Re: [libvirt] [PATCHv3 2/6] qemu: use query-cpus-fast in JSON monitor
On 04/04/2018 10:45 AM, Viktor Mihajlovski wrote:
Use query-cpus-fast instead of query-cpus if supported by QEMU.
Based on the QEMU_CAPS_QUERY_CPUS_FAST capability.
Signed-off-by: Viktor Mihajlovski <mihajlov(a)linux.vnet.ibm.com>
---
src/qemu/qemu_domain.c | 14 +++++++++++---
src/qemu/qemu_monitor.c | 30 ++++++++++++++++++------------
src/qemu/qemu_monitor.h | 7 +++++--
src/qemu/qemu_monitor_json.c | 37 +++++++++++++++++++++++++++----------
src/qemu/qemu_monitor_json.h | 3 ++-
tests/qemumonitorjsontest.c | 4 ++--
6 files changed, 65 insertions(+), 30 deletions(-)
diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c
index 9d1c33b..662937b 100644
--- a/src/qemu/qemu_domain.c
+++ b/src/qemu/qemu_domain.c
@@ -9006,7 +9006,12 @@ qemuDomainRefreshVcpuInfo(virQEMUDriverPtr driver,
if (qemuDomainObjEnterMonitorAsync(driver, vm, asyncJob) < 0)
return -1;
Count me as one of those that would prefer to see:
bool fast;
...
fast = virQEMUCapsGet(QEMU_DOMAIN_PRIVATE(vm)->qemuCaps,
QEMU_CAPS_QUERY_CPUS_FAST);
...
- rc = qemuMonitorGetCPUInfo(qemuDomainGetMonitor(vm), &info,
maxvcpus, hotplug);
+ rc = qemuMonitorGetCPUInfo(qemuDomainGetMonitor(vm),
+ &info,
+ maxvcpus,
+ hotplug,
+ virQEMUCapsGet(QEMU_DOMAIN_PRIVATE(vm)->qemuCaps,
+ QEMU_CAPS_QUERY_CPUS_FAST));
rc = qemuMonitorGetCPUInfo(qemuDomainGetMonitor(vm), &info, maxvcpus,
hotplug, fast);
if (qemuDomainObjExitMonitor(driver, vm) < 0)
goto cleanup;
@@ -9025,7 +9030,7 @@ qemuDomainRefreshVcpuInfo(virQEMUDriverPtr driver,
* thread, but it runs every vCPU in that same thread. So it
* is impossible to setup different affinity per thread.
*
- * What's more the 'query-cpus' command returns bizarre
+ * What's more the 'query-cpus[-fast]' command returns bizarre
* data for the threads. It gives the TCG thread for the
* vCPU 0, but for vCPUs 1-> N, it actually replies with
* the main process thread ID.
@@ -9126,7 +9131,10 @@ qemuDomainRefreshVcpuHalted(virQEMUDriverPtr driver,
if (qemuDomainObjEnterMonitorAsync(driver, vm, asyncJob) < 0)
return -1;
- haltedmap = qemuMonitorGetCpuHalted(qemuDomainGetMonitor(vm), maxvcpus);
+ haltedmap = qemuMonitorGetCpuHalted(qemuDomainGetMonitor(vm),
+ maxvcpus,
+
virQEMUCapsGet(QEMU_DOMAIN_PRIVATE(vm)->qemuCaps,
+ QEMU_CAPS_QUERY_CPUS_FAST));
Likewise,
bool fast;
...
fast = virQEMUCapsGet(QEMU_DOMAIN_PRIVATE(vm)->qemuCaps,
QEMU_CAPS_QUERY_CPUS_FAST);
haltedmap = qemuMonitorGetCpuHalted(qemuDomainGetMonitor(vm), maxvcpus,
fast);
if (qemuDomainObjExitMonitor(driver, vm) < 0 || !haltedmap)
goto cleanup;
[...]
John
If you're OK with this I can make the alterations... What you have
works - it's just one of those preferential things related to seeing
and/or relying on function calls within function calls.