- Devel - libvirt List Archives

[libvirt] [PATCH] qemu: avoid text monitor null deref
by Eric Blake 13 Oct '11

13 Oct '11

Detected by Coverity. If, for some reason, our text monitor input does not match our assumptions, we end up incrementing p while it is NULL, then dereferencing the pointer 0x1, which will fault. * src/qemu/qemu_monitor_text.c (qemuMonitorTextGetBlockStatsParamsNumber): Rewrite to avoid deref of strchr failure. Fix indentation. --- src/qemu/qemu_monitor_text.c | 33 +++++++++++++++------------------ 1 files changed, 15 insertions(+), 18 deletions(-) diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index 51e8c5c..1eb9846 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -1036,26 +1036,23 @@ int qemuMonitorTextGetBlockStatsParamsNumber(qemuMonitorPtr mon, * "floppy0: ") */ p = strchr(p, ' '); - p++; - while (*p) { - if (STRPREFIX (p, "rd_bytes=") || - STRPREFIX (p, "wr_bytes=") || - STRPREFIX (p, "rd_operations=") || - STRPREFIX (p, "wr_operations=") || - STRPREFIX (p, "rd_total_times_ns=") || - STRPREFIX (p, "wr_total_times_ns=") || - STRPREFIX (p, "flush_operations=") || - STRPREFIX (p, "flush_total_times_ns=")) { - num++; - } else { - VIR_DEBUG ("unknown block stat near %s", p); - } + while (p && p < eol) { + if (STRPREFIX (p, " rd_bytes=") || + STRPREFIX (p, " wr_bytes=") || + STRPREFIX (p, " rd_operations=") || + STRPREFIX (p, " wr_operations=") || + STRPREFIX (p, " rd_total_times_ns=") || + STRPREFIX (p, " wr_total_times_ns=") || + STRPREFIX (p, " flush_operations=") || + STRPREFIX (p, " flush_total_times_ns=")) { + num++; + } else { + VIR_DEBUG ("unknown block stat near %s", p); + } - /* Skip to next label. */ - p = strchr (p, ' '); - if (!p || p >= eol) break; - p++; + /* Skip to next label. */ + p = strchr(p + 1, ' '); } *nparams = num; -- 1.7.4.4

2 2

[libvirt] [PATCH] qemu: check for json allocation failure
by Eric Blake 13 Oct '11

13 Oct '11

Detected by Coverity. Introduced in commit b1b5b51. * src/qemu/qemu_monitor_json.c (qemuMonitorJSONGetBlockInfo): Avoid null dereference on OOM. --- src/qemu/qemu_monitor_json.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 18d1c9f..cd8f1e5 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -1364,6 +1364,9 @@ int qemuMonitorJSONGetBlockInfo(qemuMonitorPtr mon, virJSONValuePtr reply = NULL; virJSONValuePtr devices; + if (!cmd) + return -1; + ret = qemuMonitorJSONCommand(mon, cmd, &reply); if (ret == 0) ret = qemuMonitorJSONCheckError(cmd, reply); -- 1.7.4.4

2 2

[libvirt] [PATCH] virFDStream: close also given errfd (fd leak)
by Marc-André Lureau 13 Oct '11

13 Oct '11

In virFDStreamOpenFileInternal(), a errfd pipe is opened by virCommandRunAsync() and given to virFDStreamOpenInternal(). It seems virFDStream should close errfd, just like the other fd it is given. This fixes screenshots leaking FDs: http://bugzilla.redhat.com/show_bug.cgi?id=745761 --- src/fdstream.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/src/fdstream.c b/src/fdstream.c index b60162c..841f979 100644 --- a/src/fdstream.c +++ b/src/fdstream.c @@ -268,6 +268,10 @@ virFDStreamClose(virStreamPtr st) virCommandFree(fdst->cmd); fdst->cmd = NULL; } + + if (VIR_CLOSE(fdst->errfd) < 0) + VIR_DEBUG("ignoring failed close on fd %d", fdst->errfd); + st->privateData = NULL; if (fdst->dispatching) { -- 1.7.6.2

2 1

[libvirt] [PATCH RESEND] command: avoid fd leak on failure
by Eric Blake 13 Oct '11

13 Oct '11

virCommandTransferFD promises that the fd is no longer owned by the caller. Normally, we want the fd to remain open until the child runs, but in error situations, we must close it earlier. * src/util/command.c (virCommandTransferFD): Close fd now if we can't track it to close later. (virCommandKeepFD): Adjust helper to make this easier. --- This leak can only happen on OOM or other extreme error conditions, but ought to be plugged. When I originally posted this: https://www.redhat.com/archives/libvir-list/2011-July/msg00674.html DV was worried that callers might abuse things and use fd even after this function closed it; but I proved to myself in writing a (non-working) v2 that all callers were already safe, and that this v1 was indeed a smaller solution. src/util/command.c | 17 +++++++++++------ 1 files changed, 11 insertions(+), 6 deletions(-) diff --git a/src/util/command.c b/src/util/command.c index 699ba2d..c3ce361 100644 --- a/src/util/command.c +++ b/src/util/command.c @@ -730,23 +730,26 @@ virCommandNewArgList(const char *binary, ...) * closing it. FD must not be one of the three standard streams. If * transfer is true, then fd will be closed in the parent after a call * to Run/RunAsync/Free, otherwise caller is still responsible for fd. + * Returns true if a transferring caller should close FD now, and + * false if the transfer is successfully recorded. */ -static void +static bool virCommandKeepFD(virCommandPtr cmd, int fd, bool transfer) { if (!cmd) - return; + return fd > STDERR_FILENO; if (fd <= STDERR_FILENO || FD_SETSIZE <= fd) { if (!cmd->has_error) cmd->has_error = -1; VIR_DEBUG("cannot preserve %d", fd); - return; + return fd > STDERR_FILENO; } FD_SET(fd, &cmd->preserve); if (transfer) FD_SET(fd, &cmd->transfer); + return false; } /** @@ -761,7 +764,7 @@ virCommandKeepFD(virCommandPtr cmd, int fd, bool transfer) void virCommandPreserveFD(virCommandPtr cmd, int fd) { - return virCommandKeepFD(cmd, fd, false); + virCommandKeepFD(cmd, fd, false); } /** @@ -771,12 +774,14 @@ virCommandPreserveFD(virCommandPtr cmd, int fd) * * Transfer the specified file descriptor * to the child, instead of closing it on exec. - * Close the fd in the parent during Run/RunAsync/Free. + * The parent should no longer use fd, and the parent's copy will + * be automatically closed no later than during Run/RunAsync/Free. */ void virCommandTransferFD(virCommandPtr cmd, int fd) { - return virCommandKeepFD(cmd, fd, true); + if (virCommandKeepFD(cmd, fd, true)) + VIR_FORCE_CLOSE(fd); } -- 1.7.4.4

2 2

[libvirt] ALERT Virtualization Automatic Builder
by berrange＠redhat.com 13 Oct '11

13 Oct '11

Overall status: failed Start date: Thu Oct 13 2011 Start time: 11:05:02 UTC / 07:05:02 EDT Build counter: 1318503902 Build timestamp: 1318503902 URL: http://builder.virt-tools.org/index.html Module: libvirt Status: failed URL: http://builder.virt-tools.org/module-libvirt.html

3 2

[libvirt] [PATCH RESEND] spec: mingw cleanups
by Eric Blake 13 Oct '11

13 Oct '11

* libvirt.spec.in (%configure): Drop unused %{one} macro. * mingw32-libvirt.spec.in (%{rhel}): Compile ESX but not HyperV on mingw build for RHEL. (%build): Make configure honor spec conditionals. Reorder to match libvirt.spec. * autobuild.sh (mingw): Update list to match. Suggested by Daniel P. Berrange. --- No comment when buried in a thread, so I'll resend: https://www.redhat.com/archives/libvir-list/2011-September/msg01072.html autobuild.sh | 16 +++++++----- libvirt.spec.in | 3 +- mingw32-libvirt.spec.in | 58 ++++++++++++++++++++++++++++++++++++++++------ 3 files changed, 60 insertions(+), 17 deletions(-) diff --git a/autobuild.sh b/autobuild.sh index 491f1b8..9853287 100755 --- a/autobuild.sh +++ b/autobuild.sh @@ -71,21 +71,23 @@ if [ -x /usr/bin/i686-pc-mingw32-gcc ]; then --host=i686-pc-mingw32 \ --prefix="$AUTOBUILD_INSTALL_ROOT/i686-pc-mingw32/sys-root/mingw" \ --enable-compile-warnings=error \ - --without-sasl \ - --without-avahi \ - --without-polkit \ - --without-python \ --without-xen \ --without-qemu \ + --without-openvz \ --without-lxc \ - --without-uml \ --without-vbox \ - --without-openvz \ + --without-xenapi \ + --without-uml \ + --without-sasl \ + --without-avahi \ + --without-polkit \ + --without-python \ + --without-libvirtd \ --without-phyp \ + --without-hyperv \ --without-netcf \ --without-audit \ --without-dtrace \ - --without-libvirtd make make install diff --git a/libvirt.spec.in b/libvirt.spec.in index 7c63710..03bd7c7 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -47,7 +47,7 @@ %define with_libxl 0%{!?_without_libxl:%{server_drivers}} %define with_vmware 0%{!?_without_vmware:%{server_drivers}} -# Then the hypervisor drivers that talk a native remote protocol +# Then the hypervisor drivers that talk via a native remote protocol %define with_phyp 0%{!?_without_phyp:1} %define with_esx 0%{!?_without_esx:1} %define with_hyperv 0%{!?_without_hyperv:1} @@ -711,7 +711,6 @@ of recent versions of Linux (and other OSes). %{?_without_python} \ %{?_without_libvirtd} \ %{?_without_uml} \ - %{?_without_one} \ %{?_without_phyp} \ %{?_without_esx} \ %{?_without_hyperv} \ diff --git a/mingw32-libvirt.spec.in b/mingw32-libvirt.spec.in index f651d11..57c67ae 100644 --- a/mingw32-libvirt.spec.in +++ b/mingw32-libvirt.spec.in @@ -5,6 +5,21 @@ %define __find_provides %{_mingw32_findprovides} %define __debug_install_post %{_mingw32_debug_install_post} +# The mingw build is client only. Set up defaults for hypervisor drivers +# that talk via a native remote protocol, and for which prereq mingw +# libraries exist. +%define with_phyp 0%{!?_without_phyp:1} +%define with_esx 0%{!?_without_esx:1} +%define with_hyperv 0%{!?_without_hyperv:0} # missing libwsman +%define with_xenapi 0%{!?_without_xenapi:1} + +# RHEL ships ESX but not PowerHypervisor, HyperV, or libxenserver (xenapi) +%if 0%{?rhel} +%define with_phyp 0 +%define with_xenapi 0 +%define with_hyperv 0 +%endif + Name: mingw32-libvirt Version: @VERSION@ Release: 3%{?dist}%{?extra_release} @@ -29,6 +44,13 @@ BuildRequires: pkgconfig # Need native version for msgfmt BuildRequires: gettext +%if %{with_phyp} +BuildRequires: mingw32-libssh2 +%endif +%if %{with_esx} +BuildRequires: mingw32-curl +%endif + BuildArch: noarch %{?_mingw32_debug_package} @@ -42,23 +64,43 @@ MinGW Windows libvirt virtualization library. %build +%if ! %{with_phyp} +%define _without_phyp --without-phyp +%endif + +%if ! %{with_esx} +%define _without_esx --without-esx +%endif + +%if ! %{with_hyperv} +%define _without_hyperv --without-hyperv +%endif + +%if ! %{with_xenapi} +%define _without_xenapi --without-xenapi +%endif + # XXX enable SASL in future %{_mingw32_configure} \ + --without-xen \ + --without-qemu \ + --without-openvz \ + --without-lxc \ + --without-vbox \ + %{?_without_xenapi} \ --without-sasl \ --without-avahi \ --without-polkit \ --without-python \ - --without-xen \ - --without-qemu \ - --without-lxc \ + --without-libvirtd \ --without-uml \ - --without-vbox \ - --without-openvz \ - --without-phyp \ + %{?_without_phyp} \ + %{?_without_esx} \ + %{?_without_hyperv} \ + --without-vmware \ --without-netcf \ --without-audit \ - --without-dtrace \ - --without-libvirtd + --without-dtrace make -- 1.7.4.4

2 2

[libvirt] [PATCH] build: update to latest gnulib
by Eric Blake 13 Oct '11

13 Oct '11

358 changes; most probably have no impact on libvirt, but we might as well stay current. * .gnulib: Update to latest. --- * .gnulib da1717b...2394a60 (358): > atanl: Simplify for platforms where 'long double' == 'double'. > acosl: Simplify for platforms where 'long double' == 'double'. > asinl: Simplify for platforms where 'long double' == 'double'. > tanl: Simplify for platforms where 'long double' == 'double'. > cosl: Simplify for platforms where 'long double' == 'double'. > sinl: Simplify for platforms where 'long double' == 'double'. > logl: Simplify for platforms where 'long double' == 'double'. > expl: Simplify for platforms where 'long double' == 'double'. > sqrtl: Simplify for platforms where 'long double' == 'double'. > ldexpl: Simplify for platforms where 'long double' == 'double'. > ffsll: set correct witness > printf-frexpl: Simplify for platforms where 'long double' == 'double'. > frexpl: Simplify for platforms where 'long double' == 'double'. > test-renameat: don't leave behind a temporary file > rint: Determine RINT_LIBM correctly on AIX 7. > mathfunc.m4: Refactor. > rintl: Simplify for platforms where 'long double' == 'double'. > roundl: Simplify for platforms where 'long double' == 'double'. > truncl: Simplify for platforms where 'long double' == 'double'. > ceill: Simplify for platforms where 'long double' == 'double'. > floorl: Simplify for platforms where 'long double' == 'double'. > rint: Fix ordering constraints. > copysignl: Simplify for platforms where 'long double' == 'double'. > Tests for module 'rintl'. > New module 'rintl'. > Tests for module 'rintf'. > New module 'rintf'. > rint: Support for MSVC. > rint tests: More tests. > autoupdate > Tests for module 'copysignl'. > New module 'copysignl'. > Tests for module 'copysignf'. > New module 'copysignf'. > Ensure that HAVE_* variables are set to 1 before they are set to 0. > poll: Make macro safer. > copysign: Provide replacement. > isinf: Ensure macro on AIX 5.1. > *printf-posix tests: Fix for platforms where 'long double' == 'double'. > *printf: Fix for platforms where 'long double' == 'double'. > isnanl[-nolibm]: Fix for platforms where 'long double' == 'double'. > isinf: Fix for platforms where 'long double' == 'double'. > isfinite: Fix for platforms where 'long double' == 'double'. > isfinite, isinf, isnan tests: Avoid syntax error on MSVC 9. > Tests for module 'tanhf'. > New module 'tanhf'. > tanh: Use a .m4 file. > Tests for module 'coshf'. > New module 'coshf'. > cosh: Use a .m4 file. > Tests for module 'sinhf'. > New module 'sinhf'. > sinh: Use a .m4 file. > Tests for module 'atan2f'. > New module 'atan2f'. > atan2: Use a .m4 file. > Tests for module 'atanf'. > New module 'atanf'. > atan: Use a .m4 file. > Tests for module 'acosf'. > New module 'acosf'. > acos: Use a .m4 file. > Tests for module 'asinf'. > New module 'asinf'. > asin: Use a .m4 file. > Tests for module 'tanf'. > New module 'tanf'. > tan: Use a .m4 file. > Tests for module 'cosf'. > New module 'cosf'. > cos: Use a .m4 file. > Tests for module 'sinf'. > New module 'sinf'. > sin: Use a .m4 file. > Tests for module 'powf'. > New module 'powf'. > pow: Use a .m4 file. > Tests for module 'log10f'. > New module 'log10f'. > log10: Use a .m4 file. > Tests for module 'logf'. > New module 'logf'. > log: Use a .m4 file. > Tests for module 'expf'. > New module 'expf'. > exp: Use a .m4 file. > Tests for module 'sqrtf'. > New module 'sqrtf'. > Tests: Avoid link failures w.r.t. libintl. > pow tests: Defeat compiler optimizations. > gnulib-tool: Improve last commit. > gnulib-tool: don't follow dependencies to avoided modules > gnulib-tool: fix typo that broke Emacs on powerpc-apple-darwin9.8.0. > Fix documentation about math functions on MacOS X. > modff: Drop assumption about library that defines modff. > raise tests: Avoid a GCC warning. > Tests for module 'ldexpf'. > New module 'ldexpf'. > Tweak last commit. > frexpf: Work around problems on IRIX and mingw. > fabsf: Drop assumption about library that defines fabsf. > frexpf: Drop assumption about library that defines frexpf. > Tests for module 'frexpf'. > New module 'frexpf'. > math: Sort function declarations of math.in.h. > Tests for module 'modff'. > New module 'modff'. > modf tests: Make test sharper. > modf: Use a .m4 file. > Tweak new module 'fmodf'. > Tweak new module 'fmodf'. > Tests for module 'fmodf'. > New module 'fmodf'. > fmod: Use a .m4 file. > New module 'fabsf'. > Tests for module 'fabsf'. > New module 'fabsf'. > fabs: Use a .m4 file. > file-has-acl: revert both recent changes, 80af92af and 95f7c57f > file-has-acl: revert unintended change in behavior of ls -L > poll: Avoid link errors on MSVC. > sys_select tests: Fix link error on MSVC 9. > sys_select: Fix compilation error on mingw. > wmemset: Support for MSVC. > wmemmove: Support for MSVC. > wmemcpy: Support for MSVC. > wmemcmp: Support for MSVC. > wmemchr: Support for MSVC. > glthread/*, strsignal: Support for MSVC. > nonblocking tests: Fix test failure on OpenBSD/SPARC64. > acl: Fix specification. > relocatable-lib[-lgpl]: Avoid expensive /proc access on Linux, Cygwin. > autoupdate > maint.mk: adjust a release-related rule not to require use of gzip > test-linkat: don't leave behind a temporary file > float, math: add omitted file > isinf: Fix for OpenBSD/x86. > isfinite: Fix syntax error in configure test. > isfinite: Fix typo. > nonblocking tests: Fix test failure on Linux/IA-64. > mkfifoat tests: Fix a test failure on mingw. > float, math: Fix 'int' to 'long double' conversion on Linux/SPARC64. > nonblocking tests: Fix test failure on Linux/SPARC (32-bit and 64-bit). > gnulib-tool: Improve suggestion where to put gl_EARLY invocation. > Centralize C99 requirement. > float: Fix LDBL_MAX value on Linux/PowerPC. > Tweak last ChangeLog entry. > doc: Improve doc about gl_EARLY. > fgetc, fputc, fread, fwrite tests: Fix link error. > fputc, fwrite tests: Avoid test failure on MSVC. > raise: Fix double declaration with modules 'sigprocmask' and 'sigpipe'. > autoupdate > base64-tests: avoid memory leak > base32: new module > gnulib: use more-standard license notice wording > dup2: minor simplification > strings: Fix compilation error on MSVC. > fflush et al.: Document limitation on MSVC. > msvc-inval: Allow three invalid parameter handling modes. > msvc-inval: Make handler multithread-safe. > msvc-inval: Install handler globally. > strerror_r-posix: Fix for MSVC 9. > ftruncate: Support for MSVC 9. > New module 'fstat'. > linkat: Fix compilation on MSVC 9. > fclose: Support for MSVC 9. > dup2: minor simplifications > test-futimens: avoid a warning from gcc -Wshadow > fdopen: Support for MSVC 9. > fgetc, fputc, fread, fwrite tests: Avoid compilation error on MSVC. > dup: Simplify autoconf test. > Tests for function fwrite(). > Tests for function fread(). > Activate fputc tests. > Enhance fgetc, fputc tests. > write: Support for MSVC 9. > read: Fix last commit. > dup2: Fix last commit. > sigprocmask: Make code safer. > dup2: clarify by coalescing Windows-specific material > sigprocmask: move #include directive > read: Support for MSVC 9. > close: Support for MSVC 9. > msvc-nothrow: Avoid a compiler warning. > New module 'dup'. > getdtablesize: Support for MSVC 9. > signal-h: Rename from signal. > sigprocmask: Avoid crashes through signal() on MSVC 9. > Tests for module 'raise'. > raise: Support for MSVC. > pipe2: Fix compilation on pre-C99 compilers. > dup2: Fix typo in last commit. > New module 'msvc-nothrow'. Makes _get_osfhandle safe on MSVC 9. > dup2: Make code more maintainable. > msvc-inval: Fix compilation error. > update from texinfo > mkdir: Tweak for MSVC 9. > Tests for module 'chdir'. > New module 'chdir'. > rmdir: Support for mingw, MSVC 9. > getcwd: Tweak for MSVC 9. > strerror_r-posix: Avoid a link error on MSVC. > select: Avoid link errors on MSVC. > select: Avoid compilation error on MSVC. > Consolidate all uses of PATH_MAX in *.m4 files. > ftruncate: Un-deprecate, concentrate on Win32 support. > Add dependencies to new dirent related modules. > opendir: Avoid compilation error on mingw. > ftruncate tests: Avoid a test failure on mingw. > select tests: Avoid test failures on OSF/1 5.1 and mingw. > New module 'fdopen'. > unlockpt tests: Avoid test failure on NetBSD 5.1. > getlogin, getlogin_r tests: Avoid test failure on Linux/SPARC. > time tests: Don't require pid_t. > time tests: Don't require pid_t. > fsync tests: Avoid a test failure on mingw. > euidaccess: Update comments. > Ensure EBADF returns for socket functions on mingw. > select tests: EBADF tests. > Tests for module 'posix_spawn_file_actions_addopen. > Tests for module 'posix_spawn_file_actions_adddup2'. > Tests for module 'posix_spawn_file_actions_addclose'. > Tests for module 'unlockpt'. > Tests for module 'grantpt'. > freopen tests: EBADF tests. > fclose tests: EBADF tests. > fflush tests: EBADF tests. > ftello tests: EBADF tests. > fseeko tests: EBADF tests. > Tests for function fputc(). > Tests for function fgetc(). > Tests for function fdopen(). > Tests for module 'vdprintf'. > Tests for module 'dprintf'. > Tests for module 'ioctl'. > fcntl tests: EBADF tests. > utimensat tests: EBADF tests. > renameat tests: EBADF tests. > mkfifoat tests: EBADF tests. > readlinkat tests: EBADF tests. > symlinkat tests: EBADF tests. > linkat tests: EBADF tests. > Tests for module 'faccessat'. > fdopendir tests: EBADF tests. > openat tests: EBADF tests. > utimens, futimens, fdutimensat tests: EBADF tests. > Tests for function fstat(). > test-ttyname_r tests: EBADF tests. > Tests for module 'isatty'. > Tests for module 'write'. > Tests for module 'read'. > pwrite tests: EBADF tests. > pread tests: EBADF tests. > lseek tests: EBADF tests. > Tests for module 'ftruncate'. > fsync tests: EBADF tests. > fdatasync tests: EBADF tests. > Tests for module 'fchown'. > Tests for module 'fchmod'. > fchdir tests: EBADF tests. > dup2 tests: EBADF tests. > Tests for module 'dup'. > Tests for module 'close'. > Tests for module 'shutdown'. > Tests for module 'setsockopt'. > Tests for module 'sendto'. > Tests for module 'send'. > Tests for module 'recvfrom'. > Tests for module 'recv'. > Tests for module 'listen'. > Tests for module 'getsockopt'. > Tests for module 'getsockname'. > Tests for module 'getpeername'. > Tests for module 'connect'. > Tests for module 'bind'. > accept4 tests: Fix for native Windows. > accept tests: Fix for native Windows. > msvc-inval: Require a semicolon after DONE_MSVC_INVAL. > sched: Ensure pid_t is defined. > msvc-inval: Ensure the entire expansion is a single statement. > tests: use printf, not echo in init.sh's warn_ function > futimens: enhance test > date: accept 'hence' as opposite of 'ago' > getcwd: don't fail in a deep directory on a system without openat > maint.mk: avoid warnings from perl about missing files > * lib/stat.c: Move comment. > stat: use pathmax.h only if needed > utimens: remove dependency on dup2 > inet_ntop, inet_pton: Look for it also in libresolv. > accept, accept4 tests: Avoid link error on Solaris. > accept4: Avoid link error on Solaris. > socket functions: Avoid link errors on Solaris. > ptsname tests: Let the test fail rather than hang (e.g. on AIX 5.1). > posix_spawn_file_actions_add*: Fix module dependencies. > rename, renameat tests: Avoid test failures on FreeBSD 6.4. > Tests for module 'accept4'. > accept tests: Fix last commit. > Tests for module 'accept'. > dup2: Support for MSVC. > New module 'msvc-inval'. > Tests for module 'pclose'. > New module 'pclose'. > popen: Support for MSVC. > isnanl, isnand, isnanf: Work around MSVC bug. > sys_socket tests: Fix recent mistake. > putenv: Support for MSVC. > math: Avoid macro redefinition warnings on MSVC. > doc: Mention functions which are declared as macros. > Add dependencies to new dirent related modules. > inet_pton: Fix mistake in last commit. > inet_pton: Support for MSVC on Windows Vista or newer. > inet_ntop: Support for MSVC on Windows Vista or newer. > test-fsync: yet another enhancement > Enhance fsync, fdatasync tests. > Fix last ChangeLog entry. > Support for MSVC compiler: Ensure mode_t gets defined. > sys_stat: Support for MSVC. > Support for MSVC compiler: Ensure off_t gets defined. > fdatasync: port to Solaris > fdatasync: port to MacOS X 10.7 > fdatasync: minor improvements > unistd: update refs to newer POSIX > fdatasync: new module > test-fsync: enhance tests > Support for MSVC compiler: Ensure ssize_t gets defined. > Avoid using #, the m4 comment starter character, near brackets. > Properly quote AC_CHECK_DECLS' 4th argument. > Properly quote AC_CHECK_DECL's 4th argument. > opendir: avoid compile warning > maint.mk: sc_tight_scope: propagate failure from sub-make > tempname: Support for MSVC. > sys_time: Support for MSVC. > stat: Support for MSVC. > pathmax: Support for native Windows. > New modules 'opendir', 'readdir', 'rewinddir', 'closedir'. > asm-underscore.m4: Support for MSVC. > Expand range of versions of glibc for which _GNU_SOURCE is needed to get crypt. > doc: Update regarding MSVC 9. > unistd et al.: Don't assume <unistd.h> exists. > strcase: Support for MSVC. > strings: Don't assume <strings.h> exists. > dirent: Don't assume <dirent.h> exists. > Fix wint_t on MSVC. > sys_types: Fix typo. > autoupdate > Support for MSVC compiler: Ensure size_t gets defined. > C++ tests for module 'sys_types'. > Tests for module 'sys_types'. > New module 'sys_types'. > Support for MSVC compiler: Avoid division by a literal 0. > Ensure pid_t gets defined. > acl: Fix compilation on Solaris 10 (older version). > iconv, unsetenv: Add support for MSVC compiler. > *printf: Add support for MSVC compiler. > nocrash: Add support for native Windows. > absolute-header, include-next: Add support for MSVC compiler. > ChangeLog: change each leading sequence of 8 spaces to a TAB > maint.mk: mark the post-release commit log with "maint: " prefix .gnulib | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/.gnulib b/.gnulib index da1717b..2394a60 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit da1717b7f93b77469f980ea5e13178e4e9e7ae09 +Subproject commit 2394a603e7586e671226478e5b15d924c3841f42 -- 1.7.4.4

2 2

[libvirt] [RFC PATCH 1/5] Add new API virDomainBlockIoThrottle
by Lei Li 13 Oct '11

13 Oct '11

Signed-off-by: Zhi Yong Wu <wuzhy(a)linux.vnet.ibm.com> --- include/libvirt/libvirt.h.in | 22 ++++++++++++ src/conf/domain_conf.c | 77 ++++++++++++++++++++++++++++++++++++++++++ src/conf/domain_conf.h | 11 ++++++ src/driver.h | 11 ++++++ src/libvirt.c | 66 ++++++++++++++++++++++++++++++++++++ src/libvirt_public.syms | 1 + src/util/xml.h | 3 ++ 7 files changed, 191 insertions(+), 0 deletions(-) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index 07617be..f7b892d 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -1573,6 +1573,28 @@ int virDomainBlockJobSetSpeed(virDomainPtr dom, const char *path, int virDomainBlockPull(virDomainPtr dom, const char *path, unsigned long bandwidth, unsigned int flags); +/* + * Block I/O throttling support + */ + +typedef unsigned long long virDomainBlockIoThrottleUnits; + +typedef struct _virDomainBlockIoThrottleInfo virDomainBlockIoThrottleInfo; +struct _virDomainBlockIoThrottleInfo { + virDomainBlockIoThrottleUnits bps; + virDomainBlockIoThrottleUnits bps_rd; + virDomainBlockIoThrottleUnits bps_wr; + virDomainBlockIoThrottleUnits iops; + virDomainBlockIoThrottleUnits iops_rd; + virDomainBlockIoThrottleUnits iops_wr; +}; +typedef virDomainBlockIoThrottleInfo *virDomainBlockIoThrottleInfoPtr; + +int virDomainBlockIoThrottle(virDomainPtr dom, + const char *disk, + virDomainBlockIoThrottleInfoPtr info, + virDomainBlockIoThrottleInfoPtr reply, + unsigned int flags); /* * NUMA support diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 944cfa9..d0ba07e 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -2422,6 +2422,42 @@ virDomainDiskDefParseXML(virCapsPtr caps, iotag = virXMLPropString(cur, "io"); ioeventfd = virXMLPropString(cur, "ioeventfd"); event_idx = virXMLPropString(cur, "event_idx"); + } else if (xmlStrEqual(cur->name, BAD_CAST "iothrottle")) { + char *io_throttle = NULL; + io_throttle = virXMLPropString(cur, "bps"); + if (io_throttle) { + def->blkiothrottle.bps = strtoull(io_throttle, NULL, 10); + VIR_FREE(io_throttle); + } + + io_throttle = virXMLPropString(cur, "bps_rd"); + if (io_throttle) { + def->blkiothrottle.bps_rd = strtoull(io_throttle, NULL, 10); + VIR_FREE(io_throttle); + } + + io_throttle = virXMLPropString(cur, "bps_wr"); + if (io_throttle) { + def->blkiothrottle.bps_wr = strtoull(io_throttle, NULL, 10); + VIR_FREE(io_throttle); + } + + io_throttle = virXMLPropString(cur, "iops"); + if (io_throttle) { + def->blkiothrottle.iops = strtoull(io_throttle, NULL, 10); + VIR_FREE(io_throttle); + } + + io_throttle = virXMLPropString(cur, "iops_rd"); + if (io_throttle) { + def->blkiothrottle.iops_rd = strtoull(io_throttle, NULL, 10); + VIR_FREE(io_throttle); + } + + io_throttle = virXMLPropString(cur, "iops_wr"); + if (io_throttle) { + def->blkiothrottle.iops_wr = strtoull(io_throttle, NULL, 10); + } } else if (xmlStrEqual(cur->name, BAD_CAST "readonly")) { def->readonly = 1; } else if (xmlStrEqual(cur->name, BAD_CAST "shareable")) { @@ -9249,6 +9285,47 @@ virDomainDiskDefFormat(virBufferPtr buf, virBufferAsprintf(buf, " <target dev='%s' bus='%s'/>\n", def->dst, bus); + /*disk I/O throttling*/ + if (def->blkiothrottle.bps + || def->blkiothrottle.bps_rd + || def->blkiothrottle.bps_wr + || def->blkiothrottle.iops + || def->blkiothrottle.iops_rd + || def->blkiothrottle.iops_wr) { + virBufferAsprintf(buf, " <iothrottle"); + if (def->blkiothrottle.bps) { + virBufferAsprintf(buf, " bps='%llu'", + def->blkiothrottle.bps); + } + + if (def->blkiothrottle.bps_rd) { + virBufferAsprintf(buf, " bps_rd='%llu'", + def->blkiothrottle.bps_rd); + } + + if (def->blkiothrottle.bps_wr) { + virBufferAsprintf(buf, " bps_wr='%llu'", + def->blkiothrottle.bps_wr); + } + + if (def->blkiothrottle.iops) { + virBufferAsprintf(buf, " iops='%llu'", + def->blkiothrottle.iops); + } + + if (def->blkiothrottle.iops_rd) { + virBufferAsprintf(buf, " iops_rd='%llu'", + def->blkiothrottle.iops_rd); + } + + if (def->blkiothrottle.iops_wr) { + virBufferAsprintf(buf, " iops_wr='%llu'", + def->blkiothrottle.iops_wr); + } + + virBufferAsprintf(buf, "/>\n"); + } + if (def->bootIndex) virBufferAsprintf(buf, " <boot order='%d'/>\n", def->bootIndex); if (def->readonly) diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index e07fd2f..b60a6de 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -283,6 +283,17 @@ struct _virDomainDiskDef { virDomainDiskHostDefPtr hosts; char *driverName; char *driverType; + + /*disk I/O throttling*/ + struct { + unsigned long long bps; + unsigned long long bps_rd; + unsigned long long bps_wr; + unsigned long long iops; + unsigned long long iops_rd; + unsigned long long iops_wr; + } blkiothrottle; + char *serial; int cachemode; int error_policy; /* enum virDomainDiskErrorPolicy */ diff --git a/src/driver.h b/src/driver.h index f85a1b1..741e196 100644 --- a/src/driver.h +++ b/src/driver.h @@ -725,6 +725,16 @@ typedef int (*virDrvDomainBlockPull)(virDomainPtr dom, const char *path, unsigned long bandwidth, unsigned int flags); +/* + * Block I/O throttling support + */ + +typedef int + (*virDrvDomainBlockIoThrottle)(virDomainPtr dom, + const char *disk, + virDomainBlockIoThrottleInfoPtr info, + virDomainBlockIoThrottleInfoPtr reply, + unsigned int flags); /** * _virDriver: @@ -881,6 +891,7 @@ struct _virDriver { virDrvDomainGetBlockJobInfo domainGetBlockJobInfo; virDrvDomainBlockJobSetSpeed domainBlockJobSetSpeed; virDrvDomainBlockPull domainBlockPull; + virDrvDomainBlockIoThrottle domainBlockIoThrottle; }; typedef int diff --git a/src/libvirt.c b/src/libvirt.c index 182e031..5f4514c 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -16706,3 +16706,69 @@ error: virDispatchError(dom->conn); return -1; } + +/** + * virDomainBlockIoThrottle: + * @dom: pointer to domain object + * @disk: Fully-qualified disk name + * @info: specify block I/O limits in bytes + * @reply: store block I/O limits setting + * @flags: indicate if it set or display block I/O limits info + * + * This function is mainly to enable Block I/O throttling function in libvirt. + * It is used to set or display block I/O throttling setting for specified domain. + * + * Returns 0 if the operation has started, -1 on failure. + */ +int virDomainBlockIoThrottle(virDomainPtr dom, + const char *disk, + virDomainBlockIoThrottleInfoPtr info, + virDomainBlockIoThrottleInfoPtr reply, + unsigned int flags) +{ + virConnectPtr conn; + + VIR_DOMAIN_DEBUG(dom, "disk=%p, info=%p, reply=%p,flags=%x", + disk, info, reply, flags); + + virResetLastError(); + + if (!VIR_IS_CONNECTED_DOMAIN (dom)) { + virLibDomainError(VIR_ERR_INVALID_DOMAIN, __FUNCTION__); + virDispatchError(NULL); + return -1; + } + conn = dom->conn; + + if (dom->conn->flags & VIR_CONNECT_RO) { + virLibDomainError(VIR_ERR_OPERATION_DENIED, __FUNCTION__); + goto error; + } + + if (!disk) { + virLibDomainError(VIR_ERR_INVALID_ARG, + _("disk name is NULL")); + goto error; + } + + if (!reply) { + virLibDomainError(VIR_ERR_INVALID_ARG, + _("reply is NULL")); + goto error; + } + + if (conn->driver->domainBlockIoThrottle) { + int ret; + ret = conn->driver->domainBlockIoThrottle(dom, disk, info, reply, flags); + if (ret < 0) + goto error; + return ret; + } + + virLibDomainError(VIR_ERR_NO_SUPPORT, __FUNCTION__); + +error: + virDispatchError(dom->conn); + return -1; +} + diff --git a/src/libvirt_public.syms b/src/libvirt_public.syms index afea29b..0a79167 100644 --- a/src/libvirt_public.syms +++ b/src/libvirt_public.syms @@ -493,6 +493,7 @@ LIBVIRT_0.9.7 { global: virDomainReset; virDomainSnapshotGetParent; + virDomainBlockIoThrottle; } LIBVIRT_0.9.5; # .... define new API here using predicted next version number .... diff --git a/src/util/xml.h b/src/util/xml.h index d30e066..14b35b2 100644 --- a/src/util/xml.h +++ b/src/util/xml.h @@ -50,6 +50,9 @@ xmlNodePtr virXPathNode(const char *xpath, int virXPathNodeSet(const char *xpath, xmlXPathContextPtr ctxt, xmlNodePtr **list); +int virXMLStringToULongLong (const char *stringval, + unsigned long long *value); + char * virXMLPropString(xmlNodePtr node, const char *name); -- 1.7.1

5 15

[libvirt] [PATCH v2 0/2] Allow to make disk optional on migration
by Michal Privoznik 13 Oct '11

13 Oct '11

If user wants to migrate a machine, it's all or nothing. Either he/she has all storage accessible under same paths, or migration will fail. But there are some cases, where allowing migration to drop some disks, may be useful. E.g. when host gets itself to a state where domains need to be migrated so host can be say rebooted. Therefore we might want to give user possibility to mark some cdroms/floppy as optional, meaning if destination cannot access them, they get ejected (on destination). This idea is implemented via <migration> element, which basically says what to do with (currently) disk on migration. Right now it contains only one attribute 'optional' accepting values 'require', 'optional' and 'drop'. Then, if destination cannot access a path and corresponding disk is marked as: a) 'require' - migration will fail b) 'optional' - disk gets force ejected (source is free()'d), regardless tray being locked c) 'drop' - disk gets force ejected even if destination could access the path. diff to v1: -change xml representation from 'yes' & 'no' to 'require', 'optional' & 'drop' Michal Privoznik (2): migration: Introduce <migration> element for cdrom and floppy qemu: Implement migration optional disk docs/formatdomain.html.in | 23 ++++++ docs/schemas/domaincommon.rng | 17 ++++ src/conf/domain_conf.c | 78 +++++++++++++++++++- src/conf/domain_conf.h | 17 ++++ src/libvirt_private.syms | 2 + src/qemu/qemu_migration.c | 51 ++++++++++++- .../domain-qemu-complex-migration.xml | 68 +++++++++++++++++ 7 files changed, 253 insertions(+), 3 deletions(-) create mode 100644 tests/domainschemadata/domain-qemu-complex-migration.xml -- 1.7.3.4

3 5

[libvirt] [test-API][PATCH 0/5] Update migrate module
by Wayne Sun 13 Oct '11

13 Oct '11

Current migrate module include the remote_guest_define function which has problem with transient guest created on local, it can't find persistent domain xml on local and copy to remote. This series fixes aim to solve this problem. First, remote_guest_define function in migrate is removed, then add support for define domain on remote in define module. New xml element uuid is added into xmlgenerator, and new variable static_uuid is introduced into env.cfg, this will help domain created on local and defined on remote get the same uuid. At last, all related conf files are updated. [test-API][PATCH 1/5] Remove remote_guest_define function in migrate [test-API][PATCH 2/5] Add new variable static_uuid into env.cfg [test-API][PATCH 3/5] Update add_option_value function in parser [test-API][PATCH 4/5] Add new element uuid in xmlgenerator [test-API][PATCH 5/5] Update the migration case conf file .../migration/ssh_persistent_paused_with_dst.conf | 192 ------------------- .../migration/ssh_persistent_running_with_dst.conf | 192 ------------------- cases/migration/ssh_transient_paused_with_dst.conf | 200 -------------------- .../migration/ssh_transient_running_with_dst.conf | 198 ------------------- .../migration/tcp_persistent_paused_with_dst.conf | 192 ------------------- .../migration/tcp_persistent_running_with_dst.conf | 192 ------------------- .../tcp_sasl_persistent_paused_with_dst.conf | 64 ------- .../tcp_sasl_persistent_running_with_dst.conf | 64 ------- .../tcp_sasl_transient_paused_with_dst.conf | 64 ------- .../tcp_sasl_transient_running_with_dst.conf | 64 ------- cases/migration/tcp_transient_paused_with_dst.conf | 200 -------------------- .../migration/tcp_transient_running_with_dst.conf | 198 ------------------- .../migration/tls_persistent_paused_with_dst.conf | 192 ------------------- .../migration/tls_persistent_running_with_dst.conf | 192 ------------------- .../tls_sasl_persistent_paused_with_dst.conf | 64 ------- .../tls_sasl_persistent_running_with_dst.conf | 64 ------- .../tls_sasl_transient_paused_with_dst.conf | 64 ------- .../tls_sasl_transient_running_with_dst.conf | 64 ------- cases/migration/tls_transient_paused_with_dst.conf | 200 -------------------- .../migration/tls_transient_running_with_dst.conf | 198 ------------------- env.cfg | 3 - parser.py | 12 +- repos/domain/migrate.py | 31 +++- utils/Python/xmlgenerator.py | 7 - 24 files changed, 33 insertions(+), 2878 deletions(-)

2 10

[libvirt] [PATCH] [libvirt-tck] Allow to pass --timer to prove
by Guido Günther 13 Oct '11

13 Oct '11

so execution times can be reported too. --- bin/libvirt-tck | 11 ++++++++++- 1 files changed, 10 insertions(+), 1 deletions(-) diff --git a/bin/libvirt-tck b/bin/libvirt-tck index dcb053c..816234b 100644 --- a/bin/libvirt-tck +++ b/bin/libvirt-tck @@ -109,6 +109,10 @@ Generate an archive containing all the raw test results. The filename given should end in either C<.bz>, C<.tar.gz> or C<.tgz> +=item --timer + +Print elapsed time after each test. + =back =cut @@ -136,6 +140,7 @@ my $quiet = 0; my $help = 0; my $force = 0; my $debug = 0; +my $timer = 0; my $archive; my $config = catfile($confdir, "default.cfg"); my $format = "text"; @@ -149,7 +154,8 @@ if (!GetOptions("verbose" => \$verbose, "config=s" => \$config, "force" => \$force, "format=s" => \$format, - "testdir=s" => \$testdir) || $help) { + "testdir=s" => \$testdir, + "timer" => \$timer) || $help) { pod2usage(-verbose => $help, -output => $help ? \*STDOUT : \*STDERR, -exitval => $help ? 0 : 1); @@ -204,6 +210,9 @@ if ($format eq "xml") { -output => \*STDERR); } +if ($timer) { + push @newargv, "--timer"; +} # This env variable is the only way to pass config into # the Sys::Virt::TCK module from here -- 1.7.6.3

2 2

[libvirt] [libvirt-glib] Don't leak virStreams, take the reference
by Marc-André Lureau 13 Oct '11

13 Oct '11

Without this change, after virStreamNew() and g_value_dup_boxed() it will have refcnt = 2 and will leak when GVirStream is destroyed. --- libvirt-gobject/libvirt-gobject-stream.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/libvirt-gobject/libvirt-gobject-stream.c b/libvirt-gobject/libvirt-gobject-stream.c index 88e3a40..4aeb182 100644 --- a/libvirt-gobject/libvirt-gobject-stream.c +++ b/libvirt-gobject/libvirt-gobject-stream.c @@ -160,7 +160,7 @@ static void gvir_stream_set_property(GObject *object, case PROP_HANDLE: if (priv->handle) virStreamFree(priv->handle); - priv->handle = g_value_dup_boxed(value); + priv->handle = g_value_get_boxed(value); break; default: -- 1.7.6.2

2 1

[libvirt] RPM build failure on Redhat6
by Wayne Xia 13 Oct '11

13 Oct '11

Hi, I want to build rpm for latest libvit version on Redhat6, but when I type: ./autogen.sh --system make rpm -j4 error happens: sitemap.html.tmp index.html.tmp archdomain.html.tmp downloads.html.tmp auth.html.tmp internals/locking.html.tmp archnode.html.tmp hvsupport.html.tmp bugs.html.tmp java.html.tmp locking.html.tmp logging.html.tmp testapi.html.tmp formatnetwork.html.tmp bindings.html.tmp uri.html.tmp drvtest.html.tmp contact.html.tmp formatnwfilter.html.tmp intro.html.tmp drvxen.html.tmp drvesx.html.tmp drvqemu.html.tmp drvvmware.html.tmp testsuites.html.tmp devguide.html.tmp firewall.html.tmp architecture.html.tmp todo.html.tmp formatsnapshot.html.tmp formatstorage.html.tmp deployment.html.tmp internals.html.tmp api_extension.html.tmp drvuml.html.tmp virshcmdref.html.tmp csharp.html.tmp python.html.tmp drvremote.html.tmp archstorage.html.tmp remote.html.tmp drivers.html.tmp formatnode.html.tmp make[2]: Leaving directory `/home/xiawenc/WorkDir/Source/libvirt/libvirt/docs' make[1]: *** [distdir] Error 1 make[1]: Leaving directory `/home/xiawenc/WorkDir/Source/libvirt/libvirt' make: *** [rpm] Error 2 does anyone knows what is happening? -- Best Regards Wayne Xia mail:xiawenc@linux.vnet.ibm.com tel:86-010-82450803

3 4

[libvirt] [PATCH] qemu: Check for domain being active on successful job acquire
by Michal Privoznik 13 Oct '11

13 Oct '11

As this is needed. Although some functions check for domain being active before obtaining job, we need to check it after, because obtaining job unlocks domain object, during which a state of domain can be changed. --- src/qemu/qemu_driver.c | 75 ++++++++++++++++++++++++++++++++++++++++++++---- 1 files changed, 69 insertions(+), 6 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index ec01cd5..1efc66b 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1948,9 +1948,18 @@ static int qemuDomainInjectNMI(virDomainPtr domain, unsigned int flags) if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; + + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + qemuDomainObjEnterMonitorWithDriver(driver, vm); ret = qemuMonitorInjectNMI(priv->mon); qemuDomainObjExitMonitorWithDriver(driver, vm); + +endjob: if (qemuDomainObjEndJob(driver, vm) == 0) { vm = NULL; goto cleanup; @@ -4387,7 +4396,7 @@ static char *qemuDomainGetXMLDesc(virDomainPtr dom, virDomainObjPtr vm; char *ret = NULL; unsigned long balloon; - int err; + int err = 0; /* Flags checked by virDomainDefFormat */ @@ -4413,9 +4422,17 @@ static char *qemuDomainGetXMLDesc(virDomainPtr dom, if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_QUERY) < 0) goto cleanup; + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + qemuDomainObjEnterMonitorWithDriver(driver, vm); err = qemuMonitorGetBalloonInfo(priv->mon, &balloon); qemuDomainObjExitMonitorWithDriver(driver, vm); + +endjob: if (qemuDomainObjEndJob(driver, vm) == 0) { vm = NULL; goto cleanup; @@ -7153,6 +7170,12 @@ qemudDomainBlockStatsFlags (virDomainPtr dom, if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_QUERY) < 0) goto cleanup; + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + qemuDomainObjEnterMonitor(driver, vm); tmp = *nparams; ret = qemuMonitorGetBlockStatsParamsNumber(priv->mon, nparams); @@ -8672,6 +8695,12 @@ qemuDomainMigrateSetMaxSpeed(virDomainPtr dom, if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MIGRATION_OP) < 0) goto cleanup; + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + VIR_DEBUG("Setting migration bandwidth to %luMbs", bandwidth); qemuDomainObjEnterMonitor(driver, vm); ret = qemuMonitorSetMigrationSpeed(priv->mon, bandwidth); @@ -8680,6 +8709,7 @@ qemuDomainMigrateSetMaxSpeed(virDomainPtr dom, if (ret == 0) priv->migMaxBandwidth = bandwidth; +endjob: if (qemuDomainObjEndJob(driver, vm) == 0) vm = NULL; } else { @@ -8774,6 +8804,12 @@ qemuDomainSnapshotCreateActive(virConnectPtr conn, if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) return -1; + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) { /* savevm monitor command pauses the domain emitting an event which * confuses libvirt since it's not notified when qemu resumes the @@ -8823,6 +8859,7 @@ cleanup: _("resuming after snapshot failed")); } +endjob: if (vm && qemuDomainObjEndJob(driver, vm) == 0) { /* Only possible if a transient vm quit while our locks were down, * in which case we don't want to save snapshot metadata. */ @@ -9045,6 +9082,13 @@ qemuDomainSnapshotCreateDiskActive(virConnectPtr conn, if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) return -1; + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + + if (virDomainObjGetState(vm, NULL) == VIR_DOMAIN_RUNNING) { /* In qemu, snapshot_blkdev on a single disk will pause cpus, * but this confuses libvirt since notifications are not given @@ -9127,12 +9171,14 @@ cleanup: (persist && virDomainSaveConfig(driver->configDir, vm->newDef) < 0)) ret = -1; - if (qemuDomainObjEndJob(driver, vm) == 0) { + } + +endjob: + if (vm && (qemuDomainObjEndJob(driver, vm) == 0)) { /* Only possible if a transient vm quit while our locks were down, * in which case we don't want to save snapshot metadata. */ *vmptr = NULL; ret = -1; - } } return ret; @@ -10272,20 +10318,28 @@ static int qemuDomainMonitorCommand(virDomainPtr domain, const char *cmd, goto cleanup; } + if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) + goto cleanup; + + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", + _("domain is not running")); + goto endjob; + } + priv = vm->privateData; qemuDomainObjTaint(driver, vm, VIR_DOMAIN_TAINT_CUSTOM_MONITOR, -1); hmp = !!(flags & VIR_DOMAIN_QEMU_MONITOR_COMMAND_HMP); - if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) - goto cleanup; qemuDomainObjEnterMonitorWithDriver(driver, vm); ret = qemuMonitorArbitraryCommand(priv->mon, cmd, result, hmp); qemuDomainObjExitMonitorWithDriver(driver, vm); + +endjob: if (qemuDomainObjEndJob(driver, vm) == 0) { vm = NULL; - goto cleanup; } cleanup: @@ -10522,10 +10576,19 @@ qemuDomainBlockJobImpl(virDomainPtr dom, const char *path, if (qemuDomainObjBeginJobWithDriver(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; + + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", + _("domain is not running")); + goto endjob; + } + qemuDomainObjEnterMonitorWithDriver(driver, vm); priv = vm->privateData; ret = qemuMonitorBlockJob(priv->mon, device, bandwidth, info, mode); qemuDomainObjExitMonitorWithDriver(driver, vm); + +endjob: if (qemuDomainObjEndJob(driver, vm) == 0) { vm = NULL; goto cleanup; -- 1.7.3.4

2 2

[libvirt] [PATCH] events: Propose a separate lock for event queue
by Michal Privoznik 13 Oct '11

13 Oct '11

Currently, push & pop from event queue (both server & client side) rely on lock from higher levels, e.g. on driver lock (qemu), private_data (remote), ...; This alone is not sufficient as not every function that interacts with this queue can/does lock, esp. in client where we have a different approach, "passing the buck". Therefore we need a separate lock just to protect event queue. For more info see: https://bugzilla.redhat.com/show_bug.cgi?id=743817 --- src/conf/domain_event.c | 65 +++++++++++++++++++++++++++++++++++++++++------ src/conf/domain_event.h | 1 + 2 files changed, 58 insertions(+), 8 deletions(-) diff --git a/src/conf/domain_event.c b/src/conf/domain_event.c index 3189346..6cc8168 100644 --- a/src/conf/domain_event.c +++ b/src/conf/domain_event.c @@ -547,6 +547,24 @@ virDomainEventQueuePtr virDomainEventQueueNew(void) return ret; } +static void +virDomainEventStateLock(virDomainEventStatePtr state) +{ + if (!state) + return; + + virMutexLock(&state->lock); +} + +static void +virDomainEventStateUnlock(virDomainEventStatePtr state) +{ + if (!state) + return; + + virMutexUnlock(&state->lock); +} + /** * virDomainEventStateFree: * @list: virDomainEventStatePtr to free @@ -564,6 +582,8 @@ virDomainEventStateFree(virDomainEventStatePtr state) if (state->timer != -1) virEventRemoveTimeout(state->timer); + + virMutexDestroy(&state->lock); VIR_FREE(state); } @@ -590,6 +610,13 @@ virDomainEventStateNew(virEventTimeoutCallback timeout_cb, goto error; } + if (virMutexInit(&state->lock) < 0) { + virReportSystemError(errno, "%s", + _("unable to initialize state mutex")); + VIR_FREE(state); + goto error; + } + if (VIR_ALLOC(state->callbacks) < 0) { virReportOOMError(); goto error; @@ -1161,18 +1188,26 @@ void virDomainEventStateQueue(virDomainEventStatePtr state, virDomainEventPtr event) { + int count; + if (state->timer < 0) { virDomainEventFree(event); return; } + virDomainEventStateLock(state); + if (virDomainEventQueuePush(state->queue, event) < 0) { VIR_DEBUG("Error adding event to queue"); virDomainEventFree(event); } - if (state->queue->count == 1) + count = state->queue->count; + virDomainEventStateUnlock(state); + + if (count == 1) virEventUpdateTimeout(state->timer, 0); + } void @@ -1182,6 +1217,7 @@ virDomainEventStateFlush(virDomainEventStatePtr state, { virDomainEventQueue tempQueue; + virDomainEventStateLock(state); state->isDispatching = true; /* Copy the queue, so we're reentrant safe when dispatchFunc drops the @@ -1190,6 +1226,7 @@ virDomainEventStateFlush(virDomainEventStatePtr state, tempQueue.events = state->queue->events; state->queue->count = 0; state->queue->events = NULL; + virDomainEventStateUnlock(state); virEventUpdateTimeout(state->timer, -1); virDomainEventQueueDispatch(&tempQueue, @@ -1198,9 +1235,11 @@ virDomainEventStateFlush(virDomainEventStatePtr state, opaque); /* Purge any deleted callbacks */ + virDomainEventStateLock(state); virDomainEventCallbackListPurgeMarked(state->callbacks); state->isDispatching = false; + virDomainEventStateUnlock(state); } int @@ -1208,11 +1247,16 @@ virDomainEventStateDeregister(virConnectPtr conn, virDomainEventStatePtr state, virConnectDomainEventCallback callback) { + int ret; + + virDomainEventStateLock(state); if (state->isDispatching) - return virDomainEventCallbackListMarkDelete(conn, - state->callbacks, callback); + ret = virDomainEventCallbackListMarkDelete(conn, + state->callbacks, callback); else - return virDomainEventCallbackListRemove(conn, state->callbacks, callback); + ret = virDomainEventCallbackListRemove(conn, state->callbacks, callback); + virDomainEventStateUnlock(state); + return ret; } int @@ -1220,10 +1264,15 @@ virDomainEventStateDeregisterAny(virConnectPtr conn, virDomainEventStatePtr state, int callbackID) { + int ret; + + virDomainEventStateLock(state); if (state->isDispatching) - return virDomainEventCallbackListMarkDeleteID(conn, - state->callbacks, callbackID); + ret = virDomainEventCallbackListMarkDeleteID(conn, + state->callbacks, callbackID); else - return virDomainEventCallbackListRemoveID(conn, - state->callbacks, callbackID); + ret = virDomainEventCallbackListRemoveID(conn, + state->callbacks, callbackID); + virDomainEventStateUnlock(state); + return ret; } diff --git a/src/conf/domain_event.h b/src/conf/domain_event.h index b06be16..08930ed 100644 --- a/src/conf/domain_event.h +++ b/src/conf/domain_event.h @@ -61,6 +61,7 @@ struct _virDomainEventState { int timer; /* Flag if we're in process of dispatching */ bool isDispatching; + virMutex lock; }; typedef struct _virDomainEventState virDomainEventState; typedef virDomainEventState *virDomainEventStatePtr; -- 1.7.3.4

3 5

[libvirt] [PATCH v4 0/2] Allow domain reboot after core dump
by Michal Privoznik 13 Oct '11

13 Oct '11

This is the 4th (and hopefully last) version of these patches. diff to v3: -split in 2 patches -and some other Eric's suggestions taken in Michal Privoznik (2): virDomainCoreDump: Introduce VIR_DUMP_RESET flag qemu: Implement VIR_DUMP_RESET include/libvirt/libvirt.h.in | 1 + src/libvirt.c | 15 ++++++++++++++- src/qemu/qemu_driver.c | 9 ++++++++- tools/virsh.c | 3 +++ tools/virsh.pod | 4 +++- 5 files changed, 29 insertions(+), 3 deletions(-) -- 1.7.3.4

2 8

[libvirt] Possible security hole? unprivileged user can use virsh to overwrite sensitive system file
by Hong Xiang 13 Oct '11

13 Oct '11

I found there's a way for a unprivileged user to overwrite sensitive system file with virsh, here's how: 1. (as an unprivileged user) start virsh and connect to the r/w socket of libvirtd: virsh -c qemu+unix:///system?socket=/var/run/libvirt/libvirt-sock 2. start a guest, then issue 'save' or 'dump' command, giving a sensitive system file path as the <file> parameter, for example, '/etc/passwd'; 3. the sensitive system file will be overwritten; Attached is a test log. I'm using libvirt-0.8.7 on a OpenClient for RHEL 6.1. And latest libvirt code shows the same symptom. BTW, virsh expands the <file> parameter in step to an absolute path if user-provided is not, and libvirtd interprets it as a local file. IMHO it does not look quite right, especially when the virsh-to-libvirtd connection is remote. -- Thanks. Hong Xiang

3 3

[libvirt] [PATCH] qemu: Do not reattach PCI device used by other domain when shutdown
by Osier Yang 13 Oct '11

13 Oct '11

When failing on starting a domain, it tries to reattach all the PCI devices defined in the domain conf, regardless of whether the devices are still used by other domain. This will cause the devices are deleted from the list qemu_driver->activePciHostdevs, thus the devices will be thought as usable even if it's not true. And following commands nodedev-{reattach,reset} will be successful. How to reproduce: 1) Define two domains with same PCI device defined in the confs. 2) # virsh start domain1 3) # virsh start domain2 4) # virsh nodedev-reattach $pci_device You will see the device will be reattached to host successfully. As pciDeviceReattach just check if the device is still used by other domain via checking if the device is in list driver->activePciHostdevs, however, the device is deleted from the list by step 2). This patch is to prohibit the bug by: 1) Prohibit a domain starting or device attachment right at preparation period (qemuPrepareHostdevPCIDevices) if the device is in list driver->activePciHostdevs, which means it's used by other domain. 2) Introduces a new field for struct _pciDevice, (char *used_by), it will be set as the domain name at preparation period, (qemuPrepareHostdevPCIDevices). Thus we can prohibit deleting the device from driver->activePciHostdevs if it's still used by other domain when stopping the domain process. * src/pci.h (define two internal functions, pciDeviceSetUsedBy and pciDevceGetUsedBy) * src/pci.c (new field "char *used_by" for struct _pciDevice, implementations for the two new functions) * src/libvirt_private.syms (Add the two new internal functions) * src/qemu_hostdev.h (Modify the definition of functions qemuPrepareHostdevPCIDevices, and qemuDomainReAttachHostdevDevices) * src/qemu_hostdev.c (Prohibit preparation and don't delete the device from activePciHostdevs list if it's still used by other domain) * src/qemu_hotplug.c (Update function usage, as the definitions are changed) --- src/libvirt_private.syms | 2 ++ src/qemu/qemu_hostdev.c | 31 ++++++++++++++++++++++++++++--- src/qemu/qemu_hostdev.h | 2 ++ src/qemu/qemu_hotplug.c | 4 ++-- src/util/pci.c | 22 ++++++++++++++++++++++ src/util/pci.h | 3 +++ 6 files changed, 59 insertions(+), 5 deletions(-) diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index 8235ea1..a5c5e6c 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -872,6 +872,7 @@ virNWFilterHashTableRemoveEntry; pciDettachDevice; pciDeviceFileIterate; pciDeviceGetManaged; +pciDeviceGetUsedBy; pciDeviceIsAssignable; pciDeviceIsVirtualFunction; pciDeviceListAdd; @@ -884,6 +885,7 @@ pciDeviceListSteal; pciDeviceNetName; pciDeviceReAttachInit; pciDeviceSetManaged; +pciDeviceSetUsedBy; pciFreeDevice; pciGetDevice; pciGetPhysicalFunction; diff --git a/src/qemu/qemu_hostdev.c b/src/qemu/qemu_hostdev.c index 6f77717..ef9e3b7 100644 --- a/src/qemu/qemu_hostdev.c +++ b/src/qemu/qemu_hostdev.c @@ -101,6 +101,7 @@ cleanup: int qemuPrepareHostdevPCIDevices(struct qemud_driver *driver, + const char *name, virDomainHostdevDefPtr *hostdevs, int nhostdevs) { @@ -126,7 +127,10 @@ int qemuPrepareHostdevPCIDevices(struct qemud_driver *driver, for (i = 0; i < pciDeviceListCount(pcidevs); i++) { pciDevice *dev = pciDeviceListGet(pcidevs, i); if (!pciDeviceIsAssignable(dev, !driver->relaxedACS)) - goto reattachdevs; + goto cleanup; + + if (pciDeviceListFind(driver->activePciHostdevs, dev)) + goto cleanup; if (pciDeviceGetManaged(dev) && pciDettachDevice(dev, driver->activePciHostdevs) < 0) @@ -156,6 +160,14 @@ int qemuPrepareHostdevPCIDevices(struct qemud_driver *driver, pciDeviceListSteal(pcidevs, dev); } + /* Now set the used_by_domain of the device in driver->activePciHostdevs + * as domain name. + */ + for (i = 0; i < pciDeviceListCount(driver->activePciHostdevs); i++) { + pciDevice * dev = pciDeviceListGet(driver->activePciHostdevs, i); + pciDeviceSetUsedBy(dev, name); + } + ret = 0; goto cleanup; @@ -183,7 +195,7 @@ static int qemuPrepareHostPCIDevices(struct qemud_driver *driver, virDomainDefPtr def) { - return qemuPrepareHostdevPCIDevices(driver, def->hostdevs, def->nhostdevs); + return qemuPrepareHostdevPCIDevices(driver, def->name, def->hostdevs, def->nhostdevs); } @@ -258,11 +270,13 @@ void qemuReattachPciDevice(pciDevice *dev, struct qemud_driver *driver) void qemuDomainReAttachHostdevDevices(struct qemud_driver *driver, + const char *name, virDomainHostdevDefPtr *hostdevs, int nhostdevs) { pciDeviceList *pcidevs; int i; + const char *used_by = NULL; if (!(pcidevs = qemuGetPciHostDeviceList(hostdevs, nhostdevs))) { virErrorPtr err = virGetLastError(); @@ -277,6 +291,17 @@ void qemuDomainReAttachHostdevDevices(struct qemud_driver *driver, for (i = 0; i < pciDeviceListCount(pcidevs); i++) { pciDevice *dev = pciDeviceListGet(pcidevs, i); + pciDevice *activeDev = NULL; + + /* Never delete the dev from list driver->activePciHostdevs + * if it's used by other domain. + */ + activeDev = pciDeviceListFind(driver->activePciHostdevs, dev); + if (activeDev && + (used_by = pciDeviceGetUsedBy(activeDev)) && + STRNEQ(used_by, name)) + continue; + pciDeviceListDel(driver->activePciHostdevs, dev); } @@ -305,5 +330,5 @@ void qemuDomainReAttachHostDevices(struct qemud_driver *driver, if (!def->nhostdevs) return; - qemuDomainReAttachHostdevDevices(driver, def->hostdevs, def->nhostdevs); + qemuDomainReAttachHostdevDevices(driver, def->name, def->hostdevs, def->nhostdevs); } diff --git a/src/qemu/qemu_hostdev.h b/src/qemu/qemu_hostdev.h index 1f3d1bc..07d7de2 100644 --- a/src/qemu/qemu_hostdev.h +++ b/src/qemu/qemu_hostdev.h @@ -30,12 +30,14 @@ int qemuUpdateActivePciHostdevs(struct qemud_driver *driver, virDomainDefPtr def); int qemuPrepareHostdevPCIDevices(struct qemud_driver *driver, + const char *name, virDomainHostdevDefPtr *hostdevs, int nhostdevs); int qemuPrepareHostDevices(struct qemud_driver *driver, virDomainDefPtr def); void qemuReattachPciDevice(pciDevice *dev, struct qemud_driver *driver); void qemuDomainReAttachHostdevDevices(struct qemud_driver *driver, + const char *name, virDomainHostdevDefPtr *hostdevs, int nhostdevs); void qemuDomainReAttachHostDevices(struct qemud_driver *driver, diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 6cfe392..dc920e7 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -859,7 +859,7 @@ int qemuDomainAttachHostPciDevice(struct qemud_driver *driver, return -1; } - if (qemuPrepareHostdevPCIDevices(driver, &hostdev, 1) < 0) + if (qemuPrepareHostdevPCIDevices(driver, vm->def->name, &hostdev, 1) < 0) return -1; if (qemuCapsGet(priv->qemuCaps, QEMU_CAPS_DEVICE)) { @@ -925,7 +925,7 @@ error: hostdev->info.addr.pci.slot) < 0) VIR_WARN("Unable to release PCI address on host device"); - qemuDomainReAttachHostdevDevices(driver, &hostdev, 1); + qemuDomainReAttachHostdevDevices(driver, vm->def->name, &hostdev, 1); VIR_FREE(devstr); VIR_FREE(configfd_name); diff --git a/src/util/pci.c b/src/util/pci.c index 8d8e157..38548c7 100644 --- a/src/util/pci.c +++ b/src/util/pci.c @@ -62,6 +62,7 @@ struct _pciDevice { char name[PCI_ADDR_LEN]; /* domain:bus:slot.function */ char id[PCI_ID_LEN]; /* product vendor */ char *path; + char *used_by; /* The domain which uses the device */ int fd; unsigned initted; @@ -1312,6 +1313,7 @@ pciGetDevice(unsigned domain, dev->bus = bus; dev->slot = slot; dev->function = function; + dev->used_by = NULL; if (snprintf(dev->name, sizeof(dev->name), "%.4x:%.2x:%.2x.%.1x", dev->domain, dev->bus, dev->slot, @@ -1374,6 +1376,7 @@ pciFreeDevice(pciDevice *dev) VIR_DEBUG("%s %s: freeing", dev->id, dev->name); pciCloseConfig(dev); VIR_FREE(dev->path); + VIR_FREE(dev->used_by); VIR_FREE(dev); } @@ -1387,6 +1390,25 @@ unsigned pciDeviceGetManaged(pciDevice *dev) return dev->managed; } +int +pciDeviceSetUsedBy(pciDevice *dev, const char *name) +{ + dev->used_by = strdup(name); + + if (!dev->used_by) { + virReportOOMError(); + return -1; + } + + return 0; +} + +const char * +pciDeviceGetUsedBy(pciDevice *dev) +{ + return dev->used_by; +} + void pciDeviceReAttachInit(pciDevice *pci) { pci->unbind_from_stub = 1; diff --git a/src/util/pci.h b/src/util/pci.h index a1600fe..c9d8227 100644 --- a/src/util/pci.h +++ b/src/util/pci.h @@ -47,6 +47,9 @@ int pciResetDevice (pciDevice *dev, void pciDeviceSetManaged(pciDevice *dev, unsigned managed); unsigned pciDeviceGetManaged(pciDevice *dev); +int pciDeviceSetUsedBy(pciDevice *dev, + const char *used_by); +const char *pciDeviceGetUsedBy(pciDevice *dev); void pciDeviceReAttachInit(pciDevice *dev); pciDeviceList *pciDeviceListNew (void); -- 1.7.6

2 3

Re: [libvirt] [virt-tools-list] virt-manager/libvirt backwards compatibility problem?
by Eric Blake 13 Oct '11

13 Oct '11

[adding libvir-list] On 07/27/2011 10:28 AM, Whit Blauvelt wrote: > Looks like my real problem may be not incorporating a Debian/Ubuntu patch > before building 0.9.x, since netcat differs: > > https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/517478 Reading that bug report, it looks like upstream libvirt.git should be patched to add a runtime test on whether a remote nc supports the -q option, rather than blindly assuming that 'nc -q' exists on the remote side. -- Eric Blake eblake(a)redhat.com +1-801-349-2682 Libvirt virtualization library http://libvirt.org

4 9

[libvirt] [PATCH 0/3] example: event-test.py
by Philipp Hahn 13 Oct '11

13 Oct '11

Some minor improvements to the event-test Python example. Philipp Hahn (3): example: Fix argument handling example: Redirect --help output to stdout/stderr example: Support debug output and loop switch examples/domain-events/events-python/event-test.py | 25 +++++++++++++------ 1 files changed, 17 insertions(+), 8 deletions(-)

2 6

[libvirt] [PATCH 0/4] Xen fixes
by Philipp Hahn 12 Oct '11

12 Oct '11

Here are several patches relevant to Xen: The first two patches fix the test-suite when building as root in a Xen-dom0 domain. The other two patches fix problems when doing "virsh edit" on Xen domains. Philipp Hahn (4): tests: Add support for skipping tests tests: Skip Xen-HVM tests for root on dom0 xen: fix PyGrub boot device order xen: Return tap2 for tap2 disks src/xenxs/xen_sxpr.c | 33 ++++++++++++----- tests/testutils.c | 4 ++ tests/xencapstest.c | 95 ++++++++++++++++++++++++++++++++++---------------- 3 files changed, 93 insertions(+), 39 deletions(-)

2 8

[libvirt] [PATCH] [libvirt-perl] Add missing $flags arguments to documentation
by Guido Günther 12 Oct '11

12 Oct '11

Hi, when looking at the DomainSnapshot docs I wondered if these also take the flags argument as the other functions so I went ahead and updated the docs after checking the source. Cheers, -- Guido --- lib/Sys/Virt/DomainSnapshot.pm | 16 ++++++++++------ 1 files changed, 10 insertions(+), 6 deletions(-) diff --git a/lib/Sys/Virt/DomainSnapshot.pm b/lib/Sys/Virt/DomainSnapshot.pm index 1c4a008..5b7204b 100644 --- a/lib/Sys/Virt/DomainSnapshot.pm +++ b/lib/Sys/Virt/DomainSnapshot.pm @@ -60,18 +60,22 @@ sub _new { return $self; } -=item my $xml = $dom->get_xml_description() +=item my $xml = $dom->get_xml_description($flags) Returns an XML document containing a complete description of -the domain's configuration +the domain's configuration. -=item $dom->delete() +=item $dom->delete($flags) -Deletes this snapshot object & its datra +Deletes this snapshot object & its data. The optional C<$flags> parameter controls +what should be deleted via the C<Sys::Virt::DomainSnapshot::DELETE_*> +constants. -=item $dom->revert_to() +=item $dom->revert_to($flags) -Revert the domain to the state associated with this snapshot +Revert the domain to the state associated with this snapshot. The optional +C<$flags> control the state of the vm after the revert via the +C<Sys::Virt::DomainSnapshot::REVERT_*> constants. =back -- 1.7.6.3

2 2

[libvirt] build: fail to 'make distcheck'
by Wenyi Gao 12 Oct '11

12 Oct '11

Hi All, I failed to 'make distcheck' on Ubuntu 11.04 with the following information: ................................. 33 OK PASS: nwfilterschematest TEST: domainsnapshotschematest ........... 11 OK PASS: domainsnapshotschematest TEST: test_conf.sh .. 2 OK PASS: test_conf.sh PASS: cpuset TEST: daemon-conf .....!!!!!!...!!!!!!!!!!!!!!!!!!!kill: 98: No such process ! 34 FAILED FAIL: daemon-conf PASS: define-dev-segfault PASS: int-overflow PASS: libvirtd-fail PASS: libvirtd-pool PASS: read-bufsiz PASS: read-non-seekable PASS: start PASS: undefine PASS: vcpupin TEST: virsh-all ........................................ 40 ........................................ 80 ........................................ 120 ........................................ 160 ............ 172 OK PASS: virsh-all PASS: virsh-optparse PASS: virsh-schedinfo PASS: virsh-synopsis TEST: xml2sexprtest ........................................ 40 ............. 53 OK PASS: xml2sexprtest TEST: sexpr2xmltest ........................................ 40 .......... 50 OK PASS: sexpr2xmltest TEST: xmconfigtest ........................................ 40 .......................... 66 OK PASS: xmconfigtest TEST: xencapstest .......... 10 OK PASS: xencapstest TEST: reconnect SKIP: reconnect TEST: statstest SKIP: statstest TEST: qemuxml2argvtest ........................................ 40 ........................................ 80 ........................................ 120 ........................................ 160 . 161 OK PASS: qemuxml2argvtest TEST: qemuxml2xmltest ........................................ 40 ...................................... 78 OK PASS: qemuxml2xmltest TEST: qemuargv2xmltest ........................................ 40 ........................ 64 OK PASS: qemuargv2xmltest TEST: qemuhelptest ............ 12 OK PASS: qemuhelptest TEST: nwfilterxml2xmltest ................................. 33 OK PASS: nwfilterxml2xmltest TEST: openvzutilstest .. 2 OK PASS: openvzutilstest TEST: vmx2xmltest ........................................ 40 ...... 46 OK PASS: vmx2xmltest TEST: xml2vmxtest ........................................ 40 . 41 OK PASS: xml2vmxtest TEST: eventtest ............... 15 OK PASS: eventtest TEST: networkxml2xmltest ............ 12 OK PASS: networkxml2xmltest TEST: networkxml2argvtest ....... 7 OK PASS: networkxml2argvtest TEST: storagevolxml2xmltest ...... 6 OK PASS: storagevolxml2xmltest TEST: storagepoolxml2xmltest ............ 12 OK PASS: storagepoolxml2xmltest TEST: nodedevxml2xmltest ............. 13 OK PASS: nodedevxml2xmltest TEST: interfacexml2xmltest ................... 19 OK PASS: interfacexml2xmltest TEST: cputest ........................................ 40 ................. 57 OK PASS: cputest ======================================= 1 of 58 tests failed (2 tests were not run) Please report to libvir-list(a)redhat.com ======================================= Does anyone experience the problem or do I miss some packages? Thanks. Regards Wenyi Gao

3 4

[libvirt] [PATCH v3] Introduce <driver> under <filesystem> to support open-by-handle.
by Harsh Prateek Bora 12 Oct '11

12 Oct '11

VirtFS allows the user to choose between path/handle based fs driver. As of now, libvirt hardcode to use path based driver only. This patch provides a solution to allow user to choose between path/handle based fs driver. Sample: <filesystem type='mount'> <driver type='handle'/> <source dir='/folder/to/share1'/> <target dir='mount_tag1'/> </filesystem> <filesystem type='mount'> <driver type='path'/> <source dir='/folder/to/share2'/> <target dir='mount_tag2'/> </filesystem> Signed-off-by: Harsh Prateek Bora <harsh(a)linux.vnet.ibm.com> v3: - use enum for fs driver type (qemuDomainFSDriver) v2: - use 'path' instead of 'local' in xml terminology - added default enum for optional driver type --- docs/schemas/domaincommon.rng | 9 +++++++++ src/conf/domain_conf.c | 24 ++++++++++++++++++++++++ src/conf/domain_conf.h | 11 +++++++++++ src/qemu/qemu_command.c | 15 ++++++++++++++- 4 files changed, 58 insertions(+), 1 deletions(-) diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 492a41d..3937393 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -1025,6 +1025,15 @@ </choice> <optional> <ref name="address"/> + <element name="driver"> + <attribute name="type"> + <choice> + <value>path</value> + <value>handle</value> + </choice> + </attribute> + <empty/> + </element> <attribute name="accessmode"> <choice> <value>passthrough</value> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index a537251..ede1f01 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -238,6 +238,11 @@ VIR_ENUM_IMPL(virDomainFS, VIR_DOMAIN_FS_TYPE_LAST, "file", "template") +VIR_ENUM_IMPL(virDomainFSDriverType, VIR_DOMAIN_FS_DRIVER_TYPE_LAST, + "default", + "path", + "handle") + VIR_ENUM_IMPL(virDomainFSAccessMode, VIR_DOMAIN_FS_ACCESSMODE_LAST, "passthrough", "mapped", @@ -2828,6 +2833,7 @@ virDomainFSDefParseXML(xmlNodePtr node, virDomainFSDefPtr def; xmlNodePtr cur; char *type = NULL; + char *fsdriver = NULL; char *source = NULL; char *target = NULL; char *accessmode = NULL; @@ -2878,11 +2884,23 @@ virDomainFSDefParseXML(xmlNodePtr node, target = virXMLPropString(cur, "dir"); } else if (xmlStrEqual(cur->name, BAD_CAST "readonly")) { def->readonly = 1; + } else if ((fsdriver == NULL) && (xmlStrEqual(cur->name, BAD_CAST "driver"))) { + fsdriver = virXMLPropString(cur, "type"); } } cur = cur->next; } + if (fsdriver) { + if ((def->fsdriver = virDomainFSDriverTypeTypeFromString(fsdriver)) < 0) { + virDomainReportError(VIR_ERR_INTERNAL_ERROR, + _("unknown fs driver type '%s'"), fsdriver); + goto error; + } + } else { + def->fsdriver = VIR_DOMAIN_FS_DRIVER_TYPE_PATH; + } + if (source == NULL) { virDomainReportError(VIR_ERR_NO_SOURCE, target ? "%s" : NULL, target); @@ -2905,6 +2923,7 @@ virDomainFSDefParseXML(xmlNodePtr node, cleanup: VIR_FREE(type); + VIR_FREE(fsdriver); VIR_FREE(target); VIR_FREE(source); VIR_FREE(accessmode); @@ -9351,6 +9370,7 @@ virDomainFSDefFormat(virBufferPtr buf, { const char *type = virDomainFSTypeToString(def->type); const char *accessmode = virDomainFSAccessModeTypeToString(def->accessmode); + const char *fsdriver = virDomainFSDriverTypeTypeToString(def->fsdriver); if (!type) { virDomainReportError(VIR_ERR_INTERNAL_ERROR, @@ -9369,6 +9389,10 @@ virDomainFSDefFormat(virBufferPtr buf, " <filesystem type='%s' accessmode='%s'>\n", type, accessmode); + if (def->fsdriver) { + virBufferAsprintf(buf, " <driver type='%s'/>\n", fsdriver); + } + if (def->src) { switch (def->type) { case VIR_DOMAIN_FS_TYPE_MOUNT: diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index e07fd2f..4f8993b 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -368,6 +368,15 @@ enum virDomainFSType { VIR_DOMAIN_FS_TYPE_LAST }; +/* Filesystem driver type */ +enum virDomainFSDriverType { + VIR_DOMAIN_FS_DRIVER_TYPE_DEFAULT = 0, + VIR_DOMAIN_FS_DRIVER_TYPE_PATH, + VIR_DOMAIN_FS_DRIVER_TYPE_HANDLE, + + VIR_DOMAIN_FS_DRIVER_TYPE_LAST +}; + /* Filesystem mount access mode */ enum virDomainFSAccessMode { VIR_DOMAIN_FS_ACCESSMODE_PASSTHROUGH, @@ -381,6 +390,7 @@ typedef struct _virDomainFSDef virDomainFSDef; typedef virDomainFSDef *virDomainFSDefPtr; struct _virDomainFSDef { int type; + int fsdriver; int accessmode; char *src; char *dst; @@ -1856,6 +1866,7 @@ VIR_ENUM_DECL(virDomainController) VIR_ENUM_DECL(virDomainControllerModelSCSI) VIR_ENUM_DECL(virDomainControllerModelUSB) VIR_ENUM_DECL(virDomainFS) +VIR_ENUM_DECL(virDomainFSDriverType) VIR_ENUM_DECL(virDomainFSAccessMode) VIR_ENUM_DECL(virDomainNet) VIR_ENUM_DECL(virDomainNetBackend) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index cf99f89..b4f4bd8 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -100,6 +100,12 @@ VIR_ENUM_IMPL(qemuControllerModelUSB, VIR_DOMAIN_CONTROLLER_MODEL_USB_LAST, "vt82c686b-usb-uhci", "pci-ohci"); +VIR_ENUM_DECL(qemuDomainFSDriver) +VIR_ENUM_IMPL(qemuDomainFSDriver, VIR_DOMAIN_FS_DRIVER_TYPE_LAST, + "local", + "local", + "handle"); + static void uname_normalize (struct utsname *ut) @@ -1811,6 +1817,7 @@ char *qemuBuildFSStr(virDomainFSDefPtr fs, virBitmapPtr qemuCaps ATTRIBUTE_UNUSED) { virBuffer opt = VIR_BUFFER_INITIALIZER; + const char *driver = qemuDomainFSDriverTypeToString(fs->fsdriver); if (fs->type != VIR_DOMAIN_FS_TYPE_MOUNT) { qemuReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", @@ -1818,7 +1825,13 @@ char *qemuBuildFSStr(virDomainFSDefPtr fs, goto error; } - virBufferAddLit(&opt, "local"); + if (!driver) { + qemuReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Filesystem driver type not supported")); + goto error; + } + virBufferAdd(&opt, driver, -1); + if (fs->accessmode == VIR_DOMAIN_FS_ACCESSMODE_MAPPED) { virBufferAddLit(&opt, ",security_model=mapped"); } else if(fs->accessmode == VIR_DOMAIN_FS_ACCESSMODE_PASSTHROUGH) { -- 1.7.1.1

3 2

[libvirt] [PATCH 0/5 v2] Improve Ceph Qemu+RBD support
by Sage Weil 12 Oct '11

12 Oct '11

The current support for qemu and Ceph RBD (rados block device) has two main deficiencies: authentication doesn't work, and it relies on environment variables (which don't work with latest upstream). This patch set addresses both those problems. The first two patches update the xml schemas and conf to add a Ceph secret type and to specify authentication information along with the rbd disk. The next two patches make some libvirt changes. We pass virConnectPtr down into the Domain{Attach,Detach} methods (needed to access secrets while building the qemu command), and add a helper that will escape arbitrary characters. The final patch replaces the current RBD qemu code and uses the new conf info to do authentication properly. (We still need to make a change there to avoid having the authentication key show up on qemu command line; I'll clean that up shortly.) Comments on this approach? Thanks! sage Changes from v1: update docs/schemas/{domain,secret}.rng Sage Weil (5): secret: add Ceph secret type storage: add authId, authDomain to virDomainDiskDef qemu: pass virConnectPtr into Domain{Attach,Detach}* buf: implement generic virBufferEscape qemu/rbd: improve rbd device specification docs/schemas/domain.rng | 6 + docs/schemas/secret.rng | 17 ++ include/libvirt/libvirt.h.in | 3 + src/conf/domain_conf.c | 43 +++- src/conf/domain_conf.h | 2 + src/conf/secret_conf.c | 45 ++++- src/conf/secret_conf.h | 1 + src/libvirt_private.syms | 1 + src/qemu/qemu_command.c | 273 +++++++++++--------- src/qemu/qemu_command.h | 3 +- src/qemu/qemu_driver.c | 17 +- src/qemu/qemu_hotplug.c | 15 +- src/qemu/qemu_hotplug.h | 9 +- src/secret/secret_driver.c | 8 + src/util/buf.c | 33 ++- src/util/buf.h | 1 + .../qemuxml2argv-disk-drive-network-rbd.args | 6 +- .../qemuxml2argv-disk-drive-network-rbd.xml | 1 + 18 files changed, 328 insertions(+), 156 deletions(-) -- 1.7.4.1 >From 498cd06b76bbb4415a2f81f9d169f267ff99329c Mon Sep 17 00:00:00 2001 From: Sage Weil <sage(a)newdream.net> Date: Thu, 15 Sep 2011 13:47:40 -0700 Subject: [PATCH 0/5] Improve Ceph Qemu+RBD support The current support for qemu and Ceph RBD (rados block device) has two main deficiencies: authentication doesn't work, and it relies on environment variables (which don't work with latest upstream). This patch set addresses both those problems, while trying to integrate as cleanly as possible with the rest of libvirt. The first few patches make some changes to libvirt itself: adding a CEPH secret type (for Ceph/RBD authentication), adding authentication fields to the XML schema, passing the virConnectPtr into the Domain{Attach,Detach} methods (needed to access secrets while building the qemu command), a helper that will escape arbitrary characters, and finally a patch that replaces the current RBD qemu code. Comments on this approach? Thanks! sage Sage Weil (5): secret: add Ceph secret type storage: add authId, authDomain to virDomainDiskDef qemu: pass virConnectPtr into Domain{Attach,Detach}* buf: implement generic virBufferEscape qemu/rbd: improve rbd device specification include/libvirt/libvirt.h.in | 3 + src/conf/domain_conf.c | 43 +++- src/conf/domain_conf.h | 2 + src/conf/secret_conf.c | 45 ++++- src/conf/secret_conf.h | 1 + src/libvirt_private.syms | 1 + src/qemu/qemu_command.c | 273 +++++++++++--------- src/qemu/qemu_command.h | 3 +- src/qemu/qemu_driver.c | 17 +- src/qemu/qemu_hotplug.c | 15 +- src/qemu/qemu_hotplug.h | 9 +- src/secret/secret_driver.c | 8 + src/util/buf.c | 33 ++- src/util/buf.h | 1 + .../qemuxml2argv-disk-drive-network-rbd.args | 6 +- .../qemuxml2argv-disk-drive-network-rbd.xml | 1 + 16 files changed, 305 insertions(+), 156 deletions(-) -- 1.7.4.1

4 16