- Devel - libvirt List Archives

[libvirt] [PATCH v2 0/7] Memory locking limit improvements
by Andrea Bolognani 17 Dec '15

17 Dec '15

During my testing, I've realized that even with the series applied there's still one case in which we're unable to restore the previous memory locking limit after removing the last PCI hostdev from the guest. If an x86 guest contains a PCI hostdev in its XML definition, then the memory locking limit will be set correctly, but virCommandGetMaxMemLock() will be used instead of virProcessGetMaxMemLock(), and the limit will be set right before calling exec() to spawn the qemu binary. In this context, we have no access to the virDomainObj or even virDomainDef instances, so I see no way of storing the original limit for later retrieval. The series is still an improvement as it covers all other cases. Still, I thought this was worth mentioning. Changes since v1[1]: * reorder commits according to John's suggestions * don't fail if we can't retrieve the current memory locking limit * small changes here and there as pointed out during review Cheers. [1] https://www.redhat.com/archives/libvir-list/2015-November/msg01021.html Andrea Bolognani (7): process: Allow virProcessPrLimit() to get current limit process: Add virProcessGetMaxMemLock() qemu: Add qemuDomainAdjustMaxMemLock() qemu: Use qemuDomainAdjustMaxMemLock() qemu: Reduce memlock limit after detaching PCI hostdev qemu: Allow qemuDomainAdjustMaxMemLock() to restore previous value qemu: Replace Mlock with MemLock in function names configure.ac | 2 +- src/conf/domain_conf.h | 3 +++ src/libvirt_private.syms | 1 + src/qemu/qemu_command.c | 4 ++-- src/qemu/qemu_domain.c | 53 ++++++++++++++++++++++++++++++++++++++--- src/qemu/qemu_domain.h | 5 ++-- src/qemu/qemu_hotplug.c | 45 ++++++++++++++--------------------- src/util/virprocess.c | 61 +++++++++++++++++++++++++++++++++++++++++++----- src/util/virprocess.h | 2 ++ 9 files changed, 135 insertions(+), 41 deletions(-) -- 2.5.0

2 9

[libvirt] [PATCH] build: disable vbox on cygwin
by Eric Blake 17 Dec '15

17 Dec '15

Cygwin cannot build the vbox driver yet: CC vbox/libvirt_driver_vbox_impl_la-vbox_glue.lo In file included from vbox/vbox_glue.c:27:0: vblox/vbox_XPCOMCGlue.c:63:3: error: #error "Port me" # error "Port me" ^ In file included from vbox/vbox_XPCOMCGlue.c:45:0, from vbox/vbox_glue.c:27: vbox/vbox_XPCOMCGlue.c: In function 'tryLoadOne': vbox/vbox_XPCOMCGlue.c:98:46: error: 'DYNLIB_NAME' undeclared (first use in this function) if (virAsprintf(&name, "%s/%s", dir, DYNLIB_NAME) < 0) ^ ./util/virstring.h:245:31: note: in definition of macro 'virAsprintf' strp, __VA_ARGS__) ^ Rather than trying to figure out how to get dynamic loading of vbox to work under cygwin (since I don't even have a working vbox setup to test whether it works), I'm going to be lazy and just default to not even trying vbox on cygwin. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Not technically a build-breaker, since I can './configure --without-vbox' to work around it, so I'll wait for a review. configure.ac | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/configure.ac b/configure.ac index 75e95b7..2c34333 100644 --- a/configure.ac +++ b/configure.ac @@ -175,12 +175,13 @@ dnl are also linux specific. The "network" and storage_fs drivers are known dnl to not work on MacOS X presently, so we also make a note if compiling dnl for that -with_linux=no with_osx=no with_freebsd=no +with_linux=no with_osx=no with_freebsd=no with_win=no with_cygwin=no case $host in *-*-linux*) with_linux=yes ;; *-*-darwin*) with_osx=yes ;; *-*-freebsd*) with_freebsd=yes ;; *-*-mingw* | *-*-msvc* ) with_win=yes ;; + *-*-cygwin*) with_cygwin=yes ;; esac if test $with_linux = no; then @@ -192,9 +193,12 @@ if test $with_linux = no; then fi if test $with_freebsd = yes; then - want_ifconfig=yes + want_ifconfig=yes + with_firewalld=no +fi - with_firewalld=no +if test $with_cygwin = yes; then + with_vbox=no fi AM_CONDITIONAL([WITH_LINUX], [test "$with_linux" = "yes"]) -- 2.4.3

2 2

[libvirt] LSN-2015-0004: CVE-2015-5313: ACL bypass using ../ to access beyond storage pool
by Eric Blake 17 Dec '15

17 Dec '15

Libvirt Security Notice: LSN-2015-0004 ====================================== Summary: ACL bypass using ../ to access beyond storage pool Reported on: 20151030 Published on: 20151211 Fixed on: 20151211 Reported by: Ossi Herrala <vulncoord(a)ficora.fi> Joonas Kuorilehto <vulncoord(a)ficora.fi> Patched by: Eric Blake <eblake(a)redhat.com> See also: CVE-2015-5313, FICORA bug #876194 Description ----------- Various virStorageVol* API operate on user-supplied volume names by concatenating the volume name to the pool location. Note that the virStoragePoolListVolumes API, when used on a storage pool backed by a directory in a file system, will only list volumes immediately in that directory (there is no traversal into subdirectories). However, other APIs such as virStorageVolCreateXML were not checking if a potential volume name represented one of the volumes that could be returned by virStoragePoolListVolumes; because they were not rejecting the use of '/' in a volume name. Impact ------ Because no checking was done on volume names, a user could supply a potential volume name of something like '../../../etc/passwd' to attempt to access a file not belonging to the storage pool. When fine-grained Access Control Lists (ACL) are in effect, a user with storage_vol:create ACL permission but lacking domain:write permssion could thus abuse virStorageVolCreateXML and similar APIs to gain access to files not normally permitted to that user. Fortunately, it appears that the only APIs that could leak information or corrupt files require read-write connection to libvirtd; and when ACLs are not in use (the default without any further configuration), a user with read-write access can already be considered to have full access to the machine, and without an escalation of privilege there is no security problem. Workaround ---------- If fine-grained ACLs must be used, administrators must consider all of the storage_vol:* permissions as equivalent to domain:write when running an impacted version of libvirt. The easiest way to prevent untrusted users from gaining unauthorized access to volumes outside of permitted pools is by disabling the use of fine-graned ACLs, and ensuring that such users do not have read-write access to libvirtd. Affected product ---------------- Name: libvirt Repository: git://libvirt.org/git/libvirt.git http://libvirt.org/git/?p=libvirt.git Branch: master Broken in: v1.1.0 Broken in: v1.1.1 Broken in: v1.1.2 Broken in: v1.1.3 Broken in: v1.1.4 Broken in: v1.2.0 Broken in: v1.2.1 Broken in: v1.2.2 Broken in: v1.2.3 Broken in: v1.2.4 Broken in: v1.2.5 Broken in: v1.2.6 Broken in: v1.2.7 Broken in: v1.2.8 Broken in: v1.2.9 Broken in: v1.2.10 Broken in: v1.2.11 Broken in: v1.2.12 Broken in: v1.2.13 Broken in: v1.2.14 Broken in: v1.2.15 Broken in: v1.2.16 Broken in: v1.2.17 Broken in: v1.2.18 Broken in: v1.2.19 Broken in: v1.2.20 Broken in: v1.2.20 Broken in: v1.3.0 Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 034e47c338b13a95cf02106a3af912c1c5f818d7 Branch: v1.1.0-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 14828a59eadc7221326198a8d7af817a6b8b8c13 Branch: v1.1.1-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 692ce509efa0a07f2811d0fe3b7202b020c874e0 Branch: v1.1.2-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: e8643ef68c99e9f5068f6ff64ea0acab94cac7f6 Branch: v1.1.3-maint Broken in: v1.1.3.1 Broken in: v1.1.3.2 Broken in: v1.1.3.3 Broken in: v1.1.3.4 Broken in: v1.1.3.5 Broken in: v1.1.3.6 Broken in: v1.1.3.7 Broken in: v1.1.3.8 Broken in: v1.1.3.9 Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: dcce665904b8ebc9ac3e5109db179a567b33e1a2 Branch: v1.1.4-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: dc2db111a9ba074589c54b90c89f33c01b1e4941 Branch: v1.2.0-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: d414ecb8e1714704e6515ab01ef9386d89b8051e Branch: v1.2.1-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 02d365dae595a3453fe0e438bc274ccf3c18e20d Branch: v1.2.2-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 6542e643024ca4272f14e9052b3786378f6eec62 Branch: v1.2.3-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 91898c606496b14e0891af31dfca7eb77ba9fee3 Branch: v1.2.4-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: c9450f4f855736ef3024dfbab403a849110d8bb5 Branch: v1.2.5-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 890fc0f1ffcc479b08b9fd01de31b62e3d9e7427 Branch: v1.2.6-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 6ae433938377e1b7e657c34cca39e52426347cb4 Branch: v1.2.7-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 4ed8074672f9b847a10464d9c6be77d428c1eb1c Branch: v1.2.8-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 54be99a717873524798d39f8baf49e45054192c8 Branch: v1.2.9-maint Broken in: v1.2.9.1 Broken in: v1.2.9.2 Broken in: v1.2.9.3 Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: b0f88836e5eb5b7156bda99c005cf4aa0456ed0d Branch: v1.2.10-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 53ae31bf4df364a2110f636d5482b21af4e4a0cc Branch: v1.2.11-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 0060c4ee9e70a9f6f297373cb4fd2ace6c187be0 Branch: v1.2.12-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: b5ddfbc0fe13a7910c2303056ddd5df749bcf8b0 Branch: v1.2.13-maint Broken in: v1.2.13.1 Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: b553ec764f7ecdf8962efbf849a0e8524bae610c Branch: v1.2.14-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 6410a22743fadc3b554b2f0866c9ab8008ff4908 Branch: v1.2.15-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 01cbfeb7d81498db3c644404980c9c1aa9cac048 Branch: v1.2.16-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 3e6b40e5aa3edf47443f017a42ec7b87855ed847 Branch: v1.2.17-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 08acad56ce2e5bcfcca8600a4e4074d3aaeb44dd Branch: v1.2.18-maint Broken in: v1.2.18.1 Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: d035796675ca42795953828d11f902f691fa6b29 Branch: v1.2.19-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 69548d200409d2b0dd6356fccfd59570fb58e23a Branch: v1.2.20-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: edeef640db625d23700011dc94adff6e29b85cd3 Branch: v1.2.21-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 29b4ce46798519b93a6a17a5e3734ea4f68ea69d Branch: v1.3.0-maint Broken by: c930410bebae0a45889b992a7932c663b06cbbcd Fixed by: 1d8bcbb7c68d3f35689daf727bc74fcf80a3a6b1 -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org

1 0

[libvirt] [PATCH 0/5] Fix issues with activating pool with invalid source
by John Ferlan 17 Dec '15

17 Dec '15

These patches resolve issues with 'FS', 'NFS', and 'LOGICAL' pools where if the pool source device didn't match the reality what was running on the host there were inconsistent results. For the file pools, a pool would be declared 'active' after restart even though a start would fail because the check code only cared that the device the pool was using was mounted. Patches 3 alters the check to not only make sure the device is mounted, but that the source for the device matches the source used to start the pool. For the logical pool, a pool could be both started and declared 'active' on restart as long as the "pool->def->source.name" was a valid volume group on the host even though the pool's source device(s) didn't match the same volume group. Usually the pool build process takes care of ensuring not only that the source device exists, but matching the device(s) to the volume group name create (via vgcreate). The reality is pool startup never checked that the volume group name being used by the pool matched the reality of the volume group on the host. Patch 5 will now ensure not only startup matches the name and device list, but that restart setting 'active' would do the same. John Ferlan (5): storage: Create helper to generate FS pool source value storage: Refactor virStorageBackendFileSystemGetPoolSource storage: Check FS pool source during virStorageBackendFileSystemIsMounted storage: Create helper for virStorageBackendLogicalFindPoolSources storage: Add helper to compare logical pool def against pvs output src/storage/storage_backend_fs.c | 73 ++++++++++++----- src/storage/storage_backend_logical.c | 148 +++++++++++++++++++++++++++++----- 2 files changed, 182 insertions(+), 39 deletions(-) -- 2.5.0

3 15

[libvirt] [PATCH 0/6] Add build options to pool create[-as] and start
by John Ferlan 17 Dec '15

17 Dec '15

https://bugzilla.redhat.com/show_bug.cgi?id=830056 Add the capability to build the pool during pool creation or startup NOTE: If patches 2-4 regarding usage of #define for repeated command options are acceptible, I can start doing the same for other virsh commands. I just found it easier and I think better than the need to continually cut-n-paste for each command option entry. If it's preferred to keep cut-n-paste'ing, then I can modify patch 5 & 6 to follow that model. NOTE: Based loosely on patches posted quite a while ago by Osier Yang before he left Red Hat, see: (Cover) https://www.redhat.com/archives/libvir-list/2012-July/msg00494.html and (libvirt API, storage driver changes) https://www.redhat.com/archives/libvir-list/2012-July/msg01328.html and (virsh changes) https://www.redhat.com/archives/libvir-list/2012-July/msg00497.html John Ferlan (6): storage: Add flags to allow building pool during create processing virsh: Create macro for "pool" option virsh: Create macro for "file" option virsh: Create macro for "overwrite" and no-overwrite" options virsh: Create a macro for pool-define-as and pool-create-as options virsh: Add build flags to pool-create[-as] and pool-start include/libvirt/libvirt-storage.h | 16 +- src/libvirt-storage.c | 4 +- src/storage/storage_driver.c | 42 ++++- tools/virsh-pool.c | 360 ++++++++++++++++++++++---------------- tools/virsh.pod | 25 ++- 5 files changed, 288 insertions(+), 159 deletions(-) -- 2.5.0

3 23

[libvirt] [PATCHv2 0/2] notify about reverting to snapshot
by Dmitry Andreev 16 Dec '15

16 Dec '15

Reverting to snapshot may change domain configuration but there will be no events about that. Lack of the event become a problem for virt-manager https://bugzilla.redhat.com/show_bug.cgi?id=1081148 This patch-set introduces new event and emits it in qemuDomainRevertToSnapshot. v2: [2/2] Reworked. John noted that in some error cases I send 'defined' event without changes in configuration. Dmitry Andreev (2): Introduce new VIR_DOMAIN_EVENT_DEFINED_FROM_SNAPSHOT event qemu: emit 'defined' event after reverted to snapshot examples/object-events/event-test.c | 2 ++ include/libvirt/libvirt-domain.h | 1 + src/qemu/qemu_driver.c | 13 +++++++++++-- tools/virsh-domain.c | 3 ++- 4 files changed, 16 insertions(+), 3 deletions(-) -- 1.8.3.1

3 7

[libvirt] [PATCH] qemu: Search all nodes for shared memory access
by Martin Kletzander 16 Dec '15

16 Dec '15

In commit 686eb7a24f7d, the break was not considered part of the condition, hence breaking after first node when searching. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- Pushed as 'trivial' and also John ACKed it earlier. src/qemu/qemu_process.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index ba8dfebd1357..f2740687f655 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -4781,9 +4781,10 @@ qemuProcessLaunch(virConnectPtr conn, if (!shmem && vm->def->mem.nhugepages) { for (i = 0; i < virDomainNumaGetNodeCount(vm->def->numa); i++) { if (virDomainNumaGetNodeMemoryAccessMode(vm->def->numa, i) == - VIR_NUMA_MEM_ACCESS_SHARED) + VIR_NUMA_MEM_ACCESS_SHARED) { shmem = true; - break; + break; + } } } -- 2.6.4

1 0

[libvirt] [PATCH LIBVIRT] libxl: Use libxentoollog in preference to libxenctrl if available.
by Ian Campbell 16 Dec '15

16 Dec '15

Upstream Xen is in the process of splitting the (stable API) xtl_* interfaces out from the (unstable API) libxenctrl library and into a new (stable API) libxentoollog. In order to be compatible with Xen both before and after this transition check for xtl_createlogger_stdiostream in a libxentoollog library and use it if present. If it is not present assume it is in libxenctrl. Compile tested on Xen 4.6 and a development tree with the split in place. Signed-off-by: Ian Campbell <ian.campbell(a)citrix.com> --- I'm waiting on applying the upstream change until downstreams are prepared for this. The latest upstream patch is http://lists.xen.org/archives/html/xen-devel/2015-12/msg00454.html which had to be reverted because I had somehow not properly checked if libvirt used this interface http://lists.xen.org/archives/html/xen-devel/2015-12/msg01153.html It might be nice to get this into 1.3.0 so that supports Xen 4.7 out of the box? Not sure what the libvirt stable backport policy is but it might also be good to eventually consider it for that? --- configure.ac | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/configure.ac b/configure.ac index 98cf210..b641cc7 100644 --- a/configure.ac +++ b/configure.ac @@ -883,7 +883,6 @@ if test "$with_libxl" != "no" ; then PKG_CHECK_MODULES([LIBXL], [xenlight], [ LIBXL_FIRMWARE_DIR=`$PKG_CONFIG --variable xenfirmwaredir xenlight` LIBXL_EXECBIN_DIR=`$PKG_CONFIG --variable libexec_bin xenlight` - LIBXL_LIBS="$LIBXL_LIBS -lxenctrl" with_libxl=yes ], [LIBXL_FOUND=no]) if test "$LIBXL_FOUND" = "no"; then @@ -896,7 +895,7 @@ if test "$with_libxl" != "no" ; then LIBS="$LIBS $LIBXL_LIBS" AC_CHECK_LIB([xenlight], [libxl_ctx_alloc], [ with_libxl=yes - LIBXL_LIBS="$LIBXL_LIBS -lxenlight -lxenctrl" + LIBXL_LIBS="$LIBXL_LIBS -lxenlight" ],[ if test "$with_libxl" = "yes"; then fail=1 @@ -924,6 +923,14 @@ if test "$with_libxl" = "yes"; then if test "x$LIBXL_EXECBIN_DIR" != "x"; then AC_DEFINE_UNQUOTED([LIBXL_EXECBIN_DIR], ["$LIBXL_EXECBIN_DIR"], [directory containing Xen libexec binaries]) fi + dnl Check if the xtl_* infrastructure is in libxentoollog + dnl (since Xen 4.7) if not then assume it is in libxenctrl + dnl (as it was for 4.6 and earler) + AC_CHECK_LIB([xentoollog], [xtl_createlogger_stdiostream], [ + LIBXL_LIBS="$LIBXL_LIBS -lxentoollog" + ],[ + LIBXL_LIBS="$LIBXL_LIBS -lxenctrl" + ]) fi AM_CONDITIONAL([WITH_LIBXL], [test "$with_libxl" = "yes"]) -- 2.1.4

3 4

[libvirt] [PATCH 0/5] UEFI loader NVRAM image in Qcow2 format
by Dmitry Andreev 16 Dec '15

16 Dec '15

Libvirt allows to specify a path to an image file that will be used as a var storage for UEFI firmware. <nvram>/var/lib/libvirt/nvram/guest_VARS.fd</nvram> By default this image is created as a copy of a master image. The master image and it's copy are stored in 'raw' format. Qemu isn't able to create snapshot for an image in 'raw' format. That makes snapshotting impossible for any UEFI configuration. If the image with UEFI nvram is converted to Qcow2 format (and qemu command is modified appropriately), snapshotting works fine. In the patch-set I introduce 'format' attribute for nvram tag that specifies nvram image file format. The patch-set doesn't contains docs and test in v1. Dmitry Andreev (5): storage: split virStorageBackendCreateExecCommand in two functions storage: refactor: split out image creating tool search function storage: add new function virStorageBackendConvertImage conf: add 'format' attribute to domain/os/nvram element qemu: add support for os.nvram 'format' attribute docs/schemas/domaincommon.rng | 8 ++ src/conf/domain_conf.c | 18 +++++ src/conf/domain_conf.h | 1 + src/qemu/qemu_command.c | 8 +- src/qemu/qemu_process.c | 17 ++++ src/storage/storage_backend.c | 181 ++++++++++++++++++++++++++---------------- src/storage/storage_backend.h | 5 ++ 7 files changed, 167 insertions(+), 71 deletions(-) -- 1.8.3.1

3 9

[libvirt] [PATCH v2] vz: implementation of domainCreateXML callback
by Mikhail Feoktistov 16 Dec '15

16 Dec '15

--- diff from v1. Remove call of vzDomainSuspend() in case of VIR_DOMAIN_START_PAUSED flag is set. Now we don't support to create instance in stopped state. src/vz/vz_driver.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/src/vz/vz_driver.c b/src/vz/vz_driver.c index ea1090a..4498e01 100644 --- a/src/vz/vz_driver.c +++ b/src/vz/vz_driver.c @@ -979,6 +979,29 @@ vzDomainUndefine(virDomainPtr domain) return vzDomainUndefineFlags(domain, 0); } +static virDomainPtr +vzDomainCreateXML(virConnectPtr conn, + const char *xml, + unsigned int flags) +{ + virDomainPtr domain; + int ret; + + virCheckFlags(0, NULL); + + domain = vzDomainDefineXMLFlags(conn, xml, 0); + if (domain == NULL) + return domain; + + ret = vzDomainCreate(domain); + if (ret != 0) { + vzDomainUndefine(domain); + return NULL; + } + + return domain; +} + static int vzDomainHasManagedSaveImage(virDomainPtr domain, unsigned int flags) { @@ -1493,6 +1516,7 @@ static virHypervisorDriver vzDriver = { .domainShutdown = vzDomainShutdown, /* 0.10.0 */ .domainCreate = vzDomainCreate, /* 0.10.0 */ .domainCreateWithFlags = vzDomainCreateWithFlags, /* 1.2.10 */ + .domainCreateXML = vzDomainCreateXML, /* 1.2.22 */ .domainReboot = vzDomainReboot, /* 1.3.0 */ .domainDefineXML = vzDomainDefineXML, /* 0.10.0 */ .domainDefineXMLFlags = vzDomainDefineXMLFlags, /* 1.2.12 */ -- 1.8.3.1

2 2

[libvirt] [PATCH v3] pci: Use virPCIDeviceAddress in virPCIDevice
by Andrea Bolognani 16 Dec '15

16 Dec '15

Instead of replicating the information (domain, bus, slot, function) inside the virPCIDevice structure, use the already-existing virPCIDeviceAddress structure. For users of the module, this means that the object returned by virPCIDeviceGetAddress() can no longer be NULL and must no longer be freed by the caller. --- Changes in v3: * don't check the return value of virPCIDeviceGetAddress(), it can no longer be NULL * don't use a variable to store the device address if it's only going to be used a single time Changes in v2: * use virPCIDeviceAddress instead of virPCIDeviceAddressPtr to avoid extra allocations src/util/virhostdev.c | 20 ++--------- src/util/virpci.c | 97 +++++++++++++++++++++++---------------------------- 2 files changed, 47 insertions(+), 70 deletions(-) diff --git a/src/util/virhostdev.c b/src/util/virhostdev.c index de029a0..4065535 100644 --- a/src/util/virhostdev.c +++ b/src/util/virhostdev.c @@ -585,15 +585,12 @@ virHostdevPreparePCIDevices(virHostdevManagerPtr hostdev_mgr, goto cleanup; } - VIR_FREE(devAddr); - if (!(devAddr = virPCIDeviceGetAddress(dev))) - goto cleanup; - /* The device is in use by other active domain if * the dev is in list activePCIHostdevs. VFIO devices * belonging to same iommu group can't be shared * across guests. */ + devAddr = virPCIDeviceGetAddress(dev); if (usesVfio) { if (virPCIDeviceAddressIOMMUGroupIterate(devAddr, virHostdevIsPCINodeDeviceUsed, @@ -728,7 +725,6 @@ virHostdevPreparePCIDevices(virHostdevManagerPtr hostdev_mgr, virObjectUnlock(hostdev_mgr->activePCIHostdevs); virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnref(pcidevs); - VIR_FREE(devAddr); return ret; } @@ -1558,7 +1554,6 @@ int virHostdevPCINodeDeviceDetach(virHostdevManagerPtr hostdev_mgr, virPCIDevicePtr pci) { - virPCIDeviceAddressPtr devAddr = NULL; struct virHostdevIsPCINodeDeviceUsedData data = { hostdev_mgr, NULL, false }; int ret = -1; @@ -1566,10 +1561,7 @@ virHostdevPCINodeDeviceDetach(virHostdevManagerPtr hostdev_mgr, virObjectLock(hostdev_mgr->activePCIHostdevs); virObjectLock(hostdev_mgr->inactivePCIHostdevs); - if (!(devAddr = virPCIDeviceGetAddress(pci))) - goto out; - - if (virHostdevIsPCINodeDeviceUsed(devAddr, &data)) + if (virHostdevIsPCINodeDeviceUsed(virPCIDeviceGetAddress(pci), &data)) goto out; if (virPCIDeviceDetach(pci, hostdev_mgr->activePCIHostdevs, @@ -1581,7 +1573,6 @@ virHostdevPCINodeDeviceDetach(virHostdevManagerPtr hostdev_mgr, out: virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnlock(hostdev_mgr->activePCIHostdevs); - VIR_FREE(devAddr); return ret; } @@ -1589,7 +1580,6 @@ int virHostdevPCINodeDeviceReAttach(virHostdevManagerPtr hostdev_mgr, virPCIDevicePtr pci) { - virPCIDeviceAddressPtr devAddr = NULL; struct virHostdevIsPCINodeDeviceUsedData data = {hostdev_mgr, NULL, false}; int ret = -1; @@ -1597,10 +1587,7 @@ virHostdevPCINodeDeviceReAttach(virHostdevManagerPtr hostdev_mgr, virObjectLock(hostdev_mgr->activePCIHostdevs); virObjectLock(hostdev_mgr->inactivePCIHostdevs); - if (!(devAddr = virPCIDeviceGetAddress(pci))) - goto out; - - if (virHostdevIsPCINodeDeviceUsed(devAddr, &data)) + if (virHostdevIsPCINodeDeviceUsed(virPCIDeviceGetAddress(pci), &data)) goto out; virPCIDeviceReattachInit(pci); @@ -1613,7 +1600,6 @@ virHostdevPCINodeDeviceReAttach(virHostdevManagerPtr hostdev_mgr, out: virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnlock(hostdev_mgr->activePCIHostdevs); - VIR_FREE(devAddr); return ret; } diff --git a/src/util/virpci.c b/src/util/virpci.c index 7ca3fef..6f0cb8c 100644 --- a/src/util/virpci.c +++ b/src/util/virpci.c @@ -56,10 +56,7 @@ VIR_ENUM_IMPL(virPCIELinkSpeed, VIR_PCIE_LINK_SPEED_LAST, "", "2.5", "5", "8") struct _virPCIDevice { - unsigned int domain; - unsigned int bus; - unsigned int slot; - unsigned int function; + virPCIDeviceAddress address; char name[PCI_ADDR_LEN]; /* domain:bus:slot.function */ char id[PCI_ID_LEN]; /* product vendor */ @@ -655,10 +652,10 @@ virPCIDeviceSharesBusWithActive(virPCIDevicePtr dev, virPCIDevicePtr check, void virPCIDeviceList *inactiveDevs = data; /* Different domain, different bus, or simply identical device */ - if (dev->domain != check->domain || - dev->bus != check->bus || - (dev->slot == check->slot && - dev->function == check->function)) + if (dev->address.domain != check->address.domain || + dev->address.bus != check->address.bus || + (dev->address.slot == check->address.slot && + dev->address.function == check->address.function)) return 0; /* same bus, but inactive, i.e. about to be assigned to guest */ @@ -689,7 +686,7 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) int ret = 0; int fd; - if (dev->domain != check->domain) + if (dev->address.domain != check->address.domain) return 0; if ((fd = virPCIDeviceConfigOpen(check, false)) < 0) @@ -713,7 +710,7 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) /* if the secondary bus exactly equals the device's bus, then we found * the direct parent. No further work is necessary */ - if (dev->bus == secondary) { + if (dev->address.bus == secondary) { ret = 1; goto cleanup; } @@ -722,10 +719,12 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) * In this case, what we need to do is look for the "best" match; i.e. * the most restrictive match that still satisfies all of the conditions. */ - if (dev->bus > secondary && dev->bus <= subordinate) { + if (dev->address.bus > secondary && dev->address.bus <= subordinate) { if (*best == NULL) { - *best = virPCIDeviceNew(check->domain, check->bus, check->slot, - check->function); + *best = virPCIDeviceNew(check->address.domain, + check->address.bus, + check->address.slot, + check->address.function); if (*best == NULL) { ret = -1; goto cleanup; @@ -745,8 +744,10 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) if (secondary > best_secondary) { virPCIDeviceFree(*best); - *best = virPCIDeviceNew(check->domain, check->bus, check->slot, - check->function); + *best = virPCIDeviceNew(check->address.domain, + check->address.bus, + check->address.slot, + check->address.function); if (*best == NULL) { ret = -1; goto cleanup; @@ -970,7 +971,7 @@ virPCIDeviceReset(virPCIDevicePtr dev, ret = virPCIDeviceTryPowerManagementReset(dev, fd); /* Bus reset is not an option with the root bus */ - if (ret < 0 && dev->bus != 0) + if (ret < 0 && dev->address.bus != 0) ret = virPCIDeviceTrySecondaryBusReset(dev, fd, inactiveDevs); if (ret < 0) { @@ -1483,8 +1484,8 @@ virPCIDeviceWaitForCleanup(virPCIDevicePtr dev, const char *matcher) virStrToLong_ui(tmp + 1, &tmp, 16, &function) < 0 || *tmp != '\n') continue; - if (domain != dev->domain || bus != dev->bus || slot != dev->slot || - function != dev->function) + if (domain != dev->address.domain || bus != dev->address.bus || + slot != dev->address.slot || function != dev->address.function) continue; in_matching_device = true; match_depth = strspn(line, " "); @@ -1560,17 +1561,16 @@ virPCIDeviceNew(unsigned int domain, if (VIR_ALLOC(dev) < 0) return NULL; - dev->domain = domain; - dev->bus = bus; - dev->slot = slot; - dev->function = function; + dev->address.domain = domain; + dev->address.bus = bus; + dev->address.slot = slot; + dev->address.function = function; if (snprintf(dev->name, sizeof(dev->name), "%.4x:%.2x:%.2x.%.1x", - dev->domain, dev->bus, dev->slot, - dev->function) >= sizeof(dev->name)) { + domain, bus, slot, function) >= sizeof(dev->name)) { virReportError(VIR_ERR_INTERNAL_ERROR, _("dev->name buffer overflow: %.4x:%.2x:%.2x.%.1x"), - dev->domain, dev->bus, dev->slot, dev->function); + domain, bus, slot, function); goto error; } if (virAsprintf(&dev->path, PCI_SYSFS "devices/%s/config", @@ -1661,25 +1661,14 @@ virPCIDeviceFree(virPCIDevicePtr dev) * @dev: device to get address from * * Take a PCI device on input and return its PCI address. The - * caller must free the returned value when no longer needed. + * returned object is owned by the device and must not be freed. * - * Returns NULL on failure, the device address on success. + * Returns: a pointer to the address, which can never be NULL. */ virPCIDeviceAddressPtr virPCIDeviceGetAddress(virPCIDevicePtr dev) { - - virPCIDeviceAddressPtr pciAddrPtr; - - if (!dev || (VIR_ALLOC(pciAddrPtr) < 0)) - return NULL; - - pciAddrPtr->domain = dev->domain; - pciAddrPtr->bus = dev->bus; - pciAddrPtr->slot = dev->slot; - pciAddrPtr->function = dev->function; - - return pciAddrPtr; + return &(dev->address); } const char * @@ -1890,12 +1879,14 @@ virPCIDeviceListFindIndex(virPCIDeviceListPtr list, virPCIDevicePtr dev) { size_t i; - for (i = 0; i < list->count; i++) - if (list->devs[i]->domain == dev->domain && - list->devs[i]->bus == dev->bus && - list->devs[i]->slot == dev->slot && - list->devs[i]->function == dev->function) + for (i = 0; i < list->count; i++) { + virPCIDevicePtr other = list->devs[i]; + if (other->address.domain == dev->address.domain && + other->address.bus == dev->address.bus && + other->address.slot == dev->address.slot && + other->address.function == dev->address.function) return i; + } return -1; } @@ -1910,10 +1901,11 @@ virPCIDeviceListFindByIDs(virPCIDeviceListPtr list, size_t i; for (i = 0; i < list->count; i++) { - if (list->devs[i]->domain == domain && - list->devs[i]->bus == bus && - list->devs[i]->slot == slot && - list->devs[i]->function == function) + virPCIDevicePtr other = list->devs[i]; + if (other->address.domain == domain && + other->address.bus == bus && + other->address.slot == slot && + other->address.function == function) return list->devs[i]; } return NULL; @@ -1944,7 +1936,8 @@ int virPCIDeviceFileIterate(virPCIDevicePtr dev, int direrr; if (virAsprintf(&pcidir, "/sys/bus/pci/devices/%04x:%02x:%02x.%x", - dev->domain, dev->bus, dev->slot, dev->function) < 0) + dev->address.domain, dev->address.bus, + dev->address.slot, dev->address.function) < 0) goto cleanup; if (!(dir = opendir(pcidir))) { @@ -2074,13 +2067,11 @@ virPCIDeviceListPtr virPCIDeviceGetIOMMUGroupList(virPCIDevicePtr dev) { virPCIDeviceListPtr groupList = virPCIDeviceListNew(); - virPCIDeviceAddress devAddr = { dev->domain, dev->bus, - dev->slot, dev->function }; if (!groupList) goto error; - if (virPCIDeviceAddressIOMMUGroupIterate(&devAddr, + if (virPCIDeviceAddressIOMMUGroupIterate(&(dev->address), virPCIDeviceGetIOMMUGroupAddOne, groupList) < 0) goto error; @@ -2294,7 +2285,7 @@ virPCIDeviceIsBehindSwitchLackingACS(virPCIDevicePtr dev) * into play since devices on the root bus can't P2P without going * through the root IOMMU. */ - if (dev->bus == 0) { + if (dev->address.bus == 0) { return 0; } else { virReportError(VIR_ERR_INTERNAL_ERROR, -- 2.5.0

2 2

[libvirt] [PATCH v3 0/8] libxl: domain statistics support
by Joao Martins 16 Dec '15

16 Dec '15

Hey, [ Apologies for resending this series before receiving comments, but I found two issues (namely on patch 3 and 4) and thus I am reposting it as v3 having those fixed. ] This series bring support for various statistics about domains regarding CPU, Memory, Network Interfaces, Block and Jobs. Not all of the statistics are implemented: qdisk support is missing in this series and some of the memory statistics aren't available. With this series we further implement 7 more functions of libvirt APIs. It is organized as follows: * Patch 1, 2: implements cpu/memory statistics. * Patch 3: implements network statistics * Patch 4: adds helper method virDiskNameParse as an extension to virDiskNameToIndex (New) * Patch 5: VBD block statistics. QDisk will follow up in a separate series regarding QEMU monitor integration. * Patch 6: implement fetching all domain statistics * Patch 7, 8: implements Job information statistics. Overall it looks big but 70% of the patch is due to #5 and #6 but doesn't add necessarily more complexity to the driver I believe. Patch #7 and #8 are of special importance because GetJobInfo and GetJobStats are now used in Openstack Kilo to monitor live-migration progress. These two patches together with the p2p migration I sent earlier let us sucessfully live-migrate with Openstack Kilo. Furthermore with this series we get to support nova diagnostics. Tested this series on 4.4.3 and 4.5 setups plus Openstack Kilo. Individual patches contain the changelog and comments addressed since v1. Thanks! Joao Martins (8): libxl: implement virDomainGetCPUStats libxl: implement virDomainMemorystats libxl: implement virDomainInterfaceStats util: add virDiskNameParse to handle disk and partition idx libxl: implement virDomainBlockStats libxl: implement virConnectGetAllDomainStats libxl: implement virDomainGetJobInfo libxl: implement virDomainGetJobStats configure.ac | 2 +- src/libvirt_private.syms | 1 + src/libxl/libxl_domain.c | 54 +++ src/libxl/libxl_domain.h | 6 + src/libxl/libxl_driver.c | 960 +++++++++++++++++++++++++++++++++++++++++++++++ src/util/virutil.c | 41 +- src/util/virutil.h | 1 + tests/utiltest.c | 56 +++ 8 files changed, 1116 insertions(+), 5 deletions(-) -- 2.1.4

3 29

[libvirt] [PATCH v2] pci: Use virPCIDeviceAddress in virPCIDevice
by Andrea Bolognani 16 Dec '15

16 Dec '15

Instead of replicating the information (domain, bus, slot, function) inside the virPCIDevice structure, use the already-existing virPCIDeviceAddress structure. Outside of the module, the only visible change is that the return value of virPCIDeviceGetAddress() must no longer be freed by the caller. --- Changes in v2: * use virPCIDeviceAddress instead of virPCIDeviceAddressPtr to avoid extra allocations src/util/virhostdev.c | 4 --- src/util/virpci.c | 95 +++++++++++++++++++++++---------------------------- 2 files changed, 43 insertions(+), 56 deletions(-) diff --git a/src/util/virhostdev.c b/src/util/virhostdev.c index de029a0..173ac58 100644 --- a/src/util/virhostdev.c +++ b/src/util/virhostdev.c @@ -585,7 +585,6 @@ virHostdevPreparePCIDevices(virHostdevManagerPtr hostdev_mgr, goto cleanup; } - VIR_FREE(devAddr); if (!(devAddr = virPCIDeviceGetAddress(dev))) goto cleanup; @@ -728,7 +727,6 @@ virHostdevPreparePCIDevices(virHostdevManagerPtr hostdev_mgr, virObjectUnlock(hostdev_mgr->activePCIHostdevs); virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnref(pcidevs); - VIR_FREE(devAddr); return ret; } @@ -1581,7 +1579,6 @@ virHostdevPCINodeDeviceDetach(virHostdevManagerPtr hostdev_mgr, out: virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnlock(hostdev_mgr->activePCIHostdevs); - VIR_FREE(devAddr); return ret; } @@ -1613,7 +1610,6 @@ virHostdevPCINodeDeviceReAttach(virHostdevManagerPtr hostdev_mgr, out: virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnlock(hostdev_mgr->activePCIHostdevs); - VIR_FREE(devAddr); return ret; } diff --git a/src/util/virpci.c b/src/util/virpci.c index bececb5..bb874ac 100644 --- a/src/util/virpci.c +++ b/src/util/virpci.c @@ -56,10 +56,7 @@ VIR_ENUM_IMPL(virPCIELinkSpeed, VIR_PCIE_LINK_SPEED_LAST, "", "2.5", "5", "8") struct _virPCIDevice { - unsigned int domain; - unsigned int bus; - unsigned int slot; - unsigned int function; + virPCIDeviceAddress address; char name[PCI_ADDR_LEN]; /* domain:bus:slot.function */ char id[PCI_ID_LEN]; /* product vendor */ @@ -655,10 +652,10 @@ virPCIDeviceSharesBusWithActive(virPCIDevicePtr dev, virPCIDevicePtr check, void virPCIDeviceList *inactiveDevs = data; /* Different domain, different bus, or simply identical device */ - if (dev->domain != check->domain || - dev->bus != check->bus || - (dev->slot == check->slot && - dev->function == check->function)) + if (dev->address.domain != check->address.domain || + dev->address.bus != check->address.bus || + (dev->address.slot == check->address.slot && + dev->address.function == check->address.function)) return 0; /* same bus, but inactive, i.e. about to be assigned to guest */ @@ -689,7 +686,7 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) int ret = 0; int fd; - if (dev->domain != check->domain) + if (dev->address.domain != check->address.domain) return 0; if ((fd = virPCIDeviceConfigOpen(check, false)) < 0) @@ -713,7 +710,7 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) /* if the secondary bus exactly equals the device's bus, then we found * the direct parent. No further work is necessary */ - if (dev->bus == secondary) { + if (dev->address.bus == secondary) { ret = 1; goto cleanup; } @@ -722,10 +719,12 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) * In this case, what we need to do is look for the "best" match; i.e. * the most restrictive match that still satisfies all of the conditions. */ - if (dev->bus > secondary && dev->bus <= subordinate) { + if (dev->address.bus > secondary && dev->address.bus <= subordinate) { if (*best == NULL) { - *best = virPCIDeviceNew(check->domain, check->bus, check->slot, - check->function); + *best = virPCIDeviceNew(check->address.domain, + check->address.bus, + check->address.slot, + check->address.function); if (*best == NULL) { ret = -1; goto cleanup; @@ -745,8 +744,10 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) if (secondary > best_secondary) { virPCIDeviceFree(*best); - *best = virPCIDeviceNew(check->domain, check->bus, check->slot, - check->function); + *best = virPCIDeviceNew(check->address.domain, + check->address.bus, + check->address.slot, + check->address.function); if (*best == NULL) { ret = -1; goto cleanup; @@ -970,7 +971,7 @@ virPCIDeviceReset(virPCIDevicePtr dev, ret = virPCIDeviceTryPowerManagementReset(dev, fd); /* Bus reset is not an option with the root bus */ - if (ret < 0 && dev->bus != 0) + if (ret < 0 && dev->address.bus != 0) ret = virPCIDeviceTrySecondaryBusReset(dev, fd, inactiveDevs); if (ret < 0) { @@ -1483,8 +1484,8 @@ virPCIDeviceWaitForCleanup(virPCIDevicePtr dev, const char *matcher) virStrToLong_ui(tmp + 1, &tmp, 16, &function) < 0 || *tmp != '\n') continue; - if (domain != dev->domain || bus != dev->bus || slot != dev->slot || - function != dev->function) + if (domain != dev->address.domain || bus != dev->address.bus || + slot != dev->address.slot || function != dev->address.function) continue; in_matching_device = true; match_depth = strspn(line, " "); @@ -1560,17 +1561,16 @@ virPCIDeviceNew(unsigned int domain, if (VIR_ALLOC(dev) < 0) return NULL; - dev->domain = domain; - dev->bus = bus; - dev->slot = slot; - dev->function = function; + dev->address.domain = domain; + dev->address.bus = bus; + dev->address.slot = slot; + dev->address.function = function; if (snprintf(dev->name, sizeof(dev->name), "%.4x:%.2x:%.2x.%.1x", - dev->domain, dev->bus, dev->slot, - dev->function) >= sizeof(dev->name)) { + domain, bus, slot, function) >= sizeof(dev->name)) { virReportError(VIR_ERR_INTERNAL_ERROR, _("dev->name buffer overflow: %.4x:%.2x:%.2x.%.1x"), - dev->domain, dev->bus, dev->slot, dev->function); + domain, bus, slot, function); goto error; } if (virAsprintf(&dev->path, PCI_SYSFS "devices/%s/config", @@ -1661,25 +1661,14 @@ virPCIDeviceFree(virPCIDevicePtr dev) * @dev: device to get address from * * Take a PCI device on input and return its PCI address. The - * caller must free the returned value when no longer needed. + * returned object is owned by the device and must not be freed. * * Returns NULL on failure, the device address on success. */ virPCIDeviceAddressPtr virPCIDeviceGetAddress(virPCIDevicePtr dev) { - - virPCIDeviceAddressPtr pciAddrPtr; - - if (!dev || (VIR_ALLOC(pciAddrPtr) < 0)) - return NULL; - - pciAddrPtr->domain = dev->domain; - pciAddrPtr->bus = dev->bus; - pciAddrPtr->slot = dev->slot; - pciAddrPtr->function = dev->function; - - return pciAddrPtr; + return &(dev->address); } const char * @@ -1890,12 +1879,14 @@ virPCIDeviceListFindIndex(virPCIDeviceListPtr list, virPCIDevicePtr dev) { size_t i; - for (i = 0; i < list->count; i++) - if (list->devs[i]->domain == dev->domain && - list->devs[i]->bus == dev->bus && - list->devs[i]->slot == dev->slot && - list->devs[i]->function == dev->function) + for (i = 0; i < list->count; i++) { + virPCIDevicePtr other = list->devs[i]; + if (other->address.domain == dev->address.domain && + other->address.bus == dev->address.bus && + other->address.slot == dev->address.slot && + other->address.function == dev->address.function) return i; + } return -1; } @@ -1910,10 +1901,11 @@ virPCIDeviceListFindByIDs(virPCIDeviceListPtr list, size_t i; for (i = 0; i < list->count; i++) { - if (list->devs[i]->domain == domain && - list->devs[i]->bus == bus && - list->devs[i]->slot == slot && - list->devs[i]->function == function) + virPCIDevicePtr other = list->devs[i]; + if (other->address.domain == domain && + other->address.bus == bus && + other->address.slot == slot && + other->address.function == function) return list->devs[i]; } return NULL; @@ -1944,7 +1936,8 @@ int virPCIDeviceFileIterate(virPCIDevicePtr dev, int direrr; if (virAsprintf(&pcidir, "/sys/bus/pci/devices/%04x:%02x:%02x.%x", - dev->domain, dev->bus, dev->slot, dev->function) < 0) + dev->address.domain, dev->address.bus, + dev->address.slot, dev->address.function) < 0) goto cleanup; if (!(dir = opendir(pcidir))) { @@ -2074,13 +2067,11 @@ virPCIDeviceListPtr virPCIDeviceGetIOMMUGroupList(virPCIDevicePtr dev) { virPCIDeviceListPtr groupList = virPCIDeviceListNew(); - virPCIDeviceAddress devAddr = { dev->domain, dev->bus, - dev->slot, dev->function }; if (!groupList) goto error; - if (virPCIDeviceAddressIOMMUGroupIterate(&devAddr, + if (virPCIDeviceAddressIOMMUGroupIterate(&(dev->address), virPCIDeviceGetIOMMUGroupAddOne, groupList) < 0) goto error; @@ -2294,7 +2285,7 @@ virPCIDeviceIsBehindSwitchLackingACS(virPCIDevicePtr dev) * into play since devices on the root bus can't P2P without going * through the root IOMMU. */ - if (dev->bus == 0) { + if (dev->address.bus == 0) { return 0; } else { virReportError(VIR_ERR_INTERNAL_ERROR, -- 2.5.0

2 3

[libvirt] [PATCH 0/2] notify about reverting to snapshot
by Dmitry Andreev 16 Dec '15

16 Dec '15

Reverting to snapshot may change domain configuration but there will be no events about that. Lack of the event become a problem for virt-manager https://bugzilla.redhat.com/show_bug.cgi?id=1081148 This patch-set introduces new event and emits it in qemuDomainRevertToSnapshot. Dmitry Andreev (2): Introduce new VIR_DOMAIN_EVENT_DEFINED_FROM_SNAPSHOT event qemu: emit 'defined' event after reverted to snapshot examples/object-events/event-test.c | 2 ++ include/libvirt/libvirt-domain.h | 1 + src/qemu/qemu_driver.c | 5 +++++ tools/virsh-domain.c | 3 ++- 4 files changed, 10 insertions(+), 1 deletion(-) -- 1.8.3.1

2 5

[libvirt] libvirt qcow2 internal snapshot without vm state
by Vasiliy Tolstov 16 Dec '15

16 Dec '15

Hi! I'm try to grep sources, but can't understand, how libvirt create internal snapshot in the same qcow2 file without memory state? As i understand libvirt via qmp/hmp runs savevm command via qemu. But i don't understand how libvirt skips memory state in this case? -- Vasiliy Tolstov, e-mail: v.tolstov(a)selfip.ru

2 2

Re: [libvirt] [Qemu-devel] [Patch V2 1/2] x86, mce: Basic support to add LMCE support to QEMU
by Eduardo Habkost 15 Dec '15

15 Dec '15

On Mon, Dec 14, 2015 at 07:17:27PM -0500, Raj, Ashok wrote: > On Mon, Dec 14, 2015 at 11:37:16PM +0100, Borislav Petkov wrote: > > On Mon, Dec 14, 2015 at 02:11:46PM -0500, Raj, Ashok wrote: > > > This is mostly harmless.. since the MCG_CAP space is shared and has no > > > conflict between vendors. Also just the CAP being set has no effect. > > > > Of course it does - we check SER_P in machine_check_poll() and when > > I emulate an AMD guest and inject errors into it, error handling is > > obviously wrong, see: > > > > https://lkml.kernel.org/r/20151123150355.GE5134@pd.tnic > > > > I can see how this hurts.. since the poller isn't doing cpu model specific > stuff..? > > in the LMCE case, even if you advertise MCG_LMCE_P in MCG_CAP, the guest kernel > wont call intel_init_lmce() only from mce_intel.c.. so the same problem > won't happen. > > but the issue Eduardo mentioned seems like the following. > > New QEMU_LMCE + New KVM_LMCE + New_GUEST_LMCE - No problem > > but if you were to migrage the Guest_LMCE to a non-LMCE supported KVM host > we could run into an issue.. > > is this the compatibility issue that you were looking to fix Eduardo? If I understood you correctly, yes. Also, note that currently kvm_arch_init_vcpu() simply warns about missing capabilities, instead of preventing the VM from running/migrating (as it should). We need to change that, and figure out a good way to report "feature FOO can't be enabled in this host" errors to management software[1]. The main problem is that we don't even have a QMP console available anymore if machine initialization is aborted. CCing libvir-list so they get in the loop. [1] This is similar to what we need for CPUID checks, but the new MCE feature means we need something more generic (that just reports QOM property names, probably?) -- Eduardo

1 0

[libvirt] [PATCHv2 0/3] security: misc cleanups
by Ján Tomko 15 Dec '15

15 Dec '15

v1: https://www.redhat.com/archives/libvir-list/2015-November/msg00781.html new in v2: * split by security driver * more functions renamed Ján Tomko (3): security_dac: remove extra Security from function names security_selinux: remove extra Security from function names security_stack: : remove extra security from function names src/security/security_dac.c | 242 +++++++++++++------------- src/security/security_selinux.c | 375 ++++++++++++++++++++-------------------- src/security/security_stack.c | 68 ++++---- 3 files changed, 338 insertions(+), 347 deletions(-) -- 2.4.6

2 5

[libvirt] volume download/upload but for snapshots
by Vasiliy Tolstov 15 Dec '15

15 Dec '15

Hi! I'm happy with libvirt volume download/upload ability, but also i need to download snapshot, what is the best way to do that? Define volume from snapshot? Or write libvirt function for this? -- Vasiliy Tolstov, e-mail: v.tolstov(a)selfip.ru

2 3

[libvirt] [PATCH v2 0/2] Add support for zero-write detection
by Martin Kletzander 15 Dec '15

15 Dec '15

QEMU supports detect-zeroes option since version 2.1, but we never added support for it in libvirt. If was requested by Vasiliy Tolstov in the list, so I just added it. There are two discussions to be had, optionally. One is to decide whether we should disable detect_zeros='unmap' if discard is not set to 'unmap', but this is getting very hypervisor-specific, so I just documented the behaviour. The other one is the naming. I described why I made the decision for "zeros" instead of "zeroes" the decision in the patch, but I have no problem changing it to what others like better. v2: - format detect_zeroes on the command line instead of detect_zeros v1: - https://www.redhat.com/archives/libvir-list/2015-December/msg00484.html Martin Kletzander (2): conf: Add support of zero-detection for disks qemu: Add support for zero-detection writes docs/formatdomain.html.in | 10 ++++++ docs/schemas/domaincommon.rng | 12 +++++++ src/conf/domain_conf.c | 23 +++++++++++++- src/conf/domain_conf.h | 11 +++++++ src/libvirt_private.syms | 2 ++ src/qemu/qemu_capabilities.c | 2 ++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 11 +++++++ tests/qemucapabilitiesdata/caps_2.1.1-1.caps | 1 + tests/qemucapabilitiesdata/caps_2.4.0-1.caps | 1 + tests/qemucapabilitiesdata/caps_2.5.0-1.caps | 1 + .../qemuxml2argv-disk-drive-detect-zeros.args | 27 ++++++++++++++++ .../qemuxml2argv-disk-drive-detect-zeros.xml | 37 ++++++++++++++++++++++ tests/qemuxml2argvtest.c | 4 +++ tests/qemuxml2xmltest.c | 1 + 15 files changed, 143 insertions(+), 1 deletion(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-detect-zeros.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-detect-zeros.xml -- 2.6.4

3 7

[libvirt] Exclusive VM lock
by Dmitry Andreev 15 Dec '15

15 Dec '15

Hi, I'm looking for a mechanism to do a sequence of API calls as atomic operation. Is there any way for libvirt's API client to acquire an exclusive VM lock to prevent other client from changing VM state through libvirt?

3 5

[libvirt] [PATCH] qemu: Warn when using vhost-user without shared memory
by Martin Kletzander 15 Dec '15

15 Dec '15

When user configures vhost-user interface and forgets to also configure any shared memory, the search for the root cause of non-operational interface might take unpleasantly long time. Let's enhance user experience by emitting a warning in the logs. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1266982 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- src/qemu/qemu_process.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 420196264685..fb471342e790 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -4542,6 +4542,7 @@ qemuProcessLaunch(virConnectPtr conn, unsigned int hostdev_flags = 0; size_t nnicindexes = 0; int *nicindexes = NULL; + bool check_shmem = false; VIR_DEBUG("vm=%p name=%s id=%d asyncJob=%d " "incoming.launchURI=%s incoming.deferredURI=%s " @@ -4749,6 +4750,49 @@ qemuProcessLaunch(virConnectPtr conn, goto cleanup; } + VIR_DEBUG("Checking for any possible (non-fatal) issues"); + + /* + * For vhost-user to work, the domain has to have some type of + * shared memory configured. We're not the proper once to judge + * whether shared hugepages or shm are enough and will be in the + * future, so we'll just warn in case there is none of that + * configured. Moreover failing would give the false illusion + * that libvirt is really checking that everything works before + * running the domain and not only we are unable to do that, but + * it's also not our aim to do so. + */ + for (i = 0; i < vm->def->nnets; i++) { + if (virDomainNetGetActualType(vm->def->nets[i]) == + VIR_DOMAIN_NET_TYPE_VHOSTUSER) { + check_shmem = true; + break; + } + } + + if (check_shmem) { + bool shmem = vm->def->nshmems; + + /* + * This check is by no means complete. We merely check + * whetere there are *some* hugepages enabled and *some* NUMA + * nodes with shared memory access. + */ + if (!shmem && vm->def->mem.nhugepages) { + for (i = 0; i < virDomainNumaGetNodeCount(vm->def->numa); i++) { + if (virDomainNumaGetNodeMemoryAccessMode(vm->def->numa, i) == + VIR_NUMA_MEM_ACCESS_SHARED) + shmem = true; + break; + } + } + + if (!shmem) { + VIR_WARN("Detected vhost-user interface without any shared memory. " + "The interface might not be operational"); + } + } + VIR_DEBUG("Building emulator command line"); if (!(cmd = qemuBuildCommandLine(conn, driver, vm->def, priv->monConfig, priv->monJSON, priv->qemuCaps, -- 2.6.3

3 7

[libvirt] [PATCH] qemuMonitorJSONEjectMedia: don't stringify the replay at all
by Pavel Hrdina 15 Dec '15

15 Dec '15

Commit 256496e1 introduced a detection if "is locked" in error replay from qemu monitor. Commit c4073657 fixed a memory leak, but it was pointed out by Peter, that this could be done cleaner without stringifing the replay. Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- src/qemu/qemu_monitor_json.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 4ae63cf..d4b6514 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -2223,11 +2223,12 @@ int qemuMonitorJSONEjectMedia(qemuMonitorPtr mon, ret = qemuMonitorJSONCheckError(cmd, reply); if (ret < 0) { - char *replyStr = virJSONValueToString(reply, false); - - if (c_strcasestr(replyStr, "is locked")) - ret = -2; - VIR_FREE(replyStr); + virJSONValuePtr error = virJSONValueObjectGet(reply, "error"); + if (error) { + const char *errorStr = virJSONValueObjectGetString(error, "desc"); + if (errorStr && c_strcasestr(errorStr, "is locked")) + ret = -2; + } } virJSONValueFree(cmd); -- 2.6.4

2 2

[libvirt] qemu logging/tracing option in domain xml
by Dmitry Andreev 15 Dec '15

15 Dec '15

Hi all, QEMU has the '-d' command line option to configure logs/traces. As I can see, <qemu:commandline> is the only way to set it in domain xml. <qemu:commandline> makes a configuration tainted and doesn't allow to update tracing/logging configuration in runtime. The ability to update configuration in runtime will be very useful for us in Virtuozzo for QEMU problem investigation and development. So I want to ask you will you mind if I prepare a patch-set that introduces a new section for domain xml with next scheme: <element name="debug"> <zeroOrMore> <element name="log"> <attribute name="name"> <ref name="filter-param-value"/> </attribute> </element> </zeroOrMore> </element> Is this acceptable?

3 2

[libvirt] [PATCH] pci: Use 'addr' instead of 'dev' for virPCIDeviceAddressPtr
by Andrea Bolognani 15 Dec '15

15 Dec '15

The name 'dev' is more appropriate for virPCIDevicePtr. --- src/util/virpci.c | 7 ++++--- src/util/virpci.h | 4 ++-- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/src/util/virpci.c b/src/util/virpci.c index bb874ac..9cc4613 100644 --- a/src/util/virpci.c +++ b/src/util/virpci.c @@ -2655,12 +2655,13 @@ virPCIGetSysfsFile(char *virPCIDeviceName, char **pci_sysfs_device_link) } int -virPCIDeviceAddressGetSysfsFile(virPCIDeviceAddressPtr dev, +virPCIDeviceAddressGetSysfsFile(virPCIDeviceAddressPtr addr, char **pci_sysfs_device_link) { if (virAsprintf(pci_sysfs_device_link, - PCI_SYSFS "devices/%04x:%02x:%02x.%x", dev->domain, - dev->bus, dev->slot, dev->function) < 0) + PCI_SYSFS "devices/%04x:%02x:%02x.%x", + addr->domain, addr->bus, + addr->slot, addr->function) < 0) return -1; return 0; } diff --git a/src/util/virpci.h b/src/util/virpci.h index 6516f05..f3d5676 100644 --- a/src/util/virpci.h +++ b/src/util/virpci.h @@ -159,7 +159,7 @@ virPCIDeviceListPtr virPCIDeviceGetIOMMUGroupList(virPCIDevicePtr dev); int virPCIDeviceAddressGetIOMMUGroupAddresses(virPCIDeviceAddressPtr devAddr, virPCIDeviceAddressPtr **iommuGroupDevices, size_t *nIommuGroupDevices); -int virPCIDeviceAddressGetIOMMUGroupNum(virPCIDeviceAddressPtr dev); +int virPCIDeviceAddressGetIOMMUGroupNum(virPCIDeviceAddressPtr addr); char *virPCIDeviceGetIOMMUGroupDev(virPCIDevicePtr dev); int virPCIDeviceIsAssignable(virPCIDevicePtr dev, @@ -180,7 +180,7 @@ int virPCIGetVirtualFunctionIndex(const char *pf_sysfs_device_link, const char *vf_sysfs_device_link, int *vf_index); -int virPCIDeviceAddressGetSysfsFile(virPCIDeviceAddressPtr dev, +int virPCIDeviceAddressGetSysfsFile(virPCIDeviceAddressPtr addr, char **pci_sysfs_device_link); int virPCIGetNetName(char *device_link_sysfs_path, char **netname); -- 2.5.0

2 2

[libvirt] [PATCH 0/3] vz: allow to create CT from template and boot VM from cdrom
by Mikhail Feoktistov 15 Dec '15

15 Dec '15

Mikhail Feoktistov (3): vz: allow to create container based on template vz: allow to boot VM from cdrom vz: implementation of boot device order src/vz/vz_sdk.c | 174 ++++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 125 insertions(+), 49 deletions(-) -- 1.8.3.1

1 3

[libvirt] [PATCH] qemuMonitorJSONEjectMedia: Don't leak stringified reply
by Michal Privoznik 15 Dec '15

15 Dec '15

The return value of virJSONValueToString() should be freed when no longer needed. This is not the case after 256496e1. ==26902== 138 bytes in 2 blocks are definitely lost in loss record 1,051 of 1,239 ==26902== at 0x4C29F80: malloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so) ==26902== by 0xAA5F599: strdup (in /lib64/libc-2.21.so) ==26902== by 0x552BAD9: virStrdup (virstring.c:726) ==26902== by 0x54F60A7: virJSONValueToString (virjson.c:1790) ==26902== by 0x1DF6EBB9: qemuMonitorJSONEjectMedia (qemu_monitor_json.c:2225) ==26902== by 0x1DF57A4C: qemuMonitorEjectMedia (qemu_monitor.c:1985) ==26902== by 0x1DF1EF2D: qemuDomainChangeEjectableMedia (qemu_hotplug.c:199) ==26902== by 0x1DF90314: qemuDomainChangeDiskLive (qemu_driver.c:7985) ==26902== by 0x1DF90476: qemuDomainUpdateDeviceLive (qemu_driver.c:8030) ==26902== by 0x1DF91ED7: qemuDomainUpdateDeviceFlags (qemu_driver.c:8677) ==26902== by 0x561785F: virDomainUpdateDeviceFlags (libvirt-domain.c:8559) ==26902== by 0x134210: remoteDispatchDomainUpdateDeviceFlags (remote_dispatch.h:10966) ==26902== 106 bytes in 1 blocks are definitely lost in loss record 1,033 of 1,239 ==26902== at 0x4C29F80: malloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so) ==26902== by 0xAA5F599: strdup (in /lib64/libc-2.21.so) ==26902== by 0x552BAD9: virStrdup (virstring.c:726) ==26902== by 0x54F60A7: virJSONValueToString (virjson.c:1790) ==26902== by 0x1DF6EC0C: qemuMonitorJSONEjectMedia (qemu_monitor_json.c:2227) ==26902== by 0x1DF57A4C: qemuMonitorEjectMedia (qemu_monitor.c:1985) ==26902== by 0x1DF1EF2D: qemuDomainChangeEjectableMedia (qemu_hotplug.c:199) ==26902== by 0x1DF90314: qemuDomainChangeDiskLive (qemu_driver.c:7985) ==26902== by 0x1DF90476: qemuDomainUpdateDeviceLive (qemu_driver.c:8030) ==26902== by 0x1DF91ED7: qemuDomainUpdateDeviceFlags (qemu_driver.c:8677) ==26902== by 0x561785F: virDomainUpdateDeviceFlags (libvirt-domain.c:8559) ==26902== by 0x134210: remoteDispatchDomainUpdateDeviceFlags (remote_dispatch.h:10966) Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_monitor_json.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 84c0be2..4ae63cf 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -2222,10 +2222,13 @@ int qemuMonitorJSONEjectMedia(qemuMonitorPtr mon, if (ret == 0) ret = qemuMonitorJSONCheckError(cmd, reply); - VIR_DEBUG("%s", virJSONValueToString(reply, false)); + if (ret < 0) { + char *replyStr = virJSONValueToString(reply, false); - if (ret < 0 && c_strcasestr(virJSONValueToString(reply, false), "is locked")) - ret = -2; + if (c_strcasestr(replyStr, "is locked")) + ret = -2; + VIR_FREE(replyStr); + } virJSONValueFree(cmd); virJSONValueFree(reply); -- 2.4.10

3 2

[libvirt] [PATCH] pci: Use virPCIDeviceAddress in virPCIDevice
by Andrea Bolognani 15 Dec '15

15 Dec '15

Instead of replicating the information (domain, bus, slot, function) inside the virPCIDevice structure, use the already-existing virPCIDeviceAddress structure. Outside of the module, the only visible change is that the return value of virPCIDeviceGetAddress() must no longer be freed by the caller. --- Suggested by Laine[1] in the context of a patch series; since the idea is not really tied to the series, sending as a stand-alone cleanup. [1] https://www.redhat.com/archives/libvir-list/2015-December/msg00125.html src/util/virhostdev.c | 4 -- src/util/virpci.c | 182 ++++++++++++++++++++++++++++++++++---------------- src/util/virpci.h | 7 ++ 3 files changed, 133 insertions(+), 60 deletions(-) diff --git a/src/util/virhostdev.c b/src/util/virhostdev.c index de029a0..173ac58 100644 --- a/src/util/virhostdev.c +++ b/src/util/virhostdev.c @@ -585,7 +585,6 @@ virHostdevPreparePCIDevices(virHostdevManagerPtr hostdev_mgr, goto cleanup; } - VIR_FREE(devAddr); if (!(devAddr = virPCIDeviceGetAddress(dev))) goto cleanup; @@ -728,7 +727,6 @@ virHostdevPreparePCIDevices(virHostdevManagerPtr hostdev_mgr, virObjectUnlock(hostdev_mgr->activePCIHostdevs); virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnref(pcidevs); - VIR_FREE(devAddr); return ret; } @@ -1581,7 +1579,6 @@ virHostdevPCINodeDeviceDetach(virHostdevManagerPtr hostdev_mgr, out: virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnlock(hostdev_mgr->activePCIHostdevs); - VIR_FREE(devAddr); return ret; } @@ -1613,7 +1610,6 @@ virHostdevPCINodeDeviceReAttach(virHostdevManagerPtr hostdev_mgr, out: virObjectUnlock(hostdev_mgr->inactivePCIHostdevs); virObjectUnlock(hostdev_mgr->activePCIHostdevs); - VIR_FREE(devAddr); return ret; } diff --git a/src/util/virpci.c b/src/util/virpci.c index bececb5..2818bf7 100644 --- a/src/util/virpci.c +++ b/src/util/virpci.c @@ -56,10 +56,7 @@ VIR_ENUM_IMPL(virPCIELinkSpeed, VIR_PCIE_LINK_SPEED_LAST, "", "2.5", "5", "8") struct _virPCIDevice { - unsigned int domain; - unsigned int bus; - unsigned int slot; - unsigned int function; + virPCIDeviceAddressPtr address; char name[PCI_ADDR_LEN]; /* domain:bus:slot.function */ char id[PCI_ID_LEN]; /* product vendor */ @@ -653,12 +650,14 @@ static int virPCIDeviceSharesBusWithActive(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) { virPCIDeviceList *inactiveDevs = data; + virPCIDeviceAddressPtr devAddr = dev->address; + virPCIDeviceAddressPtr checkAddr = check->address; /* Different domain, different bus, or simply identical device */ - if (dev->domain != check->domain || - dev->bus != check->bus || - (dev->slot == check->slot && - dev->function == check->function)) + if (devAddr->domain != checkAddr->domain || + devAddr->bus != checkAddr->bus || + (devAddr->slot == checkAddr->slot && + devAddr->function == checkAddr->function)) return 0; /* same bus, but inactive, i.e. about to be assigned to guest */ @@ -686,10 +685,12 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) uint16_t device_class; uint8_t header_type, secondary, subordinate; virPCIDevicePtr *best = data; + virPCIDeviceAddressPtr devAddr = dev->address; + virPCIDeviceAddressPtr checkAddr = check->address; int ret = 0; int fd; - if (dev->domain != check->domain) + if (devAddr->domain != checkAddr->domain) return 0; if ((fd = virPCIDeviceConfigOpen(check, false)) < 0) @@ -713,7 +714,7 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) /* if the secondary bus exactly equals the device's bus, then we found * the direct parent. No further work is necessary */ - if (dev->bus == secondary) { + if (devAddr->bus == secondary) { ret = 1; goto cleanup; } @@ -722,10 +723,10 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) * In this case, what we need to do is look for the "best" match; i.e. * the most restrictive match that still satisfies all of the conditions. */ - if (dev->bus > secondary && dev->bus <= subordinate) { + if (devAddr->bus > secondary && devAddr->bus <= subordinate) { if (*best == NULL) { - *best = virPCIDeviceNew(check->domain, check->bus, check->slot, - check->function); + *best = virPCIDeviceNew(checkAddr->domain, checkAddr->bus, + checkAddr->slot, checkAddr->function); if (*best == NULL) { ret = -1; goto cleanup; @@ -745,8 +746,8 @@ virPCIDeviceIsParent(virPCIDevicePtr dev, virPCIDevicePtr check, void *data) if (secondary > best_secondary) { virPCIDeviceFree(*best); - *best = virPCIDeviceNew(check->domain, check->bus, check->slot, - check->function); + *best = virPCIDeviceNew(checkAddr->domain, checkAddr->bus, + checkAddr->slot, checkAddr->function); if (*best == NULL) { ret = -1; goto cleanup; @@ -925,6 +926,7 @@ virPCIDeviceReset(virPCIDevicePtr dev, char *drvName = NULL; int ret = -1; int fd = -1; + virPCIDeviceAddressPtr devAddr = dev->address; if (activeDevs && virPCIDeviceListFind(activeDevs, dev)) { virReportError(VIR_ERR_INTERNAL_ERROR, @@ -970,7 +972,7 @@ virPCIDeviceReset(virPCIDevicePtr dev, ret = virPCIDeviceTryPowerManagementReset(dev, fd); /* Bus reset is not an option with the root bus */ - if (ret < 0 && dev->bus != 0) + if (ret < 0 && devAddr->bus != 0) ret = virPCIDeviceTrySecondaryBusReset(dev, fd, inactiveDevs); if (ret < 0) { @@ -1428,6 +1430,7 @@ virPCIDeviceWaitForCleanup(virPCIDevicePtr dev, const char *matcher) bool in_matching_device; int ret; size_t match_depth; + virPCIDeviceAddressPtr devAddr = dev->address; fp = fopen("/proc/iomem", "r"); if (!fp) { @@ -1483,8 +1486,8 @@ virPCIDeviceWaitForCleanup(virPCIDevicePtr dev, const char *matcher) virStrToLong_ui(tmp + 1, &tmp, 16, &function) < 0 || *tmp != '\n') continue; - if (domain != dev->domain || bus != dev->bus || slot != dev->slot || - function != dev->function) + if (domain != devAddr->domain || bus != devAddr->bus || + slot != devAddr->slot || function != devAddr->function) continue; in_matching_device = true; match_depth = strspn(line, " "); @@ -1547,6 +1550,72 @@ virPCIGetAddrString(unsigned int domain, return ret; } +/** + * virPCIDeviceAddressNew: + * @domain: PCI domain + * @bus: PCI bus + * @slot: PCI slot + * @function: PCI function + * + * Create a new virPCIDeviceAddress object. + * + * Returns: a new address, or NULL on failure. + */ +virPCIDeviceAddressPtr +virPCIDeviceAddressNew(unsigned int domain, + unsigned int bus, + unsigned int slot, + unsigned int function) +{ + virPCIDeviceAddressPtr addr; + + if (VIR_ALLOC(addr) < 0) + return NULL; + + addr->domain = domain; + addr->bus = bus; + addr->slot = slot; + addr->function = function; + + return addr; +} + +/** + * virPCIDeviceAddressCopy: + * @addr: PCI address + * + * Create a copy of a PCI address. + * + * Returns: a copy of @addr, or NULL on failure. + */ +virPCIDeviceAddressPtr +virPCIDeviceAddressCopy(virPCIDeviceAddressPtr addr) +{ + virPCIDeviceAddressPtr copy; + + if (!addr) + return NULL; + + if (VIR_ALLOC(copy) < 0) + return NULL; + + *copy = *addr; + + return copy; +} + +/** + * virPCIDeviceAddressFree: + * @addr: PCI address + * + * Release all resources associated with a PCI address. + */ +void +virPCIDeviceAddressFree(virPCIDeviceAddressPtr addr) +{ + VIR_FREE(addr); +} + virPCIDevicePtr virPCIDeviceNew(unsigned int domain, unsigned int bus, @@ -1560,17 +1629,14 @@ virPCIDeviceNew(unsigned int domain, if (VIR_ALLOC(dev) < 0) return NULL; - dev->domain = domain; - dev->bus = bus; - dev->slot = slot; - dev->function = function; + if (!(dev->address = virPCIDeviceAddressNew(domain, bus, slot, function))) + goto error; if (snprintf(dev->name, sizeof(dev->name), "%.4x:%.2x:%.2x.%.1x", - dev->domain, dev->bus, dev->slot, - dev->function) >= sizeof(dev->name)) { + domain, bus, slot, function) >= sizeof(dev->name)) { virReportError(VIR_ERR_INTERNAL_ERROR, _("dev->name buffer overflow: %.4x:%.2x:%.2x.%.1x"), - dev->domain, dev->bus, dev->slot, dev->function); + domain, bus, slot, function); goto error; } if (virAsprintf(&dev->path, PCI_SYSFS "devices/%s/config", @@ -1627,9 +1693,13 @@ virPCIDeviceCopy(virPCIDevicePtr dev) /* shallow copy to take care of most attributes */ *copy = *dev; + + /* deep copy for the rest */ + copy->address = NULL; copy->path = copy->stubDriver = NULL; copy->used_by_drvname = copy->used_by_domname = NULL; - if (VIR_STRDUP(copy->path, dev->path) < 0 || + if (!(copy->address = virPCIDeviceAddressCopy(dev->address)) || + VIR_STRDUP(copy->path, dev->path) < 0 || VIR_STRDUP(copy->stubDriver, dev->stubDriver) < 0 || VIR_STRDUP(copy->used_by_drvname, dev->used_by_drvname) < 0 || VIR_STRDUP(copy->used_by_domname, dev->used_by_domname) < 0) { @@ -1649,6 +1719,7 @@ virPCIDeviceFree(virPCIDevicePtr dev) if (!dev) return; VIR_DEBUG("%s %s: freeing", dev->id, dev->name); + virPCIDeviceAddressFree(dev->address); VIR_FREE(dev->path); VIR_FREE(dev->stubDriver); VIR_FREE(dev->used_by_drvname); @@ -1661,25 +1732,14 @@ virPCIDeviceFree(virPCIDevicePtr dev) * @dev: device to get address from * * Take a PCI device on input and return its PCI address. The - * caller must free the returned value when no longer needed. + * returned object is owned by the device and must not be freed. * * Returns NULL on failure, the device address on success. */ virPCIDeviceAddressPtr virPCIDeviceGetAddress(virPCIDevicePtr dev) { - - virPCIDeviceAddressPtr pciAddrPtr; - - if (!dev || (VIR_ALLOC(pciAddrPtr) < 0)) - return NULL; - - pciAddrPtr->domain = dev->domain; - pciAddrPtr->bus = dev->bus; - pciAddrPtr->slot = dev->slot; - pciAddrPtr->function = dev->function; - - return pciAddrPtr; + return dev->address; } const char * @@ -1888,14 +1948,18 @@ virPCIDeviceListDel(virPCIDeviceListPtr list, int virPCIDeviceListFindIndex(virPCIDeviceListPtr list, virPCIDevicePtr dev) { + virPCIDeviceAddressPtr devAddr = dev->address; + virPCIDeviceAddressPtr tmpAddr; size_t i; - for (i = 0; i < list->count; i++) - if (list->devs[i]->domain == dev->domain && - list->devs[i]->bus == dev->bus && - list->devs[i]->slot == dev->slot && - list->devs[i]->function == dev->function) + for (i = 0; i < list->count; i++) { + tmpAddr = list->devs[i]->address; + if (tmpAddr->domain == devAddr->domain && + tmpAddr->bus == devAddr->bus && + tmpAddr->slot == devAddr->slot && + tmpAddr->function == devAddr->function) return i; + } return -1; } @@ -1907,13 +1971,16 @@ virPCIDeviceListFindByIDs(virPCIDeviceListPtr list, unsigned int slot, unsigned int function) { + virPCIDeviceAddressPtr tmpAddr; size_t i; for (i = 0; i < list->count; i++) { - if (list->devs[i]->domain == domain && - list->devs[i]->bus == bus && - list->devs[i]->slot == slot && - list->devs[i]->function == function) + tmpAddr = list->devs[i]->address; + if (tmpAddr && + tmpAddr->domain == domain && + tmpAddr->bus == bus && + tmpAddr->slot == slot && + tmpAddr->function == function) return list->devs[i]; } return NULL; @@ -1942,9 +2009,11 @@ int virPCIDeviceFileIterate(virPCIDevicePtr dev, int ret = -1; struct dirent *ent; int direrr; + virPCIDeviceAddressPtr devAddr = dev->address; if (virAsprintf(&pcidir, "/sys/bus/pci/devices/%04x:%02x:%02x.%x", - dev->domain, dev->bus, dev->slot, dev->function) < 0) + devAddr->domain, devAddr->bus, + devAddr->slot, devAddr->function) < 0) goto cleanup; if (!(dir = opendir(pcidir))) { @@ -2074,13 +2143,12 @@ virPCIDeviceListPtr virPCIDeviceGetIOMMUGroupList(virPCIDevicePtr dev) { virPCIDeviceListPtr groupList = virPCIDeviceListNew(); - virPCIDeviceAddress devAddr = { dev->domain, dev->bus, - dev->slot, dev->function }; + virPCIDeviceAddressPtr devAddr = dev->address; if (!groupList) goto error; - if (virPCIDeviceAddressIOMMUGroupIterate(&devAddr, + if (virPCIDeviceAddressIOMMUGroupIterate(devAddr, virPCIDeviceGetIOMMUGroupAddOne, groupList) < 0) goto error; @@ -2286,6 +2354,7 @@ static int virPCIDeviceIsBehindSwitchLackingACS(virPCIDevicePtr dev) { virPCIDevicePtr parent; + virPCIDeviceAddressPtr devAddr = dev->address; if (virPCIDeviceGetParent(dev, &parent) < 0) return -1; @@ -2294,7 +2363,7 @@ virPCIDeviceIsBehindSwitchLackingACS(virPCIDevicePtr dev) * into play since devices on the root bus can't P2P without going * through the root IOMMU. */ - if (dev->bus == 0) { + if (devAddr->bus == 0) { return 0; } else { virReportError(VIR_ERR_INTERNAL_ERROR, @@ -2664,12 +2733,13 @@ virPCIGetSysfsFile(char *virPCIDeviceName, char **pci_sysfs_device_link) } int -virPCIDeviceAddressGetSysfsFile(virPCIDeviceAddressPtr dev, +virPCIDeviceAddressGetSysfsFile(virPCIDeviceAddressPtr addr, char **pci_sysfs_device_link) { if (virAsprintf(pci_sysfs_device_link, - PCI_SYSFS "devices/%04x:%02x:%02x.%x", dev->domain, - dev->bus, dev->slot, dev->function) < 0) + PCI_SYSFS "devices/%04x:%02x:%02x.%x", + addr->domain, addr->bus, + addr->slot, addr->function) < 0) return -1; return 0; } diff --git a/src/util/virpci.h b/src/util/virpci.h index 6516f05..a137a9a 100644 --- a/src/util/virpci.h +++ b/src/util/virpci.h @@ -69,6 +69,13 @@ struct _virPCIEDeviceInfo { virPCIELink *link_sta; /* Actually negotiated capabilities */ }; +virPCIDeviceAddressPtr virPCIDeviceAddressNew(unsigned int domain, + unsigned int bus, + unsigned int slot, + unsigned int function); +virPCIDeviceAddressPtr virPCIDeviceAddressCopy(virPCIDeviceAddressPtr addr); +void virPCIDeviceAddressFree(virPCIDeviceAddressPtr addr); + virPCIDevicePtr virPCIDeviceNew(unsigned int domain, unsigned int bus, unsigned int slot, -- 2.5.0

2 2

[libvirt] [PATCHv3] hypervisor driver for Jailhouse
by Christian Loehle 15 Dec '15

15 Dec '15

>From 818447bdc0c7fb5048367276a5eb64108ad95e2a Mon Sep 17 00:00:00 2001 From: Christian Loehle <cloehle(a)linutronix.de> Date: Mon, 14 Dec 2015 20:13:07 +0100 Subject: [PATCH] Add Jailhouse Driver v3 --- configure.ac | 7 + include/libvirt/virterror.h | 1 + m4/virt-driver-jailhouse.m4 | 35 +++ po/POTFILES.in | 1 + src/Makefile.am | 15 ++ src/conf/domain_conf.c | 3 +- src/conf/domain_conf.h | 1 + src/jailhouse/README | 14 ++ src/jailhouse/jailhouse_driver.c | 511 +++++++++++++++++++++++++++++++++++++++ src/jailhouse/jailhouse_driver.h | 28 +++ src/libvirt.c | 10 + src/util/virerror.c | 1 + 12 files changed, 626 insertions(+), 1 deletion(-) create mode 100644 m4/virt-driver-jailhouse.m4 create mode 100644 src/jailhouse/README create mode 100644 src/jailhouse/jailhouse_driver.c create mode 100644 src/jailhouse/jailhouse_driver.h diff --git a/configure.ac b/configure.ac index 98cf210..445b2ca 100644 --- a/configure.ac +++ b/configure.ac @@ -1087,6 +1087,12 @@ dnl LIBVIRT_DRIVER_CHECK_BHYVE dnl +dnl Checks for Jailhouse driver +dnl + +LIBVIRT_DRIVER_CHECK_JAILHOUSE + +dnl dnl check for shell that understands <> redirection without truncation, dnl needed by src/qemu/qemu_monitor_{text,json}.c. dnl @@ -2830,6 +2836,7 @@ AC_MSG_NOTICE([ ESX: $with_esx]) AC_MSG_NOTICE([ Hyper-V: $with_hyperv]) LIBVIRT_DRIVER_RESULT_VZ LIBVIRT_DRIVER_RESULT_BHYVE +LIBVIRT_DRIVER_RESULT_JAILHOUSE AC_MSG_NOTICE([ Test: $with_test]) AC_MSG_NOTICE([ Remote: $with_remote]) AC_MSG_NOTICE([ Network: $with_network]) diff --git a/include/libvirt/virterror.h b/include/libvirt/virterror.h index 0539e48..bbc7e2e 100644 --- a/include/libvirt/virterror.h +++ b/include/libvirt/virterror.h @@ -128,6 +128,7 @@ typedef enum { VIR_FROM_THREAD = 61, /* Error from thread utils */ VIR_FROM_ADMIN = 62, /* Error from admin backend */ VIR_FROM_LOGGING = 63, /* Error from log manager */ + VIR_FROM_JAILHOUSE = 64, /* Error from Jailhouse driver */ # ifdef VIR_ENUM_SENTINELS VIR_ERR_DOMAIN_LAST diff --git a/m4/virt-driver-jailhouse.m4 b/m4/virt-driver-jailhouse.m4 new file mode 100644 index 0000000..6c72043 --- /dev/null +++ b/m4/virt-driver-jailhouse.m4 @@ -0,0 +1,35 @@ +dnl The Jailhouse driver +dnl +dnl Copyright (C) 2005-2015 Red Hat, Inc. +dnl +dnl This library is free software; you can redistribute it and/or +dnl modify it under the terms of the GNU Lesser General Public +dnl License as published by the Free Software Foundation; either +dnl version 2.1 of the License, or (at your option) any later version. +dnl +dnl This library is distributed in the hope that it will be useful, +dnl but WITHOUT ANY WARRANTY; without even the implied warranty of +dnl MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +dnl Lesser General Public License for more details. +dnl +dnl You should have received a copy of the GNU Lesser General Public +dnl License along with this library. If not, see +dnl <http://www.gnu.org/licenses/>. +dnl + +AC_DEFUN([LIBVIRT_DRIVER_CHECK_JAILHOUSE],[ + AC_ARG_WITH([jailhouse], + [AS_HELP_STRING([--with-jailhouse], + [add Jailhouse support @<:@default=yes@:>@])]) + m4_divert_text([DEFAULTS], [with_jailhouse=yes]) + + if test "$with_jailhouse" = "yes"; then + AC_DEFINE_UNQUOTED([WITH_JAILHOUSE], 1, [whether jailhouse driver is enabled]) + fi + + AM_CONDITIONAL([WITH_JAILHOUSE], [test "$with_jailhouse" = "yes"]) +]) + +AC_DEFUN([LIBVIRT_DRIVER_RESULT_JAILHOUSE],[ + AC_MSG_NOTICE([Jailhouse: $with_jailhouse]) +]) diff --git a/po/POTFILES.in b/po/POTFILES.in index 82e8d3e..29259f4 100644 --- a/po/POTFILES.in +++ b/po/POTFILES.in @@ -61,6 +61,7 @@ src/hyperv/hyperv_wmi.c src/interface/interface_backend_netcf.c src/interface/interface_backend_udev.c src/internal.h +src/jailhouse/jailhouse_driver.c src/libvirt.c src/libvirt-admin.c src/libvirt-domain.c diff --git a/src/Makefile.am b/src/Makefile.am index 7219f7c..af46766 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -622,6 +622,7 @@ DRIVER_SOURCE_FILES = \ $(VMWARE_DRIVER_SOURCES) \ $(XEN_DRIVER_SOURCES) \ $(XENAPI_DRIVER_SOURCES) \ + $(JAILHOUSE_DRIVER_SOURCES) \ $(NULL) STATEFUL_DRIVER_SOURCE_FILES = \ @@ -904,6 +905,11 @@ BHYVE_DRIVER_SOURCES = \ bhyve/bhyve_utils.h \ $(NULL) +JAILHOUSE_DRIVER_SOURCES = \ + jailhouse/jailhouse_driver.c \ + jailhouse/jailhouse_driver.h \ + $(NULL) + NETWORK_DRIVER_SOURCES = \ network/bridge_driver.h network/bridge_driver.c \ network/bridge_driver_platform.h \ @@ -1480,6 +1486,14 @@ libvirt_driver_vz_la_LIBADD = $(PARALLELS_SDK_LIBS) $(LIBNL_LIBS) libvirt_driver_vz_la_SOURCES = $(VZ_DRIVER_SOURCES) endif WITH_VZ +if WITH_JAILHOUSE +noinst_LTLIBRARIES += libvirt_driver_jailhouse.la +libvirt_la_BUILT_LIBADD += libvirt_driver_jailhouse.la +libvirt_driver_jailhouse_la_CFLAGS = \ + -I$(srcdir)/conf $(AM_CFLAGS) +libvirt_driver_jailhouse_la_SOURCES = $(JAILHOUSE_DRIVER_SOURCES) +endif WITH_JAILHOUSE + if WITH_BHYVE noinst_LTLIBRARIES += libvirt_driver_bhyve_impl.la libvirt_driver_bhyve_la_SOURCES = @@ -1845,6 +1859,7 @@ EXTRA_DIST += \ $(HYPERV_DRIVER_EXTRA_DIST) \ $(VZ_DRIVER_SOURCES) \ $(BHYVE_DRIVER_SOURCES) \ + $(JAILHOUSE_DRIVER_SOURCES) \ $(NETWORK_DRIVER_SOURCES) \ $(INTERFACE_DRIVER_SOURCES) \ $(STORAGE_DRIVER_SOURCES) \ diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 5200c27..926c1a0 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -107,7 +107,8 @@ VIR_ENUM_IMPL(virDomainVirt, VIR_DOMAIN_VIRT_LAST, "phyp", "parallels", "bhyve", - "vz") + "vz", + "jailhouse") VIR_ENUM_IMPL(virDomainOS, VIR_DOMAIN_OSTYPE_LAST, "hvm", diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index cec681a..271811f 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -225,6 +225,7 @@ typedef enum { VIR_DOMAIN_VIRT_PARALLELS, VIR_DOMAIN_VIRT_BHYVE, VIR_DOMAIN_VIRT_VZ, + VIR_DOMAIN_VIRT_JAILHOUSE, VIR_DOMAIN_VIRT_LAST } virDomainVirtType; diff --git a/src/jailhouse/README b/src/jailhouse/README new file mode 100644 index 0000000..02ba87d --- /dev/null +++ b/src/jailhouse/README @@ -0,0 +1,14 @@ +The jailhouse hypervisor driver for the libvirt project aims to provide +rudimentary support for managing jailhouse with the libvirt library. +The main advantage of this is the possibility to use virt-manager as a GUI to +manage Jailhouse cells. Thus the driver is mainly built around the API calls +that virt-manager uses and needs. +Due to the concept of Jailhouse a lot of libvirt functions can't be realized, +so this driver isn't as full-featured as upstream drivers of the +libvirt project. +Currently the driver relies on the Jailhouse binary, which has to be in $PATH +or passed when connecting a libvirt client to it +(e.g. virt-manager -c jailhouse:///path/to/jailhouse/tools/jailhouse). +Be aware though that the driver doesn't store any information about cells, +so most API calls use "jailhouse cell list" every time they're called to get +the current state. diff --git a/src/jailhouse/jailhouse_driver.c b/src/jailhouse/jailhouse_driver.c new file mode 100644 index 0000000..79e97f0 --- /dev/null +++ b/src/jailhouse/jailhouse_driver.c @@ -0,0 +1,511 @@ +/* + * virJailhouseDriver.c: hypervisor driver for managing Jailhouse cells + * + * Copyright (C) 2015 Linutronix GmbH + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + * Author: Christian Loehle + */ + +#include <config.h> +#include <string.h> +#include "jailhouse_driver.h" +#include "datatypes.h" +#include "virerror.h" +#include "viralloc.h" +#include "virlog.h" +#include "vircommand.h" +#include "virxml.h" +#include "configmake.h" +#include "virfile.h" +#include "virtypedparam.h" +#include "virstring.h" +#include "nodeinfo.h" +#include "capabilities.h" +#include "domain_conf.h" + +#define VIR_FROM_THIS VIR_FROM_JAILHOUSE + +#define IDLENGTH 8 +#define NAMELENGTH 24 +#define STATELENGTH 16 +#define CPULENGTH 24 +#define STATERUNNING 0 +#define STATERUNNINGSTRING "running " +#define STATERUNNINGLOCKED 1 +#define STATERUNNINGLOCKEDSTRING "running/locked " +#define STATESHUTDOWN 2 +#define STATESHUTDOWNSTRING "shut down " +#define STATEFAILED 3 +#define STATEFAILEDSTRING "failed " +#define JAILHOUSEBINARY "jailhouse" + +struct virJailhouseCell { + int id; + char name[NAMELENGTH+1]; + int state; + virBitmapPtr assignedCPUs; + virBitmapPtr failedCPUs; /* currently unused */ + unsigned char uuid[VIR_UUID_BUFLEN]; +}; +typedef struct virJailhouseCell virJailhouseCell; +typedef virJailhouseCell *virJailhouseCellPtr; + +/* + * Because virCommandRunRegex Callback gets called every line + */ +struct virJailhouseCellCallbackData { + size_t ncells; + virJailhouseCellPtr cells; +}; +typedef struct virJailhouseCellCallbackData virJailhouseCellCallbackData; +typedef virJailhouseCellCallbackData *virJailhouseCellCallbackDataPtr; + +/* + * The driver requeries the cells on most calls, it stores the result of the + * last query, so it can copy the UUIDs in the new query if the cell is the + * same(otherwise it just generates a new one). + * not preserving the UUID results in a lot of bugs in libvirts clients. + */ +struct virJailhouseDriver { + size_t lastQueryCellsCount; + virJailhouseCellPtr lastQueryCells; +}; +typedef struct virJailhouseDriver virJailhouseDriver; +typedef virJailhouseDriver *virJailhouseDriverPtr; + +static int virJailhouseParseListOutputCallback(char **const groups, void *data) +{ + virJailhouseCellCallbackDataPtr celldata = (virJailhouseCellCallbackDataPtr) data; + virJailhouseCellPtr cells = celldata->cells; + size_t count = celldata->ncells; + char* endptr = groups[0] + strlen(groups[0]) - 1; + char* state = groups[2]; + if (VIR_EXPAND_N(cells, count, 1)) + return -1; + celldata->ncells++; + + if (virStrToLong_i(groups[0], &endptr, 0, &cells[count-1].id)) + return -1; + if (!virStrcpy(cells[count-1].name, groups[1], NAMELENGTH+1)) + return -1; + if (STREQLEN(state, STATERUNNINGSTRING, STATELENGTH)) + cells[count-1].state = STATERUNNING; + else if (STREQLEN(state, STATESHUTDOWNSTRING, STATELENGTH)) + cells[count-1].state = STATESHUTDOWN; + else if (STREQLEN(state, STATERUNNINGLOCKEDSTRING, STATELENGTH)) + cells[count-1].state = STATERUNNINGLOCKED; + else + cells[count-1].state = STATEFAILED; + virBitmapParse(groups[3], 0, &cells[count-1].assignedCPUs, VIR_DOMAIN_CPUMASK_LEN); + virBitmapParse(groups[4], 0, &cells[count-1].failedCPUs, VIR_DOMAIN_CPUMASK_LEN); + celldata->cells = cells; + return 0; +} + +/* + * calls "jailhouse cell list" and parses the output in an array of virJailhouseCell + * example output: + * ID Name State Assigned CPUs Failed CPUs + * 0 QEMU-VM running 0-3 + */ +static ssize_t +virJailhouseParseListOutput(virJailhouseCellPtr *parsedCells) +{ + int nvars[] = { 5 }; + virJailhouseCellCallbackData callbackData; + const char *regex[] = { "([0-9]{1,8})\\s*([-0-9a-zA-Z]{1,24})\\s*([a-z/ ]{1,16})\\s*([0-9,-]{1,24})?\\s*([0-9,-]{1,24})?\\s*" }; + virCommandPtr cmd = virCommandNew(JAILHOUSEBINARY); + virCommandAddArg(cmd, "cell"); + virCommandAddArg(cmd, "list"); + virCommandAddEnvPassCommon(cmd); + callbackData.cells = NULL; + callbackData.ncells = 0; + if (virCommandRunRegex(cmd, 1, regex, nvars, &virJailhouseParseListOutputCallback, &callbackData, NULL) < 0) { + virCommandFree(cmd); + return -1; + } + virCommandFree(cmd); + *parsedCells = callbackData.cells; + return callbackData.ncells; +} + +/* + * Returns the libvirts equivalent of the cell state passed to it + */ +static virDomainState +virJailhouseCellToState(virJailhouseCellPtr cell) +{ + switch (cell->state) { + case STATERUNNING: return VIR_DOMAIN_RUNNING; + case STATERUNNINGLOCKED: return VIR_DOMAIN_RUNNING; + case STATESHUTDOWN: return VIR_DOMAIN_SHUTOFF; + case STATEFAILED: return VIR_DOMAIN_CRASHED; + default: return VIR_DOMAIN_NOSTATE; + } +} + +/* + * Returns a new virDomainPtr filled with the data of the virJailhouseCell + */ +static virDomainPtr +virJailhouseCellToDomainPtr(virConnectPtr conn, virJailhouseCellPtr cell) +{ + virDomainPtr dom = virGetDomain(conn, cell->name, cell->uuid); + dom->id = cell->id; + return dom; +} + +/* + * Check cells for cell and copies UUID if found, otherwise generates a new one, this is to preserve UUID in libvirt + */ +static void virJailhouseSetUUID(virJailhouseCellPtr cells, size_t count, virJailhouseCellPtr cell) +{ + size_t i; + for (i = 0; i < count; i++) { + if (strncmp(cells[i].name, cell->name, NAMELENGTH+1)) + continue; + memcpy(cell->uuid, cells[i].uuid, VIR_UUID_BUFLEN); + return; + } + virUUIDGenerate(cell->uuid); +} + +/* + * Frees the old list of cells, gets the new one and preserves UUID if cells were present in the old + */ +static int +virJailhouseGetCurrentCellList(virConnectPtr conn) +{ + virJailhouseDriverPtr driver = (virJailhouseDriverPtr)conn->privateData; + ssize_t count; + size_t i; + size_t lastCount = driver->lastQueryCellsCount; + virJailhouseCellPtr lastCells = driver->lastQueryCells; + virJailhouseCellPtr cells = NULL; + + count = virJailhouseParseListOutput(&cells); + for (i = 0; i < count; i++) + virJailhouseSetUUID(lastCells, lastCount, cells+i); + for (i = 0; i < lastCount; i++) { + virBitmapFree(lastCells[i].assignedCPUs); + virBitmapFree(lastCells[i].failedCPUs); + } + VIR_FREE(lastCells); + driver->lastQueryCells = cells; + driver->lastQueryCellsCount = count; + return count; +} + +/* + * Converts libvirts virDomainPtr to the internal virJailhouseCell by parsing the "jailhouse cell list" output + * and looking up the name of the virDomainPtr, returns NULL if cell is no longer present + */ +static virJailhouseCellPtr +virDomainPtrToCell(virDomainPtr dom) +{ + virJailhouseDriverPtr driver = dom->conn->privateData; + size_t cellsCount; + size_t i; + if (virJailhouseGetCurrentCellList(dom->conn) == -1) + return NULL; + cellsCount = driver->lastQueryCellsCount; + virJailhouseCellPtr cells = driver->lastQueryCells; + for (i = 0; i < cellsCount; i++) + if (dom->id == cells[i].id) + return cells+i; + return NULL; +} + +static virDrvOpenStatus +jailhouseConnectOpen(virConnectPtr conn, virConnectAuthPtr auth ATTRIBUTE_UNUSED, unsigned int flags) +{ + virCheckFlags(0, VIR_DRV_OPEN_ERROR); + virJailhouseDriverPtr driver; + if (conn->uri->scheme == NULL || + STRNEQ(conn->uri->scheme, "jailhouse")) + return VIR_DRV_OPEN_DECLINED; + if (conn->uri->path != NULL && STRNEQ(conn->uri->path, "/")) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Unexpected Jailhouse URI path '%s', try jailhouse:///"), + conn->uri->path); + return VIR_DRV_OPEN_ERROR; + } + if (VIR_ALLOC(driver) < 0) + return VIR_DRV_OPEN_ERROR; + driver->lastQueryCells = NULL; + driver->lastQueryCellsCount = 0; + conn->privateData = driver; + return VIR_DRV_OPEN_SUCCESS; +} + +static int +jailhouseConnectClose(virConnectPtr conn) +{ + + virJailhouseDriverPtr driver = (virJailhouseDriverPtr)conn->privateData; + size_t i; + size_t cellsCount = driver->lastQueryCellsCount; + virJailhouseCellPtr cells = driver->lastQueryCells; + for (i = 0; i < cellsCount; i++) { + virBitmapFree(cells[i].assignedCPUs); + virBitmapFree(cells[i].failedCPUs); + } + VIR_FREE(cells); + VIR_FREE(driver); + conn->privateData = NULL; + return 0; +} + +static int +jailhouseConnectNumOfDomains(virConnectPtr conn) +{ + if (virJailhouseGetCurrentCellList(conn) == -1) + return -1; + return ((virJailhouseDriverPtr)conn->privateData)->lastQueryCellsCount; +} + +static int +jailhouseConnectListDomains(virConnectPtr conn, int * ids, int maxids) +{ + virJailhouseDriverPtr driver = (virJailhouseDriverPtr)conn->privateData; + virJailhouseCellPtr cells; + size_t cellsCount; + size_t i; + if (virJailhouseGetCurrentCellList(conn) == -1) + return -1; + cellsCount = driver->lastQueryCellsCount; + cells = driver->lastQueryCells; + for (i = 0; i < maxids && i < cellsCount; i++) + ids[i] = cells[i].id; + return i; +} + +static int +jailhouseConnectListAllDomains(virConnectPtr conn, virDomainPtr ** domains, unsigned int flags) +{ + virCheckFlags(VIR_CONNECT_LIST_DOMAINS_ACTIVE, 0); + virJailhouseDriverPtr driver = (virJailhouseDriverPtr)conn->privateData; + virJailhouseCellPtr cells; + size_t cellsCount; + size_t i; + if (virJailhouseGetCurrentCellList(conn) == -1) + goto error; + + cellsCount = driver->lastQueryCellsCount; + cells = driver->lastQueryCells; + if (cellsCount == -1) + goto error; + if (VIR_ALLOC_N(*domains, cellsCount+1) < 0) + goto error; + for (i = 0; i < cellsCount; i++) + (*domains)[i] = virJailhouseCellToDomainPtr(conn, cells+i); + (*domains)[cellsCount] = NULL; + return cellsCount; + error: + *domains = NULL; + return -1; +} + +static virDomainPtr +jailhouseDomainLookupByID(virConnectPtr conn, int id) +{ + virJailhouseDriverPtr driver = (virJailhouseDriverPtr)conn->privateData; + size_t cellsCount; + size_t i; + if (virJailhouseGetCurrentCellList(conn) == -1) + return NULL; + + cellsCount = driver->lastQueryCellsCount; + if (cellsCount == -1) + return NULL; + virJailhouseCellPtr cells = driver->lastQueryCells; + for (i = 0; i < cellsCount; i++) + if (cells[i].id == id) + return virJailhouseCellToDomainPtr(conn, cells+i); + virReportError(VIR_ERR_NO_DOMAIN, NULL); + return NULL; +} + +static virDomainPtr +jailhouseDomainLookupByName(virConnectPtr conn, const char *lookupName) +{ + virJailhouseDriverPtr driver = (virJailhouseDriverPtr)conn->privateData; + size_t cellsCount; + size_t i; + if (virJailhouseGetCurrentCellList(conn) == -1) + return NULL; + + cellsCount = driver->lastQueryCellsCount; + if (cellsCount == -1) + return NULL; + virJailhouseCellPtr cells = driver->lastQueryCells; + for (i = 0; i < cellsCount; i++) + if (STREQ(cells[i].name, lookupName)) + return virJailhouseCellToDomainPtr(conn, cells+i); + virReportError(VIR_ERR_NO_DOMAIN, NULL); + return NULL; +} + +static virDomainPtr +jailhouseDomainLookupByUUID(virConnectPtr conn, const unsigned char * uuid) +{ + virJailhouseDriverPtr driver = (virJailhouseDriverPtr)conn->privateData; + size_t cellsCount; + size_t i; + if (virJailhouseGetCurrentCellList(conn) == -1) + return NULL; + cellsCount = driver->lastQueryCellsCount; + if (cellsCount == -1) + return NULL; + virJailhouseCellPtr cells = driver->lastQueryCells; + for (i = 0; i < cellsCount; i++) + if (memcmp(cells[i].uuid, (const char*)uuid, VIR_UUID_BUFLEN) == 0) + return virJailhouseCellToDomainPtr(conn, cells+i); + virReportError(VIR_ERR_NO_DOMAIN, NULL); + return NULL; +} + +/* + * There currently is no straightforward way for the driver to retrieve those, + * so maxMem, memory and cpuTime have dummy values + */ +static int +jailhouseDomainGetInfo(virDomainPtr domain, virDomainInfoPtr info) +{ + virJailhouseCellPtr cell = virDomainPtrToCell(domain); + if (cell == NULL) + return -1; + info->state = virJailhouseCellToState(cell); + info->maxMem = 0; + info->memory = 0; + info->nrVirtCpu = virBitmapCountBits(cell->assignedCPUs); + info->cpuTime = 0; + return 0; +} + +static int +jailhouseDomainGetState(virDomainPtr domain, int *state, + int *reason ATTRIBUTE_UNUSED, unsigned int flags) +{ + virCheckFlags(0, 0); + virJailhouseCellPtr cell = virDomainPtrToCell(domain); + if (cell == NULL) + return -1; + *state = virJailhouseCellToState(cell); + return 0; +} + +static int +jailhouseDomainDestroy(virDomainPtr domain) +{ + int resultcode; + virCommandPtr cmd = virCommandNew(JAILHOUSEBINARY); + virCommandAddArg(cmd, "cell"); + virCommandAddArg(cmd, "shutdown"); + virCommandAddArgFormat(cmd, "%d", domain->id); + virCommandAddEnvPassCommon(cmd); + resultcode = virCommandRun(cmd, NULL); + virCommandFree(cmd); + if (resultcode < 0) + return -1; + return 0; +} + +static int +jailhouseDomainCreate(virDomainPtr domain) +{ + int resultcode; + virCommandPtr cmd = virCommandNew(JAILHOUSEBINARY); + virCommandAddArg(cmd, "cell"); + virCommandAddArg(cmd, "start"); + virCommandAddArgFormat(cmd, "%d", domain->id); + virCommandAddEnvPassCommon(cmd); + resultcode = virCommandRun(cmd, NULL); + virCommandFree(cmd); + if (resultcode < 0) + return -1; + return 0; +} + +/* + * There currently is no reason why it shouldn't be + */ +static int +jailhouseConnectIsAlive(virConnectPtr conn ATTRIBUTE_UNUSED) +{ + return 1; +} + +static int +jailhouseNodeGetInfo(virConnectPtr conn ATTRIBUTE_UNUSED, virNodeInfoPtr info) +{ + return nodeGetInfo(NULL, info); +} + +/* + * Returns a dummy capabilities XML for virt-manager + */ +static char * +jailhouseConnectGetCapabilities(virConnectPtr conn ATTRIBUTE_UNUSED) +{ + virCapsPtr caps = virCapabilitiesNew(VIR_ARCH_NONE, false, false); + char* xml = virCapabilitiesFormatXML(caps); + virObjectUnref(caps); + return xml; +} + +static char * +jailhouseDomainGetXMLDesc(virDomainPtr domain, unsigned int flags) +{ + virCheckFlags(0, NULL); + char* xml; + virDomainDefPtr domainDef = virDomainDefNewFull(domain->name, domain->uuid, domain->id); + xml = virDomainDefFormat(domainDef, 0); + virDomainDefFree(domainDef); + return xml; +} + +static virHypervisorDriver jailhouseHypervisorDriver = { + .name = "jailhouse", + .connectOpen = jailhouseConnectOpen, /* 1.3.1 */ + .connectClose = jailhouseConnectClose, /* 1.3.1 */ + .connectGetCapabilities = jailhouseConnectGetCapabilities, /* 1.3.1 */ + .connectNumOfDomains = jailhouseConnectNumOfDomains, /* 1.3.1 */ + .connectListDomains = jailhouseConnectListDomains, /* 1.3.1 */ + .connectIsAlive = jailhouseConnectIsAlive, /* 1.3.1 */ + .connectListAllDomains = jailhouseConnectListAllDomains, /* 1.3.1 */ + .domainLookupByID = jailhouseDomainLookupByID, /* 1.3.1 */ + .domainLookupByName = jailhouseDomainLookupByName, /* 1.3.1 */ + .domainLookupByUUID = jailhouseDomainLookupByUUID, /* 1.3.1 */ + .domainGetInfo = jailhouseDomainGetInfo, /* 1.3.1 */ + .domainGetState = jailhouseDomainGetState, /* 1.3.1 */ + .domainGetXMLDesc = jailhouseDomainGetXMLDesc, /* 1.3.1 */ + .domainDestroy = jailhouseDomainDestroy, /* 1.3.1 */ + .domainCreate = jailhouseDomainCreate, /* 1.3.1 */ + .nodeGetInfo = jailhouseNodeGetInfo /* 1.3.1 */ +}; + +static virConnectDriver jailhouseConnectDriver = { + .hypervisorDriver = &jailhouseHypervisorDriver, +}; + +int +jailhouseRegister(void) +{ + return virRegisterConnectDriver(&jailhouseConnectDriver, + false); +} diff --git a/src/jailhouse/jailhouse_driver.h b/src/jailhouse/jailhouse_driver.h new file mode 100644 index 0000000..47c17e7 --- /dev/null +++ b/src/jailhouse/jailhouse_driver.h @@ -0,0 +1,28 @@ +/* + * jailhouse_driver.h: hypervisor driver for managing Jailhouse cells + * + * Copyright (C) 2015 Linutronix GmbH + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + * Author: Christian Loehle + */ + +#ifndef JAILHOUSE_DRIVER_H +# define JAILHOUSE_DRIVER_H + +int jailhouseRegister(void); + +#endif diff --git a/src/libvirt.c b/src/libvirt.c index dd58e9c..36d5994 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -98,6 +98,9 @@ #ifdef WITH_BHYVE # include "bhyve/bhyve_driver.h" #endif +#ifdef WITH_JAILHOUSE +# include "jailhouse/jailhouse_driver.h" +#endif #define VIR_FROM_THIS VIR_FROM_NONE @@ -437,6 +440,10 @@ virGlobalInit(void) if (vzRegister() == -1) goto error; # endif +#ifdef WITH_JAILHOUSE + if (jailhouseRegister() == -1) + goto error; +#endif #endif #ifdef WITH_REMOTE if (remoteRegister() == -1) @@ -1038,6 +1045,9 @@ do_open(const char *name, #ifndef WITH_VZ STRCASEEQ(ret->uri->scheme, "parallels") || #endif +#ifndef WITH_JAILHOUSE + STRCASEEQ(ret->uri->scheme, "jailhouse") || +#endif false)) { virReportErrorHelper(VIR_FROM_NONE, VIR_ERR_CONFIG_UNSUPPORTED, __FILE__, __FUNCTION__, __LINE__, diff --git a/src/util/virerror.c b/src/util/virerror.c index 098211a..06741bf 100644 --- a/src/util/virerror.c +++ b/src/util/virerror.c @@ -135,6 +135,7 @@ VIR_ENUM_IMPL(virErrorDomain, VIR_ERR_DOMAIN_LAST, "Thread jobs", "Admin Interface", "Log Manager", + "Jailhouse Driver", ) -- 2.1.4

1 0

[libvirt] [PATCH 0/2] Add support for zero-write detection
by Martin Kletzander 14 Dec '15

14 Dec '15

QEMU supports detect-zeroes option since version 2.1, but we never added support for it in libvirt. If was requested by Vasiliy Tolstov in the list, so I just added it. There are two discussions to be had, optionally. One is to decide whether we should disable detect_zeros='unmap' if discard is not set to 'unmap', but this is getting very hypervisor-specific, so I just documented the behaviour. The other one is the naming. I described why I made the decision for "zeros" instead of "zeroes" the decision in the patch, but I have no problem changing it to what others like better. Martin Kletzander (2): conf: Add support of zero-detection for disks qemu: Add support for zero-detection writes docs/formatdomain.html.in | 10 ++++++ docs/schemas/domaincommon.rng | 12 +++++++ src/conf/domain_conf.c | 23 +++++++++++++- src/conf/domain_conf.h | 11 +++++++ src/libvirt_private.syms | 2 ++ src/qemu/qemu_capabilities.c | 2 ++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 11 +++++++ tests/qemucapabilitiesdata/caps_2.1.1-1.caps | 1 + tests/qemucapabilitiesdata/caps_2.4.0-1.caps | 1 + tests/qemucapabilitiesdata/caps_2.5.0-1.caps | 1 + .../qemuxml2argv-disk-drive-detect-zeros.args | 27 ++++++++++++++++ .../qemuxml2argv-disk-drive-detect-zeros.xml | 37 ++++++++++++++++++++++ tests/qemuxml2argvtest.c | 4 +++ tests/qemuxml2xmltest.c | 1 + 15 files changed, 143 insertions(+), 1 deletion(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-detect-zeros.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-detect-zeros.xml -- 2.6.4

3 6

[libvirt] [PATCH] virNetDevMacVLanTapSetup: Drop @multiqueue argument
by Michal Privoznik 14 Dec '15

14 Dec '15

Firstly, there's a bug (or typo) in the only place where we call this function: @multiqueue is set whenever @tapfdSize is greater than zero, while in fact the condition should have been 'greater than one'. Then, secondly, since the condition depends on just one variable, that we are even passing down to the function, we can move the condition into the function and drop useless argument. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/util/virnetdevmacvlan.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) diff --git a/src/util/virnetdevmacvlan.c b/src/util/virnetdevmacvlan.c index 8fc71af..496416e 100644 --- a/src/util/virnetdevmacvlan.c +++ b/src/util/virnetdevmacvlan.c @@ -289,12 +289,11 @@ virNetDevMacVLanTapOpen(const char *ifname, * @tapfd: array of file descriptors of the macvtap tap * @tapfdSize: number of file descriptors in @tapfd * @vnet_hdr: whether to enable or disable IFF_VNET_HDR - * @multiqueue: whether to enable or disable IFF_MULTI_QUEUE * - * Turn on the IFF_VNET_HDR flag if requested and available, but make sure it's - * off otherwise. Similarly, turn on IFF_MULTI_QUEUE if requested, but if it - * can't be set, consider it a fatal error (rather than ignoring as with - * @vnet_hdr). + * Turn on the IFF_VNET_HDR flag if requested and available, but make sure + * it's off otherwise. Similarly, turn on IFF_MULTI_QUEUE if @tapfdSize is + * greater than one, but if it can't be set, consider it a fatal error + * (rather than ignoring as with @vnet_hdr). * * A fatal error is defined as the VNET_HDR flag being set but it cannot * be turned off for some reason. This is reported with -1. Other fatal @@ -304,7 +303,7 @@ virNetDevMacVLanTapOpen(const char *ifname, * Returns 0 on success, -1 in case of fatal error. */ static int -virNetDevMacVLanTapSetup(int *tapfd, size_t tapfdSize, bool vnet_hdr, bool multiqueue) +virNetDevMacVLanTapSetup(int *tapfd, size_t tapfdSize, bool vnet_hdr) { unsigned int features; struct ifreq ifreq; @@ -335,12 +334,12 @@ virNetDevMacVLanTapSetup(int *tapfd, size_t tapfdSize, bool vnet_hdr, bool multi } # ifdef IFF_MULTI_QUEUE - if (multiqueue) + if (tapfdSize > 1) new_flags |= IFF_MULTI_QUEUE; else new_flags &= ~IFF_MULTI_QUEUE; # else - if (multiqueue) { + if (tapfdSize > 1) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("Multiqueue devices are not supported on this system")); return -1; @@ -870,7 +869,7 @@ int virNetDevMacVLanCreateWithVPortProfile(const char *tgifname, if (virNetDevMacVLanTapOpen(cr_ifname, tapfd, tapfdSize, 10) < 0) goto disassociate_exit; - if (virNetDevMacVLanTapSetup(tapfd, tapfdSize, vnet_hdr, tapfdSize > 0) < 0) { + if (virNetDevMacVLanTapSetup(tapfd, tapfdSize, vnet_hdr) < 0) { VIR_FORCE_CLOSE(rc); /* sets rc to -1 */ goto disassociate_exit; } -- 2.4.10

2 1

[libvirt] [PATCH v2 0/7] Add multiqueue support for macvtaps
by Michal Privoznik 14 Dec '15

14 Dec '15

Patches 1, 2, 3, 6, 7 have been ACKed in previous round. However, I did slightly change them to reflect Laine's review suggestions. Patch 4 has not been ACKed yet, patch 5 is new. Michal Privoznik (7): virNetDevMacVLanCreateWithVPortProfile: Turn vnet_hdr into flag virNetDevMacVLanTapOpen: Slightly rework virNetDevMacVLanTapOpen: Rework to support multiple FDs virNetDevMacVLanTapSetup: Rework to support multiple FDs virNetDevMacVLanTapSetup: Allow enabling of IFF_MULTI_QUEUE virNetDevMacVLanCreateWithVPortProfile: Rework to support multiple FDs qemu: Enable multiqueue for macvtaps src/lxc/lxc_process.c | 3 +- src/qemu/qemu_command.c | 65 ++++++++++------ src/qemu/qemu_command.h | 2 + src/qemu/qemu_hotplug.c | 16 ++-- src/util/virnetdevmacvlan.c | 185 ++++++++++++++++++++++---------------------- src/util/virnetdevmacvlan.h | 7 +- 6 files changed, 153 insertions(+), 125 deletions(-) -- 2.4.10

3 12

[libvirt] how to pass qemu drive option
by Vasiliy Tolstov 14 Dec '15

14 Dec '15

I want to pass to my drive detect-zeros=on how can i do that in libvirt? I'm use lvm with virtio-scsi for sda disk. -- Vasiliy Tolstov, e-mail: v.tolstov(a)selfip.ru

2 2

[libvirt] [PATCH] qemu_agent: fix deadlock in qemuProcessHandleAgentEOF
by Wang Yufei 14 Dec '15

14 Dec '15

We shutdown a VM A by qemu agent,meanwhile an agent EOF of VM A happened, there's a chance that deadlock occurred: qemuProcessHandleAgentEOF in main thread A) priv->agent = NULL; //A happened before B //deadlock when we get agent lock which's held by worker thread qemuAgentClose(agent); qemuDomainObjExitAgent called by qemuDomainShutdownFlags in worker thread B) hasRefs = virObjectUnref(priv->agent); //priv->agent is NULL, return false if (hasRefs) virObjectUnlock(priv->agent); //agent lock will not be released here So I close agent first, then set priv->agent NULL to fix the deadlock. Signed-off-by: Wang Yufei <james.wangyufei(a)huawei.com> Reviewed-by: Ren Guannan <renguannan(a)huawei.com> --- src/qemu/qemu_process.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index f2586a1..8c9622e 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -150,11 +150,10 @@ qemuProcessHandleAgentEOF(qemuAgentPtr agent, goto unlock; } + qemuAgentClose(agent); priv->agent = NULL; virObjectUnlock(vm); - - qemuAgentClose(agent); return; unlock: -- 1.8.3.4

2 8

[libvirt] Release of libvirt-1.3.0
by Daniel Veillard 14 Dec '15

14 Dec '15

So as planned, I tagged the release in git and pushed signed tarball and rpms to the usual place: ftp://libvirt.org/libvirt I also pushed the associated python release for 1.3.0 though there was no actual code change except numbering: ftp://libvirt.org/libvirt/python The bump in version minor number comes from the addition of the administration API which nearly made it in 1.2.17 in the summer but was postponed since. That is backed up by serious improvement in virtio support and logging. As usual that release carries a sigificant amount of fixes and smaller improvements too: Features: - virt-admin and administration API (Erik Skultety, Martin Kletzander) - various improvements in virtio devices support (Ján Tomko, Marc-André Lureau) - log daemon, logging improvements and protocol (Daniel P. Berrange) Documentation: - libvirt: Update virDomainSetMemory description (Nikolay Shirokovskiy) - virt-admin: Provide a man page for virt-admin (Erik Skultety) - Enhance documentation of virDomainDetachDevice (Jiri Denemark) - qemu: monitor: Explain logic of qemuMonitorGetCPUInfo (Peter Krempa) - document virCommandRunRegex function (Christian Loehle) - libvirt-domain: Fix typo in debug message (Cole Robinson) - qemu: Explain mlock limit size more in detail (Peter Krempa) - virsh.pod: improve attach-interface section (Pavel Hrdina) - virnetdev: Fix function comments for virNetDevGetFeatures (John Ferlan) - virnetdev: Document reasons for ignoring some SIOCETHTOOL errno values (John Ferlan) Portability: - network: selectively disable -Wcast-align in virNetDevParseDadStatus (Ian Campbell) - log_manager: Include configmake.h last (Michal Privoznik) - virtlogd: Fix build without DBus (Martin Kletzander) - virtlogd: use %llu to print 64bit types (Guido Günther) Bug Fixes: - qemu: fix memory leak in opening log file (Daniel P. Berrange) - qemu: Automatic SCSI controller creation in SCSI disk hotplug broken (Boris Fiuczynski) - qemu: domain: Prevent overflows in memory alignment code (Peter Krempa) - conf: Revert some code to resolve issues for hostdev hotplug (Boris Fiuczynski) - virsh: report errors for empty strings (Ján Tomko) - bridge: check for invalid MAC in networkGetDHCPLeases (Ján Tomko) - qemu_agent: fix deadlock in qemuProcessHandleAgentEOF (Wang Yufei) - include: Install libvirt-common.h (Martin Kletzander) - tools: fix output of list with state-shutoff (Wei Jiangang) - virlogd: fix crash if log file exists and it's larger the maxlen (Pavel Hrdina) - systemd: Escape only needed characters for machined (Martin Kletzander) - logging: remove reference to non-existent augeas files (Daniel P. Berrange) - virtlockd: fix misc memory leaks and other bugs (Daniel P. Berrange) - systemd: Escape machine name for machined (Martin Kletzander) - schema: use a better regex for listen addresses (Ján Tomko) - apparmor: add missing qemu binaries (Guido Günther) - storage: Change virStorageBackendVolOpen to use virFileOpenAs (John Ferlan) - storage: Really fix setting mode for backend exec in NFS root-squash env (John Ferlan) - qemu: Add ppc64-specific math to qemuDomainGetMlockLimitBytes() (Andrea Bolognani) - libxl: don't unlock virDomainObj if refcnt is 0 (Jim Fehlig) - libxl: unref libxlDriverConfig object (Jim Fehlig) - qemu: Fix build error in Coverity environment (John Ferlan) - virSetUIDGID: Don't leak supplementary groups (Richard Weinberger) - locking: Add io_timeout to sanlock (Michal Privoznik) - libvirt-guests: Disable shutdown timeout (Guido Günther) - tpm: adapt sysfs cancel path for new TPM driver (Stefan Berger) - bhyve: monitor: do not override domain's privateData (Roman Bogorodskiy) - storage: Don't assume storage pool exists for FC/SCSI refresh thread (John Ferlan) - domain-conf: reorder usb controllers so the master is first (Pavel Hrdina) - qemu: fix parsing of -sdl arg (Daniel P. Berrange) - qemu: handle floppy disk bus when parsing command line argv (Daniel P. Berrange) - qemu: hotplug: Fix mlock limit handling on memory hotplug (Peter Krempa) - Revert "utils: Remove the logging of errors from virNetDevSendEthtoolIoctl" (Daniel P. Berrange) - qemu: migration: Actually error out on unsupported migration flag (Peter Krempa) - qemu: migration: Properly parse memory hotplug migration flag (Peter Krempa) - network: Remove extraneous ATTRIBUTE_NONNULL for virNetDevWaitDadFinish (John Ferlan) - virnetdev: Check correct return value for virNetDevFeatureAvailable (John Ferlan) - storage: On 'buildVol' failure don't delete the volume (John Ferlan) - Revert "storage: Prior to creating a volume, refresh the pool" (John Ferlan) - qemu: Fix memory leak in qemuProcessStart (Jiri Denemark) - qemu: Use correct type when calling qemuPrepareNVRAM (Jiri Denemark) Improvements: - Revert "libxl: implement virDomainInterfaceStats" (Jim Fehlig) - rpm: explicitly enable & start virtlogd on install (Daniel P. Berrange) - libvirtd: enable virtlockd/virtlogd socket activation on install (Daniel P. Berrange) - logging: validate flags passed from client in virtlogd (Daniel P. Berrange) - logging: change log protocol to be more reusable (Daniel P. Berrange) - logging: preserve driver, dom name & uuid against log file (Daniel P. Berrange) - qemu: include hostname in QEMU log files (Daniel P. Berrange) - rotatingfile: mark log files as close-on-exec (Daniel P. Berrange) - libvirtd: require virtlogd to start before libvirtd (Guido Günther) - schema: Allow > UINT_MAX KiB of memory for NUMA nodes (Peter Krempa) - virsh: remove custom error for cpulist from cmdIOThreadPin (Ján Tomko) - libxl: implement virDomainInterfaceStats (Joao Martins) - tests: Run virnetdaemontest iff WITH_YAJL (Michal Privoznik) - admin: Distribute libvirt-admin.conf (Martin Kletzander) - admin: Rename virAdmConnect to virAdmDaemon (Martin Kletzander) - spec: Temporarily disable new admin-related files (Martin Kletzander) - admin: Include admin_remote.c in the dist package (Martin Kletzander) - build: Create needed folders without dependency tracking (Martin Kletzander) - util: Avoid variable named 'truncate' shadowing global declaration (Martin Kletzander) - conf: Split virDomainObjList into a separate file (Michal Privoznik) - qemu: build command line for virtio-input-host device (Ján Tomko) - qemu: add passed-through input devs to cgroup ACL (Ján Tomko) - security: label the evdev for input device passthrough (Ján Tomko) - conf: add XML for input device passthrough (Ján Tomko) - qemu: add capability for virtio-input-host-device (Ján Tomko) - qemu: build command line for virtio input devices (Ján Tomko) - conf: parse and format virtio input bus in domain XML (Ján Tomko) - qemu: add capabilities for virtio input devices (Ján Tomko) - admin: Introduce virAdmConnectGetLibVersion (Erik Skultety) - admin: Add support for connection close callbacks (Erik Skultety) - admin: Add support for URI aliases (Erik Skultety) - livirt: Move URI alias matching to util (Erik Skultety) - admin: Add URI support and introduce virAdmGetDefaultURI (Erik Skultety) - admin: Do not generate remoteAdminConnect{Open,Close} (Erik Skultety) - admin: Move remote admin API version to a separate module (Erik Skultety) - admin: Introduce virAdmConnectIsAlive (Erik Skultety) - virt-admin: Introduce first working skeleton (Erik Skultety) - admin: introduce virAdmGetVersion (Erik Skultety) - libvirt: Move config getters to util (Erik Skultety) - admin: Introduce libvirt-admin.conf (Erik Skultety) - libvirt: introduce libvirt/libvirt-common.h.in (Erik Skultety) - qemu: add virtio-gpu virgl support (Marc-André Lureau) - qemu: add virtio video device (Marc-André Lureau) - domain: replace bool accel{2d, 3d} with a tristate (Marc-André Lureau) - Replace support{2d,3d} with accel{2d,3d} (Marc-André Lureau) - logging: avoid variables called 'daemon' due to function clash (Daniel P. Berrange) - logging: inhibit virtlogd shutdown while log files are open (Daniel P. Berrange) - qemu: add support for sending QEMU stdout/stderr to virtlogd (Daniel P. Berrange) - qemu: convert monitor to use qemuDomainLogContextPtr indirectly (Daniel P. Berrange) - qemu: convert process stop/attach to use qemuDomainLogContextPtr (Daniel P. Berrange) - qemu: convert qemuLogOperation to take a qemuDomainLogContextPtr (Daniel P. Berrange) - qemu: change qemuDomainTaint APIs to accept qemuDomainLogContextPtr (Daniel P. Berrange) - qemu: convert log file creation to use qemuDomainLogContextPtr (Daniel P. Berrange) - qemu: introduce a qemuDomainLogContext object (Daniel P. Berrange) - qemu: unify code for reporting errors from QEMU log files (Daniel P. Berrange) - qemu: remove writing to QEMU log file for rename operation (Daniel P. Berrange) - logging: add client for virtlogd daemon (Daniel P. Berrange) - logging: introduce log handling protocol (Daniel P. Berrange) - Import stripped down virtlockd code as basis of virtlogd (Daniel P. Berrange) - util: add APIs for reading/writing from/to rotating files (Daniel P. Berrange) - virsh: Try to keep printed XML pretty with change-media (Martin Kletzander) - qemu: Use qemuProcessLaunch in migration Prepare phase (Jiri Denemark) - qemu: Skip starting NBD servers for offline migration (Jiri Denemark) - qemu: Kill QEMU process if Prepare phase fails (Jiri Denemark) - qemu: Separate incoming URI generation from qemuMigrationPrepareAny (Jiri Denemark) - qemu: Introduce qemuProcessFinishStartup (Jiri Denemark) - qemu: Introduce qemuProcessLaunch (Jiri Denemark) - qemu: Introduce qemuProcessInit (Jiri Denemark) - conf: reject multiple panic devices of same model (Dmitry Andreev) - Allow multiple panic devices (Dmitry Andreev) - qemu: add support for hv_crash feature as a panic device (Dmitry Andreev) - tests: add tests for the new panic device attribute - 'model' (Dmitry Andreev) - conf: add 'model' attribute for panic device with values isa, pseries, hyperv (Dmitry Andreev) - conf: refactor code for checking ABI stability of panic device (Dmitry Andreev) - nodedev: report maxCount for virtual_functions capability (Laine Stump) - conf: support reporting maxCount attribute for virtual_functions cap (Laine Stump) - Post-release version bump to 1.3.0 (Pavel Hrdina) - conf: Drop useless check when parsing cpu scheduler info (Peter Krempa) - qemu: pass the asyncJob to qemuProcessStartCPUs (Ján Tomko) - xenapi: Refactor extraction of vcpu count (Peter Krempa) - phyp: Refactor extraction of vcpu count (Peter Krempa) - openvz: Refactor extraction of vcpu count (Peter Krempa) - hyperv: Allocate 'def' via virDomainDefNew (Peter Krempa) - qemuSetupChrSourceCgroup: rename dev to source (Ján Tomko) - Simplify qemuSetupChrSourceCgroup and its callers (Ján Tomko) - rename qemuSetupHostdevCGroup to qemuSetupHostdevCgroup (Ján Tomko) - qemu: handle more machines with a single builtin IDE controller (Guido Günther) - qemu: Always set locked memory limit for ppc64 domains (Andrea Bolognani) - qemu: Use qemuDomainRequiresMlock() when attaching PCI hostdev (Andrea Bolognani) - qemu: Use qemuDomainRequiresMlock() in qemuBuildCommandLine() (Andrea Bolognani) - process: Log when limiting the amount of locked memory (Andrea Bolognani) - vz: implementation of domainReboot callback (Mikhail Feoktistov) - vz: allow only en-us keymap for VNC (Mikhail Feoktistov) - qemu: Close logfd when closing monitor (Jiri Denemark) - qemu: Do not infer flags from other qemuProcessStart arguments (Jiri Denemark) - qemu: Introduce qemuProcessMakeDir (Jiri Denemark) - qemu: Separate balloon code from qemuProcessStart (Jiri Denemark) - qemu: Enter monitor within qemuProcessSetLinkStates (Jiri Denemark) - qemu: Separate raw IO code from qemuProcessStart (Jiri Denemark) - qemu: Separate graphics handling code from qemuProcessStart (Jiri Denemark) - qemu: Separate hook handling code from qemuProcessStart (Jiri Denemark) - qemu: Rename stdin_{fd,path} in qemuProcessStart (Jiri Denemark) - qemu: Use -incoming defer for migrations (Jiri Denemark) - qemu: Add APIs for migrate-incoming QMP command (Jiri Denemark) - qemu: Always set async job when starting a domain (Jiri Denemark) - qemu: Introduce qemuProcessIncomingDef (Jiri Denemark) - qemu: Move incoming URI code to qemu_migration (Jiri Denemark) - qemu: Don't generate migration URI in qemuBuildCommandLine (Jiri Denemark) - qemu: Refactor the code to build -incoming command line (Jiri Denemark) - qemu: Refactor waiting for completed migration on destination (Jiri Denemark) - util: add virDiskNameParse to handle disk and partition idx (Joao Martins) - libxl: implement virDomainMemorystats (Joao Martins) - lxc: Bind mount container TTYs (Richard Weinberger) - lxc: Don't make container's TTY a controlling TTY (Richard Weinberger) - qemu: ppc64: Support memory hotplug without NUMA enabled (Peter Krempa) - qemu: command: Prepare memory device def formatter for missing target node (Peter Krempa) - conf: Prepare making memory device target node optional (Peter Krempa) - qemu: command: Move dimm device checks from formatter to checker (Peter Krempa) - qemu: domain: Add common function to perform memory hotplug checks (Peter Krempa) - qemu: command: Always execute memory device formatter (Peter Krempa) - qemu: command: Make qemuBuildMemoryBackendStr usable without NUMA (Peter Krempa) - libxl: implement virDomainGetCPUStats (Joao Martins) - syntax-check: Add prohibit_space_in_label rule (Andrea Bolognani) - util: remove unnecessary needSize (Chen Hanxiao) - storage: Introduce virStoragePoolObjFindPoolByUUID (John Ferlan) - storage: Change cbdata scsi refresh thread field name (John Ferlan) - storage: Make active boolean (John Ferlan) - qemu: domain: Restructurate control flow in qemuDomainGetMlockLimitBytes (Peter Krempa) - qemu: Fix job entry debug message (Jiri Denemark) - tests: Add QEMU 2.4.0 capabilities (Jiri Denemark) - tests: Remove qemuxmlnstest (Jiri Denemark) - qemu: Fix style in qemuProcessStart (Jiri Denemark) - security: Cleanup DAC driver (Jiri Denemark) - domain-conf: cleanup controller insert function (Pavel Hrdina) - virsh-domain: update attach-interface to support type=hostdev (Pavel Hrdina) - vz: support cpu time in driver's domainGetInfo (Nikolay Shirokovskiy) - qemu: assume various QEMU 0.10 features are always available (Daniel P. Berrange) - qemu: assume -vga is always available (Daniel P. Berrange) - qemu: assume -drive format is always available (Daniel P. Berrange) - qemu: assume -drive cache always uses v2 option names (Daniel P. Berrange) - qemu: assume support for all migration protocols except rdma (Daniel P. Berrange) - qemu: assume vnet-hdr feature is always available (Daniel P. Berrange) - qemu: really remove last traces of Xenner support (Daniel P. Berrange) - qemu: assume -uuid is always available (Daniel P. Berrange) - qemu: assume -name is always available (Daniel P. Berrange) - qemu: assume -drive argument is always available (Daniel P. Berrange) - qemu: handle USB bus in qemuAssignDeviceDiskAliasFixed() (Daniel P. Berrange) - qemu: assume -no-reboot is always available (Daniel P. Berrange) - qemu: assume 'info chardev' is always available (Daniel P. Berrange) - qemu: assume -vnc arg always takes a ':' (Daniel P. Berrange) - qemu: remove all support for kQEMU (Daniel P. Berrange) - qemu: mandate QEMU version 0.12.0 or newer (Daniel P. Berrange) - qemu: hotplug: Reject VFIO hotplug if setting RLIMIT_MEMLOCK fails (Peter Krempa) - qemu: Extract logic to determine the mlock limit size for VFIO (Peter Krempa) - conf: Make @def const in virDomainDefGetMemoryInitial (Peter Krempa) - tests: redo test argv file line wrapping (Daniel P. Berrange) - virnetdev: Use virNetDevSetupControl in virNetDevSendEthtoolIoctl (John Ferlan) - virnetdev: Check for root in virNetDevGetFeatures (John Ferlan) - qemu: add /usr/lib to AC_PATH_PROG for qemu-bridge-helper (Michel Normand) - storage: Pull volume removal from pool in storageVolDeleteInternal (John Ferlan) - storage: Cleanup failures in virStorageBackendCreateRaw (John Ferlan) - storage: Cleanup failures virStorageBackendCreateExecCommand (John Ferlan) - storage: Fix setting mode in virStorageBackendCreateExecCommand (John Ferlan) - Remove new lines from log messages (Jiri Denemark) - qemu: Introduce cleanup label in qemuProcessStart (Jiri Denemark) - qemu: Rename ret variable in qemuProcessStart (Jiri Denemark) - qemu: Rename cleanup label in qemuProcessStart (Jiri Denemark) So thanks everybody for helping with this release, be it with patches, ideas, reports, documentation, etc... As a reminder next release supposedly 1.3.1 will come mid-January due to the end of year slowdown, and then 1.3.2 at the end or February. Enjoy ! Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

2 1

[libvirt] [PATCH] virNetDevMacVLanTapSetup: Work around older systems
by Michal Privoznik 12 Dec '15

12 Dec '15

Some older systems, e.g. RHEL-6 do not have IFF_MULTI_QUEUE flag which we use to enable multiqueue feature. Therefore one gets the following compile error there: CC util/libvirt_util_la-virnetdevmacvlan.lo util/virnetdevmacvlan.c: In function 'virNetDevMacVLanTapSetup': util/virnetdevmacvlan.c:338: error: 'IFF_MULTI_QUEUE' undeclared (first use in this function) util/virnetdevmacvlan.c:338: error: (Each undeclared identifier is reported only once util/virnetdevmacvlan.c:338: error: for each function it appears in.) make[3]: *** [util/libvirt_util_la-virnetdevmacvlan.lo] Error 1 So, whenever user wants us to enable the feature on such systems, we will just throw a runtime error instead. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/util/virnetdevmacvlan.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/src/util/virnetdevmacvlan.c b/src/util/virnetdevmacvlan.c index d8d1d90..28c9f22 100644 --- a/src/util/virnetdevmacvlan.c +++ b/src/util/virnetdevmacvlan.c @@ -307,49 +307,57 @@ static int virNetDevMacVLanTapSetup(int *tapfd, size_t tapfdSize, bool vnet_hdr, bool multiqueue) { unsigned int features; struct ifreq ifreq; short new_flags = 0; size_t i; for (i = 0; i < tapfdSize; i++) { memset(&ifreq, 0, sizeof(ifreq)); if (ioctl(tapfd[i], TUNGETIFF, &ifreq) < 0) { virReportSystemError(errno, "%s", _("cannot get interface flags on macvtap tap")); return -1; } new_flags = ifreq.ifr_flags; if (vnet_hdr) { if (ioctl(tapfd[i], TUNGETFEATURES, &features) < 0) { virReportSystemError(errno, "%s", _("cannot get feature flags on macvtap tap")); return -1; } if (features & IFF_VNET_HDR) new_flags |= IFF_VNET_HDR; } else { new_flags &= ~IFF_VNET_HDR; } +#ifdef IFF_MULTI_QUEUE if (multiqueue) new_flags |= IFF_MULTI_QUEUE; else new_flags &= ~IFF_MULTI_QUEUE; +#else + if (multiqueue) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Multiqueue devices are not supported on this system")); + return -1; + } +#endif if (new_flags != ifreq.ifr_flags) { ifreq.ifr_flags = new_flags; if (ioctl(tapfd[i], TUNSETIFF, &ifreq) < 0) { virReportSystemError(errno, "%s", _("unable to set vnet or multiqueue flags on macvtap")); return -1; } } } return 0; } -- 2.4.10

2 1

[libvirt] [PATCH] CVE-2015-5313: storage: don't allow '/' in filesystem volume names
by Eric Blake 12 Dec '15

12 Dec '15

The libvirt file system storage driver determines what file to act on by concatenating the pool location with the volume name. If a user is able to pick names like "../../../etc/passwd", then they can escape the bounds of the pool. For that matter, virStoragePoolListVolumes() doesn't descend into subdirectories, so a user really shouldn't use a name with a slash. Normally, only privileged users can coerce libvirt into creating or opening existing files using the virStorageVol APIs; and such users already have full privilege to create any domain XML (so it is not an escalation of privilege). But in the case of fine-grained ACLs, it is feasible that a user can be granted storage_vol:create but not domain:write, and it violates assumptions if such a user can abuse libvirt to access files outside of the storage pool. Therefore, prevent all use of volume names that contain "/", whether or not such a name is actually attempting to escape the pool. This changes things from: $ virsh vol-create-as default ../../../../../../etc/haha --capacity 128 Vol ../../../../../../etc/haha created $ rm /etc/haha to: $ virsh vol-create-as default ../../../../../../etc/haha --capacity 128 error: Failed to create vol ../../../../../../etc/haha error: Requested operation is not valid: volume name '../../../../../../etc/haha' cannot contain '/' Signed-off-by: Eric Blake <eblake(a)redhat.com> --- This has been reviewed on the libvirt security list, where it was assigned a CVE. Fortunately, this could only be used for an escalation of privileges under fine-grained ACLs (which is not an out-of-the-box config). I will go ahead and push this to master as well as all the active maint branches back to the introduction of ACLs. src/storage/storage_backend_fs.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index c71c724..bb3b62a 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -1,7 +1,7 @@ /* * storage_backend_fs.c: storage backend for FS and directory handling * - * Copyright (C) 2007-2014 Red Hat, Inc. + * Copyright (C) 2007-2015 Red Hat, Inc. * Copyright (C) 2007-2008 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -1057,6 +1057,14 @@ virStorageBackendFileSystemVolCreate(virConnectPtr conn ATTRIBUTE_UNUSED, else vol->type = VIR_STORAGE_VOL_FILE; + /* Volumes within a directory pools are not recursive; do not + * allow escape to ../ or a subdir */ + if (strchr(vol->name, '/')) { + virReportError(VIR_ERR_OPERATION_INVALID, + _("volume name '%s' cannot contain '/'"), vol->name); + return -1; + } + VIR_FREE(vol->target.path); if (virAsprintf(&vol->target.path, "%s/%s", pool->def->target.path, -- 2.4.3

1 0

[libvirt] [PATCH] storage: Attempt to refresh volume after successful wipe volume
by John Ferlan 12 Dec '15

12 Dec '15

https://bugzilla.redhat.com/show_bug.cgi?id=1270709 When a volume wipe is successful, a volume refresh should be done afterwards to update any volume data that may be used in future volume commands, such as volume resize. For a raw file volume, a wipe would truncate the file and a followup volume resize the capacity may fail because the volume target allocation isn't updated to reflect the wipe activity. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/storage/storage_driver.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/src/storage/storage_driver.c b/src/storage/storage_driver.c index bbf21f6..2e59e39 100644 --- a/src/storage/storage_driver.c +++ b/src/storage/storage_driver.c @@ -2436,7 +2436,19 @@ storageVolWipePattern(virStorageVolPtr obj, goto cleanup; } - ret = backend->wipeVol(obj->conn, pool, vol, algorithm, flags); + if ((ret = backend->wipeVol(obj->conn, pool, vol, algorithm, flags)) < 0) + goto cleanup; + + /* Best effort to refresh the volume data. If unsuccessful, we've already + * wiped the data so there's no going back on that. Best we can do is + * provide some details over what happened and move on + */ + if (backend->refreshVol && + backend->refreshVol(obj->conn, pool, vol) < 0) { + VIR_WARN("failed to refresh volume '%s' info after volume wipe", + vol->name); + virResetLastError(); + } cleanup: virStoragePoolObjUnlock(pool); -- 2.5.0

2 2

[libvirt] [PATCH 00/10] VFIO fixes for PCI devices
by Andrea Bolognani 12 Dec '15

12 Dec '15

This series is my attempt at fixing https://bugzilla.redhat.com/show_bug.cgi?id=1272300 In its current state, it's missing test cases covering the new functionality[1] and it's known not to handle properly one situation[2], but I'd like to get some feedback on my current work and now that I have something to show for it feels like a good time. I'm already working on the missing bits and they will either be included in the next revision or sent as separate series later on. The problem being solved is that, when using VFIO, IOMMU group ownership can't be shared, eg. two devices that are in the same IOMMU group can't be assigned to different guests, or to the host and a guest. If that happens, the host will probably crash. The series deals with this issue by making sure safety conditions are met before detaching devices from the host or reattaching them to the host. In praticular, when we're asked to reattach a device to the host but doing so would lead to sharing IOMMU group ownership, we delay the operation until we can guarantee this will not cause problems. As a nice side effect of the changes we check for this when starting a guest too, instead of assuming it will work and having QEMU error out immediately afterwards. Patches are organized as follows: 1-2: Minor cleanups that make implicit / confusing stuff explicit / less confusing 3: Convert a string field used as an enumeration to a proper enumeration 4: Introduce a simple helper function used later on 5-6: Rewrite the checks used when detaching devices from the host. With this patches applied, the behaviour is basically the same as before, except for the nice little extra detailed above 7-9: Implement the delay when reattaching devices to the host, thus preventing the crash and fixing the bug 10: Spit and polish Cheers. [1] Luckily, it doesn't break the existing tests either [2] If you call 'virsh nodedev-reattach' on a device that's assigned to a guest, libvirt won't stop you and you will end up crashing your system Andrea Bolognani (10): pci: Remove redundant parameter from virPCIDeviceBindToStub() pci: Remove 'reprobe' parameter from virPCIDeviceUnbind() pci: Introduce virPCIStubDriver enumeration pci: Introduce virPCIDeviceIOMMUGroupIterate() hostdev: Simplify virHostdevIsPCIDeviceUsed() hostdev: Check for safety before detaching VFIO devices hostdev: Delay reattach of VFIO devices hostdev: Clean up delayed VFIO devices hostdev: Devices have already been marked as inactive hostdev: Tidy up after changes to VFIO device handling src/libvirt_private.syms | 3 + src/libxl/libxl_driver.c | 3 +- src/qemu/qemu_driver.c | 6 +- src/util/virhostdev.c | 410 ++++++++++++++++++++++++++++++++++------------- src/util/virpci.c | 125 +++++++++------ src/util/virpci.h | 26 ++- src/xen/xen_driver.c | 3 +- tests/virhostdevtest.c | 5 +- tests/virpcitest.c | 35 ++-- 9 files changed, 427 insertions(+), 189 deletions(-) -- 2.5.0

3 17

[libvirt] [PATCH 0/6] Memory locking limit improvements
by Andrea Bolognani 11 Dec '15

11 Dec '15

As noticed by Peter[1], the memory locking limit for the QEMU process is increased before assigning a VFIO device to a guest, but it might not be decreased when returning said device to the host. This series fixes this inconsistent behaviour and cleans up the code a little bit along the way. The idea is to introduce a new, smarter function called qemuDomainAdjustMaxMemLock() that does The Right Thing™ and increases the limit when required, while at the same time storing the original value. This way, when memory locking is no longer needed, it can restore it. I've tested this both on x86 and ppc64, both by removing devices that were assigned in the domain XML and devices that I had hotplugged. Patches 1-2 lay some groundwork by allowing retrieval of the memory locking limit for a process. Patches 3-4 add qemuDomainAdjustMaxMemLock() and use it where appropriate in the existing code. Patch 5 adds one more use of the function, after a PCI hostdev has been detached from the guest. Patch 6 replaces the use of Mlock with MemLock. Suggestions on how to further improve the names of those functions is very welcome, this is just a first step in the right direction. Cheers. [1] https://www.redhat.com/archives/libvir-list/2015-November/msg00642.html Andrea Bolognani (6): process: Allow virProcessPrLimit() to get current limit process: Add virProcessGetMaxMemLock() qemu: Add qemuDomainAdjustMaxMemLock() qemu: Use qemuDomainAdjustMaxMemLock() qemu: Reduce memlock limit after detaching hostdev qemu: Replace Mlock with MemLock in function names configure.ac | 2 +- src/conf/domain_conf.h | 3 +++ src/libvirt_private.syms | 1 + src/qemu/qemu_command.c | 4 ++-- src/qemu/qemu_domain.c | 56 +++++++++++++++++++++++++++++++++++++++++++--- src/qemu/qemu_domain.h | 5 +++-- src/qemu/qemu_hotplug.c | 46 ++++++++++++++++---------------------- src/util/virprocess.c | 58 +++++++++++++++++++++++++++++++++++++++++++----- src/util/virprocess.h | 2 ++ 9 files changed, 136 insertions(+), 41 deletions(-) -- 2.5.0

2 25

[libvirt] [PATCH] hypervisor driver for Jailhouse
by Christian Loehle 11 Dec '15

11 Dec '15

>From README: The jailhouse hypervisor driver for the libvirt project aims to provide rudimentary support for managing jailhouse with the libvirt library. The main advantage of this is the possibility to use virt-manager as a GUI to manage Jailhouse cells. Thus the driver is mainly built around the API calls that virt-manager uses and needs. Due to the concept of Jailhouse a lot of libvirt functions can't be realized, so this driver isn't as full-featured as upstream drivers of the libvirt project. Currently the driver relies on the Jailhouse binary, which has to be passed when connecting a libvirt client to it(e.g. virt-manager -c jailhouse:///path/to/jailhouse/tools/jailhouse). This has the advantage that remote support can be easily done by not passing the original Jailhouse binary, but an executable that redirects its parameters through ssh to the real Jailhouse binary and outputs that output. Be aware though that the driver doesn't store any information about cells, so most API calls use "jailhouse cell list" every time they're called to get the current state. I would like to get Jailhouse support upstream, any feedback is greatly appreciated. -- Christian Loehle diff --git a/configure.ac b/configure.ac index f481c50..8b68828 100644 --- a/configure.ac +++ b/configure.ac @@ -563,6 +563,10 @@ AC_ARG_WITH([hyperv], [AS_HELP_STRING([--with-hyperv], [add Hyper-V support @<:@default=check@:>@])]) m4_divert_text([DEFAULTS], [with_hyperv=check]) +AC_ARG_WITH([jailhouse], + [AS_HELP_STRING([--with-jailhouse], + [add Jailhouse support @<:@default=yes@:>@])]) +m4_divert_text([DEFAULTS], [with_jailhouse=yes]) AC_ARG_WITH([test], [AS_HELP_STRING([--with-test], [add test driver support @<:@default=yes@:>@])]) @@ -722,6 +726,16 @@ AM_CONDITIONAL([WITH_VMWARE], [test "$with_vmware" = "yes"]) dnl +dnl Checks for the Jailhouse driver +dnl + +if test "$with_jailhouse" = "yes"; then + AC_DEFINE_UNQUOTED([WITH_JAILHOUSE], 1, [whether Jailhouse driver is enabled]) +fi +AM_CONDITIONAL([WITH_JAILHOUSE], [test "$with_jailhouse" = "yes"]) + + +dnl dnl check for XDR dnl @@ -1087,6 +1101,12 @@ dnl LIBVIRT_DRIVER_CHECK_BHYVE dnl +dnl Checks for Jailhouse driver +dnl + +AM_CONDITIONAL([WITH_JAILHOUSE], [test "$with_jailhouse" = "yes"]) + +dnl dnl check for shell that understands <> redirection without truncation, dnl needed by src/qemu/qemu_monitor_{text,json}.c. dnl @@ -2830,6 +2850,7 @@ AC_MSG_NOTICE([ ESX: $with_esx]) AC_MSG_NOTICE([ Hyper-V: $with_hyperv]) LIBVIRT_DRIVER_RESULT_VZ LIBVIRT_DRIVER_RESULT_BHYVE +AC_MSG_NOTICE([Jailhouse: $with_jailhouse]) AC_MSG_NOTICE([ Test: $with_test]) AC_MSG_NOTICE([ Remote: $with_remote]) AC_MSG_NOTICE([ Network: $with_network]) diff --git a/include/libvirt/virterror.h b/include/libvirt/virterror.h index f716cb9..c8fe2d3 100644 --- a/include/libvirt/virterror.h +++ b/include/libvirt/virterror.h @@ -127,6 +127,7 @@ typedef enum { VIR_FROM_POLKIT = 60, /* Error from polkit code */ VIR_FROM_THREAD = 61, /* Error from thread utils */ VIR_FROM_ADMIN = 62, /* Error from admin backend */ + VIR_FROM_JAILHOUSE = 63, /* Error from Jailhouse driver */ # ifdef VIR_ENUM_SENTINELS VIR_ERR_DOMAIN_LAST diff --git a/po/POTFILES.in b/po/POTFILES.in index 0cc5b99..2b144bf 100644 --- a/po/POTFILES.in +++ b/po/POTFILES.in @@ -59,6 +59,7 @@ src/hyperv/hyperv_wmi.c src/interface/interface_backend_netcf.c src/interface/interface_backend_udev.c src/internal.h +src/jailhouse/jailhouse_driver.c src/libvirt.c src/libvirt-admin.c src/libvirt-domain.c diff --git a/src/Makefile.am b/src/Makefile.am index 99b4993..10d59de 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -578,6 +578,7 @@ DRIVER_SOURCE_FILES = \ $(VMWARE_DRIVER_SOURCES) \ $(XEN_DRIVER_SOURCES) \ $(XENAPI_DRIVER_SOURCES) \ + $(JAILHOUSE_DRIVER_SOURCES) \ $(NULL) STATEFUL_DRIVER_SOURCE_FILES = \ @@ -860,6 +861,11 @@ BHYVE_DRIVER_SOURCES = \ bhyve/bhyve_utils.h \ $(NULL) +JAILHOUSE_DRIVER_SOURCES = \ + jailhouse/jailhouse_driver.c \ + jailhouse/jailhouse_driver.h \ + $(NULL) + NETWORK_DRIVER_SOURCES = \ network/bridge_driver.h network/bridge_driver.c \ network/bridge_driver_platform.h \ @@ -1436,6 +1442,14 @@ libvirt_driver_vz_la_LIBADD = $(PARALLELS_SDK_LIBS) $(LIBNL_LIBS) libvirt_driver_vz_la_SOURCES = $(VZ_DRIVER_SOURCES) endif WITH_VZ +if WITH_JAILHOUSE +noinst_LTLIBRARIES += libvirt_driver_jailhouse.la +libvirt_la_BUILT_LIBADD += libvirt_driver_jailhouse.la +libvirt_driver_jailhouse_la_CFLAGS = \ + -I$(srcdir)/conf $(AM_CFLAGS) +libvirt_driver_jailhouse_la_SOURCES = $(JAILHOUSE_DRIVER_SOURCES) +endif WITH_JAILHOUSE + if WITH_BHYVE noinst_LTLIBRARIES += libvirt_driver_bhyve_impl.la libvirt_driver_bhyve_la_SOURCES = @@ -1801,6 +1815,7 @@ EXTRA_DIST += \ $(HYPERV_DRIVER_EXTRA_DIST) \ $(VZ_DRIVER_SOURCES) \ $(BHYVE_DRIVER_SOURCES) \ + $(JAILHOUSE_DRIVER_SOURCES) \ $(NETWORK_DRIVER_SOURCES) \ $(INTERFACE_DRIVER_SOURCES) \ $(STORAGE_DRIVER_SOURCES) \ diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 2edf123..00d17e9 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -121,7 +121,8 @@ VIR_ENUM_IMPL(virDomainVirt, VIR_DOMAIN_VIRT_LAST, "phyp", "parallels", "bhyve", - "vz") + "vz", + "jailhouse") VIR_ENUM_IMPL(virDomainOS, VIR_DOMAIN_OSTYPE_LAST, "hvm", diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index f10b534..27beef0 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -225,6 +225,7 @@ typedef enum { VIR_DOMAIN_VIRT_PARALLELS, VIR_DOMAIN_VIRT_BHYVE, VIR_DOMAIN_VIRT_VZ, + VIR_DOMAIN_VIRT_JAILHOUSE, VIR_DOMAIN_VIRT_LAST } virDomainVirtType; diff --git a/src/jailhouse/README b/src/jailhouse/README new file mode 100644 index 0000000..564cfbd --- /dev/null +++ b/src/jailhouse/README @@ -0,0 +1,3 @@ +The jailhouse hypervisor driver for the libvirt project aims to provide rudimentary support for managing jailhouse with the libvirt library. The main advantage of this is the possibility to use virt-manager as a GUI to manage Jailhouse cells. Thus the driver is mainly built around the API calls that virt-manager uses and needs. +Due to the concept of Jailhouse a lot of libvirt functions can't be realized, so this driver isn't as full-featured as upstream drivers of the libvirt project. +Currently the driver relies on the Jailhouse binary, which has to be passed when connecting a libvirt client to it(e.g. virt-manager -c jailhouse:///path/to/jailhouse/tools/jailhouse). This has the advantage that remote support can be easily done by not passing the original Jailhouse binary, but an executable that redirects its parameters through ssh to the real Jailhouse binary and outputs that output. Be aware though that the driver doesn't store any information about cells, so most API calls use "jailhouse cell list" every time they're called to get the current state. diff --git a/src/jailhouse/jailhouse_driver.c b/src/jailhouse/jailhouse_driver.c new file mode 100644 index 0000000..21acbba --- /dev/null +++ b/src/jailhouse/jailhouse_driver.c @@ -0,0 +1,614 @@ +/* + * jailhouse_driver.c: hypervisor driver for managing Jailhouse cells + * + * Copyright (C) 2015 Linutronix GmbH + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + * Author: Christian Loehle + */ + +#include <config.h> +#include <string.h> +#include "jailhouse_driver.h" +#include "datatypes.h" +#include "virerror.h" +#include "viralloc.h" +#include "virlog.h" +#include "vircommand.h" +#include "virxml.h" +#include "configmake.h" +#include "virfile.h" +#include "virtypedparam.h" +#include "virstring.h" +#include "nodeinfo.h" + +#define VIR_FROM_THIS VIR_FROM_JAILHOUSE + +#define IDLENGTH 8 +#define NAMELENGTH 24 +#define STATELENGTH 16 +#define CPULENGTH 24 +#define STATERUNNING 0 +#define STATERUNNINGSTRING "running " +#define STATERUNNINGLOCKED 1 +#define STATERUNNINGLOCKEDSTRING "running/locked " +#define STATESHUTDOWN 2 +#define STATESHUTDOWNSTRING "shut down " +#define STATEFAILED 3 +#define STATEFAILEDSTRING "failed " +#define JAILHOUSEVERSIONOUTPUT "Jailhouse management tool" + +/* + * The driver requeries the cells on most calls, it stores the result of the last query, so it can copy the UUIDs in the new query if the cell is the same(otherwise it just generates a new one) + * not preserving the UUID results in a lot of bugs in libvirts clients. + */ +struct jailhouse_driver { + char *binary; + size_t lastQueryCellsCount; + struct jailhouse_cell* lastQueryCells; +}; + +/* + * CPUs are currently unused but this might change + */ +struct jailhouse_cell { + int id; + char name[NAMELENGTH+1]; + int state; + int *assignedCPUs; //Don't use cpumask because remote system might have different # of cpus + int assignedCPUsLength; + int *failedCPUs; + int failedCPUsLength; + unsigned char uuid[VIR_UUID_BUFLEN]; +}; + +/* + * helper function that returns the number as an integer and sets i to be the first char after the number + */ +static int +charsToInt(char* chars, size_t *i) +{ + int result = 0; + while (chars[*i] != ',' && chars[*i] != '-' && chars[*i] != ' ') { + result *= 10; + result += chars[*i] - '0'; + (*i)++; + } + return result; +} + +/* + * Takes a string in the format of "jailhouse cell list" as input, + * allocates an int array in which every CPU is explicitly listed and saves a pointer in cpusptr + */ +static size_t +parseCPUs(char* output, int **cpusptr) +{ + size_t i; + size_t count = 1; + int number; + int* cpus; + if (output[0] == ' ') { + *cpusptr = NULL; + return 0; + } + for (i = 0; i<CPULENGTH; i++) { + number = charsToInt(output, &i); + if (output[i] == ',') { + count++; + } else if (output[i] == '-') { + i++; + count += charsToInt(output, &i) - number; + } + } + if (VIR_ALLOC_N(cpus, count)) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Failed to allocate CPUs array of size %zu"), count); + return 0; + } + size_t j = 0; + i = 0; + while (output[i] != ' ') { + number = charsToInt(output, &i); + if (output[i] == ',' || output[i] == ' ') { + cpus[j++] = number; + } else if (output[i] == '-') { + i++; + int nextNumber = charsToInt(output, &i); + for (; number <= nextNumber; number++) cpus[j++] = number; + } + i++; + } + *cpusptr = cpus; + return count; +} + +/* + * calls "jailhouse cell list" and parses the output in an array of jailhouse_cell + */ +static size_t +parseListOutput(virConnectPtr conn, struct jailhouse_cell **parsedOutput) +{ + virCommandPtr cmd = virCommandNew(((struct jailhouse_driver *)conn->privateData)->binary); + virCommandAddArg(cmd, "cell"); + virCommandAddArg(cmd, "list"); + virCommandAddEnvPassCommon(cmd); + char *output; + virCommandSetOutputBuffer(cmd, &output); + size_t count = -1; // Don't count table header line + size_t i = 0; + if (virCommandRun(cmd, NULL) < 0) + goto error; + while (output[i] != '\0') { + if (output[i] == '\n') count++; + i++; + } + if (VIR_ALLOC_N(*parsedOutput, count)) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Failed to allocate jailhouse_cell array of size %zu"), count); + goto error; + } + if (*parsedOutput == NULL) + goto error; + i = 0; + size_t j; + while (output[i++] != '\n'); // Skip table header line + for (j = 0; j < count; j++) { + size_t k; + for (k = 0; k <= IDLENGTH; k++) // char after number needs to be NUL for virStrToLong + if (output[i+k] == ' ') { + output[i+k] = '\0'; + break; + } + char c = output[i+IDLENGTH]; + output[i+IDLENGTH] = '\0'; // in case ID is 8 chars long, so beginning of name won't get parsed + if (virStrToLong_i(output+i, NULL, 0, &(*parsedOutput)[j].id)) + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Failed to parse id to long: %s"), output+i); + output[i+IDLENGTH] = c; + i += IDLENGTH; + if (virStrncpy((*parsedOutput)[j].name, output+i, NAMELENGTH, NAMELENGTH+1) == NULL) + // should never happen + goto error; + (*parsedOutput)[j].name[NAMELENGTH] = '\0'; + for (k = 0; k < NAMELENGTH; k++) + if ((*parsedOutput)[j].name[k] == ' ') + break; + (*parsedOutput)[j].name[k] = '\0'; + i += NAMELENGTH; + if (STREQLEN(output+i, STATERUNNINGSTRING, STATELENGTH)) (*parsedOutput)[j].state = STATERUNNING; + else if (STREQLEN(output+i, STATESHUTDOWNSTRING, STATELENGTH)) (*parsedOutput)[j].state = STATESHUTDOWN; + else if (STREQLEN(output+i, STATEFAILEDSTRING, STATELENGTH)) (*parsedOutput)[j].state = STATEFAILED; + else if (STREQLEN(output+i, STATERUNNINGLOCKEDSTRING, STATELENGTH)) (*parsedOutput)[j].state = STATERUNNINGLOCKED; + i += STATELENGTH; + (*parsedOutput)[j].assignedCPUsLength = parseCPUs(output+i, &((*parsedOutput)[j].assignedCPUs)); + i += CPULENGTH; + (*parsedOutput)[j].failedCPUsLength = parseCPUs(output+i, &((*parsedOutput)[j].failedCPUs)); + i += CPULENGTH; + i++; // skip \n + } + VIR_FREE(output); + return count; + error: + for (i = 0; i < count; i++) { + VIR_FREE((*parsedOutput)[i].assignedCPUs); + VIR_FREE((*parsedOutput)[i].failedCPUs); + } + VIR_FREE(*parsedOutput); + *parsedOutput = NULL; + VIR_FREE(output); + output = NULL; + return -1; +} + +/* + * Returns the libvirts equivalent of the cell state passed to it + */ +static virDomainState +cellToVirDomainState(struct jailhouse_cell *cell) +{ + switch (cell->state) { + case STATERUNNING: return VIR_DOMAIN_RUNNING; + case STATERUNNINGLOCKED: return VIR_DOMAIN_RUNNING; + case STATESHUTDOWN: return VIR_DOMAIN_SHUTOFF; + case STATEFAILED: return VIR_DOMAIN_CRASHED; + default: return VIR_DOMAIN_NOSTATE; + } +} + +/* + * Returns a new virDomainPtr filled with the data of the jailhouse_cell + */ +static virDomainPtr +cellToVirDomainPtr(virConnectPtr conn, struct jailhouse_cell *cell) +{ + virDomainPtr dom = virGetDomain(conn, cell->name, cell->uuid); + dom->id = cell->id; + return dom; +} + +/* + * Check cells for cell and copies UUID if found, otherwise generates a new one, this is to preserve UUID in libvirt + */ +static void setUUID(struct jailhouse_cell *cells, size_t count, struct jailhouse_cell* cell) { + size_t i; + for (i = 0; i < count; i++) { + if (strncmp(cells[i].name, cell->name, NAMELENGTH+1)) + continue; + memcpy(cell->uuid, cells[i].uuid, VIR_UUID_BUFLEN); + return; + } + virUUIDGenerate(cell->uuid); +} + +/* + * Frees the old list of cells, gets the new one and preserves UUID if cells were present in the old + */ +static void +getCurrentCellList(virConnectPtr conn) +{ + size_t lastCount = ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount; + struct jailhouse_cell *lastCells = ((struct jailhouse_driver *)conn->privateData)->lastQueryCells; + struct jailhouse_cell *cells = NULL; + size_t i; + size_t count = parseListOutput(conn, &cells); + for (i = 0; i < count; i++) + setUUID(lastCells, lastCount, cells+i); + for (i = 0; i < lastCount; i++) { + VIR_FREE(lastCells[i].assignedCPUs); + VIR_FREE(lastCells[i].failedCPUs); + } + VIR_FREE(lastCells); + ((struct jailhouse_driver *)conn->privateData)->lastQueryCells = cells; + ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount = count; +} + +/* + * Converts libvirts virDomainPtr to the internal jailhouse_cell by parsing the "jailhouse cell list" output + * and looking up the name of the virDomainPtr, returns NULL if cell is no longer present + */ +static struct jailhouse_cell * +virDomainPtrToCell(virDomainPtr dom) +{ + getCurrentCellList(dom->conn); + size_t cellsCount = ((struct jailhouse_driver *)dom->conn->privateData)->lastQueryCellsCount; + struct jailhouse_cell *cells = ((struct jailhouse_driver *)dom->conn->privateData)->lastQueryCells; + size_t i; + for (i = 0; i < cellsCount; i++) + if (dom->id == cells[i].id) + return cells+i; + return NULL; +} + +static virDrvOpenStatus +jailhouseConnectOpen(virConnectPtr conn, virConnectAuthPtr auth ATTRIBUTE_UNUSED, unsigned int flags) +{ + virCheckFlags(0, VIR_DRV_OPEN_ERROR); + if (conn->uri->scheme == NULL || + STRNEQ(conn->uri->scheme, "jailhouse")) + return VIR_DRV_OPEN_DECLINED; + char* binary; + if (conn->uri->path == NULL) { + if (VIR_STRDUP(binary, "jailhouse") != 1) + return VIR_DRV_OPEN_ERROR; + } else { + if (!virFileIsExecutable(conn->uri->path)) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Path '%s', is not a valid executable file."), + conn->uri->path); + return VIR_DRV_OPEN_ERROR; + } + if (VIR_STRDUP(binary, conn->uri->path) != 1) + return VIR_DRV_OPEN_ERROR; + } + virCommandPtr cmd = virCommandNew(binary); + virCommandAddArg(cmd, "--version"); + virCommandAddEnvPassCommon(cmd); + char *output; + virCommandSetOutputBuffer(cmd, &output); + if (virCommandRun(cmd, NULL) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Executing '%s --version' failed."), + conn->uri->path); + VIR_FREE(output); + return VIR_DRV_OPEN_ERROR; + } + if (STRNEQLEN(JAILHOUSEVERSIONOUTPUT, output, strlen(JAILHOUSEVERSIONOUTPUT))) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("%s doesn't seem to be a correct Jailhouse binary."), + conn->uri->path); + VIR_FREE(output); + return VIR_DRV_OPEN_ERROR; + } + VIR_FREE(output); + struct jailhouse_driver *driver; + if (VIR_ALLOC(driver)) + return VIR_DRV_OPEN_ERROR; + driver->binary = binary; + driver->lastQueryCells = NULL; + driver->lastQueryCellsCount = 0; + conn->privateData = driver; + return VIR_DRV_OPEN_SUCCESS; +} + +static int +jailhouseConnectClose(virConnectPtr conn) +{ + size_t i; + size_t cellsCount = ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount; + struct jailhouse_cell *cells = ((struct jailhouse_driver *)conn->privateData)->lastQueryCells; + for (i = 0; i < cellsCount; i++) { + VIR_FREE(cells[i].assignedCPUs); + VIR_FREE(cells[i].failedCPUs); + } + VIR_FREE(cells); + VIR_FREE(((struct jailhouse_driver *)conn->privateData)->binary); + VIR_FREE(conn->privateData); + conn->privateData = NULL; + return 0; +} + +static int +jailhouseConnectNumOfDomains(virConnectPtr conn) +{ + getCurrentCellList(conn); + return ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount; +} + +static int +jailhouseConnectListDomains(virConnectPtr conn, int * ids, int maxids) +{ + getCurrentCellList(conn); + size_t cellsCount = ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount; + struct jailhouse_cell *cells = ((struct jailhouse_driver *)conn->privateData)->lastQueryCells; + size_t i; + for (i = 0; i < maxids && i < cellsCount; i++) + ids[i] = cells[i].id; + return i; +} + +static int +jailhouseConnectListAllDomains(virConnectPtr conn, virDomainPtr ** domains, unsigned int flags) +{ + virCheckFlags(VIR_CONNECT_LIST_DOMAINS_ACTIVE, 0); + getCurrentCellList(conn); + size_t cellsCount = ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount; + struct jailhouse_cell *cells = ((struct jailhouse_driver *)conn->privateData)->lastQueryCells; + if (cellsCount == -1) + goto error; + if (VIR_ALLOC_N(*domains, cellsCount+1)) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Failed to allocate virDomainPtr array of size %zu"), cellsCount+1); + goto error; + } + size_t i; + for (i = 0; i < cellsCount; i++) + (*domains)[i] = cellToVirDomainPtr(conn, cells+i); + (*domains)[cellsCount] = NULL; + return cellsCount; + error: + *domains = NULL; + return -1; +} + +static virDomainPtr +jailhouseDomainLookupByID(virConnectPtr conn, int id) +{ + getCurrentCellList(conn); + size_t cellsCount = ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount; + if (cellsCount == -1) + return NULL; + struct jailhouse_cell *cells = ((struct jailhouse_driver *)conn->privateData)->lastQueryCells; + size_t i; + for (i = 0; i < cellsCount; i++) + if (cells[i].id == id) + return cellToVirDomainPtr(conn, cells+i); + virReportError(VIR_ERR_NO_DOMAIN, NULL); + return NULL; +} + +static virDomainPtr +jailhouseDomainLookupByName(virConnectPtr conn, const char *lookupName) +{ + getCurrentCellList(conn); + size_t cellsCount = ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount; + if (cellsCount == -1) + return NULL; + struct jailhouse_cell *cells = ((struct jailhouse_driver *)conn->privateData)->lastQueryCells; + size_t i; + for (i = 0; i < cellsCount; i++) + if (STREQ(cells[i].name, lookupName)) + return cellToVirDomainPtr(conn, cells+i); + virReportError(VIR_ERR_NO_DOMAIN, NULL); + return NULL; +} + +static virDomainPtr +jailhouseDomainLookupByUUID(virConnectPtr conn, const unsigned char * uuid) +{ + getCurrentCellList(conn); + size_t cellsCount = ((struct jailhouse_driver *)conn->privateData)->lastQueryCellsCount; + if (cellsCount == -1) + return NULL; + struct jailhouse_cell *cells = ((struct jailhouse_driver *)conn->privateData)->lastQueryCells; + size_t i; + for (i = 0; i < cellsCount; i++) + if (memcmp(cells[i].uuid, (const char*)uuid, VIR_UUID_BUFLEN) == 0) + return cellToVirDomainPtr(conn, cells+i); + virReportError(VIR_ERR_NO_DOMAIN, NULL); + return NULL; +} + +/* + * There currently is no straightforward way for the driver to retrieve those, + * so maxMem, memory and cpuTime have dummy values + */ +static int +jailhouseDomainGetInfo(virDomainPtr domain, virDomainInfoPtr info) +{ + struct jailhouse_cell *cell = virDomainPtrToCell(domain); + if (cell == NULL) + return -1; + info->state = cellToVirDomainState(cell); + info->maxMem = 1; + info->memory = 1; + info->nrVirtCpu = cell->assignedCPUsLength; + info->cpuTime = 1; + return 0; +} + +static int +jailhouseDomainGetState(virDomainPtr domain, int *state, + int *reason ATTRIBUTE_UNUSED, unsigned int flags) +{ + virCheckFlags(0, 0); + struct jailhouse_cell *cell = virDomainPtrToCell(domain); + if (cell == NULL) + return -1; + *state = cellToVirDomainState(cell); + return 0; +} + +static int +jailhouseDomainShutdown(virDomainPtr domain) +{ + virCommandPtr cmd = virCommandNew(((struct jailhouse_driver *)domain->conn->privateData)->binary); + virCommandAddArg(cmd, "cell"); + virCommandAddArg(cmd, "shutdown"); + char buf[IDLENGTH+1]; + snprintf(buf, IDLENGTH+1, "%d", domain->id); + virCommandAddArg(cmd, buf); + virCommandAddEnvPassCommon(cmd); + int resultcode = virCommandRun(cmd, NULL); + if (resultcode < 0) + return -1; + return 0; +} + +/* + * CAREFUL, this is the Jailhouse destroy, not the libvirt destroy, cell will be deleted and would need to be created and loaded again. + * This is implemented anyway, so libvirt clients have an option to use jailhouse destroy too. + */ +static int +jailhouseDomainDestroy(virDomainPtr domain) +{ + virCommandPtr cmd = virCommandNew(((struct jailhouse_driver *)domain->conn->privateData)->binary); + virCommandAddArg(cmd, "cell"); + virCommandAddArg(cmd, "destroy"); + char buf[IDLENGTH+1]; + snprintf(buf, IDLENGTH+1, "%d", domain->id); + virCommandAddArg(cmd, buf); + virCommandAddEnvPassCommon(cmd); + int resultcode = virCommandRun(cmd, NULL); + if (resultcode < 0) + return -1; + return 0; +} + +static int +jailhouseDomainCreate(virDomainPtr domain) +{ + virCommandPtr cmd = virCommandNew(((struct jailhouse_driver *)domain->conn->privateData)->binary); + virCommandAddArg(cmd, "cell"); + virCommandAddArg(cmd, "start"); + char buf[IDLENGTH+1]; + snprintf(buf, IDLENGTH+1, "%d", domain->id); + virCommandAddArg(cmd, buf); + virCommandAddEnvPassCommon(cmd); + int resultcode = virCommandRun(cmd, NULL); + if (resultcode < 0) + return -1; + return 0; +} + +/* + * There currently is no reason why it shouldn't be + */ +static int +jailhouseConnectIsAlive(virConnectPtr conn ATTRIBUTE_UNUSED) +{ + return 1; +} + +static int +jailhouseNodeGetInfo(virConnectPtr conn ATTRIBUTE_UNUSED, virNodeInfoPtr info) +{ + return nodeGetInfo(NULL, info); +} + +/* + * Returns a dummy capabilities XML for virt-manager + */ +static char * +jailhouseConnectGetCapabilities(virConnectPtr conn ATTRIBUTE_UNUSED) +{ + char* caps; + if (VIR_STRDUP(caps, "<capabilities></capabilities>") != 1) + return NULL; + return caps; +} + +/* + * Returns a dummy XML for virt-manager + */ +static char * +jailhouseDomainGetXMLDesc(virDomainPtr domain, unsigned int flags) +{ + virCheckFlags(0, NULL); + char buf[200]; + char uuid[VIR_UUID_STRING_BUFLEN]; + virDomainGetUUIDString(domain, uuid); + snprintf(buf, 200, "<domain type =\"jailhouse\">\n\ + <name>%s</name>\n\ + <uuid>%s</uuid>\n\ + </domain>", domain->name, uuid); + char* result; + if (VIR_STRDUP(result, buf) != 1) + return NULL; + return result; +} + +static virHypervisorDriver jailhouseHypervisorDriver = { + .name = "jailhouse", + .connectOpen = jailhouseConnectOpen, /* 1.2.22 */ + .connectClose = jailhouseConnectClose, /* 1.2.22 */ + .connectGetCapabilities = jailhouseConnectGetCapabilities, /* 1.2.22 */ + .connectNumOfDomains = jailhouseConnectNumOfDomains, /* 1.2.22 */ + .connectListDomains = jailhouseConnectListDomains, /* 1.2.22 */ + .connectIsAlive = jailhouseConnectIsAlive, /* 1.2.22 */ + .connectListAllDomains = jailhouseConnectListAllDomains, /* 1.2.22 */ + .domainLookupByID = jailhouseDomainLookupByID, /* 1.2.22 */ + .domainLookupByName = jailhouseDomainLookupByName, /* 1.2.22 */ + .domainLookupByUUID = jailhouseDomainLookupByUUID, /* 1.2.22 */ + .domainGetInfo = jailhouseDomainGetInfo, /* 1.2.22 */ + .domainGetState = jailhouseDomainGetState, /* 1.2.22 */ + .domainGetXMLDesc = jailhouseDomainGetXMLDesc, /* 1.2.22 */ + .domainShutdown = jailhouseDomainShutdown, /* 1.2.22 */ + .domainDestroy = jailhouseDomainDestroy, /* 1.2.22 */ + .domainCreate = jailhouseDomainCreate, /* 1.2.22 */ + .nodeGetInfo = jailhouseNodeGetInfo /* 1.2.22 */ +}; + +static virConnectDriver jailhouseConnectDriver = { + .hypervisorDriver = &jailhouseHypervisorDriver, +}; + +int +jailhouseRegister(void) +{ + return virRegisterConnectDriver(&jailhouseConnectDriver, + false); +} diff --git a/src/jailhouse/jailhouse_driver.h b/src/jailhouse/jailhouse_driver.h new file mode 100644 index 0000000..47c17e7 --- /dev/null +++ b/src/jailhouse/jailhouse_driver.h @@ -0,0 +1,28 @@ +/* + * jailhouse_driver.h: hypervisor driver for managing Jailhouse cells + * + * Copyright (C) 2015 Linutronix GmbH + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library. If not, see + * <http://www.gnu.org/licenses/>. + * + * Author: Christian Loehle + */ + +#ifndef JAILHOUSE_DRIVER_H +# define JAILHOUSE_DRIVER_H + +int jailhouseRegister(void); + +#endif diff --git a/src/libvirt.c b/src/libvirt.c index 25a0040..7626353 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -98,6 +98,9 @@ #ifdef WITH_BHYVE # include "bhyve/bhyve_driver.h" #endif +#ifdef WITH_JAILHOUSE +# include "jailhouse/jailhouse_driver.h" +#endif #define VIR_FROM_THIS VIR_FROM_NONE @@ -437,12 +440,17 @@ virGlobalInit(void) if (vzRegister() == -1) goto error; # endif +#ifdef WITH_JAILHOUSE + if (jailhouseRegister() == -1) + goto error; +#endif #endif #ifdef WITH_REMOTE if (remoteRegister() == -1) goto error; #endif + return; error: @@ -1167,6 +1175,9 @@ do_open(const char *name, #ifndef WITH_VZ STRCASEEQ(ret->uri->scheme, "parallels") || #endif +#ifndef WITH_JAILHOUSE + STRCASEEQ(ret->uri->scheme, "jailhouse") || +#endif false)) { virReportErrorHelper(VIR_FROM_NONE, VIR_ERR_CONFIG_UNSUPPORTED, __FILE__, __FUNCTION__, __LINE__, diff --git a/src/util/virerror.c b/src/util/virerror.c index 6dc05f4..0d480c0 100644 --- a/src/util/virerror.c +++ b/src/util/virerror.c @@ -134,6 +134,7 @@ VIR_ENUM_IMPL(virErrorDomain, VIR_ERR_DOMAIN_LAST, "Polkit", /* 60 */ "Thread jobs", "Admin Interface", + "Jailhouse Driver", )

4 13

[libvirt] How should libvirt apps enable virtio-pci for aarch64?
by Cole Robinson 11 Dec '15

11 Dec '15

Hi all, I'm trying to figure out how apps should request virtio-pci for libvirt + qemu + arm/aarch64. Let me provide some background. qemu's arm/aarch64 original virtio support is via virtio-mmio, libvirt XML <address type='virtio-mmio'/>. Currently this is what libvirt sets as the address default for all arm/aarch64 virtio devices in the XML. Long term though all arm virt will likely be using virtio-pci: it's faster, enables hotplug, is more x86 like, etc. Support for virtio-pci is newer and not as widespread. qemu has had the necessary support since 2.4 at least, but the guest side isn't well distributed yet. For example, Fedora 23 and earlier don't work out of the box with virtio-pci. Internal RHELSA (RHEL Server for Aarch64) builds have it recently working AFAIK. Libvirt has some support for enabling virtio-pci with aarch64, commits added by Pavel Fedin in v1.2.19. (See e8d55172544c1fafe31a9e09346bdebca4f0d6f9). The patches add a PCIe controller automatically to the XML (and qemu commandline) if qemu-system-aarch64 supports it. However virtio-mmio is still used as the default virtio address, given the current lack of OS support. So we are at the point where libvirt apps want to enable this, but presently there isn't a good solution; the only option is to fully allocate <address type='pci' ...> for each virtio device in the XML. This is suboptimal for 2 reasons: #1) apps need to duplicate libvirt's non-trivial address type=pci allocation logic #2) apps have to add an <address> block for every virtio device, which is less friendly than the x86 case where this is rarely required. Any XML device snippets that work for x86 likely won't give the desired result for aarch64, since they will default to virtio-mmio. Think virsh attach-device/attach-disk commands Here are some possible solutions: * Drop the current behavior of adding a PCIe controller unconditionally, and instead require apps to specify it in the XML. Then, if libvirt sees a PCIe controller in the XML, default the virtio address type to pci. Apps will know if the OS they are installing supports virtio-pci (eventually via libosinfo), so this is the way we can implicitly ask libvirt 'allocate us pci addresses' Upsides: - Solves both the stated problems. - Simplest addition for applications IMO Downsides: - Requires a libvirt behavior change, no longer adding the PCIe controller by default. But in practice I don't think it will really affect anyone, since there isn't really any OS support for virtio-pci yet, and no apps support it either AFAIK. - The PCIe controller is not strictly about virtio-pci, it's for enabling plain emulated PCI devices as well. So there is a use case for using the PCIe controller for a graphics card even while your OS doesn't yet support virtio-pci. In the big picture though this is a small time window with current OS, and users can work around it by manually requesting <address type='virtio-mmio'/>, so medium/long term this isn't a big deal IMO - The PCIe controller XML is: <controller type='pci' index='0' model='pcie-root'/> <controller type='pci' index='1' model='dmi-to-pci-bridge'/> <controller type='pci' index='2' model='pci-bridge'/> I have no idea if that's always going to be the expected XML, maybe it's not wise to hardcode that in apps. Laine? * Next idea: Users specify something like like <address type='pci'/> and libvirt fills in the address for us. Upsides: - We can stick with the current PCIe controller default and avoid some of the problems mentioned above. - An auto address feature may be useful in other contexts as well. Downsides: - Seems potentially tricky to implement in libvirt code. There's many places that check type=pci and key off that, seems like it would be easy to miss updating a check and cause regressions. Maybe we could add a new type like auto-pci to make it explicit. There's probably some implementation trick to make this safe, but at first glance it looked a little dicey. - Doesn't really solve problem #2 mentioned up above... maybe we could change the address allocation logic to default to virtio-pci if there's already a virtio-pci device in the XML. But it's more work. - More work for apps, but nothing horrible. * Change the default address type from virtio-mmio to pci, if qemu supports it. I'm listing this for completeness. In the short term this doesn't make sense as there isn't any OS releases that will work with this default. However it might be worth considering for the future, maybe keying off a particular qemu version or machine type. I suspect 2 years from now no one is going to be using virtio-mmio so long term it's not an ideal default. I think the first option is best (keying off the PCIe controller specified by the user), with a longer term plan to change the default from mmio to pci. But I'm not really sold on anything either way. So I'm interested if anyone else has ideas. Thanks, Cole

5 13

[libvirt] [PATCH] util: Fixup virnetdevmacvlan.h ATTRIBUTE_NONNULL's
by John Ferlan 11 Dec '15

11 Dec '15

Commit id '56e2171c6' removed a variable from the argument list, but neglected to update the ATTRIBUTE_NONNULL values, so when commit id '08da97bfb' added a couple of arguments, the values were off. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- Pushed as build breaker (coverity) and trivial. src/util/virnetdevmacvlan.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/util/virnetdevmacvlan.h b/src/util/virnetdevmacvlan.h index 2844a44..70e3b01 100644 --- a/src/util/virnetdevmacvlan.h +++ b/src/util/virnetdevmacvlan.h @@ -74,8 +74,8 @@ int virNetDevMacVLanCreateWithVPortProfile(const char *ifname, int *tapfd, size_t tapfdSize, unsigned int flags) - ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(3) ATTRIBUTE_NONNULL(6) - ATTRIBUTE_NONNULL(8) ATTRIBUTE_NONNULL(12) ATTRIBUTE_RETURN_CHECK; + ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(3) ATTRIBUTE_NONNULL(5) + ATTRIBUTE_NONNULL(7) ATTRIBUTE_NONNULL(9) ATTRIBUTE_RETURN_CHECK; int virNetDevMacVLanDeleteWithVPortProfile(const char *ifname, const virMacAddr *macaddress, -- 2.5.0

1 0

[libvirt] [PATCH v2 00/10] Make loading domains with invalid XML possible
by Martin Kletzander 11 Dec '15

11 Dec '15

We always had to deal with new parsing errors in a weird way. All of them needed to go into functions starting the domains. That messes up the code, it's confusing to newcomers and so on. I once had an idea that we can handle this stuff. We know what failed, we have the XML that failed parsing and if the problem is not in the domain name nor UUID, we can create a domain object out of that as well. This way we are able to do something we weren't with this series applied. Example follows. Assume "dummy" is a domain with invalid XML (I just modified the file). Now, just for the purpose of this silly example, let's say we allowed domains with archtecture x86_*, but now we've realized that x86_64 is the only one we want to allow, but there already is a domain defined that has <type arch='x86_256' .../>. With the current master, the domain would be lost, so we would need to modify the funstion starting the domain (e.g. qemuProcessStart for qemu domain). However, with this series, it behaves like this: # virsh list --all --reason Id Name State Reason --------------------------------------------------------------- - dummy shut off invalid XML # virsh domstate --reason dummy shut off (invalid XML) # virsh start dummy error: Failed to start domain dummy error: XML error: domain 'dummy' was not loaded due to an XML error (unsupported configuration: Unknown architecture x86_256), please redefine it # VISUAL='sed -i s/x86_256/x86_64/' virsh edit dummy Domain dummy XML configuration edited. # virsh domstate --reason dummy shut off (unknown) # virsh start dummy Domain dummy started This is a logical next step for us to clean and separate parsing and starting, getting rid of some old code without sacrifying compatibility and maybe generating parser code in the future. v2: - rebased on top of current master (with virdomainobjlist.c) - only disallow starting domains with invalid definitions (change done in v1), but allow re-defining them - add support for "virsh list --reason" to better excercise the feature added in this patchset v1: - rebase - don't allow starting domains with invalid state - https://www.redhat.com/archives/libvir-list/2015-November/msg01127.html RFC: https://www.redhat.com/archives/libvir-list/2015-September/msg00698.html Martin Kletzander (10): conf, virsh: Add new domain shutoff reason virsh: Refactor the table output of the list command virsh: Add support for list -reason qemu: Few whitespace cleanups conf: Extract name-parsing into its own function conf: Extract UUID parsing into its own function conf: Optionally keep domains with invalid XML, but don't allow starting them qemu: Don't lookup invalid domains unless specified otherwise qemu: Prepare basic APIs to handle invalid defs qemu: Load domains with invalid XML on start include/libvirt/libvirt-domain.h | 2 + src/bhyve/bhyve_driver.c | 2 + src/conf/domain_conf.c | 121 +++++++++++++++++++++++++++++++-------- src/conf/domain_conf.h | 7 +++ src/conf/virdomainobjlist.c | 71 +++++++++++++++++++++-- src/conf/virdomainobjlist.h | 1 + src/libvirt_private.syms | 1 + src/libxl/libxl_driver.c | 3 + src/lxc/lxc_driver.c | 3 + src/qemu/qemu_driver.c | 64 +++++++++++++++++---- src/uml/uml_driver.c | 2 + tests/virshtest.c | 30 +++++++--- tools/virsh-domain-monitor.c | 60 ++++++++++++------- tools/virsh.pod | 5 +- 14 files changed, 303 insertions(+), 69 deletions(-) -- 2.6.3

3 26

[libvirt] [PATCH v4 00/10] Few VFIO related fixes
by Shivaprasad G Bhat 11 Dec '15

11 Dec '15

The series fixes few VFIO related host crash issues. The patches 3, 4, 9 and 10 are actual fixes. Patch 7 and 8 are test changes to allow testing patch 9. Rest of the patches are mostly code movements except for patch 2. --- Changes from v3 -> v4 - Andrea's suggestion not to overload the usage of stubDriver to set it in virPCINew is addressed. Not setting it any more. Fetch it fresh from sysfs. - The test case added in old P7 is improvised - Old P7 is split into 3 patches where I have moved the virpcimock changes to support iommu into a new patch. Shivaprasad G Bhat (10): Implement virPCIIsKnownStub function Add iommu group number info to virPCIDevice Refuse to reattach from vfio if the iommu group is in use by any domain Wait for vfio-pci device cleanups before reassinging the device to host driver Split reprobe action from the virPCIUnbindFromStub into a new function Pass activeDevs and inactiveDevs to virPCIDeviceUnbindFromStub and virPCIDeviceBindToStub Change the negative test case to try pciback instead of vfio-pci Add iommu info for pci on mocked sysfs Postpone reprobing till all the devices in iommu group are unbound from vfio Wait for iommmu device to go away before reprobing the host driver src/util/virhostdev.c | 28 ++++- src/util/virpci.c | 297 +++++++++++++++++++++++++++++++++++++++++-------- tests/virpcimock.c | 193 +++++++++++++++++++++++++++++--- tests/virpcitest.c | 117 +++++++++++++++++++ 4 files changed, 566 insertions(+), 69 deletions(-) -- Signature

6 37

[libvirt] [PATCH v3] storage sheepdog: allow to specify redundancy level
by Vasiliy Tolstov 11 Dec '15

11 Dec '15

Completely fix tests and check that all works on production env Vasiliy Tolstov (1): storage sheepdog: allow to specify redundancy level docs/schemas/storagevol.rng | 3 + src/conf/storage_conf.c | 2 + src/storage/storage_backend_sheepdog.c | 143 +++++++++++++++++----------- src/util/virstoragefile.c | 4 +- src/util/virstoragefile.h | 2 + tests/storagebackendsheepdogtest.c | 104 ++++++++++---------- tests/storagevolxml2xmlin/vol-sheepdog.xml | 1 + tests/storagevolxml2xmlout/vol-sheepdog.xml | 1 + 8 files changed, 153 insertions(+), 107 deletions(-) -- 2.5.0

2 3

[libvirt] [PATCH] test: qemuxml2argv: Mock virMemoryMaxValue to remove 32/64 bit difference
by Peter Krempa 10 Dec '15

10 Dec '15

Always return LLONG_MAX even on 32 bit systems. The limitation originates from our use of "unsigned long" in several APIs. The internal data type is unsigned long long. Make the test suite deterministic by removing the architecture difference. Flaw was introduced in 645881139b3d2c86acf9d644c3a1471520bc9e57 where I've added a test that uses too large numbers. --- I've tested this by changing the mock value to the number returned on 32-bit systems and got the same error. I didn't test it on an actual 32 bit system though. src/util/virutil.c | 2 ++ tests/qemuxml2argvmock.c | 10 ++++++++++ 2 files changed, 12 insertions(+) diff --git a/src/util/virutil.c b/src/util/virutil.c index fe65faf..9d56d66 100644 --- a/src/util/virutil.c +++ b/src/util/virutil.c @@ -2639,6 +2639,8 @@ virMemoryLimitIsSet(unsigned long long value) * @capped: whether the value must fit into unsigned long * (long long is assumed otherwise) * + * Note: This function is mocked in tests/qemuxml2argvmock.c for test stability + * * Returns the maximum possible memory value in bytes. */ unsigned long long diff --git a/tests/qemuxml2argvmock.c b/tests/qemuxml2argvmock.c index e58b8ce..8426108 100644 --- a/tests/qemuxml2argvmock.c +++ b/tests/qemuxml2argvmock.c @@ -74,3 +74,13 @@ virTPMCreateCancelPath(const char *devpath) return path; } + +/** + * Large values for memory would fail on 32 bit systems, despite having + * variables that support it. + */ +unsigned long long +virMemoryMaxValue(bool capped ATTRIBUTE_UNUSED) +{ + return LLONG_MAX; +} -- 2.6.2

1 0

[libvirt] [PATCH] vz: Allow to create container based on template
by Mikhail Feoktistov 10 Dec '15

10 Dec '15

We shouldn't delete disk from default config if we create container based on template, because we don't have the new disk from XML, only template name. And don't add template section from XML as new filesystem, we use PrlVmCfg_SetOsTemplate function to set template name. --- src/vz/vz_sdk.c | 33 +++++++++++++++++++++++++++++---- 1 file changed, 29 insertions(+), 4 deletions(-) diff --git a/src/vz/vz_sdk.c b/src/vz/vz_sdk.c index 89c9e89..865cabe 100644 --- a/src/vz/vz_sdk.c +++ b/src/vz/vz_sdk.c @@ -2096,12 +2096,14 @@ prlsdkCheckUnsupportedParams(PRL_HANDLE sdkdom, virDomainDefPtr def) return 0; } -static int prlsdkClearDevices(PRL_HANDLE sdkdom) +static int prlsdkClearDevices(PRL_HANDLE sdkdom, bool skipdisk) { PRL_RESULT pret; PRL_UINT32 n, i; PRL_HANDLE devList; PRL_HANDLE dev; + PRL_DEVICE_TYPE devType; + PRL_VM_DEV_EMULATION_TYPE emul; int ret = -1; pret = PrlVmCfg_SetVNCMode(sdkdom, PRD_DISABLED); @@ -2117,6 +2119,18 @@ static int prlsdkClearDevices(PRL_HANDLE sdkdom) pret = PrlHndlList_GetItem(devList, i, &dev); prlsdkCheckRetGoto(pret, cleanup); + if (skipdisk) { + pret = PrlVmDev_GetType(dev, &devType); + prlsdkCheckRetGoto(pret, cleanup); + + pret = PrlVmDev_GetEmulatedType(dev, &emul); + prlsdkCheckRetGoto(pret, cleanup); + + if (devType == PDE_HARD_DISK) { + PrlHandle_Free(dev); + continue; + } + } pret = PrlVmDev_Remove(dev); PrlHandle_Free(dev); } @@ -3465,6 +3479,7 @@ prlsdkDoApplyConfig(virConnectPtr conn, char uuidstr[VIR_UUID_STRING_BUFLEN + 2]; bool needBoot = true; char *mask = NULL; + bool skipdisk = false; if (prlsdkCheckUnsupportedParams(sdkdom, def) < 0) return -1; @@ -3514,7 +3529,11 @@ prlsdkDoApplyConfig(virConnectPtr conn, } prlsdkCheckRetGoto(pret, error); - if (prlsdkClearDevices(sdkdom) < 0) + if (def->nfss == 1 && + def->fss[0]->type == VIR_DOMAIN_FS_TYPE_TEMPLATE) + skipdisk = true; + + if (prlsdkClearDevices(sdkdom, skipdisk) < 0) goto error; if (prlsdkRemoveBootDevices(sdkdom) < 0) @@ -3544,6 +3563,8 @@ prlsdkDoApplyConfig(virConnectPtr conn, for (i = 0; i < def->nfss; i++) { if (STREQ(def->fss[i]->dst, "/")) needBoot = false; + if (def->fss[i]->type == VIR_DOMAIN_FS_TYPE_TEMPLATE) + continue; if (prlsdkAddFS(sdkdom, def->fss[i]) < 0) goto error; } @@ -3655,6 +3676,7 @@ prlsdkCreateCt(virConnectPtr conn, virDomainDefPtr def) int ret = -1; int useTemplate = 0; size_t i; + PRL_UINT32 flags = 0; if (def->nfss > 1) { /* Check all filesystems */ @@ -3696,8 +3718,11 @@ prlsdkCreateCt(virConnectPtr conn, virDomainDefPtr def) if (ret) goto cleanup; - job = PrlVm_RegEx(sdkdom, "", - PACF_NON_INTERACTIVE_MODE | PRNVM_PRESERVE_DISK); + flags = PACF_NON_INTERACTIVE_MODE; + if (!useTemplate) + flags = flags | PRNVM_PRESERVE_DISK; + + job = PrlVm_RegEx(sdkdom, "", flags); if (PRL_FAILED(waitJob(job))) ret = -1; -- 1.8.3.1

2 2

[libvirt] [PATCH 0/7] cleanup the input device code
by Pavel Hrdina 10 Dec '15

10 Dec '15

This patch series is an attempt to make the code for input devices cleaner and readable and also follow some good practices whit rule that one function should do only one thing. We are really bad at this and our code is sometimes really hard to follow. Pavel Hrdina (7): vmware/vmx: use virDomainDefPostParse after parsing vmx config lxc: use virDomainDefPostParse for parsing LXC config string xen: use virDomainDefPostParse for parsing XM/XL/SEXPR cofings xen: fix timer bug found by updated test libxl: copy persistent domain definition while starting a guest device: cleanup input device code qemu_domain: cleanup default input device code src/Makefile.am | 2 +- src/conf/domain_conf.c | 77 +--------------------- src/esx/esx_driver.c | 4 +- src/libxl/libxl_domain.c | 8 +++ src/libxl/libxl_driver.c | 8 ++- src/lxc/lxc_driver.c | 5 +- src/lxc/lxc_native.c | 8 ++- src/lxc/lxc_native.h | 4 +- src/qemu/qemu_domain.c | 47 ++++++++----- src/qemu/qemu_domain.h | 1 + src/qemu/qemu_process.c | 3 + src/vmware/vmware_conf.c | 3 +- src/vmware/vmware_driver.c | 2 +- src/vmx/vmx.c | 5 ++ src/vmx/vmx.h | 1 + src/xen/xen_driver.c | 5 +- src/xen/xend_internal.c | 10 +++ src/xen/xm_internal.c | 2 +- src/xenapi/xenapi_driver.c | 8 +++ src/xenconfig/xen_common.c | 24 +++++++ src/xenconfig/xen_common.h | 2 + src/xenconfig/xen_sxpr.c | 18 ++++- src/xenconfig/xen_sxpr.h | 8 ++- src/xenconfig/xen_xl.c | 9 ++- src/xenconfig/xen_xl.h | 4 +- src/xenconfig/xen_xm.c | 7 +- src/xenconfig/xen_xm.h | 2 +- tests/Makefile.am | 2 +- tests/lxcconf2xmldata/lxcconf2xml-blkiotune.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-cpusettune.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-cputune.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-idmap.xml | 3 +- .../lxcconf2xmldata/lxcconf2xml-macvlannetwork.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-memtune.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-nonenetwork.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-nonetwork.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-physnetwork.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-simple.xml | 3 +- tests/lxcconf2xmldata/lxcconf2xml-vlannetwork.xml | 3 +- tests/lxcconf2xmltest.c | 18 ++++- .../qemuxml2argv-graphics-listen-network.xml | 2 - .../qemuxml2argv-graphics-listen-network2.xml | 2 - .../qemuxml2argv-graphics-sdl-fullscreen.xml | 2 - .../qemuxml2argvdata/qemuxml2argv-graphics-sdl.xml | 2 - ...qemuxml2argv-graphics-spice-agent-file-xfer.xml | 1 - .../qemuxml2argv-graphics-spice-compression.xml | 2 - .../qemuxml2argv-graphics-spice-listen-network.xml | 1 - .../qemuxml2argv-graphics-spice-qxl-vga.xml | 2 - .../qemuxml2argv-graphics-spice-sasl.xml | 1 - .../qemuxml2argv-graphics-spice-timeout.xml | 2 - .../qemuxml2argv-graphics-spice.xml | 2 - .../qemuxml2argv-graphics-vnc-policy.xml | 2 - .../qemuxml2argv-graphics-vnc-sasl.xml | 2 - .../qemuxml2argv-graphics-vnc-socket.xml | 2 - .../qemuxml2argv-graphics-vnc-tls.xml | 2 - .../qemuxml2argv-graphics-vnc-websocket.xml | 2 - .../qemuxml2argvdata/qemuxml2argv-graphics-vnc.xml | 2 - .../qemuxml2argv-pci-autoadd-addr.xml | 1 - .../qemuxml2argv-pci-autoadd-idx.xml | 1 - .../qemuxml2argvdata/qemuxml2argv-pseries-disk.xml | 2 +- .../qemuxml2xmlout-graphics-listen-network2.xml | 2 - .../qemuxml2xmlout-graphics-spice-timeout.xml | 2 - tests/sexpr2xmldata/sexpr2xml-boot-grub.xml | 1 + tests/sexpr2xmldata/sexpr2xml-bridge-ipaddr.xml | 1 + tests/sexpr2xmldata/sexpr2xml-curmem.xml | 3 +- .../sexpr2xml-disk-block-shareable.xml | 1 + tests/sexpr2xmldata/sexpr2xml-disk-block.xml | 1 + .../sexpr2xml-disk-drv-blktap-qcow.xml | 1 + .../sexpr2xml-disk-drv-blktap-raw.xml | 1 + .../sexpr2xml-disk-drv-blktap2-raw.xml | 1 + tests/sexpr2xmldata/sexpr2xml-disk-file.xml | 1 + tests/sexpr2xmldata/sexpr2xml-fv-autoport.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-empty-kernel.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-force-hpet.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-force-nohpet.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-kernel.xml | 1 + tests/sexpr2xmldata/sexpr2xml-fv-legacy-vfb.xml | 2 + tests/sexpr2xmldata/sexpr2xml-fv-localtime.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-net-ioemu.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-net-netfront.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-parallel-tcp.xml | 5 +- .../sexpr2xml-fv-serial-dev-2-ports.xml | 5 +- .../sexpr2xml-fv-serial-dev-2nd-port.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-serial-file.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-serial-null.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-serial-pipe.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-serial-pty.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-serial-stdio.xml | 5 +- .../sexpr2xml-fv-serial-tcp-telnet.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-serial-tcp.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-serial-udp.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-serial-unix.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-sound-all.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-sound.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-usbmouse.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-usbtablet.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-utc.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv-v2.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-fv.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-net-bridged.xml | 1 + tests/sexpr2xmldata/sexpr2xml-net-e1000.xml | 1 + tests/sexpr2xmldata/sexpr2xml-net-routed.xml | 1 + tests/sexpr2xmldata/sexpr2xml-no-source-cdrom.xml | 5 +- tests/sexpr2xmldata/sexpr2xml-pci-devs.xml | 1 + .../sexpr2xml-pv-bootloader-cmdline.xml | 1 + tests/sexpr2xmldata/sexpr2xml-pv-bootloader.xml | 1 + tests/sexpr2xmldata/sexpr2xml-pv-localtime.xml | 1 + tests/sexpr2xmldata/sexpr2xml-pv-vcpus.xml | 1 + .../sexpr2xml-pv-vfb-new-vncdisplay.xml | 3 +- tests/sexpr2xmldata/sexpr2xml-pv-vfb-new.xml | 3 +- tests/sexpr2xmldata/sexpr2xml-pv-vfb-orig.xml | 3 +- .../sexpr2xmldata/sexpr2xml-pv-vfb-type-crash.xml | 3 +- tests/sexpr2xmldata/sexpr2xml-pv.xml | 1 + tests/sexpr2xmltest.c | 10 ++- tests/vmx2xmldata/vmx2xml-graphics-vnc.xml | 2 - tests/vmx2xmltest.c | 2 +- .../test-fullvirt-direct-kernel-boot.xml | 4 +- tests/xlconfigdata/test-fullvirt-multiusb.xml | 4 +- tests/xlconfigdata/test-new-disk.xml | 5 +- tests/xlconfigdata/test-spice-features.xml | 4 +- tests/xlconfigdata/test-spice.xml | 4 +- tests/xlconfigtest.c | 2 +- tests/xmconfigdata/test-escape-paths.xml | 6 +- .../xmconfigdata/test-fullvirt-default-feature.xml | 5 +- tests/xmconfigdata/test-fullvirt-force-hpet.xml | 5 +- tests/xmconfigdata/test-fullvirt-force-nohpet.xml | 5 +- tests/xmconfigdata/test-fullvirt-localtime.xml | 5 +- tests/xmconfigdata/test-fullvirt-net-ioemu.xml | 5 +- tests/xmconfigdata/test-fullvirt-net-netfront.xml | 5 +- tests/xmconfigdata/test-fullvirt-new-cdrom.xml | 5 +- tests/xmconfigdata/test-fullvirt-old-cdrom.xml | 5 +- tests/xmconfigdata/test-fullvirt-parallel-tcp.xml | 5 +- .../test-fullvirt-serial-dev-2-ports.xml | 5 +- .../test-fullvirt-serial-dev-2nd-port.xml | 5 +- tests/xmconfigdata/test-fullvirt-serial-file.xml | 5 +- tests/xmconfigdata/test-fullvirt-serial-null.xml | 5 +- tests/xmconfigdata/test-fullvirt-serial-pipe.xml | 5 +- tests/xmconfigdata/test-fullvirt-serial-pty.xml | 5 +- tests/xmconfigdata/test-fullvirt-serial-stdio.xml | 5 +- .../test-fullvirt-serial-tcp-telnet.xml | 5 +- tests/xmconfigdata/test-fullvirt-serial-tcp.xml | 5 +- tests/xmconfigdata/test-fullvirt-serial-udp.xml | 5 +- tests/xmconfigdata/test-fullvirt-serial-unix.xml | 5 +- tests/xmconfigdata/test-fullvirt-sound.xml | 5 +- tests/xmconfigdata/test-fullvirt-usbmouse.xml | 5 +- .../test-fullvirt-usbtablet-no-bus.xml | 4 +- tests/xmconfigdata/test-fullvirt-usbtablet.xml | 5 +- tests/xmconfigdata/test-fullvirt-utc.xml | 5 +- tests/xmconfigdata/test-no-source-cdrom.xml | 5 +- tests/xmconfigdata/test-paravirt-net-e1000.xml | 3 +- tests/xmconfigdata/test-paravirt-net-vifname.xml | 3 +- .../test-paravirt-new-pvfb-vncdisplay.xml | 3 +- tests/xmconfigdata/test-paravirt-new-pvfb.xml | 3 +- .../test-paravirt-old-pvfb-vncdisplay.xml | 3 +- tests/xmconfigdata/test-paravirt-old-pvfb.xml | 3 +- tests/xmconfigdata/test-paravirt-vcpu.xml | 1 + tests/xmconfigdata/test-pci-devs.xml | 5 +- tests/xmconfigtest.c | 3 +- 158 files changed, 429 insertions(+), 309 deletions(-) -- 2.6.3

4 19

[libvirt] [PATCH 0/7] tests: Always fake root directory
by Andrea Bolognani 10 Dec '15

10 Dec '15

When mocking filesystem access in the test suite, we have always assumed it to be some subdirectory of /sys depending on the needs of the specific test case. This limits our flexibility, and will be a problem once we need to start mocking eg. /dev as well, or simply two different parts of the /sys filesystem at the same time[1]. Solve the issue by always using the temporary directory as the root directory for the mocked filesystem. The series is organized as follows: 1-2: Tiny cleanups 3-4: Change the mock libraries so that they build the proper directory structure 5: Change the name of the environment variable used to pass the temporary directory name to the mock libraries 6: Make it so we'll be able to mock different parts of the /sys filesystem at the same time 7: Update scsihosttest to use the same directory structure Cheers. [1] https://www.redhat.com/archives/libvir-list/2015-November/msg00532.html is an example of why we would need to do that Andrea Bolognani (7): tests: scsihost: Don't set LIBVIRT_FAKE_SYSFS_DIR tests: pcimock: Remove check for fakesysfsdir tests: pcimock: Use the temporary directory as fake root tests: cgroupmock: Use the temporary directory as fake root tests: Rename LIBVIRT_FAKE_SYSFS_DIR to LIBVIRT_FAKE_ROOT_DIR tests: Use more specific names for variables tests: scsihost: Use fakerootdir instead of fakesysfsdir tests/scsihosttest.c | 17 ++++++--- tests/vircgroupmock.c | 99 ++++++++++++++++++++++++++++---------------------- tests/vircgrouptest.c | 16 ++++---- tests/virhostdevtest.c | 16 ++++---- tests/virpcimock.c | 69 +++++++++++++++++++---------------- tests/virpcitest.c | 16 ++++---- 6 files changed, 129 insertions(+), 104 deletions(-) -- 2.5.0

2 10

[libvirt] [PATCH v2 0/8] Add perf and Intel CMT feature support
by Qiaowei Ren 09 Dec '15

09 Dec '15

The series mainly adds Intel CMT feature support into libvirt. CMT is new introduced PQos (Platform Qos) feature to monitor the usage of cache by applications running on the platform. Currently CMT patches has been merged into Linux kernel mainline. The CMT implementation in Linux kernel is based on perf mechanism and there is no support for perf in libvirt, and so this series firstly add perf support into libvirt, including two public API and a set of util interfaces. And based on these APIs and interfaces, thie series implements CMT perf event support. TODO: 1. add support for new APIs into libvirt-python library. Changes since v1: * change perf APIs implementation to match new style of the API. * add new xml element * reenable all perf events previously enabled when libvirtd daemon restart. * redesign perf util functions. Qiaowei Ren (8): perf: add new public APIs for perf event perf: implement the remote protocol for perf event perf: implement a set of util functions for perf event qemu_driver: add support to perf event perf: add new xml element perf: reenable perf events when libvirtd restart virsh: implement new command to support perf virsh: extend domstats command daemon/remote.c | 47 ++++++ docs/schemas/domaincommon.rng | 27 ++++ include/libvirt/libvirt-domain.h | 19 +++ include/libvirt/virterror.h | 1 + src/Makefile.am | 1 + src/conf/domain_conf.c | 37 +++++ src/conf/domain_conf.h | 10 ++ src/driver-hypervisor.h | 12 ++ src/libvirt-domain.c | 93 ++++++++++++ src/libvirt_private.syms | 12 ++ src/libvirt_public.syms | 6 + src/qemu/qemu_domain.h | 3 + src/qemu/qemu_driver.c | 206 +++++++++++++++++++++++++++ src/qemu/qemu_process.c | 9 ++ src/remote/remote_driver.c | 39 +++++ src/remote/remote_protocol.x | 30 +++- src/remote_protocol-structs | 18 +++ src/util/virerror.c | 1 + src/util/virperf.c | 298 +++++++++++++++++++++++++++++++++++++++ src/util/virperf.h | 61 ++++++++ tools/virsh-domain-monitor.c | 7 + tools/virsh-domain.c | 125 ++++++++++++++++ tools/virsh.pod | 25 +++- 23 files changed, 1084 insertions(+), 3 deletions(-) create mode 100644 src/util/virperf.c create mode 100644 src/util/virperf.h -- 1.9.1

3 18