November 2015 - Devel - libvirt List Archives

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

11 Dec '15

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

2 25

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

11 Dec '15

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

4 13

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

11 Dec '15

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

6 37

[libvirt] [PATCH v3] storage sheepdog: allow to specify redundancy level
by Vasiliy Tolstov 10 Dec '15

10 Dec '15

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

2 3

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

10 Dec '15

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

2 2

[libvirt] [PATCH v2 00/27] Prepare for specific vcpu hot(un)plug - part 1
by Peter Krempa 09 Dec '15

09 Dec '15

This series is getting rather big. The target is to refactor the way libvirt stores info about vCPUs into a single structure (okay, two structures for the qemu driver. Part 1 is not yet completely there, well, not even halfway. Future work will involve fully allocating priv->vcpupids to the maxcpus size and moving around few other bits of data in cputune and other parts to the new structure. Yet another follow up work is then to add new APIs for vCPU hotplug, which will enable adding vCPUs sparsely (useful if you have NUMA). Since this refactor will result in tracking all vcpu-related data in one struct, the result will automagically fix a few bugs where we'd end up with invalid config after vcpu unplug or other operations. Version 2 does not contain already pushed patches and incorporates feedback from John's review. Since I've changed quite a few things I'm reposting this. Peter Krempa (27): conf: Replace writes to def->maxvcpus with accessor conf: Use local copy of maxvcpus in virDomainVcpuParse conf: Extract update of vcpu count if maxvcpus is decreased conf: Add helper to check whether domain has offline vCPUs conf: Replace read access to def->maxvcpus with accessor conf: Replace writes to def->vcpus with accessor conf: Move vcpu count check into helper conf: Replace read accesses to def->vcpus with accessor conf: Turn def->maxvcpus into size_t qemu: domain: Add helper to access vm->privateData->agent qemu: Extract vCPU onlining/offlining via agent into a separate function qemu: qemuDomainSetVcpusAgent: re-check agent before calling it the again qemu: Split up vCPU hotplug and hotunplug qemu: cpu hotplug: Fix error handling logic qemu: monitor: Remove weird return values from qemuMonitorSetCPU qemu: cpu hotplug: Move loops to qemuDomainSetVcpusFlags qemu: Refactor qemuDomainHotplugVcpus qemu: refactor qemuDomainHotunplugVcpus conf: turn def->vcpus into a structure conf: ABI: Split up and improve vcpu info ABI checking conf: Add helper to get pointer to a certain vCPU definition qemu: cgroup: Remove now unreachable check qemu: Drop checking vcpu threads in emulator bandwidth getter/setter qemu: Replace checking for vcpu<->pid mapping availability with a helper qemu: Add helper to retrieve vCPU pid qemu: driver: Refactor qemuDomainHelperGetVcpus qemu: cgroup: Don't use priv->ncpupids to iterate domain vCPUs src/bhyve/bhyve_command.c | 2 +- src/bhyve/bhyve_driver.c | 2 +- src/conf/domain_audit.c | 2 +- src/conf/domain_conf.c | 190 ++++++++++++--- src/conf/domain_conf.h | 20 +- src/hyperv/hyperv_driver.c | 10 +- src/libvirt_private.syms | 6 + src/libxl/libxl_conf.c | 6 +- src/libxl/libxl_driver.c | 40 ++-- src/lxc/lxc_controller.c | 2 +- src/lxc/lxc_driver.c | 2 +- src/lxc/lxc_native.c | 7 +- src/openvz/openvz_conf.c | 7 +- src/openvz/openvz_driver.c | 19 +- src/phyp/phyp_driver.c | 12 +- src/qemu/qemu_cgroup.c | 42 ++-- src/qemu/qemu_command.c | 29 ++- src/qemu/qemu_domain.c | 47 ++++ src/qemu/qemu_domain.h | 4 + src/qemu/qemu_driver.c | 542 ++++++++++++++++++++++--------------------- src/qemu/qemu_monitor.c | 3 + src/qemu/qemu_monitor_json.c | 8 - src/qemu/qemu_monitor_text.c | 23 +- src/qemu/qemu_process.c | 22 +- src/test/test_driver.c | 38 +-- src/uml/uml_driver.c | 2 +- src/vbox/vbox_common.c | 19 +- src/vmware/vmware_driver.c | 2 +- src/vmx/vmx.c | 38 +-- src/vz/vz_driver.c | 8 +- src/vz/vz_sdk.c | 13 +- src/xen/xm_internal.c | 19 +- src/xenapi/xenapi_driver.c | 7 +- src/xenapi/xenapi_utils.c | 6 +- src/xenconfig/xen_common.c | 16 +- src/xenconfig/xen_sxpr.c | 27 ++- 36 files changed, 767 insertions(+), 475 deletions(-) -- 2.6.2

1 28

[libvirt] [PATCH v2] guest-agent-socket: don't generate default path to config XML
by Pavel Hrdina 09 Dec '15

09 Dec '15

While we started using for all unix sockets as default one common directory based on a guest name it introduced several issues, for example with renaming the guest or cloning it. In general it's not entirely bad, but in this case it would be best to hide the auto-generated socket path from user and don't export it in the config XML. Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- Version 2: - removed unnecessary changes src/qemu/qemu_command.c | 28 ++++++++++++++++++++++++++++ src/qemu/qemu_command.h | 1 + src/qemu/qemu_domain.c | 16 ---------------- src/qemu/qemu_driver.c | 3 +++ src/qemu/qemu_process.c | 3 +++ tests/qemuhotplugtest.c | 5 +++++ tests/qemuxml2argvtest.c | 5 +++++ 7 files changed, 45 insertions(+), 16 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 2a9fab5..f1bb621 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -1262,6 +1262,34 @@ qemuAssignDeviceAliases(virDomainDefPtr def, virQEMUCapsPtr qemuCaps) } +int +qemuUnixSocketGenerate(virDomainDefPtr def, + virQEMUDriverConfigPtr cfg) +{ + size_t i; + + for (i = 0; i < def->nchannels; i++) { + virDomainChrDefPtr channel = def->channels[i]; + + if (channel->deviceType == VIR_DOMAIN_CHR_DEVICE_TYPE_CHANNEL && + channel->targetType == VIR_DOMAIN_CHR_CHANNEL_TARGET_TYPE_VIRTIO && + channel->source.type == VIR_DOMAIN_CHR_TYPE_UNIX && + !channel->source.data.nix.path) { + if (virAsprintf(&channel->source.data.nix.path, + "%s/domain-%s/%s", + cfg->channelTargetDir, def->name, + channel->target.name ? channel->target.name + : "unknown.sock") < 0) + return -1; + + channel->source.data.nix.listen = true; + } + } + + return 0; +} + + static void qemuDomainPrimeVirtioDeviceAddresses(virDomainDefPtr def, virDomainDeviceAddressType type) diff --git a/src/qemu/qemu_command.h b/src/qemu/qemu_command.h index bebdd27..0512a23 100644 --- a/src/qemu/qemu_command.h +++ b/src/qemu/qemu_command.h @@ -283,6 +283,7 @@ int qemuAssignDevicePCISlots(virDomainDefPtr def, virDomainPCIAddressSetPtr addrs); int qemuAssignDeviceAliases(virDomainDefPtr def, virQEMUCapsPtr qemuCaps); +int qemuUnixSocketGenerate(virDomainDefPtr def, virQEMUDriverConfigPtr cfg); int qemuDomainNetVLAN(virDomainNetDefPtr def); int qemuAssignDeviceNetAlias(virDomainDefPtr def, virDomainNetDefPtr net, int idx); int qemuAssignDeviceDiskAlias(virDomainDefPtr vmdef, diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index ed21245..7e05289 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -1329,22 +1329,6 @@ qemuDomainDeviceDefPostParse(virDomainDeviceDefPtr dev, ARCH_IS_S390(def->os.arch)) dev->data.controller->model = VIR_DOMAIN_CONTROLLER_MODEL_SCSI_VIRTIO_SCSI; - /* auto generate unix socket path */ - if (dev->type == VIR_DOMAIN_DEVICE_CHR && - dev->data.chr->deviceType == VIR_DOMAIN_CHR_DEVICE_TYPE_CHANNEL && - dev->data.chr->targetType == VIR_DOMAIN_CHR_CHANNEL_TARGET_TYPE_VIRTIO && - dev->data.chr->source.type == VIR_DOMAIN_CHR_TYPE_UNIX && - !dev->data.chr->source.data.nix.path) { - if (virAsprintf(&dev->data.chr->source.data.nix.path, - "%s/domain-%s/%s", - cfg->channelTargetDir, def->name, - dev->data.chr->target.name ? dev->data.chr->target.name - : "unknown.sock") < 0) - goto cleanup; - - dev->data.chr->source.data.nix.listen = true; - } - /* forbid capabilities mode hostdev in this kind of hypervisor */ if (dev->type == VIR_DOMAIN_DEVICE_HOSTDEV && dev->data.hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_CAPABILITIES) { diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index ae1d8e7..39f2f03 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -7254,6 +7254,9 @@ static char *qemuConnectDomainXMLToNative(virConnectPtr conn, if (qemuAssignDeviceAliases(def, qemuCaps) < 0) goto cleanup; + if (qemuUnixSocketGenerate(def, cfg) < 0) + goto cleanup; + if (qemuDomainAssignAddresses(def, qemuCaps, NULL) < 0) goto cleanup; diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 192730c..29cf965 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -4663,6 +4663,9 @@ qemuProcessLaunch(virConnectPtr conn, if (qemuAssignDeviceAliases(vm->def, priv->qemuCaps) < 0) goto cleanup; + if (qemuUnixSocketGenerate(vm->def, cfg) < 0) + goto cleanup; + /* Get the advisory nodeset from numad if 'placement' of * either <vcpu> or <numatune> is 'auto'. */ diff --git a/tests/qemuhotplugtest.c b/tests/qemuhotplugtest.c index 102e052..b17cca2 100644 --- a/tests/qemuhotplugtest.c +++ b/tests/qemuhotplugtest.c @@ -61,6 +61,7 @@ qemuHotplugCreateObjects(virDomainXMLOptionPtr xmlopt, { int ret = -1; qemuDomainObjPrivatePtr priv = NULL; + virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(&driver); if (!(*vm = virDomainObjNew(xmlopt))) goto cleanup; @@ -94,10 +95,14 @@ qemuHotplugCreateObjects(virDomainXMLOptionPtr xmlopt, if (qemuAssignDeviceAliases((*vm)->def, priv->qemuCaps) < 0) goto cleanup; + if (qemuUnixSocketGenerate((*vm)->def, cfg) < 0) + goto cleanup; + (*vm)->def->id = QEMU_HOTPLUG_TEST_DOMAIN_ID; ret = 0; cleanup: + virObjectUnref(cfg); return ret; } diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index f7596a0..f6083db 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -262,6 +262,7 @@ static int testCompareXMLToArgvFiles(const char *xml, virCommandPtr cmd = NULL; size_t i; virBitmapPtr nodeset = NULL; + virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(&driver); if (!(conn = virGetConnect())) goto out; @@ -324,6 +325,9 @@ static int testCompareXMLToArgvFiles(const char *xml, if (qemuAssignDeviceAliases(vmdef, extraFlags) < 0) goto out; + if (qemuUnixSocketGenerate(vmdef, cfg) < 0) + goto out; + for (i = 0; i < vmdef->nhostdevs; i++) { virDomainHostdevDefPtr hostdev = vmdef->hostdevs[i]; @@ -387,6 +391,7 @@ static int testCompareXMLToArgvFiles(const char *xml, virCommandFree(cmd); virDomainDefFree(vmdef); virObjectUnref(conn); + virObjectUnref(cfg); virBitmapFree(nodeset); return ret; } -- 2.6.3

2 1

[libvirt] libvirt-php 0.5.1 uneeded files
by Remi Collet 07 Dec '15

07 Dec '15

Installing the new 0.5.1 version: libvirt-php.la libvirt-php.so -> libvirt-php.so.0.0.0 libvirt-php.so.0 -> libvirt-php.so.0.0.0 libvirt-php.so.0.0.0 This is not a versioned library but a PHP extension so having those have no sense. Only libvirt-php.so is needed. Notice: I think it will be simpler to switch to the standard PHP extension build system (phpize; configure; make) Especially, this will allow to use the --with-php-config option (an thus, build for not standard PHP, e.g. TS build or SCL build) Remi.

3 8

[libvirt] [PATCH] doc: extend virDomainSetMemory semantics
by Nikolay Shirokovskiy 04 Dec '15

04 Dec '15

virDomainSetMemory is documented to change only runtime configuration of running domain. This can't be implemented for vz driver as it always change persistent state too. Let's extend semantics of this function for changes to be hyperviror specific. As a result 'virsh setmem' without parameters (which ends up in virDomainSetMemory call) would be possible for a vz domain. By the way, virsh documentation for this case states that behaviour is hypervisor specific already. Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy(a)virtuozzo.com> --- src/libvirt-domain.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/src/libvirt-domain.c b/src/libvirt-domain.c index de7eb04..56ab907 100644 --- a/src/libvirt-domain.c +++ b/src/libvirt-domain.c @@ -1882,8 +1882,9 @@ virDomainSetMaxMemory(virDomainPtr domain, unsigned long memory) * to Domain0 i.e. the domain where the application runs. * This function may require privileged access to the hypervisor. * - * This command only changes the runtime configuration of the domain, - * so can only be called on an active domain. + * This command is hypervisor-specific for whether active, persistent, + * or both configurations are changed; for more control, use + * virDomainSetMemoryFlags(). * * Returns 0 in case of success and -1 in case of failure. */ -- 1.7.1

2 1

[libvirt] [PATCH 0/3] libvirt: Fix automatic SCSI controller create in hotplug
by Boris Fiuczynski 04 Dec '15

04 Dec '15

Trying to hotplug a SCSI device to a domain without the required SCSI controller fails. For hostdev SCSI device hotplug commits 0d8b24f6 and 0785966d rearranged the automatic creation of the required SCSI controllers from the parsing xml code into the post parsing code section of virDomainDefParseXML. In doing so the code will create but not hotplug the missing SCSI controller and on the hotplug path thru the same code the SCSI controller will not get hotplugged since it already exits. Then the SCSI hostdev device is hotplugged and runs into the following internal error error: Failed to attach device from scsidev.xml error: internal error: Device alias was not set for scsi controller with index 0 For disk SCSI device hotplug commit 0260506c added in method qemuBuildDriveDevStr a lookup of the controller alias. The internal error error: Failed to attach device from scsi_disk.xml error: internal error: Could not find scsi controller with index 0 required for device occurs because in method qemuDomainAttachSCSIDisk the automatic creation of the missing SCSI controller occurs after calling qemuBuildDriveDevStr. Boris Fiuczynski (3): conf: Reposition adding SCSI controller for SCSI hostdev hotplug Revert "conf: Try controller add when searching hostdev bus for unit" Automatic SCSI controller creation in SCSI disk hotplug broken src/conf/domain_conf.c | 16 ++++------------ src/qemu/qemu_hotplug.c | 12 ++++++------ 2 files changed, 10 insertions(+), 18 deletions(-) -- 2.3.0

2 16