- Devel - libvirt List Archives

[libvirt] [PATCH] docs: minor updates to to new API doc
by Jonathon Jongsma 05 Jul '19

05 Jul '19

Fix a filename and add a couple missing words. Signed-off-by: Jonathon Jongsma <jjongsma(a)redhat.com> --- docs/api_extension.html.in | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/api_extension.html.in b/docs/api_extension.html.in index d7fbbd6e90..3bbaf4893c 100644 --- a/docs/api_extension.html.in +++ b/docs/api_extension.html.in @@ -229,7 +229,7 @@ <h3><a name='rpcclient'>Implement the RPC client</a></h3> <p> - Implementing the uses the rpcgen generated .h files. The remote + Implementing the RPC client uses the rpcgen generated .h files. The remote method calls go in: </p> @@ -256,7 +256,7 @@ The server side dispatchers are implemented in: </p> - <p><code>src/remote/daemon_dispatch.c</code></p> + <p><code>src/remote/remote_daemon_dispatch.c</code></p> <p>Again, this step uses the .h files generated by make rpcgen.</p> -- 2.20.1

2 1

[libvirt] [PATCH] tests: fix mocking of virFileGetXAttrQuiet on FreeBSD
by Daniel P. Berrangé 05 Jul '19

05 Jul '19

The qemusecuritytest is failing on FreeBSD 11/12, reporting that files are not correctly restored. Debugging code printfs show that the virFileGetXAttrQuiet mock is returning 0, but the virFileGetXAttr function is seeing -1 as the return value. Essentially there appears to be some kind of optimization between the real virFileGetXAttrQuiet and the real virFileGetXAttr, which breaks when we mock virFileGetXAttrQuiet. Rather than trying to figure out how to avoid this, it is simpler to just mock virFileGetXAttr too since it is very short code. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Pushed as a CI build fix tests/qemusecuritymock.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/tests/qemusecuritymock.c b/tests/qemusecuritymock.c index 2a9095e1bf..a15eef29c9 100644 --- a/tests/qemusecuritymock.c +++ b/tests/qemusecuritymock.c @@ -156,6 +156,34 @@ virFileGetXAttrQuiet(const char *path, } +/* + * This may look redundant but is needed to work around an + * compiler quirk. The call from the real virFileGetXAttr + * to the real virFileGetXAttrQuiet has a quirk where the + * return value from virFileGetXAttrQuiet gets scrambled + * if we mock virFileGetXAttrQuiet, returning -1 instead + * of 0 despite succeeding. This happens on FreeBSD 11/12 + * hosts with CLang, and is suspected to be some kind of + * compiler optimization. By mocking this function too we + * can workaround it. + */ +int +virFileGetXAttr(const char *path, + const char *name, + char **value) +{ + int ret; + + if ((ret = virFileGetXAttrQuiet(path, name, value)) < 0) { + virReportSystemError(errno, + "Unable to get XATTR %s on %s", + name, path); + } + + return ret; +} + + int virFileSetXAttr(const char *path, const char *name, const char *value) -- 2.21.0

1 0

Re: [libvirt] [libvirt-users] UDP broadcasts vs. nat Masquerading issue
by Nikolai Zhubr 05 Jul '19

05 Jul '19

Hi Daniel and Laine, [...] >> -A POSTROUTING -o br0 -j MASQUERADE >> -A POSTROUTING -o enp0s25 -j MASQUERADE >> -A POSTROUTING -o virbr2_nic -j MASQUERADE >> -A POSTROUTING -o vnet0 -j MASQUERADE > > *None* of those rules were added by libvirt (unless your build of [...] > You can verify my "counter-claim" by running "virsh net-destroy" for all > of your libvirt networks, and seeing that the offending rules haven't > been removed. > > In short, you need to look elsewhere for the culprit. Yes, found it. You were both right, essentially. The offending rules were added by a firewall in response to new interfaces created by libvirt dynamically, due to some dubious relict settings left in the firewall. (Silly me.) So this it not an issue of libvirt indeed! Tons of thanks for the quick and precise hit! Regards, Nikolai >> >> Here, virbr2_nic and vnet0 are used by libvirt for arranging network >> configurations for VMs, ok. However, br0 is a main interface of this >> host with primary ip address, with enp0s25 being a physical nic of >> this host, and it is used for all sorts of regular (unrelated to >> virtualization) communications. Also, br0 is used for attaching >> bridged (as opposed to NATed) VMs managed by libvirt. >> >> Clearly, libvirt somehow chooses to set up masquerading for literally >> all existing network interfaces here (except lo), > > It's clear that the rules are there. It's not clear that they were added > by libvirt. > >> but I can't see a real reason for the first two rules in the list >> above. Furthermore, they corrupt UDP broadcats coming from outside and >> reaching this host (through enp0s25/br0) such that source address gets >> replaced by this hosts primary address (as per masquerading). I've >> verified this by arranging a hand-crafted UDP listener and printing >> the respective source addresses as seen by normal userspace. >> >> Now I've discovered that I can "eliminate" the problem by either: >> >> 1. Removing "-A POSTROUTING -o br0 -j MASQUERADE" (manually) >> 2. Inserting "-A POSTROUTING -s 192.168.0.0/24 -d 192.168.0.255/32 -j >> ACCEPT" >> (Of course correcting rules by hand is not a solution, just a test) >> >> So question is, how the correct rules should ideally look like? And, >> is this issue known/fixed in most current libvirt? > > Except for putting the libvirt-added rules in their own private chains > (appearing in libvirt 5.1.0, released on Feb 1, 2019), the iptables > rules added by libvirt to support its virtual networks didn't materially > change in > 10 years. Your email is the first time I've ever seen such > rules attributed to libvirt so, as I said above, I think you need to > take a deeper dive into your host system's config. > > > Good luck! >

2 1

[libvirt] [PATCH 0/3] remote: misc cleanups to client driver
by Daniel P. Berrangé 05 Jul '19

05 Jul '19

A few bits of cleanup identified during work on the daemon splitting patches sent now since they are independent. Daniel P. Berrangé (3): remote: stop declaring variables in the middle of a function remote: use autofree for many string variables remote: conditionally declare tty variable for non-Win32 platforms src/remote/remote_driver.c | 61 ++++++++++++++------------------------ 1 file changed, 23 insertions(+), 38 deletions(-) -- 2.21.0

2 4

[libvirt] [PATCH rust 0/5] ci related plus some small updates
by Sahid Orentino Ferdjaoui 05 Jul '19

05 Jul '19

Hello Daniel, When you have a moment, can you please merge this serie. It fixes CI issue, switch to bionic, and updates the tested versions. Results: https://travis-ci.org/sahid/libvirt-rust/builds/554170698 Thanks, s. Sahid Orentino Ferdjaoui (5): fix bug integration test with rust multithreading make lookup_by_id() test more robust update tested versions from 2.5.0 to 5.5.0 switch to the last ubuntu lts bionic fix code formating in README .travis.yml | 10 +++++----- README.md | 6 +++++- tests/domain.rs | 19 +++++++++---------- tests/integration_qemu.rs | 15 +++++++++++---- tests/libvirtd.sasl | 2 ++ 5 files changed, 32 insertions(+), 20 deletions(-) create mode 100644 tests/libvirtd.sasl -- 2.20.1

2 14

[libvirt] [PATCH v2 0/4] test_driver: implement some DomainGetParametersAPIs
by Ilias Stamatis 04 Jul '19

04 Jul '19

Changes since v1: - undefine the macro introduced in 1/4 when done with it's usage - simplified 4/4 by using the virNetDevBandwidthRate struct Ilias Stamatis (4): test_driver: introduce TEST_SET_PARAM macro test_driver: implement virDomainGetMemoryParameters test_driver: implement virDomainGetNumaParameters test_driver: implement testDomainGetInterfaceParameters src/test/test_driver.c | 154 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 154 insertions(+) -- 2.22.0

2 7

[libvirt] [PATCH] test_driver: avoid using nerrors unitialized in testDomainGetDiskErrors
by Ilias Stamatis 04 Jul '19

04 Jul '19

Right now nerrors can be used unitialized in the last for loop causing a bug. Signed-off-by: Ilias Stamatis <stamatis.iliass(a)gmail.com> --- src/test/test_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index ce4ff1a582..2371581f51 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -3275,7 +3275,7 @@ static int testDomainGetDiskErrors(virDomainPtr dom, virDomainObjPtr vm = NULL; int ret = -1; size_t i; - size_t nerrors; + size_t nerrors = 0; virCheckFlags(0, -1); -- 2.22.0

2 1

[libvirt] [PATCH] test: ensure nerrors variable is initialized
by Daniel P. Berrangé 04 Jul '19

04 Jul '19

There is an error path that jumps over the initialization of nerrors, and the jump target reads the variable contents. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Pushed as a build fix src/test/test_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index ce4ff1a582..2371581f51 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -3275,7 +3275,7 @@ static int testDomainGetDiskErrors(virDomainPtr dom, virDomainObjPtr vm = NULL; int ret = -1; size_t i; - size_t nerrors; + size_t nerrors = 0; virCheckFlags(0, -1); -- 2.21.0

1 0

[libvirt] [PATCH] test_driver: sanitize user-provided array in testDomainGetDiskErrors
by Ilias Stamatis 04 Jul '19

04 Jul '19

Zero out the user provided memory in order to avoid passing random pointers to VIR_FREE later. Signed-off-by: Ilias Stamatis <stamatis.iliass(a)gmail.com> --- src/test/test_driver.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 7c58d2c8ce..8ee1f0cad6 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -3275,6 +3275,7 @@ static int testDomainGetDiskErrors(virDomainPtr dom, virDomainObjPtr vm = NULL; int ret = -1; size_t i; + size_t nerrors; virCheckFlags(0, -1); @@ -3284,8 +3285,13 @@ static int testDomainGetDiskErrors(virDomainPtr dom, if (virDomainObjCheckActive(vm) < 0) goto cleanup; + nerrors = MIN(vm->def->ndisks, maxerrors); + if (errors) { - for (i = 0; i < MIN(vm->def->ndisks, maxerrors); i++) { + /* sanitize input */ + memset(errors, 0, sizeof(virDomainDiskError) * nerrors); + + for (i = 0; i < nerrors; i++) { if (VIR_STRDUP(errors[i].disk, vm->def->disks[i]->dst) < 0) goto cleanup; errors[i].error = (i % (VIR_DOMAIN_DISK_ERROR_LAST - 1)) + 1; @@ -3297,7 +3303,7 @@ static int testDomainGetDiskErrors(virDomainPtr dom, cleanup: if (ret < 0) { - for (i = 0; i < MIN(vm->def->ndisks, maxerrors); i++) + for (i = 0; i < nerrors; i++) VIR_FREE(errors[i].disk); } virDomainObjEndAPI(&vm); -- 2.22.0

3 2

[libvirt] [PATCH] test_driver: Implement virDomainSetPerfEvents
by Ilias Stamatis 04 Jul '19

04 Jul '19

Signed-off-by: Ilias Stamatis <stamatis.iliass(a)gmail.com> --- src/test/test_driver.c | 64 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 64 insertions(+) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 7c58d2c8ce..9974eba882 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -3372,6 +3372,69 @@ testDomainGetFSInfo(virDomainPtr dom, } +static int +testDomainSetPerfEvents(virDomainPtr dom, + virTypedParameterPtr params, + int nparams, + unsigned int flags) +{ + virDomainObjPtr vm = NULL; + virDomainDefPtr def = NULL; + size_t i; + int ret = -1; + + virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | + VIR_DOMAIN_AFFECT_CONFIG, -1); + + if (virTypedParamsValidate(params, nparams, + VIR_PERF_PARAM_CMT, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_MBMT, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_MBML, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_CPU_CYCLES, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_INSTRUCTIONS, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_CACHE_REFERENCES, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_CACHE_MISSES, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_BRANCH_INSTRUCTIONS, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_BRANCH_MISSES, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_BUS_CYCLES, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_STALLED_CYCLES_FRONTEND, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_STALLED_CYCLES_BACKEND, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_REF_CPU_CYCLES, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_CPU_CLOCK, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_TASK_CLOCK, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_PAGE_FAULTS, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_CONTEXT_SWITCHES, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_CPU_MIGRATIONS, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_PAGE_FAULTS_MIN, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_PAGE_FAULTS_MAJ, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_ALIGNMENT_FAULTS, VIR_TYPED_PARAM_BOOLEAN, + VIR_PERF_PARAM_EMULATION_FAULTS, VIR_TYPED_PARAM_BOOLEAN, + NULL) < 0) + return -1; + + if (!(vm = testDomObjFromDomain(dom))) + return -1; + + if (!(def = virDomainObjGetOneDef(vm, flags))) + goto cleanup; + + for (i = 0; i < nparams; i++) { + virTypedParameterPtr param = &params[i]; + virPerfEventType type = virPerfEventTypeFromString(param->field); + + if (param->value.b) + def->perf.events[type] = VIR_TRISTATE_BOOL_YES; + else + def->perf.events[type] = VIR_TRISTATE_BOOL_NO; + } + + ret = 0; + cleanup: + virDomainObjEndAPI(&vm); + return ret; +} + + static int testDomainGetPerfEvents(virDomainPtr dom, virTypedParameterPtr *params, @@ -7410,6 +7473,7 @@ static virHypervisorDriver testHypervisorDriver = { .domainSetAutostart = testDomainSetAutostart, /* 0.3.2 */ .domainGetDiskErrors = testDomainGetDiskErrors, /* 5.4.0 */ .domainGetFSInfo = testDomainGetFSInfo, /* 5.6.0 */ + .domainSetPerfEvents = testDomainSetPerfEvents, /* 5.6.0 */ .domainGetPerfEvents = testDomainGetPerfEvents, /* 5.6.0 */ .domainGetSchedulerType = testDomainGetSchedulerType, /* 0.3.2 */ .domainGetSchedulerParameters = testDomainGetSchedulerParameters, /* 0.3.2 */ -- 2.22.0

2 1