- Devel - libvirt List Archives

[libvirt] [PATCH RFC V2] RFC: Reimplement cache allocation for a VM
by Eli Qiao 02 May '17

02 May '17

This is a RFC patch for the reimplement of `support cache tune(CAT) in libvirt`[1]. This patch defines some structs to represent data struct in linux resctrl fs which will be used later to do cache allocation. The patch expose a private interface `virResctrlFreeSchemata`, which will be used to query the cache allocation on the host. Also added unit test cases to test this interface can works well. There are already patch sets[2] to address it, and functional works, but people doesn't like it cause it has global variable, and missing unit test case for new added capabilites, etc. Martin has proposed a test infra to do vircaps2xmltest, and I extened it on top of it to extend resctrl control[3], this is kinds of new desiged apart from [2], so I propose this RFC patch to do some rework on it. [1] https://www.redhat.com/archives/libvir-list/2017-January/msg00683.html [2] https://www.redhat.com/archives/libvir-list/2017-March/msg00181.html [3] https://www.redhat.com/archives/libvir-list/2017-April/msg00516.html --- include/libvirt/virterror.h | 1 + src/Makefile.am | 1 + src/libvirt_private.syms | 6 + src/util/virerror.c | 1 + src/util/virresctrl.c | 423 ++++++++++++++++++++++++++++++ src/util/virresctrl.h | 86 ++++++ tests/Makefile.am | 7 +- tests/virresctrldata/L3-free.schemata | 1 + tests/virresctrldata/L3CODE-free.schemata | 1 + tests/virresctrldata/L3DATA-free.schemata | 1 + tests/virresctrltest.c | 117 +++++++++ 11 files changed, 644 insertions(+), 1 deletion(-) create mode 100644 src/util/virresctrl.c create mode 100644 src/util/virresctrl.h create mode 100644 tests/virresctrldata/L3-free.schemata create mode 100644 tests/virresctrldata/L3CODE-free.schemata create mode 100644 tests/virresctrldata/L3DATA-free.schemata create mode 100644 tests/virresctrltest.c diff --git a/include/libvirt/virterror.h b/include/libvirt/virterror.h index 2efee8f..4bc0c74 100644 --- a/include/libvirt/virterror.h +++ b/include/libvirt/virterror.h @@ -132,6 +132,7 @@ typedef enum { VIR_FROM_PERF = 65, /* Error from perf */ VIR_FROM_LIBSSH = 66, /* Error from libssh connection transport */ + VIR_FROM_RESCTRL = 67, /* Error from resctrl */ # ifdef VIR_ENUM_SENTINELS VIR_ERR_DOMAIN_LAST diff --git a/src/Makefile.am b/src/Makefile.am index 60eba37..0ae2af5 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -165,6 +165,7 @@ UTIL_SOURCES = \ util/virprocess.c util/virprocess.h \ util/virqemu.c util/virqemu.h \ util/virrandom.h util/virrandom.c \ + util/virresctrl.h util/virresctrl.c \ util/virrotatingfile.h util/virrotatingfile.c \ util/virscsi.c util/virscsi.h \ util/virscsihost.c util/virscsihost.h \ diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index 9d7760d..b7225fe 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -2396,6 +2396,12 @@ virRandomGenerateWWN; virRandomInt; +# util/virresctrl.h +virResctrlFreeSchemata; +virResctrlGetFreeCache; +virResctrlTypeToString; + + # util/virrotatingfile.h virRotatingFileReaderConsume; virRotatingFileReaderFree; diff --git a/src/util/virerror.c b/src/util/virerror.c index ef17fb5..02fabcc 100644 --- a/src/util/virerror.c +++ b/src/util/virerror.c @@ -139,6 +139,7 @@ VIR_ENUM_IMPL(virErrorDomain, VIR_ERR_DOMAIN_LAST, "Perf", /* 65 */ "Libssh transport layer", + "Resource Control", ) diff --git a/src/util/virresctrl.c b/src/util/virresctrl.c new file mode 100644 index 0000000..778c2ec --- /dev/null +++ b/src/util/virresctrl.c @@ -0,0 +1,423 @@ +/* + * virresctrl.c: methods for managing resource control + * + * Copyright (C) 2017 Intel, Inc. + * + * 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/>. + * + * Authors: + * Eli Qiao <liyong.qiao(a)intel.com> + */ + +#include <config.h> +#include <fcntl.h> +#include <sys/file.h> +#include <sys/stat.h> +#include <sys/types.h> + +#include "virresctrl.h" +#include "virerror.h" +#include "virlog.h" +#include "viralloc.h" +#include "virstring.h" +#include "virfile.h" + +VIR_LOG_INIT("util.resctrl"); + +#define VIR_FROM_THIS VIR_FROM_RESCTRL +#define SYSFS_RESCTRL_PATH "/sys/fs/resctrl/" + +VIR_ENUM_IMPL(virResctrl, VIR_RESCTRL_TYPE_LAST, + "L3", + "L3CODE", + "L3DATA", + "L2") + +/** + * a virResctrlDomain represents a resource control group, it's a directory + * under /sys/fs/resctrl. + * eg: /sys/fs/resctrl/CG1 + * |-- cpus + * |-- schemata + * `-- tasks + * # cat schemata + * L3DATA:0=fffff;1=fffff + * L3CODE:0=fffff;1=fffff + * + * Besides, it can also represent the default resource control group of the + * host. + */ + +typedef struct _virResctrlGroup virResctrlGroup; +typedef virResctrlGroup *virResctrlGroupPtr; +struct _virResctrlGroup { + char *name; /* resource group name, eg: CG1. If it represent host's + default resource group name, should be a NULL pointer */ + size_t n_tasks; /* number of task assigned to the resource group */ + char **tasks; /* task list which contains task id eg: 77454 */ + + size_t n_schematas; /* number of schemata the resource group contains, + eg: 2 */ + virResctrlSchemataPtr *schematas; /* scheamta list */ +}; + +/* All resource control groups on this host, including default resource group */ +typedef struct _virResctrlDomain virResctrlDomain; +typedef virResctrlDomain *virResctrlDomainPtr; +struct _virResctrlDomain { + size_t n_groups; /* number of resource control group */ + virResctrlGroupPtr *groups; /* list of resource control group */ +}; + +void +virResctrlFreeSchemata(virResctrlSchemataPtr ptr) +{ + size_t i; + + if (!ptr) + return; + + for (i = 0; i < ptr->n_schemata_items; i++) + VIR_FREE(ptr->schemata_items[i]); +} + +static void +virResctrlFreeGroup(virResctrlGroupPtr ptr) +{ + size_t i; + + if (!ptr) + return; + + for (i = 0; i < ptr->n_tasks; i++) + VIR_FREE(ptr->tasks[i]); + + for (i = 0; i < ptr->n_schematas; i++) { + virResctrlFreeSchemata(ptr->schematas[i]); + VIR_FREE(ptr->schematas[i]); + } +} + +static void +virResctrlFreeDomain(virResctrlDomainPtr ptr) +{ + size_t i; + + if (!ptr) + return; + + for (i = 0; i < ptr->n_groups; i++) { + virResctrlFreeGroup(ptr->groups[i]); + VIR_FREE(ptr->groups[i]); + } +} + +static int +virResctrlCopySchemata(virResctrlSchemataPtr src, + virResctrlSchemataPtr *dst) +{ + size_t i; + virResctrlSchemataItemPtr schemataitem; + virResctrlSchemataPtr schemata; + + if (VIR_ALLOC(schemata) < 0) + return -1; + + schemata->type = src->type; + + for (i = 0; i < src->n_schemata_items; i++) { + if (VIR_ALLOC(schemataitem) < 0) + goto error; + + schemataitem->cache_id = src->schemata_items[i]->cache_id; + schemataitem->continuous_schemata = src->schemata_items[i]->continuous_schemata; + schemataitem->schemata = src->schemata_items[i]->schemata; + schemataitem->size = src->schemata_items[i]->size; + + if (VIR_APPEND_ELEMENT(schemata->schemata_items, + schemata->n_schemata_items, + schemataitem) < 0) + goto error; + } + + *dst = schemata; + + return 0; + + error: + virResctrlFreeSchemata(schemata); + return -1; +} + +static int +virResctrlGetSchemataString(virResctrlType type, + const char *name, + char **schemata) +{ + int rc = -1; + char *tmp = NULL; + char *end = NULL; + char *buf = NULL; + char *type_suffix = NULL; + + if (virFileReadValueString(&buf, + SYSFS_RESCTRL_PATH "%s/schemata", + name ? name : "") < 0) + return -1; + + if (virAsprintf(&type_suffix, + "%s:", + virResctrlTypeToString(type)) < 0) + goto cleanup; + + tmp = strstr(buf, type_suffix); + + if (!tmp) + goto cleanup; + + end = strchr(tmp, '\n'); + if (end != NULL) + *end = '\0'; + + if (VIR_STRDUP(*schemata, tmp) < 0) + goto cleanup; + + rc = 0; + + cleanup: + VIR_FREE(buf); + VIR_FREE(type_suffix); + return rc; +} + +static int +virResctrlLoadSchemata(const char* schemata_str, + virResctrlSchemataPtr schemata) +{ + VIR_DEBUG("%s, %p\n", schemata_str, schemata); + + int ret = -1; + char **lists = NULL; + char **sms = NULL; + char **sis = NULL; + size_t i; + virResctrlSchemataItemPtr si; + + /* parse L3:0=fffff;1=f */ + lists = virStringSplit(schemata_str, ":", 2); + + if ((!lists) || (!lists[1])) + goto cleanup; + + /* parse 0=fffff;1=f */ + sms = virStringSplit(lists[1], ";", 0); + if (!sms) + goto cleanup; + + for (i = 0; sms[i] != NULL; i++) { + /* parse 0=fffff */ + sis = virStringSplit(sms[i], "=", 2); + if (!sis) + goto cleanup; + + if (VIR_ALLOC(si) < 0) + goto cleanup; + + if (virStrToLong_ui(sis[0], NULL, 10, &si->cache_id) < 0) + goto cleanup; + + if (virStrToLong_ui(sis[1], NULL, 16, &si->continuous_schemata) < 0) + goto cleanup; + + si->schemata = si->continuous_schemata; + + if (VIR_APPEND_ELEMENT(schemata->schemata_items, + schemata->n_schemata_items, + si) < 0) + goto cleanup; + + } + + ret = 0; + + cleanup: + VIR_FREE(si); + virStringListFree(lists); + virStringListFree(sms); + virStringListFree(sis); + return ret; +} + +static int +virResctrlLoadGroup(const char *name, + virResctrlDomainPtr dom) +{ + VIR_DEBUG("%s, %p\n", name, dom); + + int ret = -1; + char *path = NULL; + char *schemata_str; + virResctrlType i; + int rv; + virResctrlGroupPtr grp; + virResctrlSchemataPtr schemata; + + if ((virAsprintf(&path, "%s/%s", SYSFS_RESCTRL_PATH, name)) < 0) + return -1; + + if (!virFileExists(path)) + goto cleanup; + + if (VIR_ALLOC(grp) < 0) + goto cleanup; + + if (VIR_STRDUP(grp->name, name) < 0) + goto cleanup; + + for (i = 0; i < VIR_RESCTRL_TYPE_LAST; i++) { + rv = virResctrlGetSchemataString(i, name, &schemata_str); + if (rv < 0) + continue; + + if (VIR_ALLOC(schemata) < 0) + goto cleanup; + + schemata->type = i; + + if (virResctrlLoadSchemata(schemata_str, schemata) < 0) + goto cleanup; + + VIR_FREE(schemata_str); + + if (VIR_APPEND_ELEMENT(grp->schematas, + grp->n_schematas, + schemata) < 0) + goto cleanup; + + virResctrlFreeSchemata(schemata); + } + + if (VIR_APPEND_ELEMENT(dom->groups, + dom->n_groups, + grp) < 0) + goto cleanup; + + ret = 0; + + cleanup: + VIR_FREE(path); + virResctrlFreeGroup(grp); + return ret; +} + +static int +virResctrlLoadDomain(virResctrlDomainPtr dom) +{ + int ret = -1; + int rv = -1; + DIR *dirp = NULL; + char *path = NULL; + struct dirent *ent; + + VIR_DEBUG("%s, %p\n", "", dom); + + rv = virDirOpenIfExists(&dirp, SYSFS_RESCTRL_PATH); + + if (rv < 0) + goto cleanup; + + /* load default resctrl group */ + if (virResctrlLoadGroup("", dom) < 0) + goto cleanup; + + while ((rv = virDirRead(dirp, &ent, path)) > 0) { + /* only read directory in resctrl */ + if ((ent->d_type != DT_DIR) || STREQ(ent->d_name, "info")) + continue; + + if (virResctrlLoadGroup(ent->d_name, dom) < 0) + goto cleanup; + } + + ret = 0; + + cleanup: + virDirClose(&dirp); + return ret; +} + +static void +virResctrlRefreshDom(virResctrlDomainPtr dom, virResctrlType type) +{ + size_t i; + size_t j; + size_t k; + + virResctrlGroupPtr default_grp = NULL; + virResctrlGroupPtr grp = NULL; + virResctrlSchemataPtr schemata = NULL; + virResctrlSchemataItemPtr schemataitem = NULL; + + default_grp = dom->groups[0]; + + /* We are sure that the first group is the default one */ + for (i = 1; i < dom->n_groups; i++) { + grp = dom->groups[i]; + for (j = 0; j < grp->n_schematas; j++) { + schemata = grp->schematas[j]; + /* we can only calculate one type of schemata */ + if (schemata->type != type) + continue; + for (k = 0; k < schemata->n_schemata_items; k++) { + schemataitem = schemata->schemata_items[k]; + /* if the schemata = 1, ignore it */ + if (schemataitem->continuous_schemata > 1) + /* calculate default schemata, it can be non-continuous */ + default_grp->schematas[j]->schemata_items[k]->schemata &= ~(schemataitem->continuous_schemata); + } + } + } +} + +int virResctrlGetFreeCache(virResctrlType type, + virResctrlSchemataPtr *schemata) +{ + VIR_DEBUG("%d, %p\n", type, schemata); + int ret = -1; + size_t i; + virResctrlDomainPtr dom = NULL; + virResctrlGroupPtr grp = NULL; + + if (VIR_ALLOC(dom) < 0) + return -1; + + if (virResctrlLoadDomain(dom) < 0) + goto cleanup; + + virResctrlRefreshDom(dom, type); + grp = dom->groups[0]; + + for (i = 0; i < grp->n_schematas; i ++) + if (grp->schematas[i]->type == type) + if (virResctrlCopySchemata(grp->schematas[i], schemata) < 0) + goto cleanup; + + if (schemata != NULL) + ret = 0; + + cleanup: + virResctrlFreeDomain(dom); + return ret; +} diff --git a/src/util/virresctrl.h b/src/util/virresctrl.h new file mode 100644 index 0000000..1b040d4 --- /dev/null +++ b/src/util/virresctrl.h @@ -0,0 +1,86 @@ +/* + * virresctrl.h: header for managing resctrl control + * + * Copyright (C) 2017 Intel, Inc. + * + * 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/>. + * + * Authors: + * Eli Qiao <liyong.qiao(a)intel.com> + */ + +#ifndef __VIR_RESCTRL_H__ +# define __VIR_RESCTRL_H__ + +#include "virutil.h" + +typedef enum { + VIR_RESCTRL_TYPE_L3, + VIR_RESCTRL_TYPE_L3_CODE, + VIR_RESCTRL_TYPE_L3_DATA, + VIR_RESCTRL_TYPE_L2, + + VIR_RESCTRL_TYPE_LAST +} virResctrlType; + +VIR_ENUM_DECL(virResctrl); + +/* + * a virResctrlSchemataItem represents one of schemata object in a + * resource control group. + * eg: 0=f + */ +typedef struct _virResctrlSchemataItem virResctrlSchemataItem; +typedef virResctrlSchemataItem *virResctrlSchemataItemPtr; +struct _virResctrlSchemataItem { + unsigned int cache_id; /* cache resource id, eg: 0 */ + unsigned int continuous_schemata; /* schemata, should be a continuous bits, + eg: f, this schemata can be persisted + to sysfs */ + unsigned int schemata; /* schemata eg: f0f, a schemata which is calculated + at running time */ + unsigned long long size; /* the cache size schemata represented in B, + eg: (min * bits of continuous_schemata) */ +}; + +/* + * a virResctrlSchemata represents schemata objects of specific type of + * resource in a resource control group. + * eg: L3:0=f,1=ff + */ +typedef struct _virResctrlSchemata virResctrlSchemata; +typedef virResctrlSchemata *virResctrlSchemataPtr; +struct _virResctrlSchemata { + virResctrlType type; /* resource control type, eg: L3 */ + size_t n_schemata_items; /* number of schemata item, eg: 2 */ + virResctrlSchemataItemPtr *schemata_items; /* pointer list of schemata item */ +}; + +/* Get free cache of the host, result saved in schemata */ +int virResctrlGetFreeCache(virResctrlType type, + virResctrlSchemataPtr *schemata); + + +/* TODO Need to first define virDomainCachetunePtr */ +/* Set cache allocation for a VM domain */ +// int virResctrlSetCacheBanks(virDomainCachetunePtr cachetune, +// unsigned char *group_name, +// size_t n_pids, +// pid_t *pids); +// +/* remove cache allocation for a VM domain */ +// int virResctrlRemoveCacheBanks(unsigned char *group_name); +void virResctrlFreeSchemata(virResctrlSchemataPtr ptr); +#endif diff --git a/tests/Makefile.am b/tests/Makefile.am index 3cc828d..0e09e43 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -229,6 +229,7 @@ if WITH_LINUX test_programs += fchosttest test_programs += scsihosttest test_programs += vircaps2xmltest +test_programs += virresctrltest test_libraries += virusbmock.la \ virnetdevbandwidthmock.la \ virnumamock.la \ @@ -1150,6 +1151,10 @@ vircaps2xmltest_SOURCES = \ vircaps2xmltest.c testutils.h testutils.c virfilemock.c vircaps2xmltest_LDADD = $(LDADDS) +virresctrltest_SOURCES = \ + virresctrltest.c testutils.h testutils.c virfilemock.c +virresctrltest_LDADD = $(LDADDS) + virnumamock_la_SOURCES = \ virnumamock.c virnumamock_la_CFLAGS = $(AM_CFLAGS) @@ -1157,7 +1162,7 @@ virnumamock_la_LDFLAGS = $(MOCKLIBS_LDFLAGS) virnumamock_la_LIBADD = $(MOCKLIBS_LIBS) else ! WITH_LINUX -EXTRA_DIST += vircaps2xmltest.c virnumamock.c +EXTRA_DIST += vircaps2xmltest.c virresctrltest.c virnumamock.c endif ! WITH_LINUX if WITH_NSS diff --git a/tests/virresctrldata/L3-free.schemata b/tests/virresctrldata/L3-free.schemata new file mode 100644 index 0000000..9b47d25 --- /dev/null +++ b/tests/virresctrldata/L3-free.schemata @@ -0,0 +1 @@ +L3:0=1ffff;1=1ffff diff --git a/tests/virresctrldata/L3CODE-free.schemata b/tests/virresctrldata/L3CODE-free.schemata new file mode 100644 index 0000000..7039c45 --- /dev/null +++ b/tests/virresctrldata/L3CODE-free.schemata @@ -0,0 +1 @@ +L3CODE:0=cffff;1=cffff diff --git a/tests/virresctrldata/L3DATA-free.schemata b/tests/virresctrldata/L3DATA-free.schemata new file mode 100644 index 0000000..30f1cbd --- /dev/null +++ b/tests/virresctrldata/L3DATA-free.schemata @@ -0,0 +1 @@ +L3DATA:0=3ffff;1=3ffff diff --git a/tests/virresctrltest.c b/tests/virresctrltest.c new file mode 100644 index 0000000..4926468 --- /dev/null +++ b/tests/virresctrltest.c @@ -0,0 +1,117 @@ +/* + * Copyright (C) Intel, Inc. 2017 + * + * 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/>. + * + * Authors: + * Eli Qiao <liyong.qiao(a)intel.com> + */ + +#include <config.h> +#include <stdlib.h> + +#include "testutils.h" +#include "virbitmap.h" +#include "virfilemock.h" +#include "virresctrl.h" + + +#define VIR_FROM_THIS VIR_FROM_NONE + +struct virResctrlData { + const char *filename; + virResctrlType type; +}; + +static void +GetSchemataStr(virResctrlSchemataPtr schemata, char **str) +{ + size_t i; + virBuffer buf = VIR_BUFFER_INITIALIZER; + virBufferAsprintf(&buf, "%s:%u=%x", + virResctrlTypeToString(schemata->type), + schemata->schemata_items[0]->cache_id, + schemata->schemata_items[0]->schemata); + + for (i = 1; i < schemata->n_schemata_items; i ++) + virBufferAsprintf(&buf, ";%u=%x", + schemata->schemata_items[i]->cache_id, + schemata->schemata_items[i]->schemata); + + *str = virBufferContentAndReset(&buf); +} + +static int +test_virResctrl(const void *opaque) +{ + struct virResctrlData *data = (struct virResctrlData *) opaque; + char *dir = NULL; + char *resctrl = NULL; + int ret = -1; + virResctrlSchemataPtr schemata = NULL; + char *schemata_str; + char *schemata_file; + + if (virAsprintf(&resctrl, "%s/virresctrldata/linux-%s/resctrl", + abs_srcdir, data->filename) < 0) + goto cleanup; + + if (virAsprintf(&schemata_file, "%s/virresctrldata/%s-free.schemata", + abs_srcdir, virResctrlTypeToString(data->type)) < 0) + goto cleanup; + + virFileMockAddPrefix("/sys/fs/resctrl", resctrl); + + if (virResctrlGetFreeCache(data->type, &schemata) < 0) + goto cleanup; + + GetSchemataStr(schemata, &schemata_str); + + if (virTestCompareToFile(schemata_str, schemata_file) < 0) + goto cleanup; + + virFileMockClearPrefixes(); + + ret = 0; + + cleanup: + VIR_FREE(dir); + VIR_FREE(resctrl); + VIR_FREE(schemata_str); + virResctrlFreeSchemata(schemata); + return ret; +} + +static int +mymain(void) +{ + int ret = 0; + +#define DO_TEST_FULL(filename, type) \ + do { \ + struct virResctrlData data = {filename, \ + type}; \ + if (virTestRun(filename, test_virResctrl, &data) < 0) \ + ret = -1; \ + } while (0) + + DO_TEST_FULL("resctrl", VIR_RESCTRL_TYPE_L3); + DO_TEST_FULL("resctrl-cdp", VIR_RESCTRL_TYPE_L3_CODE); + DO_TEST_FULL("resctrl-cdp", VIR_RESCTRL_TYPE_L3_DATA); + + return ret; +} + +VIR_TEST_MAIN_PRELOAD(mymain, abs_builddir "/.libs/virnumamock.so") -- 1.9.1

3 3

[libvirt] Making panic great again
by Ed Swierk 02 May '17

02 May '17

The panic device is currently documented as a way for "libvirt to receive panic notification from a QEMU guest". This is true, but not the whole story. When a guest triggers the panic device, QEMU pauses the guest, and libvirt takes the action specified by on_crash. This can interfere with the guest's own crash handling actions (e.g. writing a dump file and rebooting itself) if the guest triggers the panic device first (as Windows does). None of this is an obvious side effect of a notification mechanism, so the panic device documentation should mention it. (I'll send a documentation patch shortly.) Nor is this a desirable side effect, for guests that are configured to deal with crashes themselves. Sure, you can avoid using the panic device with such guests, but then virsh list or another application using the libvirt API to monitor domain state won't notice guest crashes. And if you still want libvirt to take action on guests that don't do it themselves, then you have to be careful to include the panic device only for those domains. Ideally libvirt would offer (1) a state indicating "this guest crashed and needs help" independently of triggering an action, and (2) a way to trigger an action only when needed to recover from the crash, excluding guests that deal with their own crashes. Sadly pvpanic and the HyperV crash MSR convey only that the guest crashed, not whether the guest is configured to take some action on its own. So there's no way to know precisely that a crashed (and not paused) guest is in need of assistance. But a state indicating "this guest crashed N minutes ago and hasn't rebooted itself" would be a useful approximation. And triggering an action N minutes after a guest crash if it hasn't rebooted itself in the meantime would make it easy to cap the downtime of crashed domains. Both could be implemented without changing either QEMU or panic device semantics. Does this seem useful to anyone else? --Ed

4 7

[libvirt] [PATCH] virsh-domain-monitor: add human readable output for 'domblkinfo'.
by Julio Faracco 02 May '17

02 May '17

The virsh command 'domblkinfo' returns the capacity, allocation and phisycal size of the devices attached in a domain. Usually, this sizes are very big and hard to understand and calculate. This commits introduce a human readable support to check the size of each field easilly. For example, the command today: virsh # domblkinfo my_domain hda Capacity: 21474836480 Allocation: 14875545600 Physical: 21474836480 After this patch: virsh # domblkinfo my_domain hda --human Capacity: 20.0G Allocation: 13.9G Physical: 20.0G This commit supports Bytes, KiB and MiB too. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1330940 Signed-off-by: Julio Faracco <jcfaracco(a)gmail.com> --- tools/virsh-domain-monitor.c | 41 ++++++++++++++++++++++++++++++++++++++--- tools/virsh.pod | 5 +++-- 2 files changed, 41 insertions(+), 5 deletions(-) diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index 3db4795..755d740 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -396,6 +396,10 @@ static const vshCmdOptDef opts_domblkinfo[] = { .flags = VSH_OFLAG_REQ, .help = N_("block device") }, + {.name = "human", + .type = VSH_OT_BOOL, + .help = N_("Human readable output") + }, {.name = NULL} }; @@ -405,6 +409,7 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) virDomainBlockInfo info; virDomainPtr dom; bool ret = false; + bool human = false; const char *device = NULL; if (!(dom = virshCommandOptDomain(ctl, cmd, NULL))) @@ -416,9 +421,39 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) if (virDomainGetBlockInfo(dom, device, &info, 0) < 0) goto cleanup; - vshPrint(ctl, "%-15s %llu\n", _("Capacity:"), info.capacity); - vshPrint(ctl, "%-15s %llu\n", _("Allocation:"), info.allocation); - vshPrint(ctl, "%-15s %llu\n", _("Physical:"), info.physical); + human = vshCommandOptBool(cmd, "human"); + + if (!human) { + vshPrint(ctl, "%-15s %llu\n", _("Capacity:"), info.capacity); + vshPrint(ctl, "%-15s %llu\n", _("Allocation:"), info.allocation); + vshPrint(ctl, "%-15s %llu\n", _("Physical:"), info.physical); + } else { + double sizeCapacity = 1; + const char *sizeStr = "B"; + + /* Check if capacity can be in K. */ + if (info.capacity >= 1024) { + sizeCapacity = 1024.0; + sizeStr = "K"; + } + /* Check if capacity can be in M. */ + if (info.capacity >= 1048576) { + sizeCapacity = 1048576.0; + sizeStr = "M"; + } + /* Check if capacity can be in G. */ + if (info.capacity >= 1073741824) { + sizeCapacity = 1073741824.0; + sizeStr = "G"; + } + + vshPrint(ctl, "%-15s %.1f%s\n", _("Capacity:"), + info.capacity/sizeCapacity, sizeStr); + vshPrint(ctl, "%-15s %.1f%s\n", _("Allocation:"), + info.allocation/sizeCapacity, sizeStr); + vshPrint(ctl, "%-15s %.1f%s\n", _("Physical:"), + info.physical/sizeCapacity, sizeStr); + } ret = true; diff --git a/tools/virsh.pod b/tools/virsh.pod index e16f62f..9f67de2 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -841,12 +841,13 @@ B<domstate> command says that a domain was paused due to I/O error. The B<domblkerror> command lists all block devices in error state and the error seen on each of them. -=item B<domblkinfo> I<domain> I<block-device> +=item B<domblkinfo> I<domain> I<block-device> [I<--human>] Get block device size info for a domain. A I<block-device> corresponds to a unique target name (<target dev='name'/>) or source file (<source file='name'/>) for one of the disk devices attached to I<domain> (see -also B<domblklist> for listing these names). +also B<domblklist> for listing these names). If I<--human> is set, the +output will have a human readable design. =item B<domblklist> I<domain> [I<--inactive>] [I<--details>] -- 2.7.4

2 1

[libvirt] [PATCH] qemu: Adding 'downscript' feature for QEMU network interfaces.
by Julio Faracco 01 May '17

01 May '17

This commit adds the support for 'downscript' feature: - For QEMU command line with the option: '-net downscript=/etc/qemu-ifdown,...'. - For Domains with a network interface description: '<interface type='ethernet'> ... <downscript path='/etc/qemu-ifdown'/> ... </interface>' The options 'script' and 'downscript' accept the argument 'no' to disable the script executions. The way that the code was implemented, the XML file accepts '<[down]script path='no'>' to solve this problem. This commit updates the tests too. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=825939 Signed-off-by: Julio Faracco <jcfaracco(a)gmail.com> --- docs/schemas/domaincommon.rng | 8 ++++++++ src/conf/domain_conf.c | 13 +++++++++++++ src/conf/domain_conf.h | 1 + src/qemu/qemu_parse_command.c | 4 ++++ tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.args | 2 +- tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.xml | 1 + tests/qemuargv2xmldata/qemuargv2xml-net-eth.args | 2 +- tests/qemuargv2xmldata/qemuargv2xml-net-eth.xml | 1 + tests/qemuxml2argvdata/qemuxml2argv-net-eth-ifname.xml | 1 + tests/qemuxml2argvdata/qemuxml2argv-net-eth.xml | 1 + tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth-ifname.xml | 1 + tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth.xml | 1 + 12 files changed, 34 insertions(+), 2 deletions(-) diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 281309e..2f88dda 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -2609,6 +2609,14 @@ </element> </optional> <optional> + <element name="downscript"> + <attribute name="path"> + <ref name="filePath"/> + </attribute> + <empty/> + </element> + </optional> + <optional> <element name="backenddomain"> <attribute name="name"> <ref name="domainName"/> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 0ff216e..32d5720 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -1935,6 +1935,7 @@ virDomainNetDefClear(virDomainNetDefPtr def) VIR_FREE(def->backend.vhost); VIR_FREE(def->virtPortProfile); VIR_FREE(def->script); + VIR_FREE(def->downscript); VIR_FREE(def->domain_name); VIR_FREE(def->ifname); VIR_FREE(def->ifname_guest); @@ -9589,6 +9590,7 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, char *ifname_guest = NULL; char *ifname_guest_actual = NULL; char *script = NULL; + char *downscript = NULL; char *address = NULL; char *port = NULL; char *localaddr = NULL; @@ -9761,6 +9763,9 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, } else if (!script && xmlStrEqual(cur->name, BAD_CAST "script")) { script = virXMLPropString(cur, "path"); + } else if (!downscript && + xmlStrEqual(cur->name, BAD_CAST "downscript")) { + downscript = virXMLPropString(cur, "path"); } else if (!domain_name && xmlStrEqual(cur->name, BAD_CAST "backenddomain")) { domain_name = virXMLPropString(cur, "name"); @@ -10074,6 +10079,10 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, def->script = script; script = NULL; } + if (downscript != NULL) { + def->downscript = downscript; + downscript = NULL; + } if (domain_name != NULL) { def->domain_name = domain_name; domain_name = NULL; @@ -10356,6 +10365,7 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, VIR_FREE(dev); virDomainActualNetDefFree(actual); VIR_FREE(script); + VIR_FREE(downscript); VIR_FREE(bridge); VIR_FREE(model); VIR_FREE(backend); @@ -22158,6 +22168,9 @@ virDomainNetDefFormat(virBufferPtr buf, virBufferEscapeString(buf, "<script path='%s'/>\n", def->script); + if (def->downscript) + virBufferEscapeString(buf, "<downscript path='%s'/>\n", + def->downscript); virBufferEscapeString(buf, "<backenddomain name='%s'/>\n", def->domain_name); if (def->ifname && diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 09fb7aa..9deca76 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1024,6 +1024,7 @@ struct _virDomainNetDef { unsigned long sndbuf; } tune; char *script; + char *downscript; char *domain_name; /* backend domain name */ char *ifname; /* interface name on the host (<target dev='x'/>) */ virNetDevIPInfo hostIP; diff --git a/src/qemu/qemu_parse_command.c b/src/qemu/qemu_parse_command.c index af9063c..d773917 100644 --- a/src/qemu/qemu_parse_command.c +++ b/src/qemu/qemu_parse_command.c @@ -1060,6 +1060,10 @@ qemuParseCommandLineNet(virDomainXMLOptionPtr xmlopt, def->script = values[i]; values[i] = NULL; } else if (def->type == VIR_DOMAIN_NET_TYPE_ETHERNET && + STREQ(keywords[i], "downscript") && STRNEQ(values[i], "")) { + def->downscript = values[i]; + values[i] = NULL; + } else if (def->type == VIR_DOMAIN_NET_TYPE_ETHERNET && STREQ(keywords[i], "ifname")) { def->ifname = values[i]; values[i] = NULL; diff --git a/tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.args b/tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.args index 4d74ae4..5d7129c 100644 --- a/tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.args +++ b/tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.args @@ -18,6 +18,6 @@ QEMU_AUDIO_DRV=none \ -usb \ -drive file=/dev/HostVG/QEMUGuest1,format=raw,if=ide,bus=0,unit=0 \ -net nic,macaddr=00:11:22:33:44:55,vlan=0,model=rtl8139,name=net0 \ --net tap,ifname=nic02,script=/etc/qemu-ifup,vlan=0,name=hostnet0 \ +-net tap,ifname=nic02,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown,vlan=0,name=hostnet0 \ -serial none \ -parallel none diff --git a/tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.xml b/tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.xml index fa9a892..8e04efb 100644 --- a/tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.xml +++ b/tests/qemuargv2xmldata/qemuargv2xml-net-eth-ifname.xml @@ -30,6 +30,7 @@ <interface type='ethernet'> <mac address='00:11:22:33:44:55'/> <script path='/etc/qemu-ifup'/> + <downscript path='/etc/qemu-ifdown'/> <target dev='nic02'/> <model type='rtl8139'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> diff --git a/tests/qemuargv2xmldata/qemuargv2xml-net-eth.args b/tests/qemuargv2xmldata/qemuargv2xml-net-eth.args index 89eb4c1..0e3fa21 100644 --- a/tests/qemuargv2xmldata/qemuargv2xml-net-eth.args +++ b/tests/qemuargv2xmldata/qemuargv2xml-net-eth.args @@ -18,6 +18,6 @@ QEMU_AUDIO_DRV=none \ -usb \ -drive file=/dev/HostVG/QEMUGuest1,format=raw,if=ide,bus=0,unit=0 \ -net nic,macaddr=00:11:22:33:44:55,vlan=0,model=rtl8139,name=net0 \ --net tap,script=/etc/qemu-ifup,vlan=0,name=hostnet0 \ +-net tap,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown,vlan=0,name=hostnet0 \ -serial none \ -parallel none diff --git a/tests/qemuargv2xmldata/qemuargv2xml-net-eth.xml b/tests/qemuargv2xmldata/qemuargv2xml-net-eth.xml index 57c4be8..d177ca8 100644 --- a/tests/qemuargv2xmldata/qemuargv2xml-net-eth.xml +++ b/tests/qemuargv2xmldata/qemuargv2xml-net-eth.xml @@ -30,6 +30,7 @@ <interface type='ethernet'> <mac address='00:11:22:33:44:55'/> <script path='/etc/qemu-ifup'/> + <downscript path='/etc/qemu-ifdown'/> <model type='rtl8139'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> diff --git a/tests/qemuxml2argvdata/qemuxml2argv-net-eth-ifname.xml b/tests/qemuxml2argvdata/qemuxml2argv-net-eth-ifname.xml index dd0d752..21d8259 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-net-eth-ifname.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-net-eth-ifname.xml @@ -26,6 +26,7 @@ <interface type='ethernet'> <mac address='00:11:22:33:44:55'/> <script path='/etc/qemu-ifup'/> + <downscript path='/etc/qemu-ifdown'/> <target dev='nic02'/> <model type='rtl8139'/> </interface> diff --git a/tests/qemuxml2argvdata/qemuxml2argv-net-eth.xml b/tests/qemuxml2argvdata/qemuxml2argv-net-eth.xml index 48acadf..9f40122 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-net-eth.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-net-eth.xml @@ -26,6 +26,7 @@ <interface type='ethernet'> <mac address='00:11:22:33:44:55'/> <script path='/etc/qemu-ifup'/> + <downscript path='/etc/qemu-ifdown'/> <model type='rtl8139'/> </interface> <input type='mouse' bus='ps2'/> diff --git a/tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth-ifname.xml b/tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth-ifname.xml index c36baa0..b71fd5a 100644 --- a/tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth-ifname.xml +++ b/tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth-ifname.xml @@ -30,6 +30,7 @@ <interface type='ethernet'> <mac address='00:11:22:33:44:55'/> <script path='/etc/qemu-ifup'/> + <downscript path='/etc/qemu-ifdown'/> <target dev='nic02'/> <model type='rtl8139'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> diff --git a/tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth.xml b/tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth.xml index 898bda6..c55cd68 100644 --- a/tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth.xml +++ b/tests/qemuxml2xmloutdata/qemuxml2xmlout-net-eth.xml @@ -30,6 +30,7 @@ <interface type='ethernet'> <mac address='00:11:22:33:44:55'/> <script path='/etc/qemu-ifup'/> + <downscript path='/etc/qemu-ifdown'/> <model type='rtl8139'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> -- 2.7.4

1 1

[libvirt] [PATCHv3 0/6] Support more options for intel-iommu
by Ján Tomko 29 Apr '17

29 Apr '17

[Adding Peter Xu to cc to see if he has any opinion on the docs] https://bugzilla.redhat.com/show_bug.cgi?id=1427005 v1: https://www.redhat.com/archives/libvir-list/2017-March/msg01072.html v2: https://www.redhat.com/archives/libvir-list/2017-April/msg00932.html new in v3: * add more devices to qemuxml2argv-intel-iommu-caching.xml to check the ordering (intel-iommu should come before other devices) * fix the capability clustering Ján Tomko (6): conf: add <irqchip mode> to <features> qemu: format kernel_irqchip on the command line conf: add <driver intremap> to <iommu> qemu: format intremap= on intel-iommu command line conf: add caching attribute to iommu device qemu: format caching-mode on iommu command line docs/formatdomain.html.in | 41 +++++++++- docs/schemas/domaincommon.rng | 30 ++++++++ src/conf/domain_conf.c | 90 ++++++++++++++++++++-- src/conf/domain_conf.h | 14 ++++ src/libvirt_private.syms | 1 + src/qemu/qemu_capabilities.c | 13 ++++ src/qemu/qemu_capabilities.h | 5 ++ src/qemu/qemu_command.c | 40 ++++++++++ tests/qemucapabilitiesdata/caps_1.5.3.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_1.6.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_1.7.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_2.1.1.x86_64.xml | 1 + .../qemucapabilitiesdata/caps_2.4.0.x86_64.replies | 22 ++++-- tests/qemucapabilitiesdata/caps_2.4.0.x86_64.xml | 1 + .../qemucapabilitiesdata/caps_2.5.0.x86_64.replies | 24 ++++-- tests/qemucapabilitiesdata/caps_2.5.0.x86_64.xml | 1 + .../caps_2.6.0-gicv2.aarch64.xml | 1 + .../caps_2.6.0-gicv3.aarch64.xml | 1 + tests/qemucapabilitiesdata/caps_2.6.0.ppc64le.xml | 1 + .../qemucapabilitiesdata/caps_2.6.0.x86_64.replies | 24 ++++-- tests/qemucapabilitiesdata/caps_2.6.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_2.7.0.s390x.xml | 1 + .../qemucapabilitiesdata/caps_2.7.0.x86_64.replies | 28 +++++-- tests/qemucapabilitiesdata/caps_2.7.0.x86_64.xml | 2 + tests/qemucapabilitiesdata/caps_2.8.0.s390x.xml | 1 + .../qemucapabilitiesdata/caps_2.8.0.x86_64.replies | 37 +++++++-- tests/qemucapabilitiesdata/caps_2.8.0.x86_64.xml | 2 + .../qemucapabilitiesdata/caps_2.9.0.x86_64.replies | 49 +++++++++--- tests/qemucapabilitiesdata/caps_2.9.0.x86_64.xml | 3 + .../qemuxml2argv-intel-iommu-caching.args | 25 ++++++ .../qemuxml2argv-intel-iommu-caching.xml | 50 ++++++++++++ .../qemuxml2argv-intel-iommu-irqchip.args | 19 +++++ .../qemuxml2argv-intel-iommu-irqchip.xml | 31 ++++++++ tests/qemuxml2argvtest.c | 16 ++++ .../qemuxml2xmlout-intel-iommu-caching.xml | 1 + .../qemuxml2xmlout-intel-iommu-irqchip.xml | 1 + tests/qemuxml2xmltest.c | 2 + 37 files changed, 533 insertions(+), 49 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-intel-iommu-caching.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-intel-iommu-caching.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-intel-iommu-irqchip.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-intel-iommu-irqchip.xml create mode 120000 tests/qemuxml2xmloutdata/qemuxml2xmlout-intel-iommu-caching.xml create mode 120000 tests/qemuxml2xmloutdata/qemuxml2xmlout-intel-iommu-irqchip.xml -- 2.10.2

4 14

[libvirt] [PATCH] docs: Add details about panic device behavior
by Ed Swierk 29 Apr '17

29 Apr '17

When a guest triggers the panic device, QEMU pauses the guest, and libvirt takes the action specified by on_crash. This can interfere with the guest's own crash handling actions (like writing a dump file and rebooting itself) if the guest triggers the panic device first (as Windows does via the Hyper-V crash CPU feature). This is not an obvious side effect of a notification mechanism, so it's worth explaining. Signed-off-by: Ed Swierk <eswierk(a)skyportsystems.com> --- docs/formatdomain.html.in | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index e31a271..1c5fdb9 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -7078,7 +7078,11 @@ qemu-kvm -net nic,model=? /dev/null <h4><a name="elementsPanic">panic device</a></h4> <p> panic device enables libvirt to receive panic notification from a QEMU - guest. + guest. When the guest triggers the panic device, QEMU pauses the guest, and + libvirt takes the action specified by the <code>on_crash</code> element. + Note that this may interfere with the guest's own crash handling actions + (like writing a dump file and rebooting itself) if the guest triggers the + panic device first (as Windows does via the Hyper-V crash CPU feature). <span class="since">Since 1.2.1, QEMU and KVM only</span> </p> <p> -- 1.9.1

1 0

[libvirt] [PATCH] news: Document libxl nested HVM support
by Jim Fehlig 28 Apr '17

28 Apr '17

Nested HVM support in the libxl driver is a news-worthy improvement for libvirt 3.3.0. Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- docs/news.xml | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/docs/news.xml b/docs/news.xml index 1f558451f..5dc8403b4 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -48,6 +48,17 @@ ease handling such breaks if they happen in the future. </description> </change> + <change> + <summary> + Add support for nested HVM domains in the libxl driver. + </summary> + <description> + Xen has supported nested HVM domains since version 4.4. The + libvirt libxl driver now supports nested HVM domains by + specifying the host-passthrough CPU mode when defining a + domain. + </description> + </change> </section> <section title="Bug fixes"> </section> -- 2.11.0

3 4

[libvirt] [PATCH] conf: format only relevant attributes for graphics based on listen type
by Pavel Hrdina 28 Apr '17

28 Apr '17

This patch changes following output: ... <graphics type='vnc' port='-1' autoport='yes'> <listen type='none'/> </graphics> ... into this output: ... <graphics type='vnc'> <listen type='none'/> </graphics> ... Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- src/conf/domain_conf.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index fe9b7c7273..88bcd6c052 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -23297,7 +23297,8 @@ virDomainGraphicsDefFormat(virBufferPtr buf, return -1; } - if (glisten->type == VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_SOCKET) { + switch (glisten->type) { + case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_SOCKET: /* To not break migration we shouldn't print the 'socket' attribute * if it's auto-generated or if it's based on config option from * qemu.conf. If the socket is provided by user we need to print it @@ -23307,7 +23308,10 @@ virDomainGraphicsDefFormat(virBufferPtr buf, (flags & VIR_DOMAIN_DEF_FORMAT_MIGRATABLE))) { virBufferEscapeString(buf, " socket='%s'", glisten->socket); } - } else { + break; + + case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_ADDRESS: + case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_NETWORK: if (def->data.vnc.port && (!def->data.vnc.autoport || !(flags & VIR_DOMAIN_DEF_FORMAT_INACTIVE))) virBufferAsprintf(buf, " port='%d'", @@ -23325,6 +23329,10 @@ virDomainGraphicsDefFormat(virBufferPtr buf, virBufferAsprintf(buf, " websocket='%d'", def->data.vnc.websocket); virDomainGraphicsListenDefFormatAddr(buf, glisten, flags); + break; + case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_NONE: + case VIR_DOMAIN_GRAPHICS_LISTEN_TYPE_LAST: + break; } if (def->data.vnc.keymap) -- 2.12.2

2 1

[libvirt] [PATCH v2 0/3] conf: don't ignore <target dev='blah'/> for macvtap interfaces
by Laine Stump 28 Apr '17

28 Apr '17

The parser had been clearing out *all* suggested device names for type='direct' (aka macvtap) interfaces. All of the code implementing macvtap allows for a user-specified device name, so we should allow it. In the case that an interface name starts with "macvtap" or "macvlan" though, we do still clear it out, just as we do with "vnet" (which is the prefix used for automatically generated tap device names), since those are the prefixes for the names we autogenerate for macvtap and macvlan devices. Resolves: https://bugzilla.redhat.com/1335798 Changes since V1: * make a single #define for each of MACVLAN_PREFIX and MACVTAP_PREFIX in virnetdevmacvlan.h rather than duplicating #defines from virnetdevmacvlan.c into domain_conf.h (as suggested by Martin * move/rename the auto-generated prefix for regular tap devices to be consistent with the macvtap/macvlan prefixes. Laine Stump (3): util: make macvtap/macvlan generated name #defines available to other files conf: don't ignore <target dev='blah'/> for macvtap interfaces util: rename/move VIR_NET_GENERATED_PREFIX to be consistent docs/formatdomain.html.in | 6 ++-- src/bhyve/bhyve_command.c | 4 +-- src/conf/domain_conf.c | 11 ++++-- src/conf/domain_conf.h | 4 --- src/interface/interface_backend_udev.c | 2 +- src/qemu/qemu_interface.c | 8 ++--- src/uml/uml_conf.c | 4 +-- src/util/virnetdev.h | 5 +++ src/util/virnetdevmacvlan.c | 65 ++++++++++++++-------------------- src/util/virnetdevmacvlan.h | 6 ++++ 10 files changed, 57 insertions(+), 58 deletions(-) -- 2.9.3

2 5

[libvirt] [PATCH v2 0/5] add support for qemu-xhci USB controller
by Pavel Hrdina 28 Apr '17

28 Apr '17

Pavel Hrdina (5): qemu: change the logic of setting default USB controller qemu: use nec-usb-xhci as a default controller for aarch64 if available qemu: introduce QEMU_CAPS_DEVICE_QEMU_XHCI qemu: add support for qemu-xhci USB controller qemu: use qemu-xhci USB controller by default for ppc64 and aarch64 docs/formatdomain.html.in | 4 +- docs/schemas/domaincommon.rng | 1 + src/conf/domain_addr.c | 1 + src/conf/domain_conf.c | 1 + src/conf/domain_conf.h | 1 + src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 8 +++- src/qemu/qemu_domain.c | 47 ++++++++++++++-------- src/qemu/qemu_domain_address.c | 1 + tests/qemucapabilitiesdata/caps_2.9.0.x86_64.xml | 1 + ...muxml2argv-aarch64-usb-controller-nec-xhci.args | 19 +++++++++ ...emuxml2argv-aarch64-usb-controller-nec-xhci.xml | 16 ++++++++ ...uxml2argv-aarch64-usb-controller-qemu-xhci.args | 19 +++++++++ ...muxml2argv-aarch64-usb-controller-qemu-xhci.xml | 16 ++++++++ ...emuxml2argv-ppc64-usb-controller-qemu-xhci.args | 19 +++++++++ ...qemuxml2argv-ppc64-usb-controller-qemu-xhci.xml | 1 + ...qemuxml2argv-usb-controller-qemu-xhci-limit.xml | 14 +++++++ ...l2argv-usb-controller-qemu-xhci-unavailable.xml | 1 + .../qemuxml2argv-usb-controller-qemu-xhci.args | 19 +++++++++ .../qemuxml2argv-usb-controller-qemu-xhci.xml | 14 +++++++ tests/qemuxml2argvtest.c | 15 +++++++ 22 files changed, 201 insertions(+), 20 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-usb-controller-nec-xhci.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-usb-controller-nec-xhci.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-usb-controller-qemu-xhci.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-aarch64-usb-controller-qemu-xhci.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-ppc64-usb-controller-qemu-xhci.args create mode 120000 tests/qemuxml2argvdata/qemuxml2argv-ppc64-usb-controller-qemu-xhci.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-controller-qemu-xhci-limit.xml create mode 120000 tests/qemuxml2argvdata/qemuxml2argv-usb-controller-qemu-xhci-unavailable.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-controller-qemu-xhci.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-controller-qemu-xhci.xml -- 2.12.2

2 17