- Devel - libvirt List Archives

[libvirt] [PATH V10 00/12] Support cache tune in libvirt
by Eli Qiao 07 Mar '17

07 Mar '17

Addressed comment from v9 -> v8 Marcelo: * Improve default schemata calculate, support allocate from low schemata bits for other APP. Addressed comment from v9 -> v8 Marcelo: * New public API to query cache usage Eli: * Fix core dump while multiple tasks are added. Addressed comment from v8 -> v7 Martin: * Patch subject prefix. * Move some of cpu related information to virhostcpu.c. * Fix some memory leak in src/utils/resctrl.c Martin & Marcelo: * Don't remove directories which are not maintained by libvirt. Addressed comment from v7 -> v6 Marcelo: * Fix flock usage while VM initialization. Addressed comment from v6 -> v5 Marcelo: * Support other APPs to operate /sys/fs/resctrl at same time Libvirt will scan /sys/fs/resctrl again before doing cache allocation. patch 10 will address this. Addressed comment from v4 -> v5: Marcelo: * Several typos * Use flock instead of virFileLock Addressed comment from v3 -> v4: Daniel & Marcelo: * Added concurrence support Addressed comment from v2 -> v3: Daniel: * Fixed coding style, passed `make check` and `make syntax-check` * Variables renaming and move from header file to c file. * For locking/mutex support, no progress. There are some discussion from mailing list, but I can not find a better way to add locking support without performance impact. I'll explain the process and please help to advice what shoud we do. VM create: 1) Get the cache left value on each bank of the host. This should be shared amount all VMs. 2) Calculate the schemata on the bank based on all created resctrl domain's schemata 3) Calculate the default schemata by scaning all domain's schemata. 4) Flush default schemata to /sys/fs/resctrl/schemata VM destroy: 1) Remove the resctrl domain of that VM 2) Recalculate default schemata 3) Flush default schemata to /sys/fs/resctrl/schemata The key point is that all VMs shares /sys/fs/resctrl/schemata, and when a VM create a resctrl domain, the schemata of that VM depends on the default schemata and all other exsited schematas. So a global mutex is reqired. Before calculate a schemata or update default schemata, libvirt should gain this global mutex. I will try to think more about how to support this gracefully in next patch set. Marcelo: * Added vcpu support for cachetune, this will allow user to define which vcpu using which cache allocation bank. <cachetune id='0' host_id=0 size='3072' unit='KiB' vcpus='0,1'/> vcpus is a cpumap, the vcpu pids will be added to tasks file * Added cdp compatible, user can specify l3 cache even host enable cdp. See patch 8. On a cdp enabled host, specify l3code/l3data by <cachetune id='0' host_id='0' type='l3' size='3072' unit='KiB'/> This will create a schemata like: L3data:0=0xff00;... L3code:0=0xff00;... * Would you please help to test if the functions work. Martin: * Xml test case, I have no time to work on this yet, would you please show me an example, would like to amend it later. This series patches are for supportting CAT featues, which also called cache tune in libvirt. First to expose cache information which could be tuned in capabilites XML. Then add new domain xml element support to add cacahe bank which will apply on this libvirt domain. This series patches add a util file `resctrl.c/h`, an interface to talk with linux kernel's system fs. There are still one TODO left: 1. Expose a new public interface to get free cache information. 2. Expose a new public interface to set cachetune lively. Some discussion about this feature support can be found from: https://www.redhat.com/archives/libvir-list/2017-January/msg00644.html Eli Qiao (12): Resctrl: Add some utils functions Resctrl: expose cache information to capabilities Resctrl: Add new xml element to support cache tune Resctrl: Add private interfaces to operate cache bank Qemu: Set cache tune while booting a new domain. Resctrl: enable l3code/l3data Resctrl: Make sure l3data/l3code are pairs Resctrl: Compatible mode for cdp enabled Resctrl: concurrence support Resctrl: Scan resctrl before doing cache allocation Resctrl: Add Public API for nodecachestats Resctrl: Add nodecachestats daemon/remote.c | 67 +++ docs/schemas/domaincommon.rng | 46 ++ include/libvirt/libvirt-host.h | 32 ++ include/libvirt/virterror.h | 1 + po/POTFILES.in | 1 + src/Makefile.am | 1 + src/conf/capabilities.c | 56 ++ src/conf/capabilities.h | 23 + src/conf/domain_conf.c | 182 +++++++ src/conf/domain_conf.h | 19 + src/driver-hypervisor.h | 7 + src/libvirt-host.c | 41 ++ src/libvirt_private.syms | 12 + src/libvirt_public.syms | 1 + src/nodeinfo.c | 64 +++ src/nodeinfo.h | 1 + src/qemu/qemu_capabilities.c | 8 + src/qemu/qemu_driver.c | 18 + src/qemu/qemu_process.c | 54 ++ src/remote/remote_driver.c | 52 ++ src/remote/remote_protocol.x | 25 +- src/remote_protocol-structs | 16 + src/util/virerror.c | 1 + src/util/virhostcpu.c | 186 ++++++- src/util/virhostcpu.h | 6 + src/util/virresctrl.c | 1098 ++++++++++++++++++++++++++++++++++++++++ src/util/virresctrl.h | 96 ++++ tools/virsh-host.c | 49 ++ 28 files changed, 2145 insertions(+), 18 deletions(-) create mode 100644 src/util/virresctrl.c create mode 100644 src/util/virresctrl.h -- 1.9.1

4 15

[libvirt] [PATCH] vz: correct error message
by Mikhail Feoktistov 06 Mar '17

06 Mar '17

--- src/vz/vz_sdk.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/src/vz/vz_sdk.c b/src/vz/vz_sdk.c index e16bb0c..84e89a5 100644 --- a/src/vz/vz_sdk.c +++ b/src/vz/vz_sdk.c @@ -2601,8 +2601,10 @@ prlsdkCheckUnsupportedParams(PRL_HANDLE sdkdom, virDomainDefPtr def) def->onCrash != VIR_DOMAIN_LIFECYCLE_CRASH_DESTROY) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", - _("on_reboot, on_poweroff and on_crash parameters " - "are not supported by vz driver")); + _("The following parameters must be " + "on_reboot = restart, on_poweroff = destroy, " + "on_crash = destroy. " + "Different actions are not supported by vz driver")); return -1; } -- 1.8.3.1

2 1

[libvirt] [PATCH] Document preferred naming conventions
by Daniel P. Berrange 06 Mar '17

06 Mar '17

This documents the preferred conventions for naming files, structs, enums, typedefs and functions. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- HACKING | 71 ++++++++++++++++++++++++++++++++++++++++++++ docs/hacking.html.in | 83 ++++++++++++++++++++++++++++++++++++++++++++++++++++ docs/hacking2.xsl | 4 +++ 3 files changed, 158 insertions(+) diff --git a/HACKING b/HACKING index fff003b..16be5cf 100644 --- a/HACKING +++ b/HACKING @@ -239,6 +239,77 @@ on the subject, on Richard Jones' guide to working with open source projects <http://people.redhat.com/rjones/how-to-supply-code-to-open-source-projects/>. +Naming conventions +================== +When reading libvirt code, a number of different naming conventions will be +evident due to various changes in thinking over the course of the project's +lifetime. The conventions documented below should be followed when creating +any entirely new files in libvirt. When working on existing files, while it is +desirable to apply these conventions, keeping a consistent style with existing +code in that particular file is generally more important. The overall guiding +rule is that every file, enum, struct, function, and typedef name must have a +'vir' or 'VIR' prefix. All local scope variable names are exempt, and global +variables are exempt, unless exported in a header file. + +*File names* + +File naming varies depending on the subdirectory. The preferred style is to +have a 'vir' prefix, followed by a name which matches the name of the +functions / objects inside the file. For example, a file containing an object +'virHashtable' is stored in files 'virhashtable.c' and 'virhashtable.h'. +Sometimes, methods which would otherwise be declared 'static' need to be +exported for use by a test suite. For this purpose a second header file should +be added with a suffix of 'priv'. e.g. 'virhashtablepriv.h'. USe of +underscores in file names is discouraged when using the 'vir' prefix style. +The 'vir' prefix naming applies to src/util, src/rpc and tests/ directories. +Most other directories do not follow this convention. + + + +*Enum type & field names* + +All enums should have a 'vir' prefix in their typedef name, and each following +word should have its first letter in uppercase. The enum name should match the +typedef name with a leading underscore. The enum member names should be in all +uppercase, and use an underscore to separate each word. The enum member name +prefix should match the enum typedef name. + + typedef enum _virSocketType virSocketType; + enum _virSocketType { + VIR_SOCKET_TYPE_IPV4, + VIR_SOCKET_TYPE_IPV6, + }; + + +*Struct type names* + +All structs should have a 'vir' prefix in their typedef name, and each +following word should have its first letter in uppercase. The struct name +should be the same as the typedef name with a leading underscore. A second +typedef should be given for a pointer to the struct with a 'Ptr' suffix. + + typedef struct _virHashTable virHashTable; + typedef virHashTable *virHashTablePtr; + struct _virHashTable { + ... + }; + + +*Function names* + +All functions should have a 'vir' prefix in their name, followed by one or +more words with first letter of each word capitalized. Underscores should not +be used in function names. If the function is operating on an object, then the +function name prefix should match the object typedef name. For example, given +an object 'virHashTable', all functions should have a name 'virHashTableXXXX' +e.g. 'virHashTableLookup'. If there is no object associated with the function, +then its name prefix should match the filename. For example, given a filename +of 'virfile.h', all functions should have a name 'virFileXXXX' e.g. +'virFileTouch'. + + + + Code indentation ================ Libvirt's C source code generally adheres to some basic code-formatting diff --git a/docs/hacking.html.in b/docs/hacking.html.in index b1bb149..028536d 100644 --- a/docs/hacking.html.in +++ b/docs/hacking.html.in @@ -314,6 +314,89 @@ Richard Jones' guide to working with open source projects</a>. </p> + <h2><a name="naming">Naming conventions</a></h2> + + <p> + When reading libvirt code, a number of different naming conventions will + be evident due to various changes in thinking over the course of the + project's lifetime. The conventions documented below should be followed + when creating any entirely new files in libvirt. When working on existing + files, while it is desirable to apply these conventions, keeping a + consistent style with existing code in that particular file is generally + more important. The overall guiding rule is that every file, enum, struct, + function, and typedef name must have a 'vir' or 'VIR' prefix. All local + scope variable names are exempt, and global variables are exempt, unless + exported in a header file. + </p> + + <dl> + <dt>File names</dt> + <dd> + <p> + File naming varies depending on the subdirectory. The preferred + style is to have a 'vir' prefix, followed by a name which matches + the name of the functions / objects inside the file. For example, + a file containing an object 'virHashtable' is stored in files + 'virhashtable.c' and 'virhashtable.h'. Sometimes, methods which + would otherwise be declared 'static' need to be exported for use + by a test suite. For this purpose a second header file should be + added with a suffix of 'priv'. e.g. 'virhashtablepriv.h'. USe of + underscores in file names is discouraged when using the 'vir' + prefix style. The 'vir' prefix naming applies to src/util, + src/rpc and tests/ directories. Most other directories do not + follow this convention. + </p> + </dd> + <dt>Enum type & field names</dt> + <dd> + <p> + All enums should have a 'vir' prefix in their typedef name, + and each following word should have its first letter in + uppercase. The enum name should match the typedef name with + a leading underscore. The enum member names should be in all + uppercase, and use an underscore to separate each word. The + enum member name prefix should match the enum typedef name. + </p> + <pre> + typedef enum _virSocketType virSocketType; + enum _virSocketType { + VIR_SOCKET_TYPE_IPV4, + VIR_SOCKET_TYPE_IPV6, + };</pre> + </dd> + <dt>Struct type names</dt> + <dd> + <p> + All structs should have a 'vir' prefix in their typedef name, + and each following word should have its first letter in + uppercase. The struct name should be the same as the typedef + name with a leading underscore. A second typedef should be + given for a pointer to the struct with a 'Ptr' suffix. + </p> + <pre> + typedef struct _virHashTable virHashTable; + typedef virHashTable *virHashTablePtr; + struct _virHashTable { + ... + };</pre> + </dd> + <dt>Function names</dt> + <dd> + <p> + All functions should have a 'vir' prefix in their name, + followed by one or more words with first letter of each + word capitalized. Underscores should not be used in function + names. If the function is operating on an object, then the + function name prefix should match the object typedef name. + For example, given an object 'virHashTable', all functions + should have a name 'virHashTableXXXX' e.g. 'virHashTableLookup'. + If there is no object associated with the function, then its + name prefix should match the filename. For example, given a + filename of 'virfile.h', all functions should have a name + 'virFileXXXX' e.g. 'virFileTouch'. + </p> + </dd> + </dl> <h2><a name="indent">Code indentation</a></h2> <p> diff --git a/docs/hacking2.xsl b/docs/hacking2.xsl index 3595b7e..7e5ac82 100644 --- a/docs/hacking2.xsl +++ b/docs/hacking2.xsl @@ -114,6 +114,10 @@ from docs/hacking.html.in! <xsl:template match="html:li/html:ul/html:li">-- <xsl:apply-templates/><xsl:value-of select="$newline"/><xsl:value-of select="$newline"/> </xsl:template> +<xsl:template match="html:dl/html:dt">*<xsl:apply-templates/>*<xsl:value-of select="$newline"/><xsl:value-of select="$newline"/> +</xsl:template> +<xsl:template match="html:dl/html:dd"><xsl:apply-templates/><xsl:value-of select="$newline"/><xsl:value-of select="$newline"/> +</xsl:template>  -- 2.9.3

6 10

[libvirt] [PATCH] make all struct typedefs comply with proposed coding conventions
by Laine Stump 06 Mar '17

06 Mar '17

Proposed formal coding conventions encourage defining typedefs for vir[Blah] and vir[Blah]Ptr separately from the associated struct named _vir[Blah]: typedef struct _virBlah virBlah; typedef virBlah *virBlahPtr; struct _virBlah { ... }; At some point in the past, I had submitted several patches using a more compact style that I prefer, and they were accepted: typedef struct _virBlah { ... } virBlah, *virBlahPtr; Since these are by far a minority among all struct definitions, this patch changes all those definitions to reflect the style prefered by the proposal so that there is 100% consistency. --- https://www.redhat.com/archives/libvir-list/2017-March/msg00215.html Now that we're talking about formalizing the format of struct definitions, since it was me who tried to make this catch on, it should be me who takes the time to fix them when they don't catch on :-P src/conf/device_conf.h | 60 ++++++++++++++++++++++++++++++--------------- src/conf/network_conf.h | 6 +++-- src/conf/node_device_conf.h | 6 +++-- src/util/virnetdev.h | 6 +++-- src/util/virnetdevip.h | 18 +++++++++----- 5 files changed, 64 insertions(+), 32 deletions(-) diff --git a/src/conf/device_conf.h b/src/conf/device_conf.h index f435fb5..a20de85 100644 --- a/src/conf/device_conf.h +++ b/src/conf/device_conf.h @@ -51,46 +51,58 @@ typedef enum { VIR_DOMAIN_DEVICE_ADDRESS_TYPE_LAST } virDomainDeviceAddressType; -typedef struct _virDomainDeviceDriveAddress { +typedef struct _virDomainDeviceDriveAddress virDomainDeviceDriveAddress; +typedef virDomainDeviceDriveAddress *virDomainDeviceDriveAddressPtr; +struct _virDomainDeviceDriveAddress { unsigned int controller; unsigned int bus; unsigned int target; unsigned int unit; -} virDomainDeviceDriveAddress, *virDomainDeviceDriveAddressPtr; +}; -typedef struct _virDomainDeviceVirtioSerialAddress { +typedef struct _virDomainDeviceVirtioSerialAddress virDomainDeviceVirtioSerialAddress; +typedef virDomainDeviceVirtioSerialAddress *virDomainDeviceVirtioSerialAddressPtr; +struct _virDomainDeviceVirtioSerialAddress { unsigned int controller; unsigned int bus; unsigned int port; -} virDomainDeviceVirtioSerialAddress, *virDomainDeviceVirtioSerialAddressPtr; +}; # define VIR_DOMAIN_DEVICE_CCW_MAX_CSSID 254 # define VIR_DOMAIN_DEVICE_CCW_MAX_SSID 3 # define VIR_DOMAIN_DEVICE_CCW_MAX_DEVNO 65535 -typedef struct _virDomainDeviceCCWAddress { +typedef struct _virDomainDeviceCCWAddress virDomainDeviceCCWAddress; +typedef virDomainDeviceCCWAddress *virDomainDeviceCCWAddressPtr; +struct _virDomainDeviceCCWAddress { unsigned int cssid; unsigned int ssid; unsigned int devno; bool assigned; -} virDomainDeviceCCWAddress, *virDomainDeviceCCWAddressPtr; +}; -typedef struct _virDomainDeviceCcidAddress { +typedef struct _virDomainDeviceCcidAddress virDomainDeviceCcidAddress; +typedef virDomainDeviceCcidAddress *virDomainDeviceCcidAddressPtr; +struct _virDomainDeviceCcidAddress { unsigned int controller; unsigned int slot; -} virDomainDeviceCcidAddress, *virDomainDeviceCcidAddressPtr; +}; # define VIR_DOMAIN_DEVICE_USB_MAX_PORT_DEPTH 4 -typedef struct _virDomainDeviceUSBAddress { +typedef struct _virDomainDeviceUSBAddress virDomainDeviceUSBAddress; +typedef virDomainDeviceUSBAddress *virDomainDeviceUSBAddressPtr; +struct _virDomainDeviceUSBAddress { unsigned int bus; unsigned int port[VIR_DOMAIN_DEVICE_USB_MAX_PORT_DEPTH]; -} virDomainDeviceUSBAddress, *virDomainDeviceUSBAddressPtr; +}; -typedef struct _virDomainDeviceSpaprVioAddress { +typedef struct _virDomainDeviceSpaprVioAddress virDomainDeviceSpaprVioAddress; +typedef virDomainDeviceSpaprVioAddress *virDomainDeviceSpaprVioAddressPtr; +struct _virDomainDeviceSpaprVioAddress { unsigned long long reg; bool has_reg; -} virDomainDeviceSpaprVioAddress, *virDomainDeviceSpaprVioAddressPtr; +}; typedef enum { VIR_DOMAIN_CONTROLLER_MASTER_NONE, @@ -99,21 +111,29 @@ typedef enum { VIR_DOMAIN_CONTROLLER_MASTER_LAST } virDomainControllerMaster; -typedef struct _virDomainDeviceUSBMaster { +typedef struct _virDomainDeviceUSBMaster virDomainDeviceUSBMaster; +typedef virDomainDeviceUSBMaster *virDomainDeviceUSBMasterPtr; +struct _virDomainDeviceUSBMaster { unsigned int startport; -} virDomainDeviceUSBMaster, *virDomainDeviceUSBMasterPtr; +}; -typedef struct _virDomainDeviceISAAddress { +typedef struct _virDomainDeviceISAAddress virDomainDeviceISAAddress; +typedef virDomainDeviceISAAddress *virDomainDeviceISAAddressPtr; +struct _virDomainDeviceISAAddress { unsigned int iobase; unsigned int irq; -} virDomainDeviceISAAddress, *virDomainDeviceISAAddressPtr; +}; -typedef struct _virDomainDeviceDimmAddress { +typedef struct _virDomainDeviceDimmAddress virDomainDeviceDimmAddress; +typedef virDomainDeviceDimmAddress *virDomainDeviceDimmAddressPtr; +struct _virDomainDeviceDimmAddress { unsigned int slot; unsigned long long base; -} virDomainDeviceDimmAddress, *virDomainDeviceDimmAddressPtr; +}; -typedef struct _virDomainDeviceInfo { +typedef struct _virDomainDeviceInfo virDomainDeviceInfo; +typedef virDomainDeviceInfo *virDomainDeviceInfoPtr; +struct _virDomainDeviceInfo { /* If adding to this struct, ensure that * virDomainDeviceInfoIsSet() is updated * to consider the new fields @@ -147,7 +167,7 @@ typedef struct _virDomainDeviceInfo { * assignment, never saved and never reported. */ int pciConnectFlags; /* enum virDomainPCIConnectFlags */ -} virDomainDeviceInfo, *virDomainDeviceInfoPtr; +}; int virPCIDeviceAddressIsValid(virPCIDeviceAddressPtr addr, diff --git a/src/conf/network_conf.h b/src/conf/network_conf.h index 9e4ae71..dec809e 100644 --- a/src/conf/network_conf.h +++ b/src/conf/network_conf.h @@ -127,10 +127,12 @@ struct _virNetworkDNSHostDef { }; -typedef struct _virNetworkDNSForwarder { +typedef struct _virNetworkDNSForwarder virNetworkDNSForwarder; +typedef virNetworkDNSForwarder *virNetworkDNSForwarderPtr; +struct _virNetworkDNSForwarder { virSocketAddr addr; char *domain; -} virNetworkDNSForwarder, *virNetworkDNSForwarderPtr; +}; typedef struct _virNetworkDNSDef virNetworkDNSDef; typedef virNetworkDNSDef *virNetworkDNSDefPtr; diff --git a/src/conf/node_device_conf.h b/src/conf/node_device_conf.h index 7aed1f4..cf51c69 100644 --- a/src/conf/node_device_conf.h +++ b/src/conf/node_device_conf.h @@ -243,7 +243,9 @@ struct _virNodeDevCapDRM { virNodeDevDRMType type; }; -typedef struct _virNodeDevCapData { +typedef struct _virNodeDevCapData virNodeDevCapData; +typedef virNodeDevCapData *virNodeDevCapDataPtr; +struct _virNodeDevCapData { virNodeDevCapType type; union { virNodeDevCapSystem system; @@ -258,7 +260,7 @@ typedef struct _virNodeDevCapData { virNodeDevCapSCSIGeneric sg; virNodeDevCapDRM drm; }; -} virNodeDevCapData, *virNodeDevCapDataPtr; +}; typedef struct _virNodeDevCapsDef virNodeDevCapsDef; typedef virNodeDevCapsDef *virNodeDevCapsDefPtr; diff --git a/src/util/virnetdev.h b/src/util/virnetdev.h index c3e8ffc..236cf83 100644 --- a/src/util/virnetdev.h +++ b/src/util/virnetdev.h @@ -86,10 +86,12 @@ typedef enum { VIR_ENUM_DECL(virNetDevIfState) -typedef struct { +typedef struct _virNetDevIfLink virNetDevIfLink; +typedef virNetDevIfLink *virNetDevIfLinkPtr; +struct _virNetDevIfLink { virNetDevIfState state; /* link state */ unsigned int speed; /* link speed in Mbits per second */ -} virNetDevIfLink, *virNetDevIfLinkPtr; +}; typedef enum { VIR_NET_DEV_FEAT_GRXCSUM, diff --git a/src/util/virnetdevip.h b/src/util/virnetdevip.h index 8277654..b7abdf9 100644 --- a/src/util/virnetdevip.h +++ b/src/util/virnetdevip.h @@ -25,13 +25,17 @@ # include "virsocketaddr.h" -typedef struct { +typedef struct _virNetDevIPAddr virNetDevIPAddr; +typedef virNetDevIPAddr *virNetDevIPAddrPtr; +struct _virNetDevIPAddr { virSocketAddr address; /* ipv4 or ipv6 address */ virSocketAddr peer; /* ipv4 or ipv6 address of peer */ unsigned int prefix; /* number of 1 bits in the netmask */ -} virNetDevIPAddr, *virNetDevIPAddrPtr; +}; -typedef struct { +typedef struct _virNetDevIPRoute virNetDevIPRoute; +typedef virNetDevIPRoute *virNetDevIPRoutePtr; +struct _virNetDevIPRoute { char *family; /* ipv4 or ipv6 - default is ipv4 */ virSocketAddr address; /* Routed Network IP address */ @@ -46,15 +50,17 @@ typedef struct { unsigned int metric; /* value for metric (defaults to 1) */ bool has_metric; /* metric= was specified */ virSocketAddr gateway; /* gateway IP address for ip-route */ -} virNetDevIPRoute, *virNetDevIPRoutePtr; +}; /* A full set of all IP config info for a network device */ -typedef struct { +typedef struct _virNetDevIPInfo virNetDevIPInfo; +typedef virNetDevIPInfo *virNetDevIPInfoPtr; + struct _virNetDevIPInfo { size_t nips; virNetDevIPAddrPtr *ips; size_t nroutes; virNetDevIPRoutePtr *routes; -} virNetDevIPInfo, *virNetDevIPInfoPtr; +}; /* manipulating/querying the netdev */ int virNetDevIPAddrAdd(const char *ifname, -- 2.9.3

2 1

[libvirt] [PATCH] qemu_capabilities: report SATA bus in domain capabilities
by Pavel Hrdina 06 Mar '17

06 Mar '17

Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- src/qemu/qemu_capabilities.c | 4 ++++ tests/domaincapsschemadata/qemu_1.7.0.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0-gicv2-virt.aarch64.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0-gicv3-virt.aarch64.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0.aarch64.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0.ppc64le.xml | 1 + tests/domaincapsschemadata/qemu_2.6.0.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.8.0-tcg.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.8.0.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.9.0-tcg.x86_64.xml | 1 + tests/domaincapsschemadata/qemu_2.9.0.x86_64.xml | 1 + 11 files changed, 14 insertions(+) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 359a0d87a1..5a3b4ac503 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -5422,6 +5422,10 @@ virQEMUCapsFillDomainDeviceDiskCaps(virQEMUCapsPtr qemuCaps, if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_DEVICE_USB_STORAGE)) VIR_DOMAIN_CAPS_ENUM_SET(disk->bus, VIR_DOMAIN_DISK_BUS_USB); + + if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_ICH9_AHCI)) + VIR_DOMAIN_CAPS_ENUM_SET(disk->bus, VIR_DOMAIN_DISK_BUS_SATA); + return 0; } diff --git a/tests/domaincapsschemadata/qemu_1.7.0.x86_64.xml b/tests/domaincapsschemadata/qemu_1.7.0.x86_64.xml index 4aa475c1e7..1eadba393f 100644 --- a/tests/domaincapsschemadata/qemu_1.7.0.x86_64.xml +++ b/tests/domaincapsschemadata/qemu_1.7.0.x86_64.xml @@ -64,6 +64,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.6.0-gicv2-virt.aarch64.xml b/tests/domaincapsschemadata/qemu_2.6.0-gicv2-virt.aarch64.xml index 76aee83ba1..1fa7f6dff8 100644 --- a/tests/domaincapsschemadata/qemu_2.6.0-gicv2-virt.aarch64.xml +++ b/tests/domaincapsschemadata/qemu_2.6.0-gicv2-virt.aarch64.xml @@ -68,6 +68,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.6.0-gicv3-virt.aarch64.xml b/tests/domaincapsschemadata/qemu_2.6.0-gicv3-virt.aarch64.xml index 104c29fabb..d60fc1df98 100644 --- a/tests/domaincapsschemadata/qemu_2.6.0-gicv3-virt.aarch64.xml +++ b/tests/domaincapsschemadata/qemu_2.6.0-gicv3-virt.aarch64.xml @@ -68,6 +68,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.6.0.aarch64.xml b/tests/domaincapsschemadata/qemu_2.6.0.aarch64.xml index ca511f48d8..fcc6f50e0e 100644 --- a/tests/domaincapsschemadata/qemu_2.6.0.aarch64.xml +++ b/tests/domaincapsschemadata/qemu_2.6.0.aarch64.xml @@ -68,6 +68,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.6.0.ppc64le.xml b/tests/domaincapsschemadata/qemu_2.6.0.ppc64le.xml index b3d2d2a7e4..755c4f4475 100644 --- a/tests/domaincapsschemadata/qemu_2.6.0.ppc64le.xml +++ b/tests/domaincapsschemadata/qemu_2.6.0.ppc64le.xml @@ -41,6 +41,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.6.0.x86_64.xml b/tests/domaincapsschemadata/qemu_2.6.0.x86_64.xml index a8975e8e53..843bdc2b73 100644 --- a/tests/domaincapsschemadata/qemu_2.6.0.x86_64.xml +++ b/tests/domaincapsschemadata/qemu_2.6.0.x86_64.xml @@ -68,6 +68,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.8.0-tcg.x86_64.xml b/tests/domaincapsschemadata/qemu_2.8.0-tcg.x86_64.xml index dd9285311c..962cd0557f 100644 --- a/tests/domaincapsschemadata/qemu_2.8.0-tcg.x86_64.xml +++ b/tests/domaincapsschemadata/qemu_2.8.0-tcg.x86_64.xml @@ -69,6 +69,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.8.0.x86_64.xml b/tests/domaincapsschemadata/qemu_2.8.0.x86_64.xml index 4c5fffcad4..7d5ac063fa 100644 --- a/tests/domaincapsschemadata/qemu_2.8.0.x86_64.xml +++ b/tests/domaincapsschemadata/qemu_2.8.0.x86_64.xml @@ -69,6 +69,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.9.0-tcg.x86_64.xml b/tests/domaincapsschemadata/qemu_2.9.0-tcg.x86_64.xml index 573eb4bb6e..413f3fa7ce 100644 --- a/tests/domaincapsschemadata/qemu_2.9.0-tcg.x86_64.xml +++ b/tests/domaincapsschemadata/qemu_2.9.0-tcg.x86_64.xml @@ -98,6 +98,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> diff --git a/tests/domaincapsschemadata/qemu_2.9.0.x86_64.xml b/tests/domaincapsschemadata/qemu_2.9.0.x86_64.xml index b361475d7e..a4ab74dcaf 100644 --- a/tests/domaincapsschemadata/qemu_2.9.0.x86_64.xml +++ b/tests/domaincapsschemadata/qemu_2.9.0.x86_64.xml @@ -77,6 +77,7 @@ <value>scsi</value> <value>virtio</value> <value>usb</value> + <value>sata</value> </enum> </disk> <graphics supported='yes'> -- 2.12.0

2 1

[libvirt] [PATCH] Add virsystemdpriv.h to UTIL_SOURCES
by Daniel P. Berrange 06 Mar '17

06 Mar '17

Ensure virsystemdpriv.h gets included in dist tarballs. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- Pushed to fix RPM builds src/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Makefile.am b/src/Makefile.am index c85927f..62c8743 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -174,7 +174,7 @@ UTIL_SOURCES = \ util/virstoragefile.c util/virstoragefile.h \ util/virstring.h util/virstring.c \ util/virsysinfo.c util/virsysinfo.h \ - util/virsystemd.c util/virsystemd.h \ + util/virsystemd.c util/virsystemd.h util/virsystemdpriv.h \ util/virthread.c util/virthread.h \ util/virthreadjob.c util/virthreadjob.h \ util/virthreadpool.c util/virthreadpool.h \ -- 2.9.3

1 0

[libvirt] [PATCH v2] Don't print extra newline in virsh domstats output
by Nitesh Konkar 06 Mar '17

06 Mar '17

Signed-off-by: Nitesh Konkar <nitkon12(a)linux.vnet.ibm.com> --- virsh domstats --perf Domain: 'Fedora123' perf.cache_misses=394986 Domain: 'Fedora' perf.cache_misses=447017 virsh domstats Domain: 'Fedora123' state.state=1 state.reason=1 cpu.time=562472167198 cpu.user=228340000000 cpu.system=303250000000 balloon.current=1048576 balloon.maximum=1048576 perf.cache_misses=634012 Domain: 'Fedora' state.state=1 state.reason=1 cpu.time=19303277561180 cpu.user=5634250000000 cpu.system=9536140000000 balloon.current=1048576 balloon.maximum=1048576 perf.cache_misses=699207 tools/virsh-domain-monitor.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index 77aa272..901a6eb 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -2047,7 +2047,6 @@ virshDomainStatsPrintRecord(vshControl *ctl ATTRIBUTE_UNUSED, VIR_FREE(param); } - vshPrint(ctl, "\n"); return true; } @@ -2145,9 +2144,13 @@ cmdDomstats(vshControl *ctl, const vshCmd *cmd) goto cleanup; } - for (next = records; *next; next++) { + next = records; + while (*next) { if (!virshDomainStatsPrintRecord(ctl, *next, raw)) goto cleanup; + + if (*(++next)) + vshPrint(ctl, "\n"); } ret = true; -- 1.9.3

2 2

[libvirt] [PATCH v2.1] qemu_capabilities: introduce QEMU_CAPS_SD_CARD to probe sd-card drivers
by Chen Hanxiao 06 Mar '17

06 Mar '17

From: Chen Hanxiao <chenhanxiao(a)gmail.com> This patch introduces QEMU_CAPS_SD_CARD for probing whether qemu support SD card by: {"execute": "device-list-properties", "arguments":{"typename":"sd-card"}} It will be helpful for apps which used cmd 'virsh domcaps` etc. Also helpful for: https://bugzilla.redhat.com/show_bug.cgi?id=1387218 Signed-off-by: Chen Hanxiao <chenhanxiao(a)gmail.com> --- v2: rebased on upstream v2.1: add doc/news.xml description rebased on upstream docs/news.xml | 9 +++++++++ src/qemu/qemu_capabilities.c | 9 +++++++-- src/qemu/qemu_capabilities.h | 1 + 3 files changed, 17 insertions(+), 2 deletions(-) diff --git a/docs/news.xml b/docs/news.xml index 584c9af..b218463 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -26,6 +26,15 @@ don't start it with unsupported features. </description> </change> + <change> + <summary> + qemu_capabilities: introduce QEMU_CAPS_SD_CARD to probe sd-card drivers + </summary> + <description> + The new QEMU capability flag will allow libvirt to probe sd-card + drivers. + </description> + </change> </section> <section title="Bug fixes"> <change/> diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 5694d0b..1828b30 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -359,6 +359,7 @@ VIR_ENUM_IMPL(virQEMUCaps, QEMU_CAPS_LAST, "query-cpu-model-expansion", /* 245 */ "virtio-net.host_mtu", "spice-rendernode", + "sd-card", ); @@ -1626,6 +1627,7 @@ struct virQEMUCapsStringFlags virQEMUCapsObjectTypes[] = { { "ivshmem-plain", QEMU_CAPS_DEVICE_IVSHMEM_PLAIN }, { "ivshmem-doorbell", QEMU_CAPS_DEVICE_IVSHMEM_DOORBELL }, { "vhost-scsi", QEMU_CAPS_DEVICE_VHOST_SCSI }, + { "sd-card", QEMU_CAPS_SD_CARD }, }; static struct virQEMUCapsStringFlags virQEMUCapsObjectPropsVirtioBalloon[] = { @@ -5425,8 +5427,7 @@ virQEMUCapsFillDomainDeviceDiskCaps(virQEMUCapsPtr qemuCaps, VIR_DOMAIN_CAPS_ENUM_SET(disk->bus, VIR_DOMAIN_DISK_BUS_IDE, VIR_DOMAIN_DISK_BUS_SCSI, - VIR_DOMAIN_DISK_BUS_VIRTIO, - /* VIR_DOMAIN_DISK_BUS_SD */); + VIR_DOMAIN_DISK_BUS_VIRTIO); /* PowerPC pseries based VMs do not support floppy device */ if (!ARCH_IS_PPC64(qemuCaps->arch) || @@ -5435,6 +5436,10 @@ virQEMUCapsFillDomainDeviceDiskCaps(virQEMUCapsPtr qemuCaps, if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_DEVICE_USB_STORAGE)) VIR_DOMAIN_CAPS_ENUM_SET(disk->bus, VIR_DOMAIN_DISK_BUS_USB); + + if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_SD_CARD)) + VIR_DOMAIN_CAPS_ENUM_SET(disk->bus, VIR_DOMAIN_DISK_BUS_SD); + return 0; } diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index cc9f46e..decca12 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -394,6 +394,7 @@ typedef enum { QEMU_CAPS_QUERY_CPU_MODEL_EXPANSION, /* qmp query-cpu-model-expansion */ QEMU_CAPS_VIRTIO_NET_HOST_MTU, /* virtio-net-*.host_mtu */ QEMU_CAPS_SPICE_RENDERNODE, /* -spice rendernode */ + QEMU_CAPS_SD_CARD, /* -sd SD-card.img */ QEMU_CAPS_LAST /* this must always be the last item */ } virQEMUCapsFlags; -- 2.7.4

2 1

[libvirt] [PATCH] qemu: Enforce qemuSecurity wrappers
by Michal Privoznik 06 Mar '17

06 Mar '17

Now that we have some qemuSecurity wrappers over virSecurityManager APIs, lets make sure everybody sticks with them. We have them for a reason and calling virSecurityManager API directly instead of wrapper may lead into accidentally labelling a file on the host instead of namespace. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- This is an alternative approach to: https://www.redhat.com/archives/libvir-list/2017-February/msg00271.html cfg.mk | 5 ++++ src/qemu/qemu_command.c | 7 +++--- src/qemu/qemu_conf.c | 9 ++++--- src/qemu/qemu_domain.c | 17 ++++++------- src/qemu/qemu_driver.c | 63 ++++++++++++++++++++++------------------------- src/qemu/qemu_hotplug.c | 4 +-- src/qemu/qemu_migration.c | 13 +++++----- src/qemu/qemu_process.c | 61 ++++++++++++++++++++++----------------------- src/qemu/qemu_security.h | 32 ++++++++++++++++++++++++ 9 files changed, 122 insertions(+), 89 deletions(-) diff --git a/cfg.mk b/cfg.mk index 69e3f3a1a..489fda8ea 100644 --- a/cfg.mk +++ b/cfg.mk @@ -983,6 +983,11 @@ sc_prohibit_sysconf_pagesize: halt='use virGetSystemPageSize[KB] instead of sysconf(_SC_PAGESIZE)' \ $(_sc_search_regexp) +sc_prohibit_virSecurity: + @grep -P 'virSecurityManager(?!Ptr)' $$($(VC_LIST_EXCEPT) | grep '^src/qemu/' | \ + grep -v '^src/qemu/qemu_security') && \ + { echo '$(ME): prefer qemuSecurity wrappers' 1>&2; exit 1; } || : + sc_prohibit_pthread_create: @prohibit='\bpthread_create\b' \ exclude='sc_prohibit_pthread_create' \ diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index c00a47a91..110540ba7 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -28,6 +28,7 @@ #include "qemu_capabilities.h" #include "qemu_interface.h" #include "qemu_alias.h" +#include "qemu_security.h" #include "cpu/cpu.h" #include "dirname.h" #include "viralloc.h" @@ -8321,8 +8322,8 @@ qemuBuildInterfaceCommandLine(virQEMUDriverPtr driver, } for (i = 0; i < tapfdSize; i++) { - if (virSecurityManagerSetTapFDLabel(driver->securityManager, - def, tapfd[i]) < 0) + if (qemuSecuritySetTapFDLabel(driver->securityManager, + def, tapfd[i]) < 0) goto cleanup; virCommandPassFD(cmd, tapfd[i], VIR_COMMAND_PASS_FD_CLOSE_PARENT); @@ -8403,7 +8404,7 @@ qemuBuildInterfaceCommandLine(virQEMUDriverPtr driver, /* NOTE: Not using const virDomainDef here since eventually a call is made - * into virSecurityManagerSetTapFDLabel which calls it's driver + * into qemuSecuritySetTapFDLabel which calls it's driver * API domainSetSecurityTapFDLabel that doesn't use the const format. */ static int diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 0223a95d2..4fc0dee39 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -38,6 +38,7 @@ #include "qemu_conf.h" #include "qemu_capabilities.h" #include "qemu_domain.h" +#include "qemu_security.h" #include "viruuid.h" #include "virbuffer.h" #include "virconf.h" @@ -904,7 +905,7 @@ virCapsPtr virQEMUDriverCreateCapabilities(virQEMUDriverPtr driver) } /* access sec drivers and create a sec model for each one */ - if (!(sec_managers = virSecurityManagerGetNested(driver->securityManager))) + if (!(sec_managers = qemuSecurityGetNested(driver->securityManager))) goto error; /* calculate length */ @@ -917,14 +918,14 @@ virCapsPtr virQEMUDriverCreateCapabilities(virQEMUDriverPtr driver) for (i = 0; sec_managers[i]; i++) { virCapsHostSecModelPtr sm = &caps->host.secModels[i]; - doi = virSecurityManagerGetDOI(sec_managers[i]); - model = virSecurityManagerGetModel(sec_managers[i]); + doi = qemuSecurityGetDOI(sec_managers[i]); + model = qemuSecurityGetModel(sec_managers[i]); if (VIR_STRDUP(sm->model, model) < 0 || VIR_STRDUP(sm->doi, doi) < 0) goto error; for (j = 0; j < ARRAY_CARDINALITY(virtTypes); j++) { - lbl = virSecurityManagerGetBaseLabel(sec_managers[i], virtTypes[j]); + lbl = qemuSecurityGetBaseLabel(sec_managers[i], virtTypes[j]); type = virDomainVirtTypeToString(virtTypes[j]); if (lbl && virCapabilitiesHostSecModelAddBaseLabel(sm, type, lbl) < 0) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index f62bf8f1d..2c827ea2c 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -588,8 +588,8 @@ qemuDomainWriteMasterKeyFile(virQEMUDriverPtr driver, goto cleanup; } - if (virSecurityManagerDomainSetPathLabel(driver->securityManager, - vm->def, path) < 0) + if (qemuSecurityDomainSetPathLabel(driver->securityManager, + vm->def, path) < 0) goto cleanup; ret = 0; @@ -2688,7 +2688,7 @@ qemuDomainDefPostParse(virDomainDefPtr def, if (qemuDomainRecheckInternalPaths(def, cfg, parseFlags) < 0) goto cleanup; - if (virSecurityManagerVerify(driver->securityManager, def) < 0) + if (qemuSecurityVerify(driver->securityManager, def) < 0) goto cleanup; if (qemuDomainDefVcpusPostParse(def) < 0) @@ -7257,8 +7257,7 @@ qemuDomainSetupDev(virQEMUDriverPtr driver, VIR_DEBUG("Setting up /dev/ for domain %s", vm->def->name); - mount_options = virSecurityManagerGetMountOptions(driver->securityManager, - vm->def); + mount_options = qemuSecurityGetMountOptions(driver->securityManager, vm->def); if (!mount_options && VIR_STRDUP(mount_options, "") < 0) @@ -7679,7 +7678,7 @@ qemuDomainAttachDeviceMknodHelper(pid_t pid ATTRIBUTE_UNUSED, bool delDevice = false; bool isLink = S_ISLNK(data->sb.st_mode); - virSecurityManagerPostFork(data->driver->securityManager); + qemuSecurityPostFork(data->driver->securityManager); if (virFileMakeParentPath(data->file) < 0) { virReportSystemError(errno, @@ -7841,16 +7840,16 @@ qemuDomainAttachDeviceMknodRecursive(virQEMUDriverPtr driver, #endif if (STRPREFIX(file, DEVPREFIX)) { - if (virSecurityManagerPreFork(driver->securityManager) < 0) + if (qemuSecurityPreFork(driver->securityManager) < 0) goto cleanup; if (virProcessRunInMountNamespace(vm->pid, qemuDomainAttachDeviceMknodHelper, &data) < 0) { - virSecurityManagerPostFork(driver->securityManager); + qemuSecurityPostFork(driver->securityManager); goto cleanup; } - virSecurityManagerPostFork(driver->securityManager); + qemuSecurityPostFork(driver->securityManager); } if (isLink && diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 89bc833de..096fe36fe 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -405,26 +405,26 @@ qemuSecurityInit(virQEMUDriverPtr driver) cfg->securityDriverNames[0]) { names = cfg->securityDriverNames; while (names && *names) { - if (!(mgr = virSecurityManagerNew(*names, - QEMU_DRIVER_NAME, - flags))) + if (!(mgr = qemuSecurityNew(*names, + QEMU_DRIVER_NAME, + flags))) goto error; if (!stack) { - if (!(stack = virSecurityManagerNewStack(mgr))) + if (!(stack = qemuSecurityNewStack(mgr))) goto error; } else { - if (virSecurityManagerStackAddNested(stack, mgr) < 0) + if (qemuSecurityStackAddNested(stack, mgr) < 0) goto error; } mgr = NULL; names++; } } else { - if (!(mgr = virSecurityManagerNew(NULL, - QEMU_DRIVER_NAME, - flags))) + if (!(mgr = qemuSecurityNew(NULL, + QEMU_DRIVER_NAME, + flags))) goto error; - if (!(stack = virSecurityManagerNewStack(mgr))) + if (!(stack = qemuSecurityNewStack(mgr))) goto error; mgr = NULL; } @@ -432,17 +432,17 @@ qemuSecurityInit(virQEMUDriverPtr driver) if (virQEMUDriverIsPrivileged(driver)) { if (cfg->dynamicOwnership) flags |= VIR_SECURITY_MANAGER_DYNAMIC_OWNERSHIP; - if (!(mgr = virSecurityManagerNewDAC(QEMU_DRIVER_NAME, - cfg->user, - cfg->group, - flags, - qemuSecurityChownCallback))) + if (!(mgr = qemuSecurityNewDAC(QEMU_DRIVER_NAME, + cfg->user, + cfg->group, + flags, + qemuSecurityChownCallback))) goto error; if (!stack) { - if (!(stack = virSecurityManagerNewStack(mgr))) + if (!(stack = qemuSecurityNewStack(mgr))) goto error; } else { - if (virSecurityManagerStackAddNested(stack, mgr) < 0) + if (qemuSecurityStackAddNested(stack, mgr) < 0) goto error; } mgr = NULL; @@ -3088,7 +3088,7 @@ qemuDomainSaveMemory(virQEMUDriverPtr driver, if (fd < 0) goto cleanup; - if (virSecurityManagerSetImageFDLabel(driver->securityManager, vm->def, fd) < 0) + if (qemuSecuritySetImageFDLabel(driver->securityManager, vm->def, fd) < 0) goto cleanup; if (!(wrapperFd = virFileWrapperFdNew(&fd, path, wrapperFlags))) @@ -3553,8 +3553,7 @@ static int qemuDumpToFd(virQEMUDriverPtr driver, virDomainObjPtr vm, return -1; } - if (virSecurityManagerSetImageFDLabel(driver->securityManager, vm->def, - fd) < 0) + if (qemuSecuritySetImageFDLabel(driver->securityManager, vm->def, fd) < 0) return -1; VIR_FREE(priv->job.current); @@ -3846,7 +3845,7 @@ qemuDomainScreenshot(virDomainPtr dom, } unlink_tmp = true; - virSecurityManagerSetSavedStateLabel(driver->securityManager, vm->def, tmp); + qemuSecuritySetSavedStateLabel(driver->securityManager, vm->def, tmp); qemuDomainObjEnterMonitor(driver, vm); if (qemuMonitorScreendump(priv->mon, tmp) < 0) { @@ -5928,8 +5927,8 @@ static int qemuDomainGetSecurityLabel(virDomainPtr dom, virSecurityLabelPtr secl * QEMU monitor hasn't seen SIGHUP/ERR on poll(). */ if (virDomainObjIsActive(vm)) { - if (virSecurityManagerGetProcessLabel(driver->securityManager, - vm->def, vm->pid, seclabel) < 0) { + if (qemuSecurityGetProcessLabel(driver->securityManager, + vm->def, vm->pid, seclabel) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Failed to get security label")); goto cleanup; @@ -5973,8 +5972,7 @@ static int qemuDomainGetSecurityLabelList(virDomainPtr dom, ret = 0; } else { int len = 0; - virSecurityManagerPtr* mgrs = virSecurityManagerGetNested( - driver->securityManager); + virSecurityManagerPtr* mgrs = qemuSecurityGetNested(driver->securityManager); if (!mgrs) goto cleanup; @@ -5990,8 +5988,8 @@ static int qemuDomainGetSecurityLabelList(virDomainPtr dom, /* Fill the array */ for (i = 0; i < len; i++) { - if (virSecurityManagerGetProcessLabel(mgrs[i], vm->def, vm->pid, - &(*seclabels)[i]) < 0) { + if (qemuSecurityGetProcessLabel(mgrs[i], vm->def, vm->pid, + &(*seclabels)[i]) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Failed to get security label")); VIR_FREE(mgrs); @@ -6369,8 +6367,8 @@ qemuDomainSaveImageStartVM(virConnectPtr conn, cleanup: virCommandFree(cmd); VIR_FREE(errbuf); - if (virSecurityManagerRestoreSavedStateLabel(driver->securityManager, - vm->def, path) < 0) + if (qemuSecurityRestoreSavedStateLabel(driver->securityManager, + vm->def, path) < 0) VIR_WARN("failed to restore save state label on %s", path); virObjectUnref(cfg); return ret; @@ -11196,7 +11194,7 @@ qemuDomainMemoryPeek(virDomainPtr dom, goto endjob; } - virSecurityManagerSetSavedStateLabel(driver->securityManager, vm->def, tmp); + qemuSecuritySetSavedStateLabel(driver->securityManager, vm->def, tmp); priv = vm->privateData; qemuDomainObjEnterMonitor(driver, vm); @@ -17064,8 +17062,7 @@ qemuDomainOpenGraphics(virDomainPtr dom, goto endjob; } - if (virSecurityManagerSetImageFDLabel(driver->securityManager, vm->def, - fd) < 0) + if (qemuSecuritySetImageFDLabel(driver->securityManager, vm->def, fd) < 0) goto endjob; qemuDomainObjEnterMonitor(driver, vm); @@ -17129,13 +17126,13 @@ qemuDomainOpenGraphicsFD(virDomainPtr dom, goto cleanup; } - if (virSecurityManagerSetSocketLabel(driver->securityManager, vm->def) < 0) + if (qemuSecuritySetSocketLabel(driver->securityManager, vm->def) < 0) goto cleanup; if (socketpair(PF_UNIX, SOCK_STREAM, 0, pair) < 0) goto cleanup; - if (virSecurityManagerClearSocketLabel(driver->securityManager, vm->def) < 0) + if (qemuSecurityClearSocketLabel(driver->securityManager, vm->def) < 0) goto cleanup; if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MODIFY) < 0) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 2f209f12b..b99b0e9fb 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1134,8 +1134,8 @@ qemuDomainAttachNetDevice(virQEMUDriverPtr driver, } for (i = 0; i < tapfdSize; i++) { - if (virSecurityManagerSetTapFDLabel(driver->securityManager, - vm->def, tapfd[i]) < 0) + if (qemuSecuritySetTapFDLabel(driver->securityManager, + vm->def, tapfd[i]) < 0) goto cleanup; } diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 0f4a6cf21..c40cb1391 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -40,6 +40,7 @@ #include "qemu_cgroup.h" #include "qemu_hotplug.h" #include "qemu_blockjob.h" +#include "qemu_security.h" #include "domain_audit.h" #include "virlog.h" @@ -4597,7 +4598,7 @@ qemuMigrationConnect(virQEMUDriverPtr driver, spec->destType = MIGRATION_DEST_FD; spec->dest.fd.qemu = -1; - if (virSecurityManagerSetSocketLabel(driver->securityManager, vm->def) < 0) + if (qemuSecuritySetSocketLabel(driver->securityManager, vm->def) < 0) goto cleanup; if (virNetSocketNewConnectTCP(host, port, AF_UNSPEC, @@ -4605,7 +4606,7 @@ qemuMigrationConnect(virQEMUDriverPtr driver, spec->dest.fd.qemu = virNetSocketDupFD(sock, true); virObjectUnref(sock); } - if (virSecurityManagerClearSocketLabel(driver->securityManager, vm->def) < 0 || + if (qemuSecurityClearSocketLabel(driver->securityManager, vm->def) < 0 || spec->dest.fd.qemu == -1) goto cleanup; @@ -5076,8 +5077,8 @@ static int doTunnelMigrate(virQEMUDriverPtr driver, spec.dest.fd.local = fds[0]; } if (spec.dest.fd.qemu == -1 || - virSecurityManagerSetImageFDLabel(driver->securityManager, vm->def, - spec.dest.fd.qemu) < 0) { + qemuSecuritySetImageFDLabel(driver->securityManager, vm->def, + spec.dest.fd.qemu) < 0) { virReportSystemError(errno, "%s", _("cannot create pipe for tunnelled migration")); goto cleanup; @@ -6463,8 +6464,8 @@ qemuMigrationToFile(virQEMUDriverPtr driver, virDomainObjPtr vm, * doesn't have to open() the file, so while we still have to * grant SELinux access, we can do it on fd and avoid cleanup * later, as well as skip futzing with cgroup. */ - if (virSecurityManagerSetImageFDLabel(driver->securityManager, vm->def, - compressor ? pipeFD[1] : fd) < 0) + if (qemuSecuritySetImageFDLabel(driver->securityManager, vm->def, + compressor ? pipeFD[1] : fd) < 0) goto cleanup; if (qemuDomainObjEnterMonitorAsync(driver, vm, asyncJob) < 0) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 92fa69b3c..5c44e565b 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -221,8 +221,7 @@ qemuConnectAgent(virQEMUDriverPtr driver, virDomainObjPtr vm) return 0; } - if (virSecurityManagerSetDaemonSocketLabel(driver->securityManager, - vm->def) < 0) { + if (qemuSecuritySetDaemonSocketLabel(driver->securityManager, vm->def) < 0) { VIR_ERROR(_("Failed to set security context for agent for %s"), vm->def->name); goto cleanup; @@ -250,8 +249,7 @@ qemuConnectAgent(virQEMUDriverPtr driver, virDomainObjPtr vm) return -1; } - if (virSecurityManagerClearSocketLabel(driver->securityManager, - vm->def) < 0) { + if (qemuSecurityClearSocketLabel(driver->securityManager, vm->def) < 0) { VIR_ERROR(_("Failed to clear security context for agent for %s"), vm->def->name); qemuAgentClose(agent); @@ -1657,8 +1655,7 @@ qemuConnectMonitor(virQEMUDriverPtr driver, virDomainObjPtr vm, int asyncJob, int ret = -1; qemuMonitorPtr mon = NULL; - if (virSecurityManagerSetDaemonSocketLabel(driver->securityManager, - vm->def) < 0) { + if (qemuSecuritySetDaemonSocketLabel(driver->securityManager, vm->def) < 0) { VIR_ERROR(_("Failed to set security context for monitor for %s"), vm->def->name); return -1; @@ -1695,7 +1692,7 @@ qemuConnectMonitor(virQEMUDriverPtr driver, virDomainObjPtr vm, int asyncJob, } priv->mon = mon; - if (virSecurityManagerClearSocketLabel(driver->securityManager, vm->def) < 0) { + if (qemuSecurityClearSocketLabel(driver->securityManager, vm->def) < 0) { VIR_ERROR(_("Failed to clear security context for monitor for %s"), vm->def->name); return -1; @@ -2638,7 +2635,7 @@ static int qemuProcessHook(void *data) * protected across fork() */ - virSecurityManagerPostFork(h->driver->securityManager); + qemuSecurityPostFork(h->driver->securityManager); /* Some later calls want pid present */ h->vm->pid = getpid(); @@ -2651,7 +2648,7 @@ static int qemuProcessHook(void *data) * sockets the lock driver opens that we don't want * labelled. So far we're ok though. */ - if (virSecurityManagerSetSocketLabel(h->driver->securityManager, h->vm->def) < 0) + if (qemuSecuritySetSocketLabel(h->driver->securityManager, h->vm->def) < 0) goto cleanup; if (virDomainLockProcessStart(h->driver->lockManager, h->cfg->uri, @@ -2660,7 +2657,7 @@ static int qemuProcessHook(void *data) true, &fd) < 0) goto cleanup; - if (virSecurityManagerClearSocketLabel(h->driver->securityManager, h->vm->def) < 0) + if (qemuSecurityClearSocketLabel(h->driver->securityManager, h->vm->def) < 0) goto cleanup; if (qemuDomainBuildNamespace(h->driver, h->vm) < 0) @@ -3260,8 +3257,8 @@ qemuProcessBuildDestroyHugepagesPath(virQEMUDriverPtr driver, goto cleanup; } - if (virSecurityManagerDomainSetPathLabel(driver->securityManager, - vm->def, hugepagePath) < 0) { + if (qemuSecurityDomainSetPathLabel(driver->securityManager, + vm->def, hugepagePath) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Unable to set huge path in security driver")); goto cleanup; @@ -3437,13 +3434,13 @@ qemuProcessReconnect(void *opaque) /* if domain requests security driver we haven't loaded, report error, but * do not kill the domain */ - ignore_value(virSecurityManagerCheckAllLabel(driver->securityManager, - obj->def)); + ignore_value(qemuSecurityCheckAllLabel(driver->securityManager, + obj->def)); if (qemuDomainRefreshVcpuInfo(driver, obj, QEMU_ASYNC_JOB_NONE, true) < 0) goto error; - if (virSecurityManagerReserveLabel(driver->securityManager, obj->def, obj->pid) < 0) + if (qemuSecurityReserveLabel(driver->securityManager, obj->def, obj->pid) < 0) goto error; if (qemuProcessNotifyNets(obj->def) < 0) @@ -4451,8 +4448,8 @@ qemuProcessMakeDir(virQEMUDriverPtr driver, goto cleanup; } - if (virSecurityManagerDomainSetPathLabel(driver->securityManager, - vm->def, path) < 0) + if (qemuSecurityDomainSetPathLabel(driver->securityManager, + vm->def, path) < 0) goto cleanup; ret = 0; @@ -4647,7 +4644,7 @@ qemuProcessStartValidate(virQEMUDriverPtr driver, } VIR_DEBUG("Checking domain and device security labels"); - if (virSecurityManagerCheckAllLabel(driver->securityManager, vm->def) < 0) + if (qemuSecurityCheckAllLabel(driver->securityManager, vm->def) < 0) return -1; } @@ -5202,7 +5199,7 @@ qemuProcessPrepareDomain(virConnectPtr conn, /* If you are using a SecurityDriver with dynamic labelling, then generate a security label for isolation */ VIR_DEBUG("Generating domain security label (if required)"); - if (virSecurityManagerGenLabel(driver->securityManager, vm->def) < 0) { + if (qemuSecurityGenLabel(driver->securityManager, vm->def) < 0) { virDomainAuditSecurityLabel(vm, false); goto cleanup; } @@ -5513,8 +5510,8 @@ qemuProcessLaunch(virConnectPtr conn, virCommandSetUmask(cmd, 0x002); VIR_DEBUG("Setting up security labelling"); - if (virSecurityManagerSetChildProcessLabel(driver->securityManager, - vm->def, cmd) < 0) + if (qemuSecuritySetChildProcessLabel(driver->securityManager, + vm->def, cmd) < 0) goto cleanup; virCommandSetOutputFD(cmd, &logfile); @@ -5524,10 +5521,10 @@ qemuProcessLaunch(virConnectPtr conn, virCommandDaemonize(cmd); virCommandRequireHandshake(cmd); - if (virSecurityManagerPreFork(driver->securityManager) < 0) + if (qemuSecurityPreFork(driver->securityManager) < 0) goto cleanup; rv = virCommandRun(cmd, NULL); - virSecurityManagerPostFork(driver->securityManager); + qemuSecurityPostFork(driver->securityManager); /* wait for qemu process to show up */ if (rv == 0) { @@ -5604,8 +5601,8 @@ qemuProcessLaunch(virConnectPtr conn, goto cleanup; } if (S_ISFIFO(stdin_sb.st_mode) && - virSecurityManagerSetImageFDLabel(driver->securityManager, - vm->def, incoming->fd) < 0) + qemuSecuritySetImageFDLabel(driver->securityManager, + vm->def, incoming->fd) < 0) goto cleanup; } @@ -6122,7 +6119,7 @@ void qemuProcessStop(virQEMUDriverPtr driver, qemuSecurityRestoreAllLabel(driver, vm, !!(flags & VIR_QEMU_PROCESS_STOP_MIGRATED)); - virSecurityManagerReleaseLabel(driver->securityManager, vm->def); + qemuSecurityReleaseLabel(driver->securityManager, vm->def); for (i = 0; i < vm->def->ndisks; i++) { virDomainDeviceDef dev; @@ -6366,13 +6363,13 @@ int qemuProcessAttach(virConnectPtr conn ATTRIBUTE_UNUSED, vm->pid = pid; VIR_DEBUG("Detect security driver config"); - sec_managers = virSecurityManagerGetNested(driver->securityManager); + sec_managers = qemuSecurityGetNested(driver->securityManager); if (sec_managers == NULL) goto error; for (i = 0; sec_managers[i]; i++) { seclabelgen = false; - model = virSecurityManagerGetModel(sec_managers[i]); + model = qemuSecurityGetModel(sec_managers[i]); seclabeldef = virDomainDefGetSecurityLabelDef(vm->def, model); if (seclabeldef == NULL) { if (!(seclabeldef = virSecurityLabelDefNew(model))) @@ -6382,8 +6379,8 @@ int qemuProcessAttach(virConnectPtr conn ATTRIBUTE_UNUSED, seclabeldef->type = VIR_DOMAIN_SECLABEL_STATIC; if (VIR_ALLOC(seclabel) < 0) goto error; - if (virSecurityManagerGetProcessLabel(sec_managers[i], - vm->def, vm->pid, seclabel) < 0) + if (qemuSecurityGetProcessLabel(sec_managers[i], vm->def, + vm->pid, seclabel) < 0) goto error; if (VIR_STRDUP(seclabeldef->model, model) < 0) @@ -6400,9 +6397,9 @@ int qemuProcessAttach(virConnectPtr conn ATTRIBUTE_UNUSED, } } - if (virSecurityManagerCheckAllLabel(driver->securityManager, vm->def) < 0) + if (qemuSecurityCheckAllLabel(driver->securityManager, vm->def) < 0) goto error; - if (virSecurityManagerGenLabel(driver->securityManager, vm->def) < 0) + if (qemuSecurityGenLabel(driver->securityManager, vm->def) < 0) goto error; if (qemuDomainPerfRestart(vm) < 0) diff --git a/src/qemu/qemu_security.h b/src/qemu/qemu_security.h index 54638908d..d86db3f6b 100644 --- a/src/qemu/qemu_security.h +++ b/src/qemu/qemu_security.h @@ -28,6 +28,7 @@ # include "qemu_conf.h" # include "domain_conf.h" +# include "security/security_manager.h" int qemuSecuritySetAllLabel(virQEMUDriverPtr driver, virDomainObjPtr vm, @@ -60,4 +61,35 @@ int qemuSecuritySetHostdevLabel(virQEMUDriverPtr driver, int qemuSecurityRestoreHostdevLabel(virQEMUDriverPtr driver, virDomainObjPtr vm, virDomainHostdevDefPtr hostdev); + +/* Please note that for these APIs there is no wrapper yet. Do NOT blindly add + * new APIs here. If an API can touch a /dev file add a proper wrapper instead. + */ +# define qemuSecurityCheckAllLabel virSecurityManagerCheckAllLabel +# define qemuSecurityClearSocketLabel virSecurityManagerClearSocketLabel +# define qemuSecurityDomainSetPathLabel virSecurityManagerDomainSetPathLabel +# define qemuSecurityGenLabel virSecurityManagerGenLabel +# define qemuSecurityGetBaseLabel virSecurityManagerGetBaseLabel +# define qemuSecurityGetDOI virSecurityManagerGetDOI +# define qemuSecurityGetModel virSecurityManagerGetModel +# define qemuSecurityGetMountOptions virSecurityManagerGetMountOptions +# define qemuSecurityGetNested virSecurityManagerGetNested +# define qemuSecurityGetProcessLabel virSecurityManagerGetProcessLabel +# define qemuSecurityNew virSecurityManagerNew +# define qemuSecurityNewDAC virSecurityManagerNewDAC +# define qemuSecurityNewStack virSecurityManagerNewStack +# define qemuSecurityPostFork virSecurityManagerPostFork +# define qemuSecurityPreFork virSecurityManagerPreFork +# define qemuSecurityReleaseLabel virSecurityManagerReleaseLabel +# define qemuSecurityReserveLabel virSecurityManagerReserveLabel +# define qemuSecurityRestoreSavedStateLabel virSecurityManagerRestoreSavedStateLabel +# define qemuSecuritySetChildProcessLabel virSecurityManagerSetChildProcessLabel +# define qemuSecuritySetDaemonSocketLabel virSecurityManagerSetDaemonSocketLabel +# define qemuSecuritySetImageFDLabel virSecurityManagerSetImageFDLabel +# define qemuSecuritySetSavedStateLabel virSecurityManagerSetSavedStateLabel +# define qemuSecuritySetSocketLabel virSecurityManagerSetSocketLabel +# define qemuSecuritySetTapFDLabel virSecurityManagerSetTapFDLabel +# define qemuSecurityStackAddNested virSecurityManagerStackAddNested +# define qemuSecurityVerify virSecurityManagerVerify + #endif /* __QEMU_SECURITY_H__ */ -- 2.11.0

4 6

[libvirt] 'make check' fails (qemuhotplugTest) with libvirt fakeroot build, starting with 3.1.0 release
by Predrag Ivanovic 06 Mar '17

06 Mar '17

Hello. I am not sure if this is a bug, but fakerot build of libvirt-3.1.0 reliably fails at 'make check' stage, always with the same error (qemuhotplugTest). 3.0.0 builds fine, 3.2.0 (git master dec6d9df5feeb768381dd1b010d8030b44ea1a9a) fails at the same spot. If I build as root, libvirt-3.1.0 'make check' passes without errors, and the build is successful. Is this the expected behaviour, bug in the fakeroot(1.21), or? Qemu is 2.8.0 test-suite.log -- ========================================= libvirt 3.1.0: tests/test-suite.log ========================================= # TOTAL: 114 # PASS: 113 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 .. contents:: :depth: 2 FAIL: qemuhotplugtest ===================== TEST: qemuhotplugtest ........................................ 40 ...!!!...!...!.. 56 FAIL FAIL qemuhotplugtest (exit status: 1) Pedja

3 6

[libvirt] [PATCH 0/4] Split out interface object into its own module
by John Ferlan 06 Mar '17

06 Mar '17

Feedback from the initial pass at RFC for making a common pool object was that it got really confusing trying to manage the multitude of changes. See patch 3 of the series: http://www.redhat.com/archives/libvir-list/2017-February/msg00519.html So while I figure out what it'll take to make a better object framework, I figure I could start splitting out things a bit to make future work a bit easier to understand. Also rather than trying to do them all in one massive series, I'm working through shorter series contained to each of the driver conf's - this one is for interface and if follows the same pattern as node device. John Ferlan (4): conf: Introduce virinterfaceobj conf: Adjust coding style for interface conf sources conf: Use consistent function name prefixes for virinterfaceobj conf: Alter coding style of interface function prototypes po/POTFILES.in | 1 + src/Makefile.am | 3 +- src/conf/interface_conf.c | 239 ++++++++------------------------ src/conf/interface_conf.h | 60 ++------ src/conf/virinterfaceobj.c | 217 +++++++++++++++++++++++++++++ src/conf/virinterfaceobj.h | 84 +++++++++++ src/interface/interface_backend_netcf.c | 1 + src/interface/interface_backend_udev.c | 1 + src/libvirt_private.syms | 19 +-- src/test/test_driver.c | 26 ++-- 10 files changed, 396 insertions(+), 255 deletions(-) create mode 100644 src/conf/virinterfaceobj.c create mode 100644 src/conf/virinterfaceobj.h -- 2.9.3

2 5

[libvirt] [PATH v9 00/12] Support cache tune in libvirt
by Eli Qiao 06 Mar '17

06 Mar '17

Addressed comment from v9 -> v8 Marcelo: * New public API to query cache usage Eli: * Fix core dump while multiple tasks are added. Addressed comment from v8 -> v7 Martin: * Patch subject prefix. * Move some of cpu related information to virhostcpu.c. * Fix some memory leak in src/utils/resctrl.c Martin & Marcelo: * Don't remove directories which are not maintained by libvirt. Addressed comment from v7 -> v6 Marcelo: * Fix flock usage while VM initialization. Addressed comment from v6 -> v5 Marcelo: * Support other APPs to operate /sys/fs/resctrl at same time Libvirt will scan /sys/fs/resctrl again before doing cache allocation. patch 10 will address this. Addressed comment from v4 -> v5: Marcelo: * Several typos * Use flock instead of virFileLock Addressed comment from v3 -> v4: Daniel & Marcelo: * Added concurrence support Addressed comment from v2 -> v3: Daniel: * Fixed coding style, passed `make check` and `make syntax-check` * Variables renaming and move from header file to c file. * For locking/mutex support, no progress. There are some discussion from mailing list, but I can not find a better way to add locking support without performance impact. I'll explain the process and please help to advice what shoud we do. VM create: 1) Get the cache left value on each bank of the host. This should be shared amount all VMs. 2) Calculate the schemata on the bank based on all created resctrl domain's schemata 3) Calculate the default schemata by scaning all domain's schemata. 4) Flush default schemata to /sys/fs/resctrl/schemata VM destroy: 1) Remove the resctrl domain of that VM 2) Recalculate default schemata 3) Flush default schemata to /sys/fs/resctrl/schemata The key point is that all VMs shares /sys/fs/resctrl/schemata, and when a VM create a resctrl domain, the schemata of that VM depends on the default schemata and all other exsited schematas. So a global mutex is reqired. Before calculate a schemata or update default schemata, libvirt should gain this global mutex. I will try to think more about how to support this gracefully in next patch set. Marcelo: * Added vcpu support for cachetune, this will allow user to define which vcpu using which cache allocation bank. <cachetune id='0' host_id=0 size='3072' unit='KiB' vcpus='0,1'/> vcpus is a cpumap, the vcpu pids will be added to tasks file * Added cdp compatible, user can specify l3 cache even host enable cdp. See patch 8. On a cdp enabled host, specify l3code/l3data by <cachetune id='0' host_id='0' type='l3' size='3072' unit='KiB'/> This will create a schemata like: L3data:0=0xff00;... L3code:0=0xff00;... * Would you please help to test if the functions work. Martin: * Xml test case, I have no time to work on this yet, would you please show me an example, would like to amend it later. This series patches are for supportting CAT featues, which also called cache tune in libvirt. First to expose cache information which could be tuned in capabilites XML. Then add new domain xml element support to add cacahe bank which will apply on this libvirt domain. This series patches add a util file `resctrl.c/h`, an interface to talk with linux kernel's system fs. There are still one TODO left: 1. Expose a new public interface to set cachetune lively. Some discussion about this feature support can be found from: https://www.redhat.com/archives/libvir-list/2017-January/msg00644.html Eli Qiao (12): Resctrl: Add some utils functions Resctrl: expose cache information to capabilities Resctrl: Add new xml element to support cache tune Resctrl: Add private interfaces to operate cache bank Qemu: Set cache tune while booting a new domain. Resctrl: enable l3code/l3data Resctrl: Make sure l3data/l3code are pairs Resctrl: Compatible mode for cdp enabled Resctrl: concurrence support Resctrl: Scan resctrl before doing cache allocation Resctrl: Add Public API for nodecachestats Resctrl: Add nodecachestats daemon/remote.c | 67 +++ docs/schemas/domaincommon.rng | 46 ++ include/libvirt/libvirt-host.h | 32 ++ include/libvirt/virterror.h | 1 + po/POTFILES.in | 1 + src/Makefile.am | 1 + src/conf/capabilities.c | 56 +++ src/conf/capabilities.h | 23 + src/conf/domain_conf.c | 182 +++++++ src/conf/domain_conf.h | 19 + src/driver-hypervisor.h | 7 + src/libvirt-host.c | 41 ++ src/libvirt_private.syms | 12 + src/libvirt_public.syms | 1 + src/nodeinfo.c | 64 +++ src/nodeinfo.h | 1 + src/qemu/qemu_capabilities.c | 8 + src/qemu/qemu_driver.c | 18 + src/qemu/qemu_process.c | 54 ++ src/remote/remote_driver.c | 52 ++ src/remote/remote_protocol.x | 25 +- src/remote_protocol-structs | 16 + src/util/virerror.c | 1 + src/util/virhostcpu.c | 186 ++++++- src/util/virhostcpu.h | 6 + src/util/virresctrl.c | 1082 ++++++++++++++++++++++++++++++++++++++++ src/util/virresctrl.h | 96 ++++ tools/virsh-host.c | 49 ++ 28 files changed, 2129 insertions(+), 18 deletions(-) create mode 100644 src/util/virresctrl.c create mode 100644 src/util/virresctrl.h -- 1.9.1

3 16

[libvirt] [PATCH 0/3] Clean
by John Ferlan 06 Mar '17

06 Mar '17

More "fallout" from the RFC series to make a common pool object: http://www.redhat.com/archives/libvir-list/2017-February/msg00519.html were changes made along the way in the test_driver code to "commonalize" the various vir*ObjFindBy{Name|UUID} callers. This series will do the same for the existing code. Not even attempted was making test_driver to have updated coding style (maybe another day for that)! John Ferlan (3): test: Make common test*ObjFindByName helpers test: Make a common testNetworkObjFindByName test: Make common test*ObjFindByUUID helpers src/test/test_driver.c | 499 ++++++++++++++++--------------------------------- 1 file changed, 166 insertions(+), 333 deletions(-) -- 2.9.3

2 5

[libvirt] [RFC PATCH 00/12] Support multiple PHBs on pSeries guests
by Andrea Bolognani 06 Mar '17

06 Mar '17

Note: if you want to try this out, you'll need to make sure your QEMU binary includes this commit[1]; moreover, that commit is missing a way for libvirt to detect whether the new naming scheme is in place, so this will have to remain an RFC until the QEMU side has been sorted out. Patches 1-3 are just setting up the stage. Patch 4 starts actually introducing the feature, by relaxing some checks that can no longer be as strict. Patches 5-10 puts all the boring bits (XML parsing and formatting, QEMU capabilities) in place. Patch 11 enables the feature at last. Patch 12 introduces a single test, a bunch more will be added before posting this for real (not as RFC). [1] https://github.com/dgibson/qemu/commit/0a6a9ba2adc48a9a5ea7406d1a5fb3c36f00… Andrea Bolognani (12): qemu: Allow qemuBuildControllerDevStr() to return NULL qemu: Tweak index number checking conf: Move index number checking to drivers qemu: Relax pci-root index requirement for pSeries guests schema: Allow <target index='...'/> schema: Add 'spapr-pci-host-bridge' controller model conf: Parse and format <target index='...'/> conf: Add 'spapr-pci-host-bridge' controller model qemu: Automatically pick index and model for pci-root controllers qemu: Introduce QEMU_CAPS_DEVICE_SPAPR_PCI_HOST_BRIDGE qemu: Use multiple PHBs for pSeries guests tests: Add tests for pSeries guests with multiple PHBs docs/schemas/domaincommon.rng | 7 ++ src/bhyve/bhyve_domain.c | 15 +++ src/conf/domain_conf.c | 34 ++++-- src/conf/domain_conf.h | 2 + src/libxl/libxl_domain.c | 14 +++ src/lxc/lxc_domain.c | 14 +++ src/openvz/openvz_driver.c | 14 +++ src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 126 +++++++++++++++++---- src/qemu/qemu_command.h | 9 +- src/qemu/qemu_domain.c | 13 +++ src/qemu/qemu_domain_address.c | 47 +++++++- src/qemu/qemu_hotplug.c | 5 +- src/uml/uml_driver.c | 14 +++ src/vz/vz_driver.c | 14 +++ src/xen/xen_driver.c | 14 +++ .../qemuargv2xmldata/qemuargv2xml-pseries-disk.xml | 5 +- .../qemuargv2xml-pseries-nvram.xml | 5 +- tests/qemucapabilitiesdata/caps_2.6.0.ppc64le.xml | 1 + .../qemuxml2argv-pseries-phb-simple.args | 26 +++++ .../qemuxml2argv-pseries-phb-simple.xml | 20 ++++ tests/qemuxml2argvtest.c | 5 + .../qemuxml2xmlout-panic-pseries.xml | 5 +- .../qemuxml2xmlout-ppc64-usb-controller-legacy.xml | 5 +- .../qemuxml2xmlout-ppc64-usb-controller.xml | 5 +- .../qemuxml2xmlout-pseries-nvram.xml | 5 +- .../qemuxml2xmlout-pseries-panic-missing.xml | 5 +- .../qemuxml2xmlout-pseries-panic-no-address.xml | 5 +- ...g.xml => qemuxml2xmlout-pseries-phb-simple.xml} | 13 ++- tests/qemuxml2xmltest.c | 4 + 31 files changed, 407 insertions(+), 47 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pseries-phb-simple.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pseries-phb-simple.xml copy tests/qemuxml2xmloutdata/{qemuxml2xmlout-pseries-panic-missing.xml => qemuxml2xmlout-pseries-phb-simple.xml} (69%) -- 2.7.4

3 17

[libvirt] [PATCH] news: Fix typo
by Jiri Denemark 06 Mar '17

06 Mar '17

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- Pushed as trivial. docs/news.xml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/news.xml b/docs/news.xml index 584c9afac..97393b3aa 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -21,7 +21,7 @@ </summary> <description> Previously, libvirt detected the host CPU model using CPUID - instruction, which cased libvirt to detect a lot of CPU features + instruction, which caused libvirt to detect a lot of CPU features that are not supported by QEMU/KVM. Asking QEMU makes sure we don't start it with unsupported features. </description> -- 2.12.0

1 0

[libvirt] [PATCH] qemu: Drop virQEMUCapsFreeStringList
by Jiri Denemark 06 Mar '17

06 Mar '17

The implementation matches virStringListFreeCount. The only difference between the two functions is the ordering of their parameters. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/qemu/qemu_capabilities.c | 31 +++++++++---------------------- 1 file changed, 9 insertions(+), 22 deletions(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 5694d0ba4..359a0d87a 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -1881,19 +1881,6 @@ virQEMUCapsProcessProps(virQEMUCapsPtr qemuCaps, } -static void -virQEMUCapsFreeStringList(size_t len, - char **values) -{ - size_t i; - if (!values) - return; - for (i = 0; i < len; i++) - VIR_FREE(values[i]); - VIR_FREE(values); -} - - #define OBJECT_TYPE_PREFIX "name \"" static int @@ -1928,7 +1915,7 @@ virQEMUCapsParseDeviceStrObjectTypes(const char *str, cleanup: if (ret < 0) - virQEMUCapsFreeStringList(ntypelist, typelist); + virStringListFreeCount(typelist, ntypelist); return ret; } @@ -1981,7 +1968,7 @@ virQEMUCapsParseDeviceStrObjectProps(const char *str, cleanup: if (ret < 0) - virQEMUCapsFreeStringList(nproplist, proplist); + virStringListFreeCount(proplist, nproplist); return ret; } @@ -1999,7 +1986,7 @@ virQEMUCapsParseDeviceStr(virQEMUCapsPtr qemuCaps, const char *str) ARRAY_CARDINALITY(virQEMUCapsObjectTypes), virQEMUCapsObjectTypes, nvalues, values); - virQEMUCapsFreeStringList(nvalues, values); + virStringListFreeCount(values, nvalues); for (i = 0; i < ARRAY_CARDINALITY(virQEMUCapsObjectProps); i++) { const char *type = virQEMUCapsObjectProps[i].type; @@ -2011,7 +1998,7 @@ virQEMUCapsParseDeviceStr(virQEMUCapsPtr qemuCaps, const char *str) virQEMUCapsObjectProps[i].nprops, virQEMUCapsObjectProps[i].props, nvalues, values); - virQEMUCapsFreeStringList(nvalues, values); + virStringListFreeCount(values, nvalues); } /* Prefer -chardev spicevmc (detected earlier) over -device spicevmc */ @@ -2665,7 +2652,7 @@ virQEMUCapsProbeQMPCommands(virQEMUCapsPtr qemuCaps, ARRAY_CARDINALITY(virQEMUCapsCommands), virQEMUCapsCommands, ncommands, commands); - virQEMUCapsFreeStringList(ncommands, commands); + virStringListFreeCount(commands, ncommands); /* QMP add-fd was introduced in 1.2, but did not support * management control of set numbering, and did not have a @@ -2707,7 +2694,7 @@ virQEMUCapsProbeQMPEvents(virQEMUCapsPtr qemuCaps, ARRAY_CARDINALITY(virQEMUCapsEvents), virQEMUCapsEvents, nevents, events); - virQEMUCapsFreeStringList(nevents, events); + virStringListFreeCount(events, nevents); return 0; } @@ -2727,7 +2714,7 @@ virQEMUCapsProbeQMPObjects(virQEMUCapsPtr qemuCaps, ARRAY_CARDINALITY(virQEMUCapsObjectTypes), virQEMUCapsObjectTypes, nvalues, values); - virQEMUCapsFreeStringList(nvalues, values); + virStringListFreeCount(values, nvalues); for (i = 0; i < ARRAY_CARDINALITY(virQEMUCapsObjectProps); i++) { const char *type = virQEMUCapsObjectProps[i].type; @@ -2743,7 +2730,7 @@ virQEMUCapsProbeQMPObjects(virQEMUCapsPtr qemuCaps, ARRAY_CARDINALITY(virQEMUCapsPropObjects), virQEMUCapsPropObjects, type, nvalues, values); - virQEMUCapsFreeStringList(nvalues, values); + virStringListFreeCount(values, nvalues); } /* Prefer -chardev spicevmc (detected earlier) over -device spicevmc */ @@ -3043,7 +3030,7 @@ virQEMUCapsProbeQMPMigrationCapabilities(virQEMUCapsPtr qemuCaps, ARRAY_CARDINALITY(virQEMUCapsMigration), virQEMUCapsMigration, ncaps, caps); - virQEMUCapsFreeStringList(ncaps, caps); + virStringListFreeCount(caps, ncaps); return 0; } -- 2.12.0

2 1

[libvirt] [PATCH] node: Replace variable named 'system' with 'syscap'
by John Ferlan 04 Mar '17

04 Mar '17

Changes in commit id 'dec6d9df' caused a compilation failure on a RHEL6 CI build environment. So just replace 'system' with 'syscap' as a name. cc1: warnings being treated as errors ../../src/conf/node_device_conf.c: In function 'virNodeDevCapSystemParseXML': ../../src/conf/node_device_conf.c:1415: error: declaration of 'system' shadows a global declaration [-Wshadow] Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- Pushed as a build breaker src/conf/node_device_conf.c | 8 ++++---- src/node_device/node_device_udev.c | 8 ++++---- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c index d60f69d..3565aec 100644 --- a/src/conf/node_device_conf.c +++ b/src/conf/node_device_conf.c @@ -1412,10 +1412,10 @@ static int virNodeDevCapSystemParseXML(xmlXPathContextPtr ctxt, virNodeDeviceDefPtr def, xmlNodePtr node, - virNodeDevCapSystemPtr system) + virNodeDevCapSystemPtr syscap) { - virNodeDevCapSystemHardwarePtr hardware = &system->hardware; - virNodeDevCapSystemFirmwarePtr firmware = &system->firmware; + virNodeDevCapSystemHardwarePtr hardware = &syscap->hardware; + virNodeDevCapSystemFirmwarePtr firmware = &syscap->firmware; xmlNodePtr orignode; int ret = -1; char *tmp; @@ -1423,7 +1423,7 @@ virNodeDevCapSystemParseXML(xmlXPathContextPtr ctxt, orignode = ctxt->node; ctxt->node = node; - system->product_name = virXPathString("string(./product[1])", ctxt); + syscap->product_name = virXPathString("string(./product[1])", ctxt); hardware->vendor_name = virXPathString("string(./hardware/vendor[1])", ctxt); hardware->version = virXPathString("string(./hardware/version[1])", ctxt); diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c index af44604..bcae444 100644 --- a/src/node_device/node_device_udev.c +++ b/src/node_device/node_device_udev.c @@ -1420,12 +1420,12 @@ static void udevEventHandleCallback(int watch ATTRIBUTE_UNUSED, /* DMI is intel-compatible specific */ #if defined(__x86_64__) || defined(__i386__) || defined(__amd64__) static void -udevGetDMIData(virNodeDevCapSystemPtr system) +udevGetDMIData(virNodeDevCapSystemPtr syscap) { struct udev *udev = NULL; struct udev_device *device = NULL; - virNodeDevCapSystemHardwarePtr hardware = &system->hardware; - virNodeDevCapSystemFirmwarePtr firmware = &system->firmware; + virNodeDevCapSystemHardwarePtr hardware = &syscap->hardware; + virNodeDevCapSystemFirmwarePtr firmware = &syscap->firmware; udev = udev_monitor_get_udev(DRV_STATE_UDEV_MONITOR(driver)); @@ -1441,7 +1441,7 @@ udevGetDMIData(virNodeDevCapSystemPtr system) } if (udevGetStringSysfsAttr(device, "product_name", - &system->product_name) < 0) + &syscap->product_name) < 0) goto cleanup; if (udevGetStringSysfsAttr(device, "sys_vendor", &hardware->vendor_name) < 0) -- 2.9.3

1 0

[libvirt] [PATCH v2] Document preferred naming conventions
by Daniel P. Berrange 03 Mar '17

03 Mar '17

This documents the preferred conventions for naming files, structs, enums, typedefs and functions. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- Changed in v2: - Fix typo - Add macro docs - Clarify that $VERB is always last in function/macro names HACKING | 81 ++++++++++++++++++++++++++++++++++++++++++++ docs/hacking.html.in | 94 ++++++++++++++++++++++++++++++++++++++++++++++++++++ docs/hacking2.xsl | 4 +++ 3 files changed, 179 insertions(+) diff --git a/HACKING b/HACKING index fff003b..5077304 100644 --- a/HACKING +++ b/HACKING @@ -239,6 +239,87 @@ on the subject, on Richard Jones' guide to working with open source projects <http://people.redhat.com/rjones/how-to-supply-code-to-open-source-projects/>. +Naming conventions +================== +When reading libvirt code, a number of different naming conventions will be +evident due to various changes in thinking over the course of the project's +lifetime. The conventions documented below should be followed when creating +any entirely new files in libvirt. When working on existing files, while it is +desirable to apply these conventions, keeping a consistent style with existing +code in that particular file is generally more important. The overall guiding +rule is that every file, enum, struct, function, and typedef name must have a +'vir' or 'VIR' prefix. All local scope variable names are exempt, and global +variables are exempt, unless exported in a header file. + +*File names* + +File naming varies depending on the subdirectory. The preferred style is to +have a 'vir' prefix, followed by a name which matches the name of the +functions / objects inside the file. For example, a file containing an object +'virHashtable' is stored in files 'virhashtable.c' and 'virhashtable.h'. +Sometimes, methods which would otherwise be declared 'static' need to be +exported for use by a test suite. For this purpose a second header file should +be added with a suffix of 'priv'. e.g. 'virhashtablepriv.h'. Use of +underscores in file names is discouraged when using the 'vir' prefix style. +The 'vir' prefix naming applies to src/util, src/rpc and tests/ directories. +Most other directories do not follow this convention. + + + +*Enum type & field names* + +All enums should have a 'vir' prefix in their typedef name, and each following +word should have its first letter in uppercase. The enum name should match the +typedef name with a leading underscore. The enum member names should be in all +uppercase, and use an underscore to separate each word. The enum member name +prefix should match the enum typedef name. + + typedef enum _virSocketType virSocketType; + enum _virSocketType { + VIR_SOCKET_TYPE_IPV4, + VIR_SOCKET_TYPE_IPV6, + }; + + +*Struct type names* + +All structs should have a 'vir' prefix in their typedef name, and each +following word should have its first letter in uppercase. The struct name +should be the same as the typedef name with a leading underscore. A second +typedef should be given for a pointer to the struct with a 'Ptr' suffix. + + typedef struct _virHashTable virHashTable; + typedef virHashTable *virHashTablePtr; + struct _virHashTable { + ... + }; + + +*Function names* + +All functions should have a 'vir' prefix in their name, followed by one or +more words with first letter of each word capitalized. Underscores should not +be used in function names. If the function is operating on an object, then the +function name prefix should match the object typedef name. When naming +functions the verb should always be the last part of the name. For example, +given an object 'virHashTable', all functions should have a name +'virHashTable$VERB' e.g. 'virHashTableLookup'. If there is no object +associated with the function, then its name prefix should match the filename. +For example, given a filename of 'virfile.h', all functions should have a name +'virFile$VERB' e.g. 'virFileTouch'. + + + +*Macro names* + +All macros should have a "VIR" prefix in their name, followed by one or more +uppercase words separated by underscores. The macro argument names should be +in lowercase. As with functions, the verb should always be the last part of +the name. For example "VIR_FILE_CLOSE". + + + + Code indentation ================ Libvirt's C source code generally adheres to some basic code-formatting diff --git a/docs/hacking.html.in b/docs/hacking.html.in index b1bb149..1001f49 100644 --- a/docs/hacking.html.in +++ b/docs/hacking.html.in @@ -314,6 +314,100 @@ Richard Jones' guide to working with open source projects</a>. </p> + <h2><a name="naming">Naming conventions</a></h2> + + <p> + When reading libvirt code, a number of different naming conventions will + be evident due to various changes in thinking over the course of the + project's lifetime. The conventions documented below should be followed + when creating any entirely new files in libvirt. When working on existing + files, while it is desirable to apply these conventions, keeping a + consistent style with existing code in that particular file is generally + more important. The overall guiding rule is that every file, enum, struct, + function, and typedef name must have a 'vir' or 'VIR' prefix. All local + scope variable names are exempt, and global variables are exempt, unless + exported in a header file. + </p> + + <dl> + <dt>File names</dt> + <dd> + <p> + File naming varies depending on the subdirectory. The preferred + style is to have a 'vir' prefix, followed by a name which matches + the name of the functions / objects inside the file. For example, + a file containing an object 'virHashtable' is stored in files + 'virhashtable.c' and 'virhashtable.h'. Sometimes, methods which + would otherwise be declared 'static' need to be exported for use + by a test suite. For this purpose a second header file should be + added with a suffix of 'priv'. e.g. 'virhashtablepriv.h'. Use of + underscores in file names is discouraged when using the 'vir' + prefix style. The 'vir' prefix naming applies to src/util, + src/rpc and tests/ directories. Most other directories do not + follow this convention. + </p> + </dd> + <dt>Enum type & field names</dt> + <dd> + <p> + All enums should have a 'vir' prefix in their typedef name, + and each following word should have its first letter in + uppercase. The enum name should match the typedef name with + a leading underscore. The enum member names should be in all + uppercase, and use an underscore to separate each word. The + enum member name prefix should match the enum typedef name. + </p> + <pre> + typedef enum _virSocketType virSocketType; + enum _virSocketType { + VIR_SOCKET_TYPE_IPV4, + VIR_SOCKET_TYPE_IPV6, + };</pre> + </dd> + <dt>Struct type names</dt> + <dd> + <p> + All structs should have a 'vir' prefix in their typedef name, + and each following word should have its first letter in + uppercase. The struct name should be the same as the typedef + name with a leading underscore. A second typedef should be + given for a pointer to the struct with a 'Ptr' suffix. + </p> + <pre> + typedef struct _virHashTable virHashTable; + typedef virHashTable *virHashTablePtr; + struct _virHashTable { + ... + };</pre> + </dd> + <dt>Function names</dt> + <dd> + <p> + All functions should have a 'vir' prefix in their name, + followed by one or more words with first letter of each + word capitalized. Underscores should not be used in function + names. If the function is operating on an object, then the + function name prefix should match the object typedef name. + When naming functions the verb should always be the last + part of the name. For example, given an object 'virHashTable', + all functions should have a name 'virHashTable$VERB' + e.g. 'virHashTableLookup'. If there is no object associated + with the function, then its name prefix should match the + filename. For example, given a filename of 'virfile.h', all + functions should have a name 'virFile$VERB' e.g. 'virFileTouch'. + </p> + </dd> + <dt>Macro names</dt> + <dd> + <p> + All macros should have a "VIR" prefix in their name, followed + by one or more uppercase words separated by underscores. The + macro argument names should be in lowercase. As with functions, + the verb should always be the last part of the name. For example + "VIR_FILE_CLOSE". + </p> + </dd> + </dl> <h2><a name="indent">Code indentation</a></h2> <p> diff --git a/docs/hacking2.xsl b/docs/hacking2.xsl index 3595b7e..7e5ac82 100644 --- a/docs/hacking2.xsl +++ b/docs/hacking2.xsl @@ -114,6 +114,10 @@ from docs/hacking.html.in! <xsl:template match="html:li/html:ul/html:li">-- <xsl:apply-templates/><xsl:value-of select="$newline"/><xsl:value-of select="$newline"/> </xsl:template> +<xsl:template match="html:dl/html:dt">*<xsl:apply-templates/>*<xsl:value-of select="$newline"/><xsl:value-of select="$newline"/> +</xsl:template> +<xsl:template match="html:dl/html:dd"><xsl:apply-templates/><xsl:value-of select="$newline"/><xsl:value-of select="$newline"/> +</xsl:template>  -- 2.9.3

3 4

[libvirt] [PATCH RESEND] qdev: Make "hotplugged" property read-only
by Eduardo Habkost 03 Mar '17

03 Mar '17

The "hotplugged" property is user visible, but it was never meant to be set by the user. There are probably multiple ways to break or crash device code by overriding the property. For example, we recently fixed a crash in rtc_set_memory() related to the property (commit 26ef65beab852caf2b1ef4976e3473f2d525164d). There has been some discussion about making management software use "hotplugged=on" on migration, to indicate devices that were hotplugged in the migration source. There were other suggestions to address this, like including the "hotplugged" field in the migration stream instead of requiring it to be set propertly. Whatever solution we choose in the future, this patch disables setting "hotplugged" explicitly in the command-line by now, because the ability to set the property is unused, untested, and undocumented. Signed-off-by: Eduardo Habkost <ehabkost(a)redhat.com> --- hw/core/qdev.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/hw/core/qdev.c b/hw/core/qdev.c index 06ba02e2a3..800c9ca23f 100644 --- a/hw/core/qdev.c +++ b/hw/core/qdev.c @@ -1016,13 +1016,6 @@ static bool device_get_hotplugged(Object *obj, Error **err) return dev->hotplugged; } -static void device_set_hotplugged(Object *obj, bool value, Error **err) -{ - DeviceState *dev = DEVICE(obj); - - dev->hotplugged = value; -} - static void device_initfn(Object *obj) { DeviceState *dev = DEVICE(obj); @@ -1042,7 +1035,7 @@ static void device_initfn(Object *obj) object_property_add_bool(obj, "hotpluggable", device_get_hotpluggable, NULL, NULL); object_property_add_bool(obj, "hotplugged", - device_get_hotplugged, device_set_hotplugged, + device_get_hotplugged, NULL, &error_abort); class = object_get_class(OBJECT(dev)); -- 2.11.0.259.g40922b1

3 4

[libvirt] [PATCH] news: Move host CPU model improvement to 3.2.0
by Jiri Denemark 03 Mar '17

03 Mar '17

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- Pushed as trivial. docs/news.xml | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) diff --git a/docs/news.xml b/docs/news.xml index e4b5a9d31..584c9afac 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -15,7 +15,17 @@ <change/> </section> <section title="Improvements"> - <change/> + <change> + <summary> + qemu: Detect host CPU model by asking QEMU on x86_64 + </summary> + <description> + Previously, libvirt detected the host CPU model using CPUID + instruction, which cased libvirt to detect a lot of CPU features + that are not supported by QEMU/KVM. Asking QEMU makes sure we + don't start it with unsupported features. + </description> + </change> </section> <section title="Bug fixes"> <change/> @@ -141,17 +151,6 @@ driver with all backends which may pull in too many dependencies. </description> </change> - <change> - <summary> - qemu: Detect host CPU model by asking QEMU on x86_64 - </summary> - <description> - Previously, libvirt detected the host CPU model using CPUID - instruction, which cased libvirt to detect a lot of CPU features - that are not supported by QEMU/KVM. Asking QEMU makes sure we - don't start it with unsupported features. - </description> - </change> </section> <section title="Bug fixes"> <change> -- 2.12.0

1 0

[libvirt] [PATCH 0/2] cpu_x86: Disable TSX on broken models
by Jiri Denemark 03 Mar '17

03 Mar '17

This series applies on top of "qemu: Detect host CPU model by asking QEMU on x86_64". https://bugzilla.redhat.com/show_bug.cgi?id=1406791 Jiri Denemark (2): cpu_x86: Disable TSX on broken models cputest: Add CPUID data for Haswell with TSX src/cpu/cpu_x86.c | 72 ++++++++++++++++++++-- tests/cputest.c | 1 + .../x86_64-cpuid-Core-i5-4670T-guest.xml | 2 +- .../x86_64-cpuid-Core-i5-4670T-host.xml | 2 +- .../x86_64-cpuid-Core-i5-4670T-json.xml | 2 +- .../x86_64-cpuid-Xeon-E7-8890-guest.xml | 32 ++++++++++ .../cputestdata/x86_64-cpuid-Xeon-E7-8890-host.xml | 32 ++++++++++ tests/cputestdata/x86_64-cpuid-Xeon-E7-8890.xml | 37 +++++++++++ 8 files changed, 172 insertions(+), 8 deletions(-) create mode 100644 tests/cputestdata/x86_64-cpuid-Xeon-E7-8890-guest.xml create mode 100644 tests/cputestdata/x86_64-cpuid-Xeon-E7-8890-host.xml create mode 100644 tests/cputestdata/x86_64-cpuid-Xeon-E7-8890.xml -- 2.11.1

2 4

[libvirt] [PATCH v3 00/28] qemu: Detect host CPU model by asking QEMU on x86_64
by Jiri Denemark 03 Mar '17

03 Mar '17

Until now host-model CPU mode tried to enable all CPU features supported by the host CPU even if QEMU/KVM did not support them. This caused a number of issues and made host-model quite unreliable. Asking QEMU for the CPU it can provide and the current host makes host-model much more robust. This series fixes the following bugs: https://bugzilla.redhat.com/show_bug.cgi?id=1018251 https://bugzilla.redhat.com/show_bug.cgi?id=1371617 https://bugzilla.redhat.com/show_bug.cgi?id=1372581 https://bugzilla.redhat.com/show_bug.cgi?id=1404627 https://bugzilla.redhat.com/show_bug.cgi?id=870071 In addition to that, the following bug should be mostly limited to cases when an unsupported feature is explicitly requested: https://bugzilla.redhat.com/show_bug.cgi?id=1335534 The series relies on features which are not in QEMU yet, but should be hopefully close enough to be pushed in 2.9.0. In the meantime, Eduardo's work/x86-query-cpu-expansion-full branch can be used to play with them. Version 3: - a few patches which were ACKed in v2 and didn't have any dependencies were pushed to make the series a bit smaller - review comments addressed (see individual patches for more details) Version 2: - properly set vendor property in converted test data files - fix cpu-parse.sh to use "x86_64" prefix for the generated files Jiri Denemark (28): qemucapstest: Update test data for QEMU 2.9.0 domaincapstest: Add test data for QEMU 2.9.0 qemu: Refactor virQEMUCapsInitHostCPUModel qemu: Fix CPU model fallback in domain capabilities docs: Update description of the host-model CPU mode qemu: Rename hostCPU/feature element in capabilities cache qemu: Prepare for more types in qemuMonitorCPUModelInfo qemu: Store more types in qemuMonitorCPUModelInfo qemu: Probe "max" CPU model in TCG cpu_x86: Drop virCPUx86MakeData and use virCPUDataNew cpu_x86: Make virCPUx86DataClear static cpu: Rework cpuDataFree cpu_x86: Make virCPUx86DataAddCPUID work with virCPUDataPtr cpu_x86: Introduce virCPUx86DataSetSignature cpu_x86: Introduce virCPUx86DataSetVendor cpu_x86: Introduce virCPUx86DataAddFeature qemu: Get host CPU model from QEMU on x86_64 qemu: Use enum for CPU model expansion type qemu: Use full CPU model expansion on x86 qemu: Make virQEMUCapsInitCPUModel testable cputest: Rename x86 data files cputest: Use virArch enum rather then strings cputest: Switch host CPU data scripts to model expansion cputest: Convert all json data files to query-cpu-model-expansion cputest: Test virQEMUCapsInitCPUModel cputest: Drop obsolete CPU test data files cputest: Drop .new suffix from CPU test data files news: Detect host CPU model by asking QEMU on x86_64 docs/formatdomain.html.in | 37 +- docs/news.xml | 11 + src/bhyve/bhyve_capabilities.c | 2 +- src/cpu/cpu.c | 21 +- src/cpu/cpu.h | 4 +- src/cpu/cpu_arm.c | 7 - src/cpu/cpu_ppc64.c | 6 +- src/cpu/cpu_s390.c | 7 - src/cpu/cpu_x86.c | 280 ++++--- src/cpu/cpu_x86.h | 13 +- src/libvirt_private.syms | 7 +- src/libxl/libxl_capabilities.c | 18 +- src/qemu/qemu_capabilities.c | 463 ++++++++--- src/qemu/qemu_capabilities.h | 3 +- src/qemu/qemu_capspriv.h | 13 +- src/qemu/qemu_command.c | 2 +- src/qemu/qemu_monitor.c | 29 +- src/qemu/qemu_monitor.h | 35 +- src/qemu/qemu_monitor_json.c | 107 ++- src/qemu/qemu_monitor_json.h | 4 +- src/qemu/qemu_parse_command.c | 2 +- src/qemu/qemu_process.c | 7 +- src/vmware/vmware_conf.c | 2 +- src/vz/vz_driver.c | 2 +- tests/cputest.c | 324 ++++---- tests/cputestdata/cpu-convert.py | 249 ++++++ tests/cputestdata/cpu-gather.sh | 39 +- tests/cputestdata/cpu-parse.sh | 5 +- tests/cputestdata/x86-cpuid-A10-5800K.json | 77 -- tests/cputestdata/x86-cpuid-Core-i5-2500.json | 88 --- tests/cputestdata/x86-cpuid-Core-i5-2540M.json | 82 -- tests/cputestdata/x86-cpuid-Core-i5-4670T.json | 77 -- tests/cputestdata/x86-cpuid-Core-i5-6600.json | 82 -- tests/cputestdata/x86-cpuid-Core-i7-2600.json | 77 -- tests/cputestdata/x86-cpuid-Core-i7-3740QM.json | 77 -- tests/cputestdata/x86-cpuid-Core-i7-3770.json | 77 -- tests/cputestdata/x86-cpuid-Core-i7-4600U.json | 82 -- tests/cputestdata/x86-cpuid-Core-i7-5600U-json.xml | 12 - tests/cputestdata/x86-cpuid-Core-i7-5600U.json | 88 --- tests/cputestdata/x86-cpuid-Core2-E6850.json | 77 -- tests/cputestdata/x86-cpuid-Opteron-2350.json | 71 -- tests/cputestdata/x86-cpuid-Opteron-6234.json | 88 --- tests/cputestdata/x86-cpuid-Phenom-B95.json | 77 -- tests/cputestdata/x86-cpuid-Xeon-E3-1245.json | 88 --- tests/cputestdata/x86-cpuid-Xeon-E5-2630.json | 77 -- tests/cputestdata/x86-cpuid-Xeon-E5-2650.json | 71 -- tests/cputestdata/x86-cpuid-Xeon-E7-4820.json | 77 -- tests/cputestdata/x86-cpuid-Xeon-W3520.json | 77 -- ...ack.xml => x86_64-Haswell-noTSX-nofallback.xml} | 0 ...-Haswell-noTSX.xml => x86_64-Haswell-noTSX.xml} | 0 .../{x86-Haswell.xml => x86_64-Haswell.xml} | 0 ...e-1-result.xml => x86_64-baseline-1-result.xml} | 0 .../{x86-baseline-1.xml => x86_64-baseline-1.xml} | 0 ...e-2-result.xml => x86_64-baseline-2-result.xml} | 0 .../{x86-baseline-2.xml => x86_64-baseline-2.xml} | 0 ...expanded.xml => x86_64-baseline-3-expanded.xml} | 0 ...e-3-result.xml => x86_64-baseline-3-result.xml} | 0 .../{x86-baseline-3.xml => x86_64-baseline-3.xml} | 0 ...expanded.xml => x86_64-baseline-4-expanded.xml} | 0 ...e-4-result.xml => x86_64-baseline-4-result.xml} | 0 .../{x86-baseline-4.xml => x86_64-baseline-4.xml} | 0 ...expanded.xml => x86_64-baseline-5-expanded.xml} | 0 ...e-5-result.xml => x86_64-baseline-5-result.xml} | 0 .../{x86-baseline-5.xml => x86_64-baseline-5.xml} | 0 ...atable.xml => x86_64-baseline-6-migratable.xml} | 0 ...e-6-result.xml => x86_64-baseline-6-result.xml} | 0 .../{x86-baseline-6.xml => x86_64-baseline-6.xml} | 0 ...e-7-result.xml => x86_64-baseline-7-result.xml} | 0 .../{x86-baseline-7.xml => x86_64-baseline-7.xml} | 0 ...e-8-result.xml => x86_64-baseline-8-result.xml} | 0 .../{x86-baseline-8.xml => x86_64-baseline-8.xml} | 0 ...ml => x86_64-baseline-incompatible-vendors.xml} | 0 ...lt.xml => x86_64-baseline-no-vendor-result.xml} | 0 ...no-vendor.xml => x86_64-baseline-no-vendor.xml} | 0 ...xml => x86_64-baseline-some-vendors-result.xml} | 0 ...endors.xml => x86_64-baseline-some-vendors.xml} | 0 ...-bogus-feature.xml => x86_64-bogus-feature.xml} | 0 ...{x86-bogus-model.xml => x86_64-bogus-model.xml} | 0 ...86-bogus-vendor.xml => x86_64-bogus-vendor.xml} | 0 ...-guest.xml => x86_64-cpuid-A10-5800K-guest.xml} | 0 ...0K-host.xml => x86_64-cpuid-A10-5800K-host.xml} | 0 ...0K-json.xml => x86_64-cpuid-A10-5800K-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-A10-5800K.json | 203 +++++ ...id-A10-5800K.xml => x86_64-cpuid-A10-5800K.xml} | 0 ...-guest.xml => x86_64-cpuid-Atom-D510-guest.xml} | 0 ...10-host.xml => x86_64-cpuid-Atom-D510-host.xml} | 0 ...id-Atom-D510.xml => x86_64-cpuid-Atom-D510.xml} | 0 ...-guest.xml => x86_64-cpuid-Atom-N450-guest.xml} | 0 ...50-host.xml => x86_64-cpuid-Atom-N450-host.xml} | 0 ...id-Atom-N450.xml => x86_64-cpuid-Atom-N450.xml} | 0 ...est.xml => x86_64-cpuid-Core-i5-2500-guest.xml} | 0 ...host.xml => x86_64-cpuid-Core-i5-2500-host.xml} | 0 ...json.xml => x86_64-cpuid-Core-i5-2500-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Core-i5-2500.json | 203 +++++ ...e-i5-2500.xml => x86_64-cpuid-Core-i5-2500.xml} | 0 ...st.xml => x86_64-cpuid-Core-i5-2540M-guest.xml} | 0 ...ost.xml => x86_64-cpuid-Core-i5-2540M-host.xml} | 0 ...son.xml => x86_64-cpuid-Core-i5-2540M-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Core-i5-2540M.json | 203 +++++ ...i5-2540M.xml => x86_64-cpuid-Core-i5-2540M.xml} | 0 ...st.xml => x86_64-cpuid-Core-i5-4670T-guest.xml} | 0 ...ost.xml => x86_64-cpuid-Core-i5-4670T-host.xml} | 0 ...son.xml => x86_64-cpuid-Core-i5-4670T-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Core-i5-4670T.json | 203 +++++ ...i5-4670T.xml => x86_64-cpuid-Core-i5-4670T.xml} | 0 ...est.xml => x86_64-cpuid-Core-i5-6600-guest.xml} | 0 ...host.xml => x86_64-cpuid-Core-i5-6600-host.xml} | 0 ...json.xml => x86_64-cpuid-Core-i5-6600-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Core-i5-6600.json | 203 +++++ ...e-i5-6600.xml => x86_64-cpuid-Core-i5-6600.xml} | 0 ...est.xml => x86_64-cpuid-Core-i7-2600-guest.xml} | 0 ...host.xml => x86_64-cpuid-Core-i7-2600-host.xml} | 0 ...json.xml => x86_64-cpuid-Core-i7-2600-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Core-i7-2600.json | 203 +++++ ...e-i7-2600.xml => x86_64-cpuid-Core-i7-2600.xml} | 0 ...st.xml => x86_64-cpuid-Core-i7-3520M-guest.xml} | 0 ...ost.xml => x86_64-cpuid-Core-i7-3520M-host.xml} | 0 ...i7-3520M.xml => x86_64-cpuid-Core-i7-3520M.xml} | 0 ...t.xml => x86_64-cpuid-Core-i7-3740QM-guest.xml} | 0 ...st.xml => x86_64-cpuid-Core-i7-3740QM-host.xml} | 0 ...on.xml => x86_64-cpuid-Core-i7-3740QM-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Core-i7-3740QM.json | 203 +++++ ...-3740QM.xml => x86_64-cpuid-Core-i7-3740QM.xml} | 0 ...est.xml => x86_64-cpuid-Core-i7-3770-guest.xml} | 0 ...host.xml => x86_64-cpuid-Core-i7-3770-host.xml} | 0 ...json.xml => x86_64-cpuid-Core-i7-3770-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Core-i7-3770.json | 203 +++++ ...e-i7-3770.xml => x86_64-cpuid-Core-i7-3770.xml} | 0 ...st.xml => x86_64-cpuid-Core-i7-4600U-guest.xml} | 0 ...ost.xml => x86_64-cpuid-Core-i7-4600U-host.xml} | 0 ...son.xml => x86_64-cpuid-Core-i7-4600U-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Core-i7-4600U.json | 203 +++++ ...i7-4600U.xml => x86_64-cpuid-Core-i7-4600U.xml} | 0 ...st.xml => x86_64-cpuid-Core-i7-5600U-guest.xml} | 0 ...ost.xml => x86_64-cpuid-Core-i7-5600U-host.xml} | 0 .../x86_64-cpuid-Core-i7-5600U-json.xml | 16 + tests/cputestdata/x86_64-cpuid-Core-i7-5600U.json | 203 +++++ ...i7-5600U.xml => x86_64-cpuid-Core-i7-5600U.xml} | 0 ...uest.xml => x86_64-cpuid-Core2-E6850-guest.xml} | 0 ...-host.xml => x86_64-cpuid-Core2-E6850-host.xml} | 0 ...-json.xml => x86_64-cpuid-Core2-E6850-json.xml} | 5 +- tests/cputestdata/x86_64-cpuid-Core2-E6850.json | 203 +++++ ...ore2-E6850.xml => x86_64-cpuid-Core2-E6850.xml} | 0 ...uest.xml => x86_64-cpuid-Core2-Q9500-guest.xml} | 0 ...-host.xml => x86_64-cpuid-Core2-Q9500-host.xml} | 0 ...ore2-Q9500.xml => x86_64-cpuid-Core2-Q9500.xml} | 0 ...50-guest.xml => x86_64-cpuid-FX-8150-guest.xml} | 0 ...8150-host.xml => x86_64-cpuid-FX-8150-host.xml} | 0 ...-cpuid-FX-8150.xml => x86_64-cpuid-FX-8150.xml} | 0 ...est.xml => x86_64-cpuid-Opteron-1352-guest.xml} | 0 ...host.xml => x86_64-cpuid-Opteron-1352-host.xml} | 0 ...eron-1352.xml => x86_64-cpuid-Opteron-1352.xml} | 0 ...est.xml => x86_64-cpuid-Opteron-2350-guest.xml} | 0 ...host.xml => x86_64-cpuid-Opteron-2350-host.xml} | 0 ...json.xml => x86_64-cpuid-Opteron-2350-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Opteron-2350.json | 203 +++++ ...eron-2350.xml => x86_64-cpuid-Opteron-2350.xml} | 0 ...est.xml => x86_64-cpuid-Opteron-6234-guest.xml} | 0 ...host.xml => x86_64-cpuid-Opteron-6234-host.xml} | 0 ...json.xml => x86_64-cpuid-Opteron-6234-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Opteron-6234.json | 203 +++++ ...eron-6234.xml => x86_64-cpuid-Opteron-6234.xml} | 0 ...est.xml => x86_64-cpuid-Opteron-6282-guest.xml} | 0 ...host.xml => x86_64-cpuid-Opteron-6282-host.xml} | 0 ...eron-6282.xml => x86_64-cpuid-Opteron-6282.xml} | 0 ...st.xml => x86_64-cpuid-Pentium-P6100-guest.xml} | 0 ...ost.xml => x86_64-cpuid-Pentium-P6100-host.xml} | 0 ...um-P6100.xml => x86_64-cpuid-Pentium-P6100.xml} | 0 ...guest.xml => x86_64-cpuid-Phenom-B95-guest.xml} | 0 ...5-host.xml => x86_64-cpuid-Phenom-B95-host.xml} | 0 ...5-json.xml => x86_64-cpuid-Phenom-B95-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Phenom-B95.json | 203 +++++ ...-Phenom-B95.xml => x86_64-cpuid-Phenom-B95.xml} | 0 ...-guest.xml => x86_64-cpuid-Xeon-5110-guest.xml} | 0 ...10-host.xml => x86_64-cpuid-Xeon-5110-host.xml} | 0 ...id-Xeon-5110.xml => x86_64-cpuid-Xeon-5110.xml} | 0 ...est.xml => x86_64-cpuid-Xeon-E3-1245-guest.xml} | 0 ...host.xml => x86_64-cpuid-Xeon-E3-1245-host.xml} | 0 ...json.xml => x86_64-cpuid-Xeon-E3-1245-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Xeon-E3-1245.json | 203 +++++ ...n-E3-1245.xml => x86_64-cpuid-Xeon-E3-1245.xml} | 0 ...est.xml => x86_64-cpuid-Xeon-E5-2630-guest.xml} | 0 ...host.xml => x86_64-cpuid-Xeon-E5-2630-host.xml} | 0 ...json.xml => x86_64-cpuid-Xeon-E5-2630-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Xeon-E5-2630.json | 203 +++++ ...n-E5-2630.xml => x86_64-cpuid-Xeon-E5-2630.xml} | 0 ...est.xml => x86_64-cpuid-Xeon-E5-2650-guest.xml} | 0 ...host.xml => x86_64-cpuid-Xeon-E5-2650-host.xml} | 0 ...json.xml => x86_64-cpuid-Xeon-E5-2650-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Xeon-E5-2650.json | 203 +++++ ...n-E5-2650.xml => x86_64-cpuid-Xeon-E5-2650.xml} | 0 ...est.xml => x86_64-cpuid-Xeon-E7-4820-guest.xml} | 0 ...host.xml => x86_64-cpuid-Xeon-E7-4820-host.xml} | 0 ...json.xml => x86_64-cpuid-Xeon-E7-4820-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Xeon-E7-4820.json | 203 +++++ ...n-E7-4820.xml => x86_64-cpuid-Xeon-E7-4820.xml} | 0 ...guest.xml => x86_64-cpuid-Xeon-W3520-guest.xml} | 0 ...0-host.xml => x86_64-cpuid-Xeon-W3520-host.xml} | 0 ...0-json.xml => x86_64-cpuid-Xeon-W3520-json.xml} | 1 + tests/cputestdata/x86_64-cpuid-Xeon-W3520.json | 203 +++++ ...-Xeon-W3520.xml => x86_64-cpuid-Xeon-W3520.xml} | 0 ...guest.xml => x86_64-cpuid-Xeon-X5460-guest.xml} | 0 ...0-host.xml => x86_64-cpuid-Xeon-X5460-host.xml} | 0 ...-Xeon-X5460.xml => x86_64-cpuid-Xeon-X5460.xml} | 0 ...le-extra.xml => x86_64-exact-disable-extra.xml} | 0 ...-exact-disable.xml => x86_64-exact-disable.xml} | 0 ...xact-disable2.xml => x86_64-exact-disable2.xml} | 0 ...bid-extra.xml => x86_64-exact-forbid-extra.xml} | 0 ...86-exact-forbid.xml => x86_64-exact-forbid.xml} | 0 ...-Haswell.xml => x86_64-exact-force-Haswell.xml} | 0 ...{x86-exact-force.xml => x86_64-exact-force.xml} | 0 ...re-extra.xml => x86_64-exact-require-extra.xml} | 0 ...-exact-require.xml => x86_64-exact-require.xml} | 0 .../{x86-exact.xml => x86_64-exact.xml} | 0 ...-nofallback.xml => x86_64-guest-nofallback.xml} | 0 .../{x86-guest.xml => x86_64-guest.xml} | 0 ...t.xml => x86_64-host+guest,model486-result.xml} | 0 ...ult.xml => x86_64-host+guest,models-result.xml} | 0 ...est-result.xml => x86_64-host+guest-result.xml} | 0 .../{x86-host+guest.xml => x86_64-host+guest.xml} | 0 ... x86_64-host+host+host-model,models-result.xml} | 0 ...k.xml => x86_64-host+host-model-nofallback.xml} | 0 ...t+host-model.xml => x86_64-host+host-model.xml} | 0 ...l => x86_64-host+host-passthrough-features.xml} | 0 ...hrough.xml => x86_64-host+host-passthrough.xml} | 0 .../{x86-host+min.xml => x86_64-host+min.xml} | 0 ...ult.xml => x86_64-host+penryn-force-result.xml} | 0 ...-host+pentium3.xml => x86_64-host+pentium3.xml} | 0 ...l => x86_64-host+strict-force-extra-result.xml} | 0 ...-host-Haswell-noTSX+Haswell,haswell-result.xml} | 0 ...Haswell-noTSX+Haswell-noTSX,haswell-result.xml} | 0 ...64-host-Haswell-noTSX+Haswell-noTSX-result.xml} | 0 ...ell-noTSX.xml => x86_64-host-Haswell-noTSX.xml} | 0 ...SandyBridge.xml => x86_64-host-SandyBridge.xml} | 0 ...-host-amd-fake.xml => x86_64-host-amd-fake.xml} | 0 .../{x86-host-amd.xml => x86_64-host-amd.xml} | 0 ....xml => x86_64-host-better+pentium3-result.xml} | 0 ...{x86-host-better.xml => x86_64-host-better.xml} | 0 ...incomp-arch.xml => x86_64-host-incomp-arch.xml} | 0 ...model.xml => x86_64-host-invtsc+host-model.xml} | 0 ...{x86-host-invtsc.xml => x86_64-host-invtsc.xml} | 0 ...llback.xml => x86_64-host-model-nofallback.xml} | 0 .../{x86-host-model.xml => x86_64-host-model.xml} | 0 ...ost-no-vendor.xml => x86_64-host-no-vendor.xml} | 0 ...es.xml => x86_64-host-passthrough-features.xml} | 0 ...passthrough.xml => x86_64-host-passthrough.xml} | 0 ...sult.xml => x86_64-host-worse+guest-result.xml} | 0 .../{x86-host-worse.xml => x86_64-host-worse.xml} | 0 .../cputestdata/{x86-host.xml => x86_64-host.xml} | 0 tests/cputestdata/{x86-min.xml => x86_64-min.xml} | 0 ...86-penryn-force.xml => x86_64-penryn-force.xml} | 0 ...86-pentium3-amd.xml => x86_64-pentium3-amd.xml} | 0 .../{x86-pentium3.xml => x86_64-pentium3.xml} | 0 ...trict-disable.xml => x86_64-strict-disable.xml} | 0 ...rce-extra.xml => x86_64-strict-force-extra.xml} | 0 ...{x86-strict-full.xml => x86_64-strict-full.xml} | 0 .../{x86-strict.xml => x86_64-strict.xml} | 0 tests/domaincapsschemadata/qemu_2.8.0.s390x.xml | 2 +- .../domaincapsschemadata/qemu_2.9.0-tcg.x86_64.xml | 145 ++++ tests/domaincapsschemadata/qemu_2.9.0.x86_64.xml | 124 +++ tests/domaincapstest.c | 8 + .../qemucapabilitiesdata/caps_2.8.0.s390x.replies | 8 + tests/qemucapabilitiesdata/caps_2.8.0.s390x.xml | 32 +- .../qemucapabilitiesdata/caps_2.9.0.x86_64.replies | 879 ++++++++++++++++++++- tests/qemucapabilitiesdata/caps_2.9.0.x86_64.xml | 465 ++++++++++- tests/qemumonitorjsontest.c | 4 +- tests/qemuxml2argvtest.c | 3 +- 267 files changed, 6731 insertions(+), 2055 deletions(-) create mode 100755 tests/cputestdata/cpu-convert.py delete mode 100644 tests/cputestdata/x86-cpuid-A10-5800K.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i5-2500.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i5-2540M.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i5-4670T.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i5-6600.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i7-2600.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i7-3740QM.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i7-3770.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i7-4600U.json delete mode 100644 tests/cputestdata/x86-cpuid-Core-i7-5600U-json.xml delete mode 100644 tests/cputestdata/x86-cpuid-Core-i7-5600U.json delete mode 100644 tests/cputestdata/x86-cpuid-Core2-E6850.json delete mode 100644 tests/cputestdata/x86-cpuid-Opteron-2350.json delete mode 100644 tests/cputestdata/x86-cpuid-Opteron-6234.json delete mode 100644 tests/cputestdata/x86-cpuid-Phenom-B95.json delete mode 100644 tests/cputestdata/x86-cpuid-Xeon-E3-1245.json delete mode 100644 tests/cputestdata/x86-cpuid-Xeon-E5-2630.json delete mode 100644 tests/cputestdata/x86-cpuid-Xeon-E5-2650.json delete mode 100644 tests/cputestdata/x86-cpuid-Xeon-E7-4820.json delete mode 100644 tests/cputestdata/x86-cpuid-Xeon-W3520.json rename tests/cputestdata/{x86-Haswell-noTSX-nofallback.xml => x86_64-Haswell-noTSX-nofallback.xml} (100%) rename tests/cputestdata/{x86-Haswell-noTSX.xml => x86_64-Haswell-noTSX.xml} (100%) rename tests/cputestdata/{x86-Haswell.xml => x86_64-Haswell.xml} (100%) rename tests/cputestdata/{x86-baseline-1-result.xml => x86_64-baseline-1-result.xml} (100%) rename tests/cputestdata/{x86-baseline-1.xml => x86_64-baseline-1.xml} (100%) rename tests/cputestdata/{x86-baseline-2-result.xml => x86_64-baseline-2-result.xml} (100%) rename tests/cputestdata/{x86-baseline-2.xml => x86_64-baseline-2.xml} (100%) rename tests/cputestdata/{x86-baseline-3-expanded.xml => x86_64-baseline-3-expanded.xml} (100%) rename tests/cputestdata/{x86-baseline-3-result.xml => x86_64-baseline-3-result.xml} (100%) rename tests/cputestdata/{x86-baseline-3.xml => x86_64-baseline-3.xml} (100%) rename tests/cputestdata/{x86-baseline-4-expanded.xml => x86_64-baseline-4-expanded.xml} (100%) rename tests/cputestdata/{x86-baseline-4-result.xml => x86_64-baseline-4-result.xml} (100%) rename tests/cputestdata/{x86-baseline-4.xml => x86_64-baseline-4.xml} (100%) rename tests/cputestdata/{x86-baseline-5-expanded.xml => x86_64-baseline-5-expanded.xml} (100%) rename tests/cputestdata/{x86-baseline-5-result.xml => x86_64-baseline-5-result.xml} (100%) rename tests/cputestdata/{x86-baseline-5.xml => x86_64-baseline-5.xml} (100%) rename tests/cputestdata/{x86-baseline-6-migratable.xml => x86_64-baseline-6-migratable.xml} (100%) rename tests/cputestdata/{x86-baseline-6-result.xml => x86_64-baseline-6-result.xml} (100%) rename tests/cputestdata/{x86-baseline-6.xml => x86_64-baseline-6.xml} (100%) rename tests/cputestdata/{x86-baseline-7-result.xml => x86_64-baseline-7-result.xml} (100%) rename tests/cputestdata/{x86-baseline-7.xml => x86_64-baseline-7.xml} (100%) rename tests/cputestdata/{x86-baseline-8-result.xml => x86_64-baseline-8-result.xml} (100%) rename tests/cputestdata/{x86-baseline-8.xml => x86_64-baseline-8.xml} (100%) rename tests/cputestdata/{x86-baseline-incompatible-vendors.xml => x86_64-baseline-incompatible-vendors.xml} (100%) rename tests/cputestdata/{x86-baseline-no-vendor-result.xml => x86_64-baseline-no-vendor-result.xml} (100%) rename tests/cputestdata/{x86-baseline-no-vendor.xml => x86_64-baseline-no-vendor.xml} (100%) rename tests/cputestdata/{x86-baseline-some-vendors-result.xml => x86_64-baseline-some-vendors-result.xml} (100%) rename tests/cputestdata/{x86-baseline-some-vendors.xml => x86_64-baseline-some-vendors.xml} (100%) rename tests/cputestdata/{x86-bogus-feature.xml => x86_64-bogus-feature.xml} (100%) rename tests/cputestdata/{x86-bogus-model.xml => x86_64-bogus-model.xml} (100%) rename tests/cputestdata/{x86-bogus-vendor.xml => x86_64-bogus-vendor.xml} (100%) rename tests/cputestdata/{x86-cpuid-A10-5800K-guest.xml => x86_64-cpuid-A10-5800K-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-A10-5800K-host.xml => x86_64-cpuid-A10-5800K-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-A10-5800K-json.xml => x86_64-cpuid-A10-5800K-json.xml} (96%) create mode 100644 tests/cputestdata/x86_64-cpuid-A10-5800K.json rename tests/cputestdata/{x86-cpuid-A10-5800K.xml => x86_64-cpuid-A10-5800K.xml} (100%) rename tests/cputestdata/{x86-cpuid-Atom-D510-guest.xml => x86_64-cpuid-Atom-D510-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Atom-D510-host.xml => x86_64-cpuid-Atom-D510-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Atom-D510.xml => x86_64-cpuid-Atom-D510.xml} (100%) rename tests/cputestdata/{x86-cpuid-Atom-N450-guest.xml => x86_64-cpuid-Atom-N450-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Atom-N450-host.xml => x86_64-cpuid-Atom-N450-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Atom-N450.xml => x86_64-cpuid-Atom-N450.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-2500-guest.xml => x86_64-cpuid-Core-i5-2500-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-2500-host.xml => x86_64-cpuid-Core-i5-2500-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-2540M-json.xml => x86_64-cpuid-Core-i5-2500-json.xml} (94%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i5-2500.json rename tests/cputestdata/{x86-cpuid-Core-i5-2500.xml => x86_64-cpuid-Core-i5-2500.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-2540M-guest.xml => x86_64-cpuid-Core-i5-2540M-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-2540M-host.xml => x86_64-cpuid-Core-i5-2540M-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-2500-json.xml => x86_64-cpuid-Core-i5-2540M-json.xml} (94%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i5-2540M.json rename tests/cputestdata/{x86-cpuid-Core-i5-2540M.xml => x86_64-cpuid-Core-i5-2540M.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-4670T-guest.xml => x86_64-cpuid-Core-i5-4670T-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-4670T-host.xml => x86_64-cpuid-Core-i5-4670T-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-4670T-json.xml => x86_64-cpuid-Core-i5-4670T-json.xml} (95%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i5-4670T.json rename tests/cputestdata/{x86-cpuid-Core-i5-4670T.xml => x86_64-cpuid-Core-i5-4670T.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-6600-guest.xml => x86_64-cpuid-Core-i5-6600-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-6600-host.xml => x86_64-cpuid-Core-i5-6600-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i5-6600-json.xml => x86_64-cpuid-Core-i5-6600-json.xml} (93%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i5-6600.json rename tests/cputestdata/{x86-cpuid-Core-i5-6600.xml => x86_64-cpuid-Core-i5-6600.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-2600-guest.xml => x86_64-cpuid-Core-i7-2600-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-2600-host.xml => x86_64-cpuid-Core-i7-2600-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-2600-json.xml => x86_64-cpuid-Core-i7-2600-json.xml} (93%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i7-2600.json rename tests/cputestdata/{x86-cpuid-Core-i7-2600.xml => x86_64-cpuid-Core-i7-2600.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3520M-guest.xml => x86_64-cpuid-Core-i7-3520M-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3520M-host.xml => x86_64-cpuid-Core-i7-3520M-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3520M.xml => x86_64-cpuid-Core-i7-3520M.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3740QM-guest.xml => x86_64-cpuid-Core-i7-3740QM-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3740QM-host.xml => x86_64-cpuid-Core-i7-3740QM-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3740QM-json.xml => x86_64-cpuid-Core-i7-3740QM-json.xml} (93%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i7-3740QM.json rename tests/cputestdata/{x86-cpuid-Core-i7-3740QM.xml => x86_64-cpuid-Core-i7-3740QM.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3770-guest.xml => x86_64-cpuid-Core-i7-3770-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3770-host.xml => x86_64-cpuid-Core-i7-3770-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-3770-json.xml => x86_64-cpuid-Core-i7-3770-json.xml} (92%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i7-3770.json rename tests/cputestdata/{x86-cpuid-Core-i7-3770.xml => x86_64-cpuid-Core-i7-3770.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-4600U-guest.xml => x86_64-cpuid-Core-i7-4600U-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-4600U-host.xml => x86_64-cpuid-Core-i7-4600U-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-4600U-json.xml => x86_64-cpuid-Core-i7-4600U-json.xml} (95%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i7-4600U.json rename tests/cputestdata/{x86-cpuid-Core-i7-4600U.xml => x86_64-cpuid-Core-i7-4600U.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-5600U-guest.xml => x86_64-cpuid-Core-i7-5600U-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core-i7-5600U-host.xml => x86_64-cpuid-Core-i7-5600U-host.xml} (100%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i7-5600U-json.xml create mode 100644 tests/cputestdata/x86_64-cpuid-Core-i7-5600U.json rename tests/cputestdata/{x86-cpuid-Core-i7-5600U.xml => x86_64-cpuid-Core-i7-5600U.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core2-E6850-guest.xml => x86_64-cpuid-Core2-E6850-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core2-E6850-host.xml => x86_64-cpuid-Core2-E6850-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core2-E6850-json.xml => x86_64-cpuid-Core2-E6850-json.xml} (75%) create mode 100644 tests/cputestdata/x86_64-cpuid-Core2-E6850.json rename tests/cputestdata/{x86-cpuid-Core2-E6850.xml => x86_64-cpuid-Core2-E6850.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core2-Q9500-guest.xml => x86_64-cpuid-Core2-Q9500-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core2-Q9500-host.xml => x86_64-cpuid-Core2-Q9500-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Core2-Q9500.xml => x86_64-cpuid-Core2-Q9500.xml} (100%) rename tests/cputestdata/{x86-cpuid-FX-8150-guest.xml => x86_64-cpuid-FX-8150-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-FX-8150-host.xml => x86_64-cpuid-FX-8150-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-FX-8150.xml => x86_64-cpuid-FX-8150.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-1352-guest.xml => x86_64-cpuid-Opteron-1352-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-1352-host.xml => x86_64-cpuid-Opteron-1352-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-1352.xml => x86_64-cpuid-Opteron-1352.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-2350-guest.xml => x86_64-cpuid-Opteron-2350-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-2350-host.xml => x86_64-cpuid-Opteron-2350-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-2350-json.xml => x86_64-cpuid-Opteron-2350-json.xml} (97%) create mode 100644 tests/cputestdata/x86_64-cpuid-Opteron-2350.json rename tests/cputestdata/{x86-cpuid-Opteron-2350.xml => x86_64-cpuid-Opteron-2350.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-6234-guest.xml => x86_64-cpuid-Opteron-6234-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-6234-host.xml => x86_64-cpuid-Opteron-6234-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-6234-json.xml => x86_64-cpuid-Opteron-6234-json.xml} (96%) create mode 100644 tests/cputestdata/x86_64-cpuid-Opteron-6234.json rename tests/cputestdata/{x86-cpuid-Opteron-6234.xml => x86_64-cpuid-Opteron-6234.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-6282-guest.xml => x86_64-cpuid-Opteron-6282-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-6282-host.xml => x86_64-cpuid-Opteron-6282-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Opteron-6282.xml => x86_64-cpuid-Opteron-6282.xml} (100%) rename tests/cputestdata/{x86-cpuid-Pentium-P6100-guest.xml => x86_64-cpuid-Pentium-P6100-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Pentium-P6100-host.xml => x86_64-cpuid-Pentium-P6100-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Pentium-P6100.xml => x86_64-cpuid-Pentium-P6100.xml} (100%) rename tests/cputestdata/{x86-cpuid-Phenom-B95-guest.xml => x86_64-cpuid-Phenom-B95-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Phenom-B95-host.xml => x86_64-cpuid-Phenom-B95-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Phenom-B95-json.xml => x86_64-cpuid-Phenom-B95-json.xml} (97%) create mode 100644 tests/cputestdata/x86_64-cpuid-Phenom-B95.json rename tests/cputestdata/{x86-cpuid-Phenom-B95.xml => x86_64-cpuid-Phenom-B95.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-5110-guest.xml => x86_64-cpuid-Xeon-5110-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-5110-host.xml => x86_64-cpuid-Xeon-5110-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-5110.xml => x86_64-cpuid-Xeon-5110.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E3-1245-guest.xml => x86_64-cpuid-Xeon-E3-1245-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E3-1245-host.xml => x86_64-cpuid-Xeon-E3-1245-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E3-1245-json.xml => x86_64-cpuid-Xeon-E3-1245-json.xml} (93%) create mode 100644 tests/cputestdata/x86_64-cpuid-Xeon-E3-1245.json rename tests/cputestdata/{x86-cpuid-Xeon-E3-1245.xml => x86_64-cpuid-Xeon-E3-1245.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E5-2630-guest.xml => x86_64-cpuid-Xeon-E5-2630-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E5-2630-host.xml => x86_64-cpuid-Xeon-E5-2630-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E5-2630-json.xml => x86_64-cpuid-Xeon-E5-2630-json.xml} (95%) create mode 100644 tests/cputestdata/x86_64-cpuid-Xeon-E5-2630.json rename tests/cputestdata/{x86-cpuid-Xeon-E5-2630.xml => x86_64-cpuid-Xeon-E5-2630.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E5-2650-guest.xml => x86_64-cpuid-Xeon-E5-2650-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E5-2650-host.xml => x86_64-cpuid-Xeon-E5-2650-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E5-2650-json.xml => x86_64-cpuid-Xeon-E5-2650-json.xml} (94%) create mode 100644 tests/cputestdata/x86_64-cpuid-Xeon-E5-2650.json rename tests/cputestdata/{x86-cpuid-Xeon-E5-2650.xml => x86_64-cpuid-Xeon-E5-2650.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E7-4820-guest.xml => x86_64-cpuid-Xeon-E7-4820-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E7-4820-host.xml => x86_64-cpuid-Xeon-E7-4820-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-E7-4820-json.xml => x86_64-cpuid-Xeon-E7-4820-json.xml} (94%) create mode 100644 tests/cputestdata/x86_64-cpuid-Xeon-E7-4820.json rename tests/cputestdata/{x86-cpuid-Xeon-E7-4820.xml => x86_64-cpuid-Xeon-E7-4820.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-W3520-guest.xml => x86_64-cpuid-Xeon-W3520-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-W3520-host.xml => x86_64-cpuid-Xeon-W3520-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-W3520-json.xml => x86_64-cpuid-Xeon-W3520-json.xml} (93%) create mode 100644 tests/cputestdata/x86_64-cpuid-Xeon-W3520.json rename tests/cputestdata/{x86-cpuid-Xeon-W3520.xml => x86_64-cpuid-Xeon-W3520.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-X5460-guest.xml => x86_64-cpuid-Xeon-X5460-guest.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-X5460-host.xml => x86_64-cpuid-Xeon-X5460-host.xml} (100%) rename tests/cputestdata/{x86-cpuid-Xeon-X5460.xml => x86_64-cpuid-Xeon-X5460.xml} (100%) rename tests/cputestdata/{x86-exact-disable-extra.xml => x86_64-exact-disable-extra.xml} (100%) rename tests/cputestdata/{x86-exact-disable.xml => x86_64-exact-disable.xml} (100%) rename tests/cputestdata/{x86-exact-disable2.xml => x86_64-exact-disable2.xml} (100%) rename tests/cputestdata/{x86-exact-forbid-extra.xml => x86_64-exact-forbid-extra.xml} (100%) rename tests/cputestdata/{x86-exact-forbid.xml => x86_64-exact-forbid.xml} (100%) rename tests/cputestdata/{x86-exact-force-Haswell.xml => x86_64-exact-force-Haswell.xml} (100%) rename tests/cputestdata/{x86-exact-force.xml => x86_64-exact-force.xml} (100%) rename tests/cputestdata/{x86-exact-require-extra.xml => x86_64-exact-require-extra.xml} (100%) rename tests/cputestdata/{x86-exact-require.xml => x86_64-exact-require.xml} (100%) rename tests/cputestdata/{x86-exact.xml => x86_64-exact.xml} (100%) rename tests/cputestdata/{x86-guest-nofallback.xml => x86_64-guest-nofallback.xml} (100%) rename tests/cputestdata/{x86-guest.xml => x86_64-guest.xml} (100%) rename tests/cputestdata/{x86-host+guest,model486-result.xml => x86_64-host+guest,model486-result.xml} (100%) rename tests/cputestdata/{x86-host+guest,models-result.xml => x86_64-host+guest,models-result.xml} (100%) rename tests/cputestdata/{x86-host+guest-result.xml => x86_64-host+guest-result.xml} (100%) rename tests/cputestdata/{x86-host+guest.xml => x86_64-host+guest.xml} (100%) rename tests/cputestdata/{x86-host+host+host-model,models-result.xml => x86_64-host+host+host-model,models-result.xml} (100%) rename tests/cputestdata/{x86-host+host-model-nofallback.xml => x86_64-host+host-model-nofallback.xml} (100%) rename tests/cputestdata/{x86-host+host-model.xml => x86_64-host+host-model.xml} (100%) rename tests/cputestdata/{x86-host+host-passthrough-features.xml => x86_64-host+host-passthrough-features.xml} (100%) rename tests/cputestdata/{x86-host+host-passthrough.xml => x86_64-host+host-passthrough.xml} (100%) rename tests/cputestdata/{x86-host+min.xml => x86_64-host+min.xml} (100%) rename tests/cputestdata/{x86-host+penryn-force-result.xml => x86_64-host+penryn-force-result.xml} (100%) rename tests/cputestdata/{x86-host+pentium3.xml => x86_64-host+pentium3.xml} (100%) rename tests/cputestdata/{x86-host+strict-force-extra-result.xml => x86_64-host+strict-force-extra-result.xml} (100%) rename tests/cputestdata/{x86-host-Haswell-noTSX+Haswell,haswell-result.xml => x86_64-host-Haswell-noTSX+Haswell,haswell-result.xml} (100%) rename tests/cputestdata/{x86-host-Haswell-noTSX+Haswell-noTSX,haswell-result.xml => x86_64-host-Haswell-noTSX+Haswell-noTSX,haswell-result.xml} (100%) rename tests/cputestdata/{x86-host-Haswell-noTSX+Haswell-noTSX-result.xml => x86_64-host-Haswell-noTSX+Haswell-noTSX-result.xml} (100%) rename tests/cputestdata/{x86-host-Haswell-noTSX.xml => x86_64-host-Haswell-noTSX.xml} (100%) rename tests/cputestdata/{x86-host-SandyBridge.xml => x86_64-host-SandyBridge.xml} (100%) rename tests/cputestdata/{x86-host-amd-fake.xml => x86_64-host-amd-fake.xml} (100%) rename tests/cputestdata/{x86-host-amd.xml => x86_64-host-amd.xml} (100%) rename tests/cputestdata/{x86-host-better+pentium3-result.xml => x86_64-host-better+pentium3-result.xml} (100%) rename tests/cputestdata/{x86-host-better.xml => x86_64-host-better.xml} (100%) rename tests/cputestdata/{x86-host-incomp-arch.xml => x86_64-host-incomp-arch.xml} (100%) rename tests/cputestdata/{x86-host-invtsc+host-model.xml => x86_64-host-invtsc+host-model.xml} (100%) rename tests/cputestdata/{x86-host-invtsc.xml => x86_64-host-invtsc.xml} (100%) rename tests/cputestdata/{x86-host-model-nofallback.xml => x86_64-host-model-nofallback.xml} (100%) rename tests/cputestdata/{x86-host-model.xml => x86_64-host-model.xml} (100%) rename tests/cputestdata/{x86-host-no-vendor.xml => x86_64-host-no-vendor.xml} (100%) rename tests/cputestdata/{x86-host-passthrough-features.xml => x86_64-host-passthrough-features.xml} (100%) rename tests/cputestdata/{x86-host-passthrough.xml => x86_64-host-passthrough.xml} (100%) rename tests/cputestdata/{x86-host-worse+guest-result.xml => x86_64-host-worse+guest-result.xml} (100%) rename tests/cputestdata/{x86-host-worse.xml => x86_64-host-worse.xml} (100%) rename tests/cputestdata/{x86-host.xml => x86_64-host.xml} (100%) rename tests/cputestdata/{x86-min.xml => x86_64-min.xml} (100%) rename tests/cputestdata/{x86-penryn-force.xml => x86_64-penryn-force.xml} (100%) rename tests/cputestdata/{x86-pentium3-amd.xml => x86_64-pentium3-amd.xml} (100%) rename tests/cputestdata/{x86-pentium3.xml => x86_64-pentium3.xml} (100%) rename tests/cputestdata/{x86-strict-disable.xml => x86_64-strict-disable.xml} (100%) rename tests/cputestdata/{x86-strict-force-extra.xml => x86_64-strict-force-extra.xml} (100%) rename tests/cputestdata/{x86-strict-full.xml => x86_64-strict-full.xml} (100%) rename tests/cputestdata/{x86-strict.xml => x86_64-strict.xml} (100%) create mode 100644 tests/domaincapsschemadata/qemu_2.9.0-tcg.x86_64.xml create mode 100644 tests/domaincapsschemadata/qemu_2.9.0.x86_64.xml -- 2.11.1

3 54

[libvirt] [PATCH] Fix location of blkid.h in include header
by Nehal J Wani 03 Mar '17

03 Mar '17

The build system for libvirt correctly detects the location of blkid using PKG_CONFIG_PATH environment variable. The file blkid.pc states that the include flags should be: 'Cflags: -I${includedir}/blkid' but libvirt searches for blkid.h inside ${includedir}/blkid/blkid, which is wrong. Until now, the compilation for libvirt succeeded because of pure luck, as it had -I/usr/include as a CFLAG. This issue was faced while compiling libvirt on Ubuntu 16.04.2 with bare minimum dev packages and a custom compiled blkid kept in a non-standard $prefix. Signed-off-by: Nehal J Wani <nehaljw.kkd1(a)gmail.com> --- src/lxc/lxc_container.c | 2 +- src/storage/storage_util.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index 99bd7e9..e306b6e 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -49,7 +49,7 @@ #endif #if WITH_BLKID -# include <blkid/blkid.h> +# include <blkid.h> #endif #if WITH_SELINUX diff --git a/src/storage/storage_util.c b/src/storage/storage_util.c index a665bac..41253b9 100644 --- a/src/storage/storage_util.c +++ b/src/storage/storage_util.c @@ -39,7 +39,7 @@ #endif #if WITH_BLKID -# include <blkid/blkid.h> +# include <blkid.h> #endif #if WITH_SELINUX -- 2.7.4

2 1

[libvirt] [PATCH v2 0/6] Split out node device object into its own module
by John Ferlan 03 Mar '17

03 Mar '17

v1: http://www.redhat.com/archives/libvir-list/2017-March/msg00046.html Differences... patch 1 and 3 are the same and both were ACK'd patch 2 removes the lower case function name changes (a/k/a "vir" removal) patch 4 is new adding modifications as requested by Pavel from patch 1 comment patch 5 & 6 are new based on a type A reaction to seeing a deeply nested structure and picking it apart to allow for more direct access to the data fields via pointers which patch6 dutify does for virNodeDevCapDataPtr John Ferlan (6): conf: Introduce virnodedeviceobj conf: Adjust coding style for nodedev conf sources conf: Use consistent function name prefixes for virnodedeviceobj conf: Alter coding style of nodedev function prototypes conf: Clean up the _virNodeDevCapData nodedev: Reduce virNodeDevCapDataPtr usage po/POTFILES.in | 1 + src/Makefile.am | 3 +- src/conf/node_device_conf.c | 757 ++++++------------------------ src/conf/node_device_conf.h | 316 ++++++------- src/conf/virnodedeviceobj.c | 558 ++++++++++++++++++++++ src/conf/virnodedeviceobj.h | 92 ++++ src/libvirt_private.syms | 23 +- src/node_device/node_device_driver.c | 30 +- src/node_device/node_device_driver.h | 2 +- src/node_device/node_device_hal.c | 14 +- src/node_device/node_device_linux_sysfs.c | 134 +++--- src/node_device/node_device_linux_sysfs.h | 4 +- src/node_device/node_device_udev.c | 198 ++++---- src/test/test_driver.c | 29 +- 14 files changed, 1164 insertions(+), 997 deletions(-) create mode 100644 src/conf/virnodedeviceobj.c create mode 100644 src/conf/virnodedeviceobj.h -- 2.9.3

3 10

[libvirt] [PATCH RFC 0/4] libxl: qemu agent support
by Joao Martins 03 Mar '17

03 Mar '17

Hey! Back when channels were introduced in libxl (in answer to Michal[0]) I suggested the idea of integrating qemu guest agent (which currently lives qemu driver). This series is an attempt at pulling qemu agent from qemu driver into util in using it in libxl in subsequent patches. What do folks think of the idea? Note that this is still all very RFC because 1) there's a lot of code we could potentially share between qemu and libxl with respect to finding guest agent config and keeping some of its state (see patch 3); 2) also we need to ignore "execute" messages to be able to query the agent and see it's returned data (patch 2). Which despite the commit not being incorrect I am not sure yet why we need it yet. As PoC I only implemented domainQemuAgentCommand/domainInterfaceAddresses, but there's a lot more driver APIs we can potentially introduce after this. Tracing all driver APIs that might require a guest agent: * domainFSThaw, domainFSFreeze, domainFSTrim, domainGetFSInfo * domainSetUserPassword * domainGetTime, domainSetTime * domainShutdown (with VIR_DOMAIN_SHUTDOWN_GUEST_AGENT) * domainReboot (with VIR_DOMAIN_REBOOT_GUEST_AGENT) * domainGetGuestVcpus * domainSetGuestVcpus * domainSetVcpusFlags (with VIR_DOMAIN_VCPU_GUEST) * domainGetVcpusFlags (with VIR_DOMAIN_VCPU_GUEST) Comments/Feedback is appreciated :) Cheers, Joao [0] https://www.spinics.net/linux/fedora/libvir/msg136685.html Joao Martins (4): qemu_agent: move agent into util qemu_agent: ignore requests echoed back by guest libxl: implement qemu-agent-command libxl: domainInterfaceAddresses agent support po/POTFILES.in | 2 +- src/Makefile.am | 2 +- src/libvirt_private.syms | 21 + src/libxl/libxl_domain.c | 239 ++++- src/libxl/libxl_domain.h | 16 + src/libxl/libxl_driver.c | 69 ++ src/qemu/qemu_agent.c | 2248 ----------------------------------------- src/qemu/qemu_agent.h | 123 --- src/qemu/qemu_domain.h | 2 +- src/qemu/qemu_driver.c | 2 +- src/util/virqemuagent.c | 2249 ++++++++++++++++++++++++++++++++++++++++++ src/util/virqemuagent.h | 123 +++ tests/qemuagenttest.c | 2 +- tests/qemumonitortestutils.c | 2 +- tests/qemumonitortestutils.h | 2 +- 15 files changed, 2723 insertions(+), 2379 deletions(-) delete mode 100644 src/qemu/qemu_agent.c delete mode 100644 src/qemu/qemu_agent.h create mode 100644 src/util/virqemuagent.c create mode 100644 src/util/virqemuagent.h -- 2.1.4

3 17