- Devel - libvirt List Archives

[libvirt] libvirt 6.0.0 make check-TESTS fails "FAIL: test-ptsname_r"
by Satheesh Rajendran 14 Jan '20

14 Jan '20

Hi, libvirt 6.0.0 (commit:bfd5f69d6038daeac03d2684fcd98aeee4ef1e24), "make rpm" fails at make check-TESTS ... .. PASS: test-memchr PASS: test-send ../../../build-aux/test-driver: line 107: 453114 Aborted (core dumped) "$@" > $log_file 2>&1 FAIL: test-ptsname_r PASS: test-setsockopt PASS: test-sigaction PASS: test-signal-h ... ------- # cat /root/rpmbuild/BUILD/libvirt-6.0.0/ppc64le-redhat-linux-gnu/gnulib/tests/test-suite.log ================================================ libvirt 6.0.0: gnulib/tests/test-suite.log ================================================ # TOTAL: 149 # PASS: 148 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 .. contents:: :depth: 2 FAIL: test-ptsname_r ==================== ../../../gnulib/tests/test-ptsname_r.c:89: assertion 'result == EINVAL' failed FAIL test-ptsname_r (exit status: 134) -------- Env: HW: Power8 ArcH: ppc64le OS: Fedora 31 GCC: 9.2.1 20190827 (Red Hat 9.2.1-1) Glibc: glibc-2.30-8.fc31.ppc64le Gdb output: # gdb /root/rpmbuild/BUILD/libvirt-6.0.0/ppc64le-redhat-linux-gnu/gnulib/tests/test-ptsname_r -c core.test-ptsname_r.0.1a9e4ce7d7ec4d7e81e295f7f67c7029.317503.1578922893000000000000 GNU gdb (GDB) Fedora 8.3.50.20190824-26.fc31 .. .. Reading symbols from /root/rpmbuild/BUILD/libvirt-6.0.0/ppc64le-redhat-linux-gnu/gnulib/tests/test-ptsname_r... [New LWP 317503] [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". Core was generated by `./test-ptsname_r '. Program terminated with signal SIGABRT, Aborted. #0 0x00007fffac447358 in __libc_signal_restore_set (set=0x7fffefdc3078) at ../sysdeps/unix/sysv/linux/internal-signals.h:84 84 return INTERNAL_SYSCALL (rt_sigprocmask, err, 4, SIG_SETMASK, set, NULL, (gdb) start Temporary breakpoint 1 at 0x11d850c88: file ../../../gnulib/tests/test-ptsname_r.c, line 94. Starting program: /root/rpmbuild/BUILD/libvirt-6.0.0/ppc64le-redhat-linux-gnu/gnulib/tests/test-ptsname_r [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". Temporary breakpoint 1, 0x0000000100000c88 in main () at ../../../gnulib/tests/test-ptsname_r.c:94 94 { (gdb) c Continuing. ../../../gnulib/tests/test-ptsname_r.c:89: assertion 'result == EINVAL' failed Program received signal SIGABRT, Aborted. 0x00007ffff7d57358 in __libc_signal_restore_set (set=0x7fffffffe438) at ../sysdeps/unix/sysv/linux/internal-signals.h:84 84 return INTERNAL_SYSCALL (rt_sigprocmask, err, 4, SIG_SETMASK, set, NULL, (gdb) print result $1 = '\000' <repeats 113 times> (gdb) Any help would be appreciated, Thanks in advance. Regards, -Satheesh.

2 2

[libvirt] [PATCH] news: News for RDT-MBM feature
by Han Han 14 Jan '20

14 Jan '20

Signed-off-by: Han Han <hhan(a)redhat.com> --- docs/news.xml | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/docs/news.xml b/docs/news.xml index acb63709da..822f89f420 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -106,6 +106,15 @@ hostname and the information stored in its TLS certificate. </description> </change> + <change> + <summary> + qemu: Support to report the stats of memory bandwidth usage + </summary> + <description> + Implement Intel RDT-MBM to libvirt. The stats can be got via + <code>virsh domstats --memory</code>. + </description> + </change> </section> <section title="Removed features"> <change> -- 2.24.0.rc1

2 2

[libvirt] [PATCH 0/7] Adapt machine name generation to new facts
by Ján Tomko 14 Jan '20

14 Jan '20

Ján Tomko (7): virbuftest: remove extra G_GNUC_UNUSED markers virbuftest: use g_autofree virbuftest: remove unnecessary labels virbuftest: declare testBufAddStrData earlier virbuftest: use field names when initalizing test info util: add virBufferTrimChars conf: do not generate machine names ending with a dash src/conf/domain_conf.c | 3 + src/libvirt_private.syms | 1 + src/util/virbuffer.c | 26 +++++++++ src/util/virbuffer.h | 1 + tests/virbuftest.c | 123 ++++++++++++++++++++++----------------- tests/virsystemdtest.c | 4 ++ 6 files changed, 103 insertions(+), 55 deletions(-) -- 2.21.0

2 14

[libvirt] [PATCH for 6.0.0] news: Document <disk type='nvme'/>
by Michal Privoznik 14 Jan '20

14 Jan '20

Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- docs/news.xml | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/docs/news.xml b/docs/news.xml index acb63709da..e9863dacf3 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -106,6 +106,22 @@ hostname and the information stored in its TLS certificate. </description> </change> + <change> + <summary> + qemu: Allow accessing NVMe disks directly + </summary> + <description> + Before this release there were two ways to configure a NVMe disk for + a domain. The first was using <disk/> with the <source/> + pointing to the <code>/dev/nvmeXXXX</code>. The other was using PCI + assignment via <hostdev/> element. Both have their + disadvantages: the former adds latency of file system and block + layers of the host kernel, the latter prohibits domain migration. In + this release the third way of configuring NVMe disk is added which + combines the advantages and drops disadvantages of the previous two + ways. It's accessible via <disk type='nvme'/>. + </description> + </change> </section> <section title="Removed features"> <change> -- 2.24.1

2 1

[libvirt] [PATCH] docs: Notes for geometry in formatdomain doc
by Han Han 14 Jan '20

14 Jan '20

Since commit b340c6c614 CHS geometry is not supported in usb and sd, CHS translation is only supported in ide. Add these notes to formatdomain doc. Signed-off-by: Han Han <hhan(a)redhat.com> --- docs/formatdomain.html.in | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 2af4960981..74bea528be 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3715,9 +3715,13 @@ <dt><code>secs</code></dt> <dd>The <code>secs</code> attribute is the number of sectors per track. </dd> + <dt>Note that <code>cyls</code>, <code>heads</code> and + <code>secs</code> are not supported in <code>usb</code> + or <code>sd</code> bus.</dt> <dt><code>trans</code></dt> <dd>The optional <code>trans</code> attribute is the - BIOS-Translation-Modus (none, lba or auto)</dd> + BIOS-Translation-Modus (none, lba or auto). Note that this + attribute is only supported in <code>ide</code> bus.</dd> </dl> </dd> <dt><code>blockio</code></dt> -- 2.24.0.rc1

1 0

[libvirt] [PATCH] domain_conf: Do not use USB by default for <input> devices on s390x
by Thomas Huth 14 Jan '20

14 Jan '20

When trying to specify an input device on s390x without bus like this: <input type='keyboard'/> ... then libvirt currently complains: error: unsupported configuration: USB is disabled for this domain, but USB devices are present in the domain XML This is somewhat confusing since the user did not specify an USB device here. Since USB is not available on s390x, we should default to the "virtio" bus here instead. Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1790189 Signed-off-by: Thomas Huth <thuth(a)redhat.com> --- src/conf/domain_conf.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 1290241923..c3761b0f45 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -13428,6 +13428,8 @@ virDomainInputDefParseXML(virDomainXMLOptionPtr xmlopt, def->type == VIR_DOMAIN_INPUT_TYPE_KBD) && (ARCH_IS_X86(dom->os.arch) || dom->os.arch == VIR_ARCH_NONE)) { def->bus = VIR_DOMAIN_INPUT_BUS_PS2; + } else if (ARCH_IS_S390(dom->os.arch)) { + def->bus = VIR_DOMAIN_INPUT_BUS_VIRTIO; } else { def->bus = VIR_DOMAIN_INPUT_BUS_USB; } -- 2.18.1

4 6

[libvirt] Entering freeze for 6.0.0
by Daniel Veillard 13 Jan '20

13 Jan '20

I just tagged it RC1 in git, and pushed signed tarball and source rpm to the usual place: https://libvirt.org/sources/ Looks fine in my very limited testing, ci seems mostly green (though I sometimes wonder how reliable that is) so looks we should be fine, but please give it some testing ! If everything goes smoothly, goal is to make rc2 on Monday and then push the release on the 15th , in the meantime, please give it a try ! thanks, Daniel -- Daniel Veillard | Red Hat Developers Tools http://developer.redhat.com/ veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 1

[libvirt] [PATCH 0/4] qemu: Don't lose domain on failed restore
by Michal Privoznik 13 Jan '20

13 Jan '20

https://bugzilla.redhat.com/show_bug.cgi?id=1718707 When restoring a domain we might leave a qemu process behind. The problem is described in 4/4 in more detail, but the digest is: we start qemu process successfully but then fail to restore its vCPUs and leave qemu process running and forget about. I thought of two possible ways to fix this: 1) Kill qemu process and claim error. 2) Don't kill qemu process and don't remove it from our internal list and claim error. I lean towards 1) because I find it clearer. If I were a mgmt app and call 'virsh restore /path/to/domain.save' then I'd much rather see either success or an error without any need for cleanup (which would be implied if we went with number 2). Michal Prívozník (4): qemuDomainSaveImageStartVM: Use VIR_AUTOCLOSE for @intermediatefd qemuDomainSaveImageStartVM: Use g_autoptr() for virCommand qemu: Use g_autoptr() for qemuDomainSaveCookie qemu: Stop domain on failed restore src/qemu/qemu_domain.c | 28 ++++++++++------------------ src/qemu/qemu_domain.h | 1 + src/qemu/qemu_driver.c | 37 +++++++++++++++++-------------------- 3 files changed, 28 insertions(+), 38 deletions(-) -- 2.24.1

2 5

[libvirt] [PATCH 0/3] news: Fix and update for libvirt 6.0.0
by Andrea Bolognani 13 Jan '20

13 Jan '20

<blurb/> Andrea Bolognani (3): news: Fix typo (Libivrt -> Libvirt) news: Rearrange a few entries news: Update for libvirt 6.0.0 docs/news.xml | 111 +++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 87 insertions(+), 24 deletions(-) -- 2.24.1

3 10

[libvirt] [PATCH for 6.0.0] vircgroupv2devices: free BPF map when replacing with new one
by Pavel Hrdina 13 Jan '20

13 Jan '20

This leaks the FD of BPF map which means it will not be freed. Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- src/util/vircgroupv2devices.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/src/util/vircgroupv2devices.c b/src/util/vircgroupv2devices.c index 402ce3f86f..445f9c53fc 100644 --- a/src/util/vircgroupv2devices.c +++ b/src/util/vircgroupv2devices.c @@ -311,6 +311,11 @@ virCgroupV2DevicesAttachProg(virCgroupPtr group, VIR_FORCE_CLOSE(group->unified.devices.progfd); } + if (group->unified.devices.mapfd > 0) { + VIR_DEBUG("Closing existing map that was replaced by new one."); + VIR_FORCE_CLOSE(group->unified.devices.mapfd); + } + group->unified.devices.progfd = progfd; group->unified.devices.mapfd = mapfd; group->unified.devices.max = max; -- 2.24.1

2 1