- Devel - libvirt List Archives

[libvirt] [PATCH 0/2] Fix a few memory leaks in libvirtd
by Jiri Denemark 27 Sep '13

27 Sep '13

Jiri Denemark (2): qemu: Don't leak reference to virQEMUDriverConfigPtr qemu: Free all driver data in qemuStateCleanup src/qemu/qemu_driver.c | 2 ++ src/qemu/qemu_process.c | 3 ++- 2 files changed, 4 insertions(+), 1 deletion(-) -- 1.8.3.2

2 4

[libvirt] [PATCH] build: Fix VPATH build error for locking daemon
by Viktor Mihajlovski 27 Sep '13

27 Sep '13

Removed superfluous/wrong srcdir prefix. Signed-off-by: Viktor Mihajlovski <mihajlov(a)linux.vnet.ibm.com> --- src/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Makefile.am b/src/Makefile.am index 4057eda..09311b7 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -196,7 +196,7 @@ BUILT_SOURCES += $(LOCK_PROTOCOL_GENERATED) MAINTAINERCLEANFILES += $(LOCK_PROTOCOL_GENERATED) LOCK_DAEMON_GENERATED = \ - $(srcdir)/locking/lock_daemon_dispatch_stubs.h + locking/lock_daemon_dispatch_stubs.h $(NULL) BUILT_SOURCES += $(LOCK_DAEMON_GENERATED) -- 1.7.9.5

2 1

[libvirt] [PATCH]lxc: don't start container when no root fs found
by Chen Hanxiao 27 Sep '13

27 Sep '13

From: Chen Hanxiao <chenhanxiao(a)cn.fujitsu.com> Currently, if we don't explicitly add root fs for container, libvirt will add one for us implicitly with "/" as src. It would be not safe. Unless user asked for it, we should not assume this. Signed-off-by: Chen Hanxiao <chenhanxiao(a)cn.fujitsu.com> --- src/lxc/lxc_process.c | 29 +++++++---------------------- 1 file changed, 7 insertions(+), 22 deletions(-) diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c index 4835bd5..4f4a906 100644 --- a/src/lxc/lxc_process.c +++ b/src/lxc/lxc_process.c @@ -911,29 +911,14 @@ virLXCProcessEnsureRootFS(virDomainObjPtr vm) { virDomainFSDefPtr root = virDomainGetRootFilesystem(vm->def); - if (root) + if (root) { return 0; - - if (VIR_ALLOC(root) < 0) - goto error; - - root->type = VIR_DOMAIN_FS_TYPE_MOUNT; - - if (VIR_STRDUP(root->src, "/") < 0 || - VIR_STRDUP(root->dst, "/") < 0) - goto error; - - if (VIR_INSERT_ELEMENT(vm->def->fss, - 0, - vm->def->nfss, - root) < 0) - goto error; - - return 0; - -error: - virDomainFSDefFree(root); - return -1; + } else { + errno = EINVAL; + virReportSystemError(errno, "%s", + _("No root fs found for container")); + return -1; + } } /** -- 1.8.2.1

2 2

[libvirt] [PATCHv2] util: recognize SMB/CIFS filesystems as shared
by Laine Stump 27 Sep '13

27 Sep '13

This should resolve: https://bugzilla.redhat.com/show_bug.cgi?id=1012085 libvirt previously recognized NFS, GFS2, OCFS2, and AFS filesystems as "shared", and thus eligible for exceptions to certain rules/actions about chowning image files before handing them off to a guest. This patch widens the definition of "shared filesystem" to include SMB and CIFS filesystems (aka "Windows file sharing"); both of these use the same protocol, but different drivers so there are different magic numbers for each. --- Change from V1 - also check for CIFS_SUPER_MAGIC, not just SMB_SUPER_MAGIC. src/util/virstoragefile.c | 16 +++++++++++++++- src/util/virstoragefile.h | 2 ++ 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c index 0b9cec3..48d5fbb 100644 --- a/src/util/virstoragefile.c +++ b/src/util/virstoragefile.c @@ -1241,6 +1241,12 @@ cleanup: # ifndef AFS_FS_MAGIC # define AFS_FS_MAGIC 0x6B414653 # endif +# ifndef SMB_SUPER_MAGIC +# define SMB_SUPER_MAGIC 0x517B +# endif +# ifndef CIFS_SUPER_MAGIC +# define CIFS_SUPER_MAGIC 0xFF534D42 +# endif int virStorageFileIsSharedFSType(const char *path, @@ -1304,6 +1310,12 @@ int virStorageFileIsSharedFSType(const char *path, if ((fstypes & VIR_STORAGE_FILE_SHFS_AFS) && (sb.f_type == AFS_FS_MAGIC)) return 1; + if ((fstypes & VIR_STORAGE_FILE_SHFS_SMB) && + (sb.f_type == SMB_SUPER_MAGIC)) + return 1; + if ((fstypes & VIR_STORAGE_FILE_SHFS_CIFS) && + (sb.f_type == CIFS_SUPER_MAGIC)) + return 1; return 0; } @@ -1322,7 +1334,9 @@ int virStorageFileIsSharedFS(const char *path) VIR_STORAGE_FILE_SHFS_NFS | VIR_STORAGE_FILE_SHFS_GFS2 | VIR_STORAGE_FILE_SHFS_OCFS | - VIR_STORAGE_FILE_SHFS_AFS); + VIR_STORAGE_FILE_SHFS_AFS | + VIR_STORAGE_FILE_SHFS_SMB | + VIR_STORAGE_FILE_SHFS_CIFS); } int virStorageFileIsClusterFS(const char *path) diff --git a/src/util/virstoragefile.h b/src/util/virstoragefile.h index 1f89839..d5effa4 100644 --- a/src/util/virstoragefile.h +++ b/src/util/virstoragefile.h @@ -112,6 +112,8 @@ enum { VIR_STORAGE_FILE_SHFS_GFS2 = (1 << 1), VIR_STORAGE_FILE_SHFS_OCFS = (1 << 2), VIR_STORAGE_FILE_SHFS_AFS = (1 << 3), + VIR_STORAGE_FILE_SHFS_SMB = (1 << 4), + VIR_STORAGE_FILE_SHFS_CIFS = (1 << 5), }; int virStorageFileIsSharedFS(const char *path); -- 1.7.11.7

2 1

[libvirt] [PATCH] build: fix testsuite building under cygwin
by Eric Blake 27 Sep '13

27 Sep '13

Similar to commit 8f34f19. * tests/Makefile.am (virnetserverclienttest_CFLAGS): Add XDR_CFLAGS. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Pushing under the build-breaker rule. tests/Makefile.am | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/Makefile.am b/tests/Makefile.am index 777758f..1a7ff4b 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -639,6 +639,7 @@ virnetsockettest_LDADD = $(LDADDS) virnetserverclienttest_SOURCES = \ virnetserverclienttest.c \ testutils.h testutils.c +virnetserverclienttest_CFLAGS = $(XDR_CFLAGS) $(AM_CFLAGS) virnetserverclienttest_LDADD = $(LDADDS) virnetserverclientmock_la_SOURCES = \ -- 1.8.3.1

1 0

[libvirt] [PATCH] rpc: Increase bound limit for virDomainGetJobStats
by Jiri Denemark 27 Sep '13

27 Sep '13

https://bugzilla.redhat.com/show_bug.cgi?id=1012818 Commit 6d7d0b1869ed293e3208d11f375cecea0129dfc5 (in 1.1.2) added bounds checking to virDomainGetJobStats. But even at that time the API was able to return 20 parameters while the limit was set to 16. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/remote/remote_protocol.x | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/remote/remote_protocol.x b/src/remote/remote_protocol.x index 8d17bad..810eeca 100644 --- a/src/remote/remote_protocol.x +++ b/src/remote/remote_protocol.x @@ -230,7 +230,7 @@ const REMOTE_NODE_MEMORY_PARAMETERS_MAX = 64; const REMOTE_DOMAIN_MIGRATE_PARAM_LIST_MAX = 64; /* Upper limit on number of job stats */ -const REMOTE_DOMAIN_JOB_STATS_MAX = 16; +const REMOTE_DOMAIN_JOB_STATS_MAX = 64; /* Upper limit on number of CPU models */ const REMOTE_CONNECT_CPU_MODELS_MAX = 8192; -- 1.8.3.2

2 2

[libvirt] [PATCH] qemu_migrate: Fix assign the same port when migrating concurrently
by Wangyufei (A) 27 Sep '13

27 Sep '13

From: WangYufei <james.wangyufei(a)huawei.com> Date: Fri, 27 Sep 2013 11:53:57 +0800 Subject: [PATCH] qemu_migrate: Fix assign the same port when migrating concurrently When we migrate vms concurrently, there's a chance that libvirtd on destination assign the same port for different migrations, which will lead to migration failed during migration prepare phase on destination. So we change the port increase to atomic operation to solve the problem. Signed-off-by: WangYufei <james.wangyufei(a)huawei.com> --- src/qemu/qemu_migration.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 3a1aab7..0f496f4 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -57,6 +57,7 @@ #include "virhook.h" #include "virstring.h" #include "virtypedparam.h" +#include "viratomic.h" #define VIR_FROM_THIS VIR_FROM_QEMU @@ -2521,7 +2522,7 @@ qemuMigrationPrepareDirect(virQEMUDriverPtr driver, * to be a correct hostname which refers to the target machine). */ if (uri_in == NULL) { - this_port = QEMUD_MIGRATION_FIRST_PORT + port++; + this_port = QEMUD_MIGRATION_FIRST_PORT + virAtomicIntInc(&port); if (port == QEMUD_MIGRATION_NUM_PORTS) port = 0; /* Get hostname */ @@ -2578,7 +2579,7 @@ qemuMigrationPrepareDirect(virQEMUDriverPtr driver, if (uri->port == 0) { /* Generate a port */ - this_port = QEMUD_MIGRATION_FIRST_PORT + port++; + this_port = QEMUD_MIGRATION_FIRST_PORT + virAtomicIntInc(&port); if (port == QEMUD_MIGRATION_NUM_PORTS) port = 0; -- 1.7.3.1.msysgit.0 Best Regards, -WangYufei

3 3

[libvirt] [libvirt-designer 1/2] Fix typo in error message
by Christophe Fergeau 27 Sep '13

27 Sep '13

--- Pushed under the trivial rule libvirt-designer/libvirt-designer-main.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libvirt-designer/libvirt-designer-main.h b/libvirt-designer/libvirt-designer-main.h index 5e82f8a..9341754 100644 --- a/libvirt-designer/libvirt-designer-main.h +++ b/libvirt-designer/libvirt-designer-main.h @@ -21,7 +21,7 @@ */ #if !defined(__LIBVIRT_DESIGNER_H__) && !defined(LIBVIRT_DESIGNER_BUILD) -#error "Only <libvirt-gdesigner/libvirt-gdesigner.h> can be included directly." +#error "Only <libvirt-designer/libvirt-designer.h> can be included directly." #endif #include <osinfo/osinfo.h> -- 1.8.3.1

2 2

[libvirt] [PATCH]LXC: fix a comment typo in lxc_controller.c
by Chen Hanxiao 27 Sep '13

27 Sep '13

From: Chen Hanxiao <chenhanxiao(a)cn.fujitsu.com> LXC: fix a comment typo in lxc_controller.c Signed-off-by: Chen Hanxiao <chenhanxiao(a)cn.fujitsu.com> --- src/lxc/lxc_controller.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/lxc/lxc_controller.c b/src/lxc/lxc_controller.c index ed73ab0..636cf6e 100644 --- a/src/lxc/lxc_controller.c +++ b/src/lxc/lxc_controller.c @@ -2011,7 +2011,7 @@ virLXCControllerEventSend(virLXCControllerPtr ctrl, virNetMessagePtr msg; if (!ctrl->client) { - VIR_WARN("Dropping event %d becuase libvirtd is not connected", procnr); + VIR_WARN("Dropping event %d because libvirtd is not connected", procnr); return; } -- 1.8.2.1

2 1

[libvirt] [PATCH] build: fix use of rpc.h on cygwin
by Eric Blake 27 Sep '13

27 Sep '13

Caused by commit 012c25e8 splitting out a convenience library. CC libvirtd_conf_la-libvirtd-config.lo In file included from ../src/rpc/virnetmessage.h:24:0, from ../src/rpc/virnetserverprogram.h:27, from ../src/rpc/virnetserver.h:32, from libvirtd-config.c:31: ../src/rpc/virnetprotocol.h:9:21: fatal error: rpc/rpc.h: No such file or directory * daemon/Makefile.am (libvirtd_conf_la_CFLAGS): Add XDR_CFLAGS. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Pushing under the build-breaker rule (well, I will once libvirt.org is responsive again - it's been having problems with the git server today) daemon/Makefile.am | 1 + 1 file changed, 1 insertion(+) diff --git a/daemon/Makefile.am b/daemon/Makefile.am index eb58de9..46c766c 100644 --- a/daemon/Makefile.am +++ b/daemon/Makefile.am @@ -100,6 +100,7 @@ noinst_LTLIBRARIES = libvirtd_conf.la libvirtd_conf_la_SOURCES = $(LIBVIRTD_CONF_SOURCES) libvirtd_conf_la_CFLAGS = \ $(LIBXML_CFLAGS) \ + $(XDR_CFLAGS) \ $(WARN_CFLAGS) $(PIE_CFLAGS) \ $(COVERAGE_CFLAGS) \ $(NULL) -- 1.8.3.1

1 0

[libvirt] [PATCH]LXC: free dst before lxcDomainAttachDeviceDiskLive return
by Chen Hanxiao 26 Sep '13

26 Sep '13

From: Chen Hanxiao <chenhanxiao(a)cn.fujitsu.com> free dst before lxcDomainAttachDeviceDiskLive return Signed-off-by: Chen Hanxiao <chenhanxiao(a)cn.fujitsu.com> --- src/lxc/lxc_driver.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 9c58f67..7906ad1 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -3053,6 +3053,7 @@ cleanup: virDomainAuditDisk(vm, NULL, def->src, "attach", ret == 0); if (dst && created && ret < 0) unlink(dst); + VIR_FREE(dst); return ret; } -- 1.8.2.1

5 5

[libvirt] Is it possible to attach a harddisk to an ESX node
by varun bhatnagar 26 Sep '13

26 Sep '13

Hi, Is it possible to attach a harddisk to an ESX node later after creating/defining it? If yes, can anyone tell me how to do it? Regards, Varun

2 2

[libvirt] [PATCH 0/3] qemu: Hotplug SCSI controllers/hostdevs
by Eric Farman 26 Sep '13

26 Sep '13

Today, hotplug of a SCSI hostdev relies on the presence of an existing SCSI controller, which can be defined in the initial domain XML. Hotplug of a SCSI controller relies on the presence of a PCI bus, which does not work in all environments. These patches correct this behavior, in order to allow hotplug (and unplug) work without incident. Reviewed-by: Viktor Mihajlovski <mihajlov(a)linux.vnet.ibm.com> Eric Farman (3): qemu: Rename controller hotplug functions to not be PCI-specific qemu: Separate calls based on controller bus type qemu: Auto-generate controller for hotplugged hostdev src/qemu/qemu_driver.c | 4 +-- src/qemu/qemu_hotplug.c | 62 ++++++++++++++++++++++++++++++++--------------- src/qemu/qemu_hotplug.h | 12 ++++----- 3 files changed, 50 insertions(+), 28 deletions(-) -- 1.7.9.5

1 3

[libvirt] [PATCH] util: recognize SMB filesystems as shared
by Laine Stump 26 Sep '13

26 Sep '13

This should resolve: https://bugzilla.redhat.com/show_bug.cgi?id=1012085 libvirt previously recognized NFS, GFS2, OCFS2, and AFS filesystems as "shared", and thus eligible for exceptions to certain rules/actions about chowning image files before handing them off to a guest. This patch widens the definition of "shared filesystem" to include the SMB filesystem (sometimes called CIFS, or "Windows file sharing"). --- src/util/virstoragefile.c | 9 ++++++++- src/util/virstoragefile.h | 1 + 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c index 0b9cec3..ed43b2b 100644 --- a/src/util/virstoragefile.c +++ b/src/util/virstoragefile.c @@ -1241,6 +1241,9 @@ cleanup: # ifndef AFS_FS_MAGIC # define AFS_FS_MAGIC 0x6B414653 # endif +# ifndef SMB_SUPER_MAGIC +# define SMB_SUPER_MAGIC 0x517B +# endif int virStorageFileIsSharedFSType(const char *path, @@ -1304,6 +1307,9 @@ int virStorageFileIsSharedFSType(const char *path, if ((fstypes & VIR_STORAGE_FILE_SHFS_AFS) && (sb.f_type == AFS_FS_MAGIC)) return 1; + if ((fstypes & VIR_STORAGE_FILE_SHFS_SMB) && + (sb.f_type == SMB_SUPER_MAGIC)) + return 1; return 0; } @@ -1322,7 +1328,8 @@ int virStorageFileIsSharedFS(const char *path) VIR_STORAGE_FILE_SHFS_NFS | VIR_STORAGE_FILE_SHFS_GFS2 | VIR_STORAGE_FILE_SHFS_OCFS | - VIR_STORAGE_FILE_SHFS_AFS); + VIR_STORAGE_FILE_SHFS_AFS | + VIR_STORAGE_FILE_SHFS_SMB); } int virStorageFileIsClusterFS(const char *path) diff --git a/src/util/virstoragefile.h b/src/util/virstoragefile.h index 1f89839..9c07212 100644 --- a/src/util/virstoragefile.h +++ b/src/util/virstoragefile.h @@ -112,6 +112,7 @@ enum { VIR_STORAGE_FILE_SHFS_GFS2 = (1 << 1), VIR_STORAGE_FILE_SHFS_OCFS = (1 << 2), VIR_STORAGE_FILE_SHFS_AFS = (1 << 3), + VIR_STORAGE_FILE_SHFS_SMB = (1 << 4), }; int virStorageFileIsSharedFS(const char *path); -- 1.7.11.7

4 4

[libvirt] [libvirt-glib] Use latest warnings.m4 from gnulib
by Christophe Fergeau 26 Sep '13

26 Sep '13

The one we were using does not work properly with clang, causing the build process to try to use -f/-W options that are not supported by clang. --- m4/warnings.m4 | 82 ++++++++++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 62 insertions(+), 20 deletions(-) diff --git a/m4/warnings.m4 b/m4/warnings.m4 index 69d05a6..e3d239b 100644 --- a/m4/warnings.m4 +++ b/m4/warnings.m4 @@ -1,5 +1,5 @@ -# warnings.m4 serial 5 -dnl Copyright (C) 2008-2012 Free Software Foundation, Inc. +# warnings.m4 serial 11 +dnl Copyright (C) 2008-2013 Free Software Foundation, Inc. dnl This file is free software; the Free Software Foundation dnl gives unlimited permission to copy and/or distribute it, dnl with or without modifications, as long as this notice is preserved. @@ -14,24 +14,66 @@ m4_ifdef([AS_VAR_APPEND], [m4_define([gl_AS_VAR_APPEND], [AS_VAR_SET([$1], [AS_VAR_GET([$1])$2])])]) -# gl_WARN_ADD(PARAMETER, [VARIABLE = WARN_CFLAGS]) -# ------------------------------------------------ -# Adds parameter to WARN_CFLAGS if the compiler supports it. For example, -# gl_WARN_ADD([-Wparentheses]). -AC_DEFUN([gl_WARN_ADD], -dnl FIXME: gl_Warn must be used unquoted until we can assume -dnl autoconf 2.64 or newer. -[AS_VAR_PUSHDEF([gl_Warn], [gl_cv_warn_$1])dnl -AC_CACHE_CHECK([whether compiler handles $1], m4_defn([gl_Warn]), [ - gl_save_CPPFLAGS="$CPPFLAGS" - CPPFLAGS="${CPPFLAGS} $1" - AC_PREPROC_IFELSE([AC_LANG_PROGRAM([])], - [AS_VAR_SET(gl_Warn, [yes])], - [AS_VAR_SET(gl_Warn, [no])]) - CPPFLAGS="$gl_save_CPPFLAGS" + +# gl_COMPILER_OPTION_IF(OPTION, [IF-SUPPORTED], [IF-NOT-SUPPORTED], +# [PROGRAM = AC_LANG_PROGRAM()]) +# ----------------------------------------------------------------- +# Check if the compiler supports OPTION when compiling PROGRAM. +# +# FIXME: gl_Warn must be used unquoted until we can assume Autoconf +# 2.64 or newer. +AC_DEFUN([gl_COMPILER_OPTION_IF], +[AS_VAR_PUSHDEF([gl_Warn], [gl_cv_warn_[]_AC_LANG_ABBREV[]_$1])dnl +AS_VAR_PUSHDEF([gl_Flags], [_AC_LANG_PREFIX[]FLAGS])dnl +AS_LITERAL_IF([$1], + [m4_pushdef([gl_Positive], m4_bpatsubst([$1], [^-Wno-], [-W]))], + [gl_positive="$1" +case $gl_positive in + -Wno-*) gl_positive=-W`expr "X$gl_positive" : 'X-Wno-$.*$'` ;; +esac +m4_pushdef([gl_Positive], [$gl_positive])])dnl +AC_CACHE_CHECK([whether _AC_LANG compiler handles $1], m4_defn([gl_Warn]), [ + gl_save_compiler_FLAGS="$gl_Flags" + gl_AS_VAR_APPEND(m4_defn([gl_Flags]), + [" $gl_unknown_warnings_are_errors ]m4_defn([gl_Positive])["]) + AC_LINK_IFELSE([m4_default([$4], [AC_LANG_PROGRAM([])])], + [AS_VAR_SET(gl_Warn, [yes])], + [AS_VAR_SET(gl_Warn, [no])]) + gl_Flags="$gl_save_compiler_FLAGS" ]) -AS_VAR_IF(gl_Warn, [yes], - [gl_AS_VAR_APPEND(m4_if([$2], [], [[WARN_CFLAGS]], [[$2]]), [" $1"])]) +AS_VAR_IF(gl_Warn, [yes], [$2], [$3]) +m4_popdef([gl_Positive])dnl +AS_VAR_POPDEF([gl_Flags])dnl AS_VAR_POPDEF([gl_Warn])dnl -m4_ifval([$2], [AS_LITERAL_IF([$2], [AC_SUBST([$2])], [])])dnl ]) + +# gl_UNKNOWN_WARNINGS_ARE_ERRORS +# ------------------------------ +# Clang doesn't complain about unknown warning options unless one also +# specifies -Wunknown-warning-option -Werror. Detect this. +AC_DEFUN([gl_UNKNOWN_WARNINGS_ARE_ERRORS], +[gl_COMPILER_OPTION_IF([-Werror -Wunknown-warning-option], + [gl_unknown_warnings_are_errors='-Wunknown-warning-option -Werror'], + [gl_unknown_warnings_are_errors=])]) + +# gl_WARN_ADD(OPTION, [VARIABLE = WARN_CFLAGS], +# [PROGRAM = AC_LANG_PROGRAM()]) +# --------------------------------------------- +# Adds parameter to WARN_CFLAGS if the compiler supports it when +# compiling PROGRAM. For example, gl_WARN_ADD([-Wparentheses]). +# +# If VARIABLE is a variable name, AC_SUBST it. +AC_DEFUN([gl_WARN_ADD], +[AC_REQUIRE([gl_UNKNOWN_WARNINGS_ARE_ERRORS]) +gl_COMPILER_OPTION_IF([$1], + [gl_AS_VAR_APPEND(m4_if([$2], [], [[WARN_CFLAGS]], [[$2]]), [" $1"])], + [], + [$3]) +m4_ifval([$2], + [AS_LITERAL_IF([$2], [AC_SUBST([$2])])], + [AC_SUBST([WARN_CFLAGS])])dnl +]) + +# Local Variables: +# mode: autoconf +# End: -- 1.8.3.1

2 2

[libvirt] [PATCH]qemu: virDomainControllerFind may return 0 if controller found
by Chen Hanxiao 26 Sep '13

26 Sep '13

From: Chen Hanxiao <chenhanxiao(a)cn.fujitsu.com> The return value of virDomainControllerFind >=0 means that the specific controller was found. But some functions invoke it and treat 0 as not found. This patch fix these incorrect invocation. Signed-off-by: Chen Hanxiao <chenhanxiao(a)cn.fujitsu.com> --- src/qemu/qemu_driver.c | 2 +- src/qemu/qemu_hotplug.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 346a8f9..ccd9cc6 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -6626,7 +6626,7 @@ qemuDomainAttachDeviceConfig(virQEMUCapsPtr qemuCaps, case VIR_DOMAIN_DEVICE_CONTROLLER: controller = dev->data.controller; if (virDomainControllerFind(vmdef, controller->type, - controller->idx) > 0) { + controller->idx) >= 0) { virReportError(VIR_ERR_OPERATION_INVALID, "%s", _("Target already exists")); return -1; diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 6cdee44..f06930e 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -358,7 +358,7 @@ int qemuDomainAttachPciControllerDevice(virQEMUDriverPtr driver, qemuDomainObjPrivatePtr priv = vm->privateData; bool releaseaddr = false; - if (virDomainControllerFind(vm->def, controller->type, controller->idx) > 0) { + if (virDomainControllerFind(vm->def, controller->type, controller->idx) >= 0) { virReportError(VIR_ERR_OPERATION_FAILED, _("target %s:%d already exists"), type, controller->idx); -- 1.8.2.1

2 1

[libvirt] [PATCH 0/2] Fix crash in chardev source parsing
by Peter Krempa 26 Sep '13

26 Sep '13

Peter Krempa (2): conf: Don't crash on invalid chardev source definition of RNGs and other conf: clean up virDomainChrSourceDefParseXML src/conf/domain_conf.c | 68 +++++++++++----------- .../qemuxml2argv-virtio-rng-egd-crash.xml | 27 +++++++++ tests/qemuxml2argvtest.c | 2 + 3 files changed, 64 insertions(+), 33 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-egd-crash.xml -- 1.8.3.2

2 4

[libvirt] [PATCH] LXC start after shutdown didn't work
by Cédric Bosdonnat 26 Sep '13

26 Sep '13

The bug is that after shutting down an lxc full-distro container, start always fails. It turns out that shutdown doesn't clean the machine on the machined side. A workaround is to machinectl terminate the container. A proper fix would be that machinectl tracks the state of the init process and cleans the machine if that one disappears. This fix is the reasonable fix we can have on the libvirt side: really try to start the container manually if starting it with systemd failed. It currently never calls the virCgroupNewMachineManual function. --- src/util/vircgroup.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/src/util/vircgroup.c b/src/util/vircgroup.c index e99caf5..466c170 100644 --- a/src/util/vircgroup.c +++ b/src/util/vircgroup.c @@ -1656,9 +1656,6 @@ virCgroupNewMachine(const char *name, group)) == 0) return 0; - if (rv == -1) - return -1; - return virCgroupNewMachineManual(name, drivername, pidleader, -- 1.8.4

2 1

[libvirt] virDomainGetInfo() returns wrong domain state
by Panday Ritesh Sharma (rpanday) 26 Sep '13

26 Sep '13

Hi Team, I have written below code to get the VM state at run time. I found, though the VM is in shut-off state, when I use the function virDomainGetInfo(); I get state as running. Could you please let me know what wrong I am doing. To know the actual VM state I used 'virsh list' and it clearly shows the VM is in shut-off state. Please find the log and code snippet below. Log from virsh: ================= [host:~]$ virsh list --all Id Name State ---------------------------------------------------- 1 calvados running 2 LCXR running 3 default-sdr--1 running - test--2 shut off Out put: ========= 04.03.06.698923264:INFO: vm_libvirt_state_to_vmm_state: state returned is 1 Note : Here 1 is actually running. Code snippet: =============== enum cidl_vmm_vm_state vm_libvirt_state_to_vmm_state(unsigned char libvirt_state) { enum cidl_vmm_vm_state state; INFO("%s: state returned is %u\n",__FUNCTION__, libvirt_state); <<<<<<<<<<<<<<<<<< if (libvirt_state == VIR_DOMAIN_RUNNING) { state = cidl_vm_state_running; } else if ((libvirt_state == VIR_DOMAIN_PAUSED) || (libvirt_state == VIR_DOMAIN_BLOCKED)) { state = cidl_vm_state_paused; } else if (libvirt_state == VIR_DOMAIN_SHUTOFF) { state = cidl_vm_state_defined; } else { state = cidl_vm_state_not_defined; } return state; } virDomainInfo res_util; virDomainPtr dom = virDomainLookupByName(virt, private_names[vm_idx]); res = virDomainGetInfo(dom, &res_util); vminfo[vm_idx].vm_state = vm_libvirt_state_to_vmm_state(res_util.state); Regards Ritesh Sharma

2 3

[libvirt] [PATCH 1/2] qemu: Add SPAPR_VFIO_BRIDGE capability for PPC platform
by Li Zhang 26 Sep '13

26 Sep '13

From: Li Zhang <zhlcindy(a)linux.vnet.ibm.com> To support VFIO for PPC, it is needed spapr-vfio-pci-host-bridge in QEMU. This patch is to add one capability for it. Signed-off-by: Li Zhang <zhlcindy(a)linux.vnet.ibm.com> --- src/qemu/qemu_capabilities.c | 2 ++ src/qemu/qemu_capabilities.h | 1 + 2 files changed, 3 insertions(+) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index d94188a..7cc02db 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -241,6 +241,7 @@ VIR_ENUM_IMPL(virQEMUCaps, QEMU_CAPS_LAST, "usb-storage", /* 155 */ "usb-storage.removable", "virtio-mmio", + "spapr-pci-vfio-host-bridge", ); struct _virQEMUCaps { @@ -1391,6 +1392,7 @@ struct virQEMUCapsStringFlags virQEMUCapsObjectTypes[] = { { "i82801b11-bridge", QEMU_CAPS_DEVICE_DMI_TO_PCI_BRIDGE }, { "usb-storage", QEMU_CAPS_DEVICE_USB_STORAGE }, { "virtio-mmio", QEMU_CAPS_DEVICE_VIRTIO_MMIO }, + { "spapr-pci-vfio-host-bridge", QEMU_CAPS_SPAPR_VFIO_BRIDGE }, }; static struct virQEMUCapsStringFlags virQEMUCapsObjectPropsVirtioBlk[] = { diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index f3c8fa8..33d303e 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -196,6 +196,7 @@ enum virQEMUCapsFlags { QEMU_CAPS_DEVICE_USB_STORAGE = 155, /* -device usb-storage */ QEMU_CAPS_USB_STORAGE_REMOVABLE = 156, /* usb-storage.removable */ QEMU_CAPS_DEVICE_VIRTIO_MMIO = 157, /* -device virtio-mmio */ + QEMU_CAPS_SPAPR_VFIO_BRIDGE = 158, /* -device spapr-pci-vfio-host-bridge */ QEMU_CAPS_LAST, /* this must always be the last item */ }; -- 1.8.1.4

2 5

[libvirt] [PATCH 00/23] Fix yet more OOM errors
by Daniel P. Berrange 26 Sep '13

26 Sep '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> A nice mix of leaks and crashes and ignoring errors. Daniel P. Berrange (23): Fix crash on OOM in xenParseSxpr Fix handling of OOM when getting Xen dom ID Don't clobber return value in virInterfaceDefParseProtoIPv6 Fix crash on OOM in virDomainSnapshotDefParse Don't clobber 'ret' in LXC XML test case Fix double free of hostdev on OOM in xenParseSxprPCI Fix crash on OOM parsing storage pool XML Add missing check for OOM with virVMXEscapeHexPipe Fix leak of comment string if virConfAddEntry fails on OOM Don't print all test suite errors to stderr in vmx2xmltest Fix leak of iterators in virDBusMessageIterEncode Fix double-free in virJSONParserHandleStartMap on OOM Fix leak of parser state in virJSONValueFromString Fix leak in virLockSpaceResourceFree Don't ignore errors parsing nwfilter rules Fix leak on OOM in qemuMonitorCommonTestNew Avoid double free in qemuMonitorCommonTestInit on OOM Avoid uninitialized data in qemuMonitorTestNew Avoid crash on OOM in virbuftest Avoid crash on OOM in virlockspacetest Avoid crash on OOM in virportallocatortest Avoid crash on OOM in virnetmessagetest Avoid use of uninitialized data in virnetmessagetest src/conf/interface_conf.c | 16 ++++++-------- src/conf/nwfilter_conf.c | 16 +++++++------- src/conf/snapshot_conf.c | 4 ++-- src/conf/storage_conf.c | 6 +++--- src/util/virconf.c | 5 ++++- src/util/virdbus.c | 24 ++++++++++++++++++--- src/util/virjson.c | 2 +- src/util/virlockspace.c | 1 + src/vmx/vmx.c | 3 ++- src/xen/xen_driver.c | 3 ++- src/xen/xend_internal.c | 10 +++++---- src/xenxs/xen_sxpr.c | 23 ++++++++++---------- src/xenxs/xen_sxpr.h | 4 ++-- tests/lxcxml2xmltest.c | 13 +++++++---- tests/nwfilterxml2xmltest.c | 14 ++++++------ tests/qemumonitortestutils.c | 7 +++++- tests/sexpr2xmltest.c | 3 ++- tests/virbuftest.c | 14 ++++++++++++ tests/virlockspacetest.c | 21 ++++++++++++------ tests/virnetmessagetest.c | 9 +++++++- tests/virportallocatortest.c | 6 ++++++ tests/vmx2xmltest.c | 51 +++++++++++++++++--------------------------- 22 files changed, 154 insertions(+), 101 deletions(-) -- 1.8.3.1

3 32

[libvirt] networking restart after redefinition
by Mihamina Rakotomandimby 26 Sep '13

26 Sep '13

Hi all, Running Fedora 18 and the bundled libvirt and virt-tools. Desktop use. As I like to access my guests with a hostname and not a numerical IP address, I fix the IP addressing in the network configuration: http://pastebin.com/rfMKn40j Unfortunately, I cant find a way to add a host without loosing the whole networking, as when I add a ahost, I destroy and redefine the network. Restarting libvirtd is not the solution. How would you recommend to have networking UP after network redefinition? -- RMA.

2 1

[libvirt] [PATCH 0/4] Rewrite the test OOM checking code
by Daniel P. Berrange 25 Sep '13

25 Sep '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> This rewrites the OOM checking code to be much much much more scalable. Instead of looping running 'main' multiple times, we loop running 'virtTestRun' multiple times. Although the overall number of mallocs to be checked is basically the same, checking them in small blocks is a massive efficiency win, because the complexity is 'n * (n + 1) / 1'. Although I've fixed many issues, there are still some big problems remaining - The APIs in virxml.h don't let the caller distinguish between "attribute not found" and other errors. So anywhere we have an attribute that is optional, we are failing to diagnose & report OOM correctly (or indeed other errors we might get from libxml) - It is desirable to validate that we get VIR_ERR_NO_MEMORY for each failure, but there are times where we won't get this. In particular with any of the APIs which have delayed error reporting such as virCommand and virBuffer. You can see an OOM from those APIs, but there's a window between that occurring & being reported, where another error may be reported. - We should not count allocs in virerror.c or virlog.c APIs when doing OOM testing, since they're intentionally non-fatal Daniel P. Berrange (4): Don't clobber 'ret' variable in testCompareXMLToXMLHelper Remove existing OOM test impl Remove test case average timing Introduce new OOM testing support tests/commandtest.c | 2 +- tests/cputest.c | 5 +- tests/domainsnapshotxml2xmltest.c | 2 +- tests/esxutilstest.c | 2 +- tests/fchosttest.c | 10 +- tests/fdstreamtest.c | 8 +- tests/interfacexml2xmltest.c | 6 +- tests/jsontest.c | 2 +- tests/libvirtdconftest.c | 2 +- tests/lxcxml2xmltest.c | 4 +- tests/metadatatest.c | 6 +- tests/networkxml2conftest.c | 2 +- tests/networkxml2xmltest.c | 2 +- tests/networkxml2xmlupdatetest.c | 2 +- tests/nodedevxml2xmltest.c | 6 +- tests/nodeinfotest.c | 2 +- tests/nwfilterxml2xmltest.c | 2 +- tests/openvzutilstest.c | 2 +- tests/qemuagenttest.c | 4 +- tests/qemuargv2xmltest.c | 18 +- tests/qemuhelptest.c | 2 +- tests/qemuhotplugtest.c | 2 +- tests/qemumonitorjsontest.c | 8 +- tests/qemumonitortest.c | 2 +- tests/qemuxml2argvtest.c | 14 +- tests/qemuxml2xmltest.c | 25 +-- tests/qemuxmlnstest.c | 20 +- tests/securityselinuxlabeltest.c | 6 +- tests/securityselinuxtest.c | 2 +- tests/sexpr2xmltest.c | 2 +- tests/sockettest.c | 16 +- tests/statstest.c | 2 +- tests/storagepoolxml2xmltest.c | 6 +- tests/storagevolxml2argvtest.c | 2 +- tests/storagevolxml2xmltest.c | 14 +- tests/sysinfotest.c | 2 +- tests/testutils.c | 386 ++++++++++++++++++++------------------ tests/testutils.h | 3 +- tests/utiltest.c | 4 +- tests/viratomictest.c | 4 +- tests/virauthconfigtest.c | 2 +- tests/virbitmaptest.c | 18 +- tests/virbuftest.c | 2 +- tests/vircgrouptest.c | 22 +-- tests/virdbustest.c | 10 +- tests/virdrivermoduletest.c | 2 +- tests/virendiantest.c | 4 +- tests/virhashtest.c | 2 +- tests/viridentitytest.c | 4 +- tests/virkeycodetest.c | 4 +- tests/virkeyfiletest.c | 2 +- tests/virlockspacetest.c | 14 +- tests/virnetmessagetest.c | 10 +- tests/virnetserverclienttest.c | 2 +- tests/virnetsockettest.c | 30 +-- tests/virnettlscontexttest.c | 2 +- tests/virnettlssessiontest.c | 4 +- tests/virportallocatortest.c | 4 +- tests/virshtest.c | 36 ++-- tests/virstoragetest.c | 2 +- tests/virstringtest.c | 8 +- tests/virsystemdtest.c | 10 +- tests/virtimetest.c | 2 +- tests/viruritest.c | 2 +- tests/vmwarevertest.c | 2 +- tests/vmx2xmltest.c | 2 +- tests/xencapstest.c | 22 +-- tests/xmconfigtest.c | 4 +- tests/xml2sexprtest.c | 2 +- tests/xml2vmxtest.c | 2 +- 70 files changed, 436 insertions(+), 407 deletions(-) -- 1.8.3.1

2 8

[libvirt] [PATCH] spec: Clean up distribution of ChangeLog (and others)
by Cole Robinson 25 Sep '13

25 Sep '13

- Move COPYING* to libvirt-client, so every package pulls them in - Move AUTHORS ChangeLog.gz NEWS README TODO from -daemon to -docs - Drop duplicate distribution of docs in -python --- libvirt.spec.in | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index f899149..d294aad 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1723,6 +1723,8 @@ fi %files docs %defattr(-, root, root) +%doc AUTHORS ChangeLog.gz NEWS README TODO + # Website %dir %{_datadir}/doc/libvirt-docs-%{version} %dir %{_datadir}/doc/libvirt-docs-%{version}/html @@ -1739,7 +1741,6 @@ fi %files daemon %defattr(-, root, root) -%doc AUTHORS ChangeLog.gz NEWS README COPYING COPYING.LESSER TODO %dir %attr(0700, root, root) %{_sysconfdir}/libvirt/ %if %{with_network} @@ -2011,7 +2012,7 @@ fi %files client -f %{name}.lang %defattr(-, root, root) -%doc AUTHORS ChangeLog.gz NEWS README COPYING COPYING.LESSER TODO +%doc COPYING COPYING.LESSER %config(noreplace) %{_sysconfdir}/libvirt/libvirt.conf %if %{with_lxc} @@ -2101,7 +2102,6 @@ fi %files python %defattr(-, root, root) -%doc AUTHORS NEWS README COPYING COPYING.LESSER %{_libdir}/python*/site-packages/libvirt.py* %{_libdir}/python*/site-packages/libvirt_qemu.py* %{_libdir}/python*/site-packages/libvirt_lxc.py* -- 1.8.3.1

2 2

[libvirt] [PATCH v2 0/9] test: Mock snapshot APIs and misc improvements
by Cole Robinson 25 Sep '13

25 Sep '13

This series implements snapshot and managed save APIs for the test driver, and adds some misc improvements, like specifying domain state in the passed in driver XML. v2: Use a separate XML namespace for the custom XML additions Add snapshot REDEFINE support, sharing the generic logic with qemu Cole Robinson (9): test: Allow specifying object runstate in driver XML test: Allow specifying object transient state in driver XML test: Wire up managed save APIs test: Implement readonly snapshot APIs test: Implement snapshot create/delete/revert APIs qemu: snapshots: Simplify REDEFINE flag check qemu: snapshot: Break out redefine preparation to shared function test: snapshot: Add REDEFINE support test: Allow specifying domainsnapshot XML src/conf/domain_conf.c | 3 +- src/conf/snapshot_conf.c | 150 ++++++ src/conf/snapshot_conf.h | 7 + src/libvirt_private.syms | 1 + src/qemu/qemu_driver.c | 145 +----- src/test/test_driver.c | 1236 +++++++++++++++++++++++++++++++++++++++++++++- tests/virshtest.c | 2 +- 7 files changed, 1399 insertions(+), 145 deletions(-) -- 1.8.3.1

2 15