECT_LIVE or * VIR_DOMAIN_AFFECT_CONFIG and not both. -- 2.45.2

2 1

[PATCH] virt-aa-helper: Drop needless comments
by Michal Privoznik 09 Jul '24

09 Jul '24

When generating paths for a domain specific AppArmor profile each path undergoes a validation where it's matched against an array of well known prefixes (among other things). Now, for OVMF/AAVMF/... images we have a list and some entries have comments to which type of image the entry belongs to. For instance: "/usr/share/OVMF/", /* for OVMF images */ "/usr/share/AAVMF/", /* for AAVMF images */ But these comments are pretty useless. The path itself already gives away the image type. Drop them. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/security/virt-aa-helper.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index a3f85d26b0..c1e89dc6cf 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -475,15 +475,15 @@ valid_path(const char *path, const bool readonly) "/initrd", "/initrd.img", "/usr/share/edk2/", - "/usr/share/edk2-ovmf/", /* for OVMF images */ - "/usr/share/OVMF/", /* for OVMF images */ - "/usr/share/ovmf/", /* for OVMF images */ - "/usr/share/AAVMF/", /* for AAVMF images */ + "/usr/share/edk2-ovmf/", + "/usr/share/OVMF/", + "/usr/share/ovmf/", + "/usr/share/AAVMF/", "/usr/share/qemu-efi/", /* for AAVMF images */ - "/usr/share/qemu-efi-aarch64/", /* for AAVMF images */ + "/usr/share/qemu-efi-aarch64/", "/usr/share/qemu/", /* SUSE path for OVMF and AAVMF images */ - "/usr/lib/u-boot/", /* u-boot loaders for qemu */ - "/usr/lib/riscv64-linux-gnu/opensbi" /* RISC-V SBI implementation */ + "/usr/lib/u-boot/", + "/usr/lib/riscv64-linux-gnu/opensbi", }; /* override the above with these */ const char * const override[] = { -- 2.44.2

2 1

[libvirt PATCHv2 0/5] qemu: do not use deprecated options for new virtiofsd (virtiofs epopee)
by Ján Tomko 09 Jul '24

09 Jul '24

Also, allow migration if we think we can. Ján Tomko (5): tests: vhostuser: add virtiofsd json descriptor tests: qemuxmlconf: adjust test case to new virtiofsd qemu: fill capabilities for virtiofsd qemu: do not use deprecated options for new virtiofsd qemu: migration: allow migration for virtiofs src/conf/domain_conf.h | 1 + src/qemu/qemu_migration.c | 9 +- src/qemu/qemu_vhost_user.c | 60 ++++++++++++ src/qemu/qemu_vhost_user.h | 12 +++ src/qemu/qemu_virtiofs.c | 92 +++++++++++++------ .../libexec/qemu/vhost-user/test-virtiofsd | 10 ++ .../qemu/vhost-user/50-qemu-virtiofsd.json | 5 + tests/qemuvhostusertest.c | 1 + .../vhost-user-fs-fd-memory.xml | 3 +- 9 files changed, 161 insertions(+), 32 deletions(-) create mode 100755 tests/qemuvhostuserdata/usr/libexec/qemu/vhost-user/test-virtiofsd create mode 100644 tests/qemuvhostuserdata/usr/share/qemu/vhost-user/50-qemu-virtiofsd.json -- 2.45.2

2 7

[PATCH v3 0/2] security_manager: Fix security manager deadlock after libvirtd fork()
by hongmianquan 09 Jul '24

09 Jul '24

We have the stack security driver, which internally manages other security drivers, just call them "top" and "nested". We call virSecurityStackPreFork() to lock the top one, and it also locks and then unlocks the nested drivers prior to fork. Then in qemuSecurityPostFork(), it unlocks the top one, but not the nested ones. Thus, if one of the nested drivers ("dac" or "selinux") is still locked, it will cause a deadlock. If we always surround nested locks with top lock, it is always secure. Because we have got top lock before fork child libvirtd. However, it is not always the case in the current code, We discovered this case: the nested list obtained through the qemuSecurityGetNested() will be locked directly for subsequent use, such as in virQEMUDriverCreateCapabilities(), where the nested list is locked using qemuSecurityGetDOI, but the top one is not locked beforehand. The problem stack is as follows: libvirtd thread1 libvirtd thread2 child libvirtd | | | | | | virsh capabilities qemuProcessLanuch | | | | | lock top | | | | lock nested | | | | | | fork------------------->|(nested lock held by thread1) | | | | | | unlock nested unlock top unlock top | | qemuSecuritySetSocketLabel | | lock nested (deadlock) v3 changes: Made modifications based on Michal's comments - ensured matching qemuSecurityStackLock() and qemuSecurityStackUnlock() - modify the correct order in libvirt_private.syms - split the code streamlining part into a separate patch hongmianquan (2): security_manager: Ensure top lock is acquired before nested locks security_manager: Remove redundant qemuSecurityGetNested() call src/libvirt_private.syms | 2 ++ src/qemu/qemu_conf.c | 13 +++++++++++-- src/qemu/qemu_driver.c | 21 +++++++++++++-------- src/qemu/qemu_security.h | 2 ++ src/security/security_manager.c | 22 ++++++++++++++++++++++ src/security/security_manager.h | 2 ++ 6 files changed, 52 insertions(+), 10 deletions(-) -- 2.20.1

3 5

[PATCH 0/1] qemuDomainChangeNet: check virtio options for non-virtio models
by Miroslav Los 09 Jul '24

09 Jul '24

Hi, in our project, we create all of our domain interfaces regardless of model with a <driver><host csum='off'/></driver>. The documentation is not clear as to which models support offloading or not, and the domain validates. In libvirtd, since <driver> is present, this causes each network device to have its virDomainVirtioOptions *virtio to be non-NULL, with all members ABSENT, which is also deemed valid. However, clients receive the interface elements without the driver. When the modified value is sent back via updateDeviceFlags, the candidate's virtio attribute is thus NULL. The validation in qemuDomainChangeNet requires both structs to be equal or both NULL, which is violated by such a request. This can be seen also with how virsh domif-setlink operates: virsh # domiflist 5 Interface Type Source Model MAC -------------------------------------------------------- - udp - e1000 52:54:00:1c:10:42 - udp - e1000 52:54:00:09:29:9c - udp - e1000 52:54:00:1d:a6:8d - udp - e1000 52:54:00:07:4a:83 52:54:00:1c:10:42 up virsh # domif-setlink 5 52:54:00:1c:10:42 down error: Failed to update interface link state error: Operation not supported: cannot modify virtio network device driver options A workaround is to add an empty <driver> in the update request, which works until libvirt is restarted. The driver element on a non-virtio interface is not present when the domain is reloaded, thus the pointer is NULL in olddev, causing the inverse issue and same error. Our current workaround is to retry with the empty driver element only if the first call fails, but we think libvirt should handle this struct more consistently. This patch expands the check to the non-virtio case, where the olddev is assumed to be valid, and newdev's virtio options are validated to be NULL or all-ABSENT. Since the driver element is normally preserved for virtio models, the stricter equality condition is left unchanged in that case. I did not investigate further if the discrepancy between NULL and all-ABSENT values for virDomainVirtioOptions can cause a problem elsewhere, or in other device kinds. Though I do wonder if the options handling should be improved. One possibility is that non-virtio devices keep the value NULL, e.g. after validating that the driver element does not contain any of the options, whereas virtio devices always create the struct even with no driver element present on input. Another option is that NULL pointer is treated the same as all-ABSENT struct, e.g. in virDomainCheckVirtioOptionsAreAbsent and virDomainVirtioOptionsCheckABIStability; the former does that while the latter does not. These functions would also need to become callable from the qemu driver. Miroslav Los (1): qemuDomainChangeNet: check virtio options for non-virtio models src/qemu/qemu_hotplug.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) -- 2.25.1

2 2