April 2019 - Devel - libvirt List Archives

[libvirt] [PATCH 0/3] maint: Drop ChangeLog*
by Andrea Bolognani 02 Apr '19

02 Apr '19

Rationale in the respective commit messages. Patch 2/3 has been snipped quite extensively so that it would make it to the mailing list: the unabridged version of this series can be fetched from https://github.com/andreabolognani/libvirt/tree/changelog Andrea Bolognani (3): maint: Stop generating ChangeLog from git maint: Drop ChangeLog-old maint: Update references to ChangeLog* .gitignore | 1 - ChangeLog | 15 + ChangeLog-old | 16699 ------------------------------------------ Makefile.am | 15 +- bootstrap.conf | 4 +- cfg.mk | 2 +- docs/news-ascii.xsl | 3 +- 7 files changed, 21 insertions(+), 16718 deletions(-) create mode 100644 ChangeLog delete mode 100644 ChangeLog-old -- 2.20.1

4 12

[libvirt] [PATCH] virDirRead: Fix d_type if needed
by Michal Privoznik 02 Apr '19

02 Apr '19

Some of our code iterates over a directory and does some decisions based on entry type (e.g. if it's not a regular file continue with next iteration). This is done by looking at ent->d_type struct member. However, in some cases (e.g. for some filesystems) the d_type can be set to DT_UNKNOWN which effectively defeats the aforementioned checks. For instance, some XFS' (depending on how they were created) might not bother filling the member. Fix this by running stat() and filling in the member ourselves. If needed. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- With Dan's patch that fixes stat() mocking, I'm passing tests successfully everywhere I tried (32bit fedora and gentoo on RPi, RHEL-7, amd64 gentoo, fedora rawhide on x86_64) src/util/virfile.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/src/util/virfile.c b/src/util/virfile.c index ec8d85929c..7d7c27a7d2 100644 --- a/src/util/virfile.c +++ b/src/util/virfile.c @@ -2893,6 +2893,39 @@ virDirOpenQuiet(DIR **dirp, const char *name) return virDirOpenInternal(dirp, name, false, true); } +static int +virDirReadFixDType(DIR *dirp, struct dirent *ent) +{ + /* DO NOT CLOSE THIS FD */ + const int fd = dirfd(dirp); + struct stat sb; + + if (fd < 0) + return -1; + + if (fstatat(fd, ent->d_name, &sb, AT_NO_AUTOMOUNT | AT_SYMLINK_NOFOLLOW) < 0) + return -1; + + if (S_ISDIR(sb.st_mode)) + ent->d_type = DT_DIR; + else if (S_ISCHR(sb.st_mode)) + ent->d_type = DT_CHR; + else if (S_ISBLK(sb.st_mode)) + ent->d_type = DT_BLK; + else if (S_ISREG(sb.st_mode)) + ent->d_type = DT_REG; + else if (S_ISFIFO(sb.st_mode)) + ent->d_type = DT_FIFO; + else if (S_ISLNK(sb.st_mode)) + ent->d_type = DT_LNK; + else if (S_ISSOCK(sb.st_mode)) + ent->d_type = DT_SOCK; + else + return -1; + + return 0; +} + /** * virDirRead: * @dirp: directory to read @@ -2926,6 +2959,17 @@ int virDirRead(DIR *dirp, struct dirent **ent, const char *name) } } while (*ent && (STREQ((*ent)->d_name, ".") || STREQ((*ent)->d_name, ".."))); + + if (*ent && + (*ent)->d_type == DT_UNKNOWN && + virDirReadFixDType(dirp, *ent) < 0) { + if (name) + virReportSystemError(errno, + _("Unable to fix d_type of '%s/%s'"), + name, (*ent)->d_name); + return -1; + } + return !!*ent; } -- 2.21.0

2 1

[libvirt] [PATCH v2] examples: fix 64-bit integer formatting on Windows
by Daniel P. Berrangé 02 Apr '19

02 Apr '19

The Windows printf functions don't support %llu/%lld for printing 64-bit integers. For most of libvirt this doesn't matter as we rely on gnulib which provides a replacement printf that is sane. The example code is designed to compile against the normal OS headers, with no use of gnulib and thus has to use the platform specific printf. To deal with this we must use the macros PRI* macros from inttypes.h to get the platform specific format string. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- examples/admin/client_info.c | 7 ++++--- examples/admin/list_clients.c | 3 ++- examples/domtop/domtop.c | 8 ++++++-- 3 files changed, 12 insertions(+), 6 deletions(-) Changed in v2: - Now actually commit the int64 casts diff --git a/examples/admin/client_info.c b/examples/admin/client_info.c index f3f62a656b..7fc6c72bbd 100644 --- a/examples/admin/client_info.c +++ b/examples/admin/client_info.c @@ -3,6 +3,7 @@ #include <stdlib.h> #include <time.h> #include <string.h> +#include <inttypes.h> #include <libvirt/libvirt-admin.h> static const char * @@ -66,11 +67,11 @@ exampleGetTypedParamValue(virTypedParameterPtr item) break; case VIR_TYPED_PARAM_LLONG: - ret = asprintf(&str, "%lld", item->value.l); + ret = asprintf(&str, "%" PRId64, (int64_t)item->value.l); break; case VIR_TYPED_PARAM_ULLONG: - ret = asprintf(&str, "%llu", item->value.ul); + ret = asprintf(&str, "%" PRIu64, (uint64_t)item->value.ul); break; case VIR_TYPED_PARAM_DOUBLE: @@ -143,7 +144,7 @@ int main(int argc, char **argv) if (!(timestr = exampleGetTimeStr(virAdmClientGetTimestamp(clnt)))) goto cleanup; - printf("%-15s: %llu\n", "id", virAdmClientGetID(clnt)); + printf("%-15s: %" PRIu64 "\n", "id", (uint64_t)virAdmClientGetID(clnt)); printf("%-15s: %s\n", "connection_time", timestr); printf("%-15s: %s\n", "transport", exampleTransportToString(virAdmClientGetTransport(clnt))); diff --git a/examples/admin/list_clients.c b/examples/admin/list_clients.c index 5cf8e4c2a6..2876637d42 100644 --- a/examples/admin/list_clients.c +++ b/examples/admin/list_clients.c @@ -1,6 +1,7 @@ #include <stdio.h> #include <stdlib.h> #include <time.h> +#include <inttypes.h> #include <libvirt/libvirt-admin.h> static const char * @@ -96,7 +97,7 @@ int main(int argc, char **argv) exampleGetTimeStr(virAdmClientGetTimestamp(client)))) goto cleanup; - printf(" %-5llu %-15s %-15s\n", id, + printf(" %-5" PRIu64 " %-15s %-15s\n", (uint64_t)id, exampleTransportToString(transport), timestr); free(timestr); } diff --git a/examples/domtop/domtop.c b/examples/domtop/domtop.c index 008065c651..e1e7fbff8b 100644 --- a/examples/domtop/domtop.c +++ b/examples/domtop/domtop.c @@ -29,6 +29,7 @@ #include <string.h> #include <sys/time.h> #include <unistd.h> +#include <inttypes.h> static bool debug; static bool run_top; @@ -226,8 +227,11 @@ print_cpu_usage(const char *dom_name, return; } - DEBUG("now_params=%llu then_params=%llu now=%llu then=%llu", - now_params[pos].value.ul, then_params[pos].value.ul, now, then); + DEBUG("now_params=%" PRIu64 " then_params=%" PRIu64 + " now=%" PRIu64 " then=%" PRIu64, + (uint64_t)now_params[pos].value.ul, + (uint64_t)then_params[pos].value.ul, + (uint64_t)now, (uint64_t)then); /* @now_params and @then_params are in nanoseconds, @now and @then are * in microseconds. In ideal world, we would translate them both into -- 2.20.1

2 1

[libvirt] [PATCH 0/2] qemu: caps: Few more cleanups
by Peter Krempa 02 Apr '19

02 Apr '19

These were prompted by the review on the original series. Both will be pushed as trivial once the release is out. Peter Krempa (2): qemu: Remove ATTRIBUTE_UNUSED from 'qemuCaps' of virQEMUCapsInitQMPMonitorTCG qemu: Remove cleanup section of virQEMUCapsInitQMPMonitorTCG src/qemu/qemu_capabilities.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) -- 2.20.1

1 2

[libvirt] [PATCH 0/5] qemu: hotplug: Media change improvements (blockdev-add saga)
by Peter Krempa 02 Apr '19

02 Apr '19

First patch is not entirely relevant in this series. Peter Krempa (5): qemu: domain: Use VIR_AUTOFREE in qemuDomainObjPrivateXMLParseBlockjobs qemu: hotplug: Remove unused copies of virQEMUDriverConfigPtr qemu: hotplug: Use VIR_AUTOUNREF for virQEMUDriverConfigPtr qemu: hotplug: Disallow media change while blockjob is active qemu: domain: Forbid copy_on_read option also for floppies src/qemu/qemu_domain.c | 23 +++++++++++++------ src/qemu/qemu_hotplug.c | 51 ++++++++++++++++------------------------- 2 files changed, 36 insertions(+), 38 deletions(-) -- 2.20.1

2 10

[libvirt] [PATCH] examples: fix 64-bit integer formatting on Windows
by Daniel P. Berrangé 02 Apr '19

02 Apr '19

The Windows printf functions don't support %llu/%lld for printing 64-bit integers. For most of libvirt this doesn't matter as we rely on gnulib which provides a replacement printf that is sane. The example code is designed to compile against the normal OS headers, with no use of gnulib and thus has to use the platform specific printf. To deal with this we must use the macros PRI* macros from inttypes.h to get the platform specific format string. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- examples/admin/client_info.c | 7 ++++--- examples/admin/list_clients.c | 3 ++- examples/domtop/domtop.c | 4 +++- 3 files changed, 9 insertions(+), 5 deletions(-) diff --git a/examples/admin/client_info.c b/examples/admin/client_info.c index f3f62a656b..d60512624d 100644 --- a/examples/admin/client_info.c +++ b/examples/admin/client_info.c @@ -3,6 +3,7 @@ #include <stdlib.h> #include <time.h> #include <string.h> +#include <inttypes.h> #include <libvirt/libvirt-admin.h> static const char * @@ -66,11 +67,11 @@ exampleGetTypedParamValue(virTypedParameterPtr item) break; case VIR_TYPED_PARAM_LLONG: - ret = asprintf(&str, "%lld", item->value.l); + ret = asprintf(&str, "%" PRId64, item->value.l); break; case VIR_TYPED_PARAM_ULLONG: - ret = asprintf(&str, "%llu", item->value.ul); + ret = asprintf(&str, "%" PRIu64, item->value.ul); break; case VIR_TYPED_PARAM_DOUBLE: @@ -143,7 +144,7 @@ int main(int argc, char **argv) if (!(timestr = exampleGetTimeStr(virAdmClientGetTimestamp(clnt)))) goto cleanup; - printf("%-15s: %llu\n", "id", virAdmClientGetID(clnt)); + printf("%-15s: %" PRIu64 "\n", "id", virAdmClientGetID(clnt)); printf("%-15s: %s\n", "connection_time", timestr); printf("%-15s: %s\n", "transport", exampleTransportToString(virAdmClientGetTransport(clnt))); diff --git a/examples/admin/list_clients.c b/examples/admin/list_clients.c index 5cf8e4c2a6..545b9cbd01 100644 --- a/examples/admin/list_clients.c +++ b/examples/admin/list_clients.c @@ -1,6 +1,7 @@ #include <stdio.h> #include <stdlib.h> #include <time.h> +#include <inttypes.h> #include <libvirt/libvirt-admin.h> static const char * @@ -96,7 +97,7 @@ int main(int argc, char **argv) exampleGetTimeStr(virAdmClientGetTimestamp(client)))) goto cleanup; - printf(" %-5llu %-15s %-15s\n", id, + printf(" %-5" PRIu64 " %-15s %-15s\n", id, exampleTransportToString(transport), timestr); free(timestr); } diff --git a/examples/domtop/domtop.c b/examples/domtop/domtop.c index 008065c651..33d97c74fe 100644 --- a/examples/domtop/domtop.c +++ b/examples/domtop/domtop.c @@ -29,6 +29,7 @@ #include <string.h> #include <sys/time.h> #include <unistd.h> +#include <inttypes.h> static bool debug; static bool run_top; @@ -226,7 +227,8 @@ print_cpu_usage(const char *dom_name, return; } - DEBUG("now_params=%llu then_params=%llu now=%llu then=%llu", + DEBUG("now_params=%" PRIu64 " then_params=%" PRIu64 + " now=%" PRIu64 " then=%" PRIu64, now_params[pos].value.ul, then_params[pos].value.ul, now, then); /* @now_params and @then_params are in nanoseconds, @now and @then are -- 2.20.1

1 0

[libvirt] [PATCH 0/3] introducing QEMU_CAPS_QUERY_CURRENT_MACHINE and QEMU_CAPS_WAKEUP_SUSPEND_SUPPORT
by Daniel Henrique Barboza 02 Apr '19

02 Apr '19

Hi, In this series, I present the Libvirt support for a QEMU API that will be introduced in QEMU 4.0, called 'query-current-machine'. More background info about the API and its purpose can be found in the commit msg of patch 1. Patch 1 contains the caps declarations (a short patch, perhaps squashable with patch 2). Patch 2 enable the QEMU_CAPS_WAKEUP_SUSPEND_SUPPORT cap by querying 'query-current-machine', if the QEMU binary supports it. Patch 3 uses the caps inside qemuDomainPMSuspendForDuration to avoid suspending a domain that can't wake up from suspend. Daniel Henrique Barboza (3): adding QEMU_CAPS_QUERY_CURRENT_MACHINE and QEMU_CAPS_WAKEUP_SUSPEND_SUPPORT setting QEMU_CAPS_WAKEUP_SUSPEND_SUPPORT qemu capability qemuDomainPMSuspendForDuration: check for QEMU_CAPS_WAKEUP_SUSPEND_SUPPORT src/qemu/qemu_capabilities.c | 5 +++++ src/qemu/qemu_capabilities.h | 4 ++++ src/qemu/qemu_driver.c | 24 +++++++++++++++++++++++ src/qemu/qemu_monitor.c | 8 ++++++++ src/qemu/qemu_monitor.h | 3 +++ src/qemu/qemu_monitor_json.c | 35 +++++++++++++++++++++++++++++++++ src/qemu/qemu_monitor_json.h | 4 ++++ src/qemu/qemu_process.c | 38 ++++++++++++++++++++++++++++++++++++ 8 files changed, 121 insertions(+) -- 2.20.1

2 6

[libvirt] Can jobs suck like qemu-pr-helper does be transfered to libvirtd?
by Zhangbo (Oscar) 02 Apr '19

02 Apr '19

Hi all: qemu-pr-helper exits to help qemu do the high-privileged scsi related jobs. LIBVIRTD is responsible to launch qemu-pr-helper and qemu, and set selinux/DAC labels for them and their socket. # # # # # ___________ # ___________|libvirtd |__________ # | |___________| | # | | | # | | | # _________|________ _____|______ ___|____ # | qemu-pr-helper |__| vm1.sock |_____| qemu1 | # |__________________| |____________| |________| There may be other jobs quite like qemu-pr-helper, shall we make them as "complex" as the qemu-pr-helper scheme? Will it be OK to just let qemu send an EVENT to libvirtd, and let libvirtd do the SCSI/other jobs? Will it be OK if we remove qemu-pr-helper and similar processes? What's the disadvantage if we let libvirtd do there high privileged jobs instead qemu-pr-helper-like processes? Thanks!

2 6

[libvirt] [PATCH] virsh: cmdConsole: make --domain optional
by Ján Tomko 01 Apr '19

01 Apr '19

In the absence of a domain parameter, console the user instead. Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- tools/virsh-domain.c | 25 ++++++++++++++++++++++++- 1 file changed, 24 insertions(+), 1 deletion(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index afcd0a5f8d..de41f6312d 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -49,6 +49,7 @@ #include "virsh-console.h" #include "virsh-domain-monitor.h" #include "virerror.h" +#include "virrandom.h" #include "virtime.h" #include "virtypedparam.h" #include "virxml.h" @@ -2972,7 +2973,7 @@ static const vshCmdInfo info_console[] = { }; static const vshCmdOptDef opts_console[] = { - VIRSH_COMMON_OPT_DOMAIN_FULL(VIR_CONNECT_LIST_DOMAINS_ACTIVE), + VIRSH_COMMON_OPT_DOMAIN_OT_STRING_FULL(0, VIR_CONNECT_LIST_DOMAINS_ACTIVE), {.name = "devname", /* sc_prohibit_devname */ .type = VSH_OT_STRING, .help = N_("character device name") @@ -3023,6 +3024,23 @@ cmdRunConsole(vshControl *ctl, virDomainPtr dom, return ret; } +static const char * +constellation_phrases[] = { + "Everything is going to be all right.", + "Things are going to work out in the end.", + "Imagine pleasant nonsense", /* https://www.instagram.com/p/BuRfmkrBigI/ */ + "Ursa Major", + "There, there.", +}; + +static void +cmdConsoleUser(vshControl *ctl) +{ + uint32_t die = virRandomInt(ARRAY_CARDINALITY(constellation_phrases)); + + vshPrintExtra(ctl, "%s", constellation_phrases[die]); +} + static bool cmdConsole(vshControl *ctl, const vshCmd *cmd) { @@ -3033,6 +3051,11 @@ cmdConsole(vshControl *ctl, const vshCmd *cmd) unsigned int flags = 0; const char *name = NULL; + if (!vshCommandOptBool(cmd, "domain")) { + cmdConsoleUser(ctl); + return true; + } + if (!(dom = virshCommandOptDomain(ctl, cmd, NULL))) return false; -- 2.20.1

2 1

[libvirt] vcpupin reports bogus vcpu affinities
by Allen, John 01 Apr '19

01 Apr '19

Sent this out to the list a few days ago, but never saw it appear on the archives. Resending--hopefully this makes it to the list. --- For pinned vcpus, vcpupin will report inaccurate affinity values on machines with high core counts (256 cores in my case). The problem is produced as follows: $ virsh vcpupin myguest 0 4 $ virsh vcpupin myguest 0 VCPU CPU Affinity --------------------------- 0 4,192,194,196-197 Running taskset on the qemu threads shows the correct affinity, so this seems to be a reporting problem. Strangely, the value "192" is significant. If I pin a cpu greater than 192, the problem no longer appears. I believe the cause of the problem in my case is that in this case in src/conf/domain_conf.c:virDomainDefGetVcpuPinInfoHelper: ... if (vcpu && vcpu->cpumask) bitmap = vcpu->cpumask; ... vcpu->cpumask is "shortened" in that it is only long enough to contain the last set bit in the mask. However, when we go to copy the mask to the buffer that is returned, we use the masklen passed to the function which is the "full" masklen with a bit for each cpu. So it seems virBitmapToDataBuf copies some extra data past the end of the bitmask. Why the "192" value is always set and I typically see similar bogus bits set is still unknown. What is the function meant to assume in this case? Is it sane to assume that the bitmask is the full length of the buffer here and it's the responsibility of the setter of vcpu->cpumask to provide the length of the bitmap we're expecting? Or should we assume that we may receive a shortened bitmask here and expand the bitmask before copying to the buffer? -John

2 1