- Devel - libvirt List Archives

[libvirt] [PATCH] Allow for CPU topology specification without model
by Jiri Denemark 19 Jan '10

19 Jan '10

Currently CPU topology may only be specified together with CPU model: <cpu match='exact'> <model>name</model> <topology sockets='1' cores='2' threads='3'/> </cpu> This patch allows for CPU topology specification without the need for also specifying CPU model: <cpu> <topology sockets='1' cores='2' threads='3'/> </cpu> 'match' attribute and 'model' element are made optional with the restriction that 'match' attribute has to be set when 'model' is present. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- docs/schemas/domain.rng | 24 ++++++++++++-------- src/conf/cpu_conf.c | 55 ++++++++++++++++++++++++++++++++-------------- 2 files changed, 52 insertions(+), 27 deletions(-) diff --git a/docs/schemas/domain.rng b/docs/schemas/domain.rng index 566b117..137bbc6 100644 --- a/docs/schemas/domain.rng +++ b/docs/schemas/domain.rng @@ -1242,17 +1242,21 @@ --> <define name="cpu"> <element name="cpu"> - <attribute name="match"> - <choice> - <value>minimum</value> - <value>exact</value> - <value>strict</value> - </choice> - </attribute> + <optional> + <attribute name="match"> + <choice> + <value>minimum</value> + <value>exact</value> + <value>strict</value> + </choice> + </attribute> + </optional> <interleave> - <element name="model"> - <text/> - </element> + <optional> + <element name="model"> + <text/> + </element> + </optional> <optional> <element name="topology"> <attribute name="sockets"> diff --git a/src/conf/cpu_conf.c b/src/conf/cpu_conf.c index bbe2cc2..e7924c1 100644 --- a/src/conf/cpu_conf.c +++ b/src/conf/cpu_conf.c @@ -1,7 +1,7 @@ /* * cpu_conf.h: CPU XML handling * - * Copyright (C) 2009 Red Hat, Inc. + * Copyright (C) 2009, 2010 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -76,7 +76,6 @@ virCPUDefParseXML(virConnectPtr conn, { virCPUDefPtr def; xmlNodePtr *nodes = NULL; - char *match; int n; unsigned int i; @@ -85,18 +84,33 @@ virCPUDefParseXML(virConnectPtr conn, return NULL; } - match = virXMLPropString(node, "match"); - - if (mode == VIR_CPU_TYPE_AUTO) - def->type = (match == NULL) ? VIR_CPU_TYPE_HOST : VIR_CPU_TYPE_GUEST; - else + if (mode == VIR_CPU_TYPE_AUTO) { + if (virXPathBoolean(conn, "boolean(./arch)", ctxt)) + def->type = VIR_CPU_TYPE_HOST; + else + def->type = VIR_CPU_TYPE_GUEST; + } else def->type = mode; if (def->type == VIR_CPU_TYPE_GUEST) { - if ((def->match = virCPUMatchTypeFromString(match)) < 0) { - virCPUReportError(conn, VIR_ERR_INTERNAL_ERROR, - "%s", _("Invalid match attribute for CPU specification")); - goto error; + char *match = virXMLPropString(node, "match"); + + if (!match) { + if (virXPathBoolean(conn, "boolean(./model)", ctxt)) { + virCPUReportError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("Missing match attribute for CPU specification")); + goto error; + } + def->match = -1; + } else { + def->match = virCPUMatchTypeFromString(match); + VIR_FREE(match); + + if (def->match < 0) { + virCPUReportError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("Invalid match attribute for CPU specification")); + goto error; + } } } @@ -109,7 +123,8 @@ virCPUDefParseXML(virConnectPtr conn, } } - if (!(def->model = virXPathString(conn, "string(./model[1])", ctxt))) { + if (!(def->model = virXPathString(conn, "string(./model[1])", ctxt)) && + def->type == VIR_CPU_TYPE_HOST) { virCPUReportError(conn, VIR_ERR_INTERNAL_ERROR, "%s", _("Missing CPU model name")); goto error; @@ -158,6 +173,12 @@ virCPUDefParseXML(virConnectPtr conn, goto error; if (n > 0) { + if (!def->model) { + virCPUReportError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("Non-empty feature list specified without CPU model")); + goto error; + } + if (VIR_ALLOC_N(def->features, n) < 0) goto no_memory; def->nfeatures = n; @@ -206,7 +227,6 @@ virCPUDefParseXML(virConnectPtr conn, } cleanup: - VIR_FREE(match); VIR_FREE(nodes); return def; @@ -263,14 +283,14 @@ virCPUDefFormatBuf(virConnectPtr conn, if (indent == NULL) indent = ""; - if (!def->model) { + if (!def->model && def->nfeatures) { virCPUReportError(conn, VIR_ERR_INTERNAL_ERROR, - "%s", _("Missing CPU model")); + "%s", _("Non-empty feature list specified without CPU model")); return -1; } if (!(flags & VIR_CPU_FORMAT_EMBEDED)) { - if (def->type == VIR_CPU_TYPE_GUEST) { + if (def->type == VIR_CPU_TYPE_GUEST && def->model) { const char *match; if (!(match = virCPUMatchTypeToString(def->match))) { virCPUReportError(conn, VIR_ERR_INTERNAL_ERROR, @@ -287,7 +307,8 @@ virCPUDefFormatBuf(virConnectPtr conn, virBufferVSprintf(buf, "%s <arch>%s</arch>\n", indent, def->arch); } - virBufferVSprintf(buf, "%s <model>%s</model>\n", indent, def->model); + if (def->model) + virBufferVSprintf(buf, "%s <model>%s</model>\n", indent, def->model); if (def->sockets && def->cores && def->threads) { virBufferVSprintf(buf, "%s <topology", indent); -- 1.6.6

3 3

[libvirt] [PATCH v2] Implement CPU topology support for QEMU driver
by Jiri Denemark 19 Jan '10

19 Jan '10

QEMU's command line equivalent for the following domain XML fragment <vcpus>2</vcpus> <cpu ...> ... <topology sockets='1' cores='2', threads='1'/> </cpu> is -smp 2,sockets=1,cores=2,threads=1 This syntax was introduced in QEMU-0.12. Version 2 changes: - -smp argument build split into a separate function - always add ",sockets=S,cores=C,threads=T" to -smp if qemu supports it - use qemuParseCommandLineKeywords for command line parsing Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/qemu/qemu_conf.c | 159 +++++++++++++++++++++++++++++++++++++++++++++----- src/qemu/qemu_conf.h | 3 +- 2 files changed, 145 insertions(+), 17 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 9f35217..09bbf7a 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -1,7 +1,7 @@ /* * qemu_conf.c: QEMU configuration management * - * Copyright (C) 2006, 2007, 2008, 2009 Red Hat, Inc. + * Copyright (C) 2006, 2007, 2008, 2009, 2010 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -1115,6 +1115,10 @@ static unsigned int qemudComputeCmdFlags(const char *help, flags |= QEMUD_CMD_FLAG_CHARDEV; if (strstr(help, "-balloon")) flags |= QEMUD_CMD_FLAG_BALLOON; + if (strstr(help, "cores=") && + strstr(help, "threads=") && + strstr(help, "sockets=")) + flags |= QEMUD_CMD_FLAG_SMP_TOPOLOGY; if (version >= 9000) flags |= QEMUD_CMD_FLAG_VNC_COLON; @@ -1865,6 +1869,39 @@ no_memory: goto cleanup; } +static char * +qemudBuildCommandLineSmp(virConnectPtr conn, + const virDomainDefPtr def, + int qemuCmdFlags) +{ + virBuffer buf = VIR_BUFFER_INITIALIZER; + + virBufferVSprintf(&buf, "%lu", def->vcpus); + + if ((qemuCmdFlags & QEMUD_CMD_FLAG_SMP_TOPOLOGY)) { + /* sockets, cores, and threads are either all zero + * or all non-zero, thus checking one of them is enough */ + if (def->cpu && def->cpu->sockets) { + virBufferVSprintf(&buf, ",sockets=%u", def->cpu->sockets); + virBufferVSprintf(&buf, ",cores=%u", def->cpu->cores); + virBufferVSprintf(&buf, ",threads=%u", def->cpu->threads); + } + else { + virBufferVSprintf(&buf, ",sockets=%lu", def->vcpus); + virBufferVSprintf(&buf, ",cores=%u", 1); + virBufferVSprintf(&buf, ",threads=%u", 1); + } + } + + if (virBufferError(&buf)) { + virBufferFreeAndReset(&buf); + virReportOOMError(conn); + return NULL; + } + + return virBufferContentAndReset(&buf); +} + #define QEMU_SERIAL_PARAM_ACCEPTED_CHARS \ "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_" @@ -1900,7 +1937,6 @@ int qemudBuildCommandLine(virConnectPtr conn, const char *migrateFrom) { int i; char memory[50]; - char vcpus[50]; char boot[VIR_DOMAIN_BOOT_LAST]; struct utsname ut; int disableKQEMU = 0; @@ -1914,6 +1950,7 @@ int qemudBuildCommandLine(virConnectPtr conn, char uuid[VIR_UUID_STRING_BUFLEN]; char domid[50]; char *cpu; + char *smp; uname_normalize(&ut); @@ -2049,7 +2086,6 @@ int qemudBuildCommandLine(virConnectPtr conn, * is not supported, then they're out of luck anyway */ snprintf(memory, sizeof(memory), "%lu", def->maxmem/1024); - snprintf(vcpus, sizeof(vcpus), "%lu", def->vcpus); snprintf(domid, sizeof(domid), "%d", def->id); ADD_ENV_LIT("LC_ALL=C"); @@ -2112,8 +2148,13 @@ int qemudBuildCommandLine(virConnectPtr conn, ADD_ARG_LIT("-mem-path"); ADD_ARG_LIT(driver->hugepage_path); } + + if (!(smp = qemudBuildCommandLineSmp(conn, def, qemuCmdFlags))) + goto error; + ADD_ARG_LIT("-smp"); - ADD_ARG_LIT(vcpus); + ADD_ARG_LIT(smp); + VIR_FREE(smp); if (qemuCmdFlags & QEMUD_CMD_FLAG_NAME) { ADD_ARG_LIT("-name"); @@ -3740,6 +3781,27 @@ error: } +static virCPUDefPtr +qemuInitGuestCPU(virConnectPtr conn, + virDomainDefPtr dom) +{ + if (!dom->cpu) { + virCPUDefPtr cpu; + + if (VIR_ALLOC(cpu) < 0) { + virReportOOMError(conn); + return NULL; + } + + cpu->type = VIR_CPU_TYPE_GUEST; + cpu->match = VIR_CPU_MATCH_EXACT; + dom->cpu = cpu; + } + + return dom->cpu; +} + + static int qemuParseCommandLineCPU(virConnectPtr conn, virDomainDefPtr dom, @@ -3749,10 +3811,8 @@ qemuParseCommandLineCPU(virConnectPtr conn, const char *p = val; const char *next; - if (VIR_ALLOC(cpu) < 0) - goto no_memory; - - cpu->type = VIR_CPU_TYPE_GUEST; + if (!(cpu = qemuInitGuestCPU(conn, dom))) + goto error; do { if (*p == '\0' || *p == ',') @@ -3796,7 +3856,6 @@ qemuParseCommandLineCPU(virConnectPtr conn, } } while ((p = next)); - dom->cpu = cpu; return 0; syntax: @@ -3807,11 +3866,84 @@ syntax: no_memory: virReportOOMError(conn); error: - virCPUDefFree(cpu); return -1; } +static int +qemuParseCommandLineSmp(virConnectPtr conn, + virDomainDefPtr dom, + const char *val) +{ + unsigned int sockets = 0; + unsigned int cores = 0; + unsigned int threads = 0; + int i; + int nkws; + char **kws; + char **vals; + int n; + char *end; + int ret; + + nkws = qemuParseCommandLineKeywords(conn, val, &kws, &vals, 1); + if (nkws < 0) + return -1; + + for (i = 0; i < nkws; i++) { + if (vals[i] == NULL) { + if (i > 0 || + virStrToLong_i(kws[i], &end, 10, &n) < 0 || + !end || *end != '\0') + goto syntax; + dom->vcpus = n; + } else { + if (virStrToLong_i(vals[i], &end, 10, &n) < 0 || + !end || *end != '\0') + goto syntax; + if (STREQ(kws[i], "sockets")) + sockets = n; + else if (STREQ(kws[i], "cores")) + cores = n; + else if (STREQ(kws[i], "threads")) + threads = n; + else + goto syntax; + } + } + + if (sockets && cores && threads) { + virCPUDefPtr cpu; + + if (!(cpu = qemuInitGuestCPU(conn, dom))) + goto error; + cpu->sockets = sockets; + cpu->cores = cores; + cpu->threads = threads; + } else if (sockets || cores || threads) + goto syntax; + + ret = 0; + +cleanup: + for (i = 0; i < nkws; i++) { + VIR_FREE(kws[i]); + VIR_FREE(vals[i]); + } + VIR_FREE(kws); + VIR_FREE(vals); + + return ret; + +syntax: + qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, + _("cannot parse CPU topology '%s'"), val); +error: + ret = -1; + goto cleanup; +} + + /* * Analyse the env and argv settings and reconstruct a * virDomainDefPtr representing these settings as closely @@ -3959,14 +4091,9 @@ virDomainDefPtr qemuParseCommandLine(virConnectPtr conn, } def->memory = def->maxmem = mem * 1024; } else if (STREQ(arg, "-smp")) { - int vcpus; WANT_VALUE(); - if (virStrToLong_i(val, NULL, 10, &vcpus) < 0) { - qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, \ - _("cannot parse CPU count '%s'"), val); + if (qemuParseCommandLineSmp(conn, def, val) < 0) goto error; - } - def->vcpus = vcpus; } else if (STREQ(arg, "-uuid")) { WANT_VALUE(); if (virUUIDParse(val, def->uuid) < 0) { diff --git a/src/qemu/qemu_conf.h b/src/qemu/qemu_conf.h index 82254ca..3f74cc9 100644 --- a/src/qemu/qemu_conf.h +++ b/src/qemu/qemu_conf.h @@ -1,7 +1,7 @@ /* * qemu_conf.h: QEMU configuration management * - * Copyright (C) 2006, 2007, 2009 Red Hat, Inc. + * Copyright (C) 2006, 2007, 2009, 2010 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -78,6 +78,7 @@ enum qemud_cmd_flags { QEMUD_CMD_FLAG_ENABLE_KVM = (1 << 23), /* Is the -enable-kvm flag available to "enable KVM full virtualization support" */ QEMUD_CMD_FLAG_MONITOR_JSON = (1 << 24), /* JSON mode for monitor */ QEMUD_CMD_FLAG_BALLOON = (1 << 25), /* -balloon available */ + QEMUD_CMD_FLAG_SMP_TOPOLOGY = (1 << 26), /* Is sockets=s,cores=c,threads=t available for -smp? */ }; /* Main driver state */ -- 1.6.6

3 6

[libvirt] [PATCH] Enhance qemuParseCommandLineKeywords
by Jiri Denemark 19 Jan '10

19 Jan '10

Current version expects name=value,... list and when an incorrect string such as "a,b,c=d" would be parsed as "a,b,c" keyword with "d" value without reporting any error, which is probably not the expected behavior. This patch adds an extra argument called allowEmptyValue, which if non-zero will permit keywords with no value; "a,b=c,,d=" will be parsed as follows: keyword value "a" NULL "b" "c" "" NULL "d" "" In case allowEmptyValue is zero, the string is required to contain name=value pairs only; retvalues is guaranteed to contain non-NULL pointers. Now, "a,b,c=d" will result in an error. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/qemu/qemu_conf.c | 51 ++++++++++++++++++++++++++++++------------------- 1 files changed, 31 insertions(+), 20 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index d3da776..9f35217 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -3066,47 +3066,58 @@ static const char *qemuFindEnv(const char **progenv, /* * Takes a string containing a set of key=value,key=value,key... * parameters and splits them up, returning two arrays with - * the individual keys and values + * the individual keys and values. If allowEmptyValue is nonzero, + * the "=value" part is optional and if a key with no value is found, + * NULL is be placed into corresponding place in retvalues. */ static int qemuParseCommandLineKeywords(virConnectPtr conn, const char *str, char ***retkeywords, - char ***retvalues) + char ***retvalues, + int allowEmptyValue) { int keywordCount = 0; int keywordAlloc = 0; char **keywords = NULL; char **values = NULL; const char *start = str; + const char *end; int i; *retkeywords = NULL; *retvalues = NULL; + end = start + strlen(str); while (start) { const char *separator; const char *endmark; char *keyword; - char *value; + char *value = NULL; - if (!(separator = strchr(start, '='))) { - qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("malformed keyword arguments in '%s'"), str); - goto error; + if (!(endmark = strchr(start, ','))) + endmark = end; + if (!(separator = strchr(start, '='))) + separator = end; + + if (separator >= endmark) { + if (!allowEmptyValue) { + qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, + _("malformed keyword arguments in '%s'"), str); + goto error; + } + separator = endmark; } + if (!(keyword = strndup(start, separator - start))) goto no_memory; - separator++; - endmark = strchr(separator, ','); - - value = endmark ? - strndup(separator, endmark - separator) : - strdup(separator); - if (!value) { - VIR_FREE(keyword); - goto no_memory; + if (separator < endmark) { + separator++; + if (!(value = strndup(separator, endmark - separator))) { + VIR_FREE(keyword); + goto no_memory; + } } if (keywordAlloc == keywordCount) { @@ -3123,7 +3134,7 @@ qemuParseCommandLineKeywords(virConnectPtr conn, values[keywordCount] = value; keywordCount++; - start = endmark ? endmark + 1 : NULL; + start = endmark < end ? endmark + 1 : NULL; } *retkeywords = keywords; @@ -3163,7 +3174,7 @@ qemuParseCommandLineDisk(virConnectPtr conn, if ((nkeywords = qemuParseCommandLineKeywords(conn, val, &keywords, - &values)) < 0) + &values, 0)) < 0) return NULL; if (VIR_ALLOC(def) < 0) { @@ -3347,7 +3358,7 @@ qemuParseCommandLineNet(virConnectPtr conn, if ((nkeywords = qemuParseCommandLineKeywords(conn, tmp+1, &keywords, - &values)) < 0) + &values, 0)) < 0) return NULL; } else { nkeywords = 0; @@ -3420,7 +3431,7 @@ qemuParseCommandLineNet(virConnectPtr conn, if ((nkeywords = qemuParseCommandLineKeywords(conn, nic + strlen("nic,"), &keywords, - &values)) < 0) { + &values, 0)) < 0) { virDomainNetDefFree(def); def = NULL; goto cleanup; -- 1.6.6

3 2

[libvirt] [PATCH] storage_conf: plug a leak on OOM error path
by Jim Meyering 18 Jan '10

18 Jan '10

Obvious, for a change... >From 964f24e0724eda6bf12d115318ea9576bbd91f10 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 18 Jan 2010 18:40:13 +0100 Subject: [PATCH] storage_conf: plug a leak on OOM error path * src/conf/storage_conf.c (virStoragePoolSourceListNewSource): Free just-allocated "source" upon VIR_REALLOC_N failure. --- src/conf/storage_conf.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index 0aefa06..ea49531 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -1,7 +1,7 @@ /* * storage_conf.c: config handling for storage driver * - * Copyright (C) 2006-2009 Red Hat, Inc. + * Copyright (C) 2006-2010 Red Hat, Inc. * Copyright (C) 2006-2008 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -1695,6 +1695,7 @@ virStoragePoolSourceListNewSource(virConnectPtr conn, } if (VIR_REALLOC_N(list->sources, list->nsources+1) < 0) { + VIR_FREE(source); virReportOOMError(conn); return NULL; } -- 1.6.6.638.g2bc54

2 1

[libvirt] Cannot use console with 0.7.5, error: internal error no assigned pty for device serial0
by Marc Haber 18 Jan '10

18 Jan '10

[my apologies for this re-post. I stupidly managed to hide the first instance of this mail away in an unrelated patch-thread and am thus afraid that the people who could have answered didn't see it] Hi, I have one test host running Debian unstable (kernel 2.6.32.3), and I would like to virtualize on it using KVM and virsh. Debian unstable has libvirt 0.7.5. On this host, I cannot start any KVM domain using console with these XML parts: <serial type='pty'> <target port='0'/> </serial> <console type='pty'> <target port='0'/> </console> When I try virsh start $domain, I get the error message: error: internal error no assigned pty for device serial0 When I remove the serial and console stanza, the machine starts up. This is the configuration documented everywhere. At first, I suspected a change in the configuration syntax and filed Debian bug #565145, But on the #virt IRC channel, people suggested that there was a change in libvirt parsing KVM's output, which may be a genuine bug breaking the console. Is this already a known issue? Can I do anything to help debugging? Greetings Marc -- ----------------------------------------------------------------------------- Marc Haber | "I don't trust Computers. They | Mailadresse im Header Mannheim, Germany | lose things." Winona Ryder | Fon: *49 621 72739834 Nordisch by Nature | How to make an American Quilt | Fax: *49 3221 2323190

5 10

[libvirt] Current cpu and memory usage for Host and Domains
by anuj rampal 18 Jan '10

18 Jan '10

Hi, Is there a way to get the Current CPU and Memory usage of the Host machine and the Guest Domains..?? Regards Anuj

4 5

[libvirt] virsh list after libvirt restart "forgets" existing guests (0.7.5)
by Thomas Sjolshagen 18 Jan '10

18 Jan '10

After the upgrade to 0.7.5 from the preview repository, I've noticed that a restart of libvirt does not appear to "re-acquire" any guests already running on the system. # virsh list Compiled against library: libvir 0.7.5 Using library: libvir 0.7.5 Using API: QEMU 0.7.5 Running hypervisor: QEMU 0.12.1 # ps -ef | grep qemu-kvm qemu 1445 1 0 Jan17 ? 00:03:32 /usr/bin/qemu-kvm -S -M pc-0.11 -enable-kvm -m 512 -smp 1 -name imap [SNIP] qemu 1516 1 0 Jan17 ? 00:04:11 /usr/bin/qemu-kvm -S -M pc-0.11 -cpu qemu32 -enable-kvm -m 512 -smp 1 -name mail [SNIP] qemu 21409 1 16 Jan16 ? 07:29:32 /usr/bin/qemu-kvm -S -M pc-0.11 -enable-kvm -m 1640 -smp 1 -name imap2-cluster [SNIP] # virsh list Id Name State ---------------------------------- 1 imap running 2 mail running Additionally, something doesn't appear to be getting either parsed right or set correctly when the guest is started as the /var/log/libvirt/qemu/<guestname>.log file contains: "Option 'ipv4: Use 'on' or 'off'" "Failed to parse "yes" for "dummy.ipv4"

2 1

[libvirt] [PATCH] xen_driver: don't leak a parsed-config buffer
by Jim Meyering 18 Jan '10

18 Jan '10

Another leak: >From 10a428a2ed1ddbe4da2b4fbd7690da5a0fe0420b Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 18 Jan 2010 18:02:17 +0100 Subject: [PATCH] xen_driver: don't leak a parsed-config buffer * src/xen/xen_driver.c (xenUnifiedDomainXMLFromNative): Also free "conf" before returning. --- src/xen/xen_driver.c | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/src/xen/xen_driver.c b/src/xen/xen_driver.c index 4911c9e..72f56ae 100644 --- a/src/xen/xen_driver.c +++ b/src/xen/xen_driver.c @@ -1,7 +1,7 @@ /* * xen_driver.c: Unified Xen driver. * - * Copyright (C) 2007, 2008, 2009 Red Hat, Inc. + * Copyright (C) 2007-2010 Red Hat, Inc. * * See COPYING.LIB for the License of this software * @@ -1199,6 +1199,8 @@ xenUnifiedDomainXMLFromNative(virConnectPtr conn, cleanup: virDomainDefFree(def); + if (conf) + virConfFree(conf); return ret; } -- 1.6.6.638.g2bc54

2 1

[libvirt] [PATCH] qemu_driver: don't leak a virDomainDeviceDef buffer
by Jim Meyering 18 Jan '10

18 Jan '10

At first I was going to call virDomainDeviceDefFree only "if (dev)", but saw that it handles a NULL "dev" just fine, so it's better to skip the test altogether, just as we do for many other free-like functions. >From ea8511d709492f5cdc152a1eaccbccd05f036648 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 18 Jan 2010 16:55:36 +0100 Subject: [PATCH] qemu_driver: don't leak a virDomainDeviceDef buffer * src/qemu/qemu_driver.c (qemudDomainAttachDevice): Don't leak "dev". --- src/qemu/qemu_driver.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 365921f..1aa8af6 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -6062,11 +6062,11 @@ cleanup: if (cgroup) virCgroupFree(&cgroup); - if (ret < 0 && dev != NULL) { + if (ret < 0 && dev != NULL) if (qemuDomainSetDeviceOwnership(dom->conn, driver, dev, 1) < 0) VIR_WARN0("Fail to restore disk device ownership"); - virDomainDeviceDefFree(dev); - } + virDomainDeviceDefFree(dev); + if (vm) virDomainObjUnlock(vm); qemuDriverUnlock(driver); -- 1.6.6.638.g2bc54

3 4

[libvirt] 0.7.5 errors in configure script
by Avi Weit 18 Jan '10

18 Jan '10

Hi, I am using Ubuntu 9.10 and downloaded and extracted libvirt.0.7.5. I have issued: ./configure --prefix=$HOME/usr --without-xen but I get: You must install device-mapper-devel/libdevmapper >= 1.0.0 to compile libvirt Below is a snippet of the final configure output: checking for showmount... no checking for pvcreate... no checking for vgcreate... no checking for lvcreate... no checking for pvremove... no checking for vgremove... no checking for lvremove... no checking for vgchange... no checking for vgscan... no checking for pvs... no checking for vgs... no checking for lvs... no checking for iscsiadm... no checking for DEVMAPPER... no checking libdevmapper.h usability... no checking libdevmapper.h presence... no checking for libdevmapper.h... no checking for dm_task_run in -ldevmapper... no configure: error: You must install device-mapper-devel/libdevmapper >= 1.0.0 to compile libvirt /dev/mapper exists on the system. Any help is appreciated. Thanks, - Avi

4 4

[libvirt] [PATCH 00/34] Add device addressing and disk controller support
by Daniel P. Berrange 18 Jan '10

18 Jan '10

This series is a merge of two previous series I posted http://www.redhat.com/archives/libvir-list/2009-December/msg00232.html http://www.redhat.com/archives/libvir-list/2009-December/msg00392.html It accomplishes quite a lot of things, having major impact on the QEMU driver, hopefully all in a postive way :-) In particular it does * Add standard XML syntax for addressing of PCI devices, and disk drives * Add support for disk controllers as a managed device in XML * Add support for disk controller hotplug/unplug * Convert everything over to use QEMU's -device flag where available * Add PCI addressing when using -device * Introduce a way to give every device a unique 'alias' name in the XML format I can't promise it works perfectly, but i've done quite alot of positive testing with it now & believe all the back comptability stuff is working right.

4 85

[libvirt] [PATCH] storage_backend_fs.c: do not ignore probe failure
by Jim Meyering 18 Jan '10

18 Jan '10

When virStorageBackendProbeTarget fails, it returns -1 or -2. The two uses below obviously intended to handle those cases differently, but due to a wrong comparison, they always treated a "real" (e.g., open) failure (-1) just like an ignorable "wrong file type" failure (-2). FYI, coverity reported that the "goto cleanup" statement was unreachable, which was true in each case, but hardly the real problem. >From c9ba79daf124afbab351741192bbaf7110a199dd Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 18 Jan 2010 10:34:53 +0100 Subject: [PATCH] storage_backend_fs.c: do not ignore probe failure * src/storage/storage_backend_fs.c (virStorageBackendFileSystemRefresh): Correct two error-handling bugs. The documented intent is to ignore non-regular files, yet due to a comparison "< 0" (rather than "!= 0") all errors were handled that way. (vboxStorageVolDelete): Likewise. (vboxDomainDefineXML): Likewise. --- src/storage/storage_backend_fs.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index 4fe40b3..fa72f9c 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -1,7 +1,7 @@ /* * storage_backend_fs.c: storage backend for FS and directory handling * - * Copyright (C) 2007-2009 Red Hat, Inc. + * Copyright (C) 2007-2010 Red Hat, Inc. * Copyright (C) 2007-2008 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -562,7 +562,7 @@ virStorageBackendFileSystemRefresh(virConnectPtr conn, &backingStore, &vol->allocation, &vol->capacity, - &vol->target.encryption) < 0)) { + &vol->target.encryption) != 0)) { if (ret == -1) goto cleanup; else { @@ -603,7 +603,7 @@ virStorageBackendFileSystemRefresh(virConnectPtr conn, if ((ret = virStorageBackendProbeTarget(conn, &vol->backingStore, NULL, NULL, NULL, - NULL)) < 0) { + NULL)) != 0) { if (ret == -1) goto cleanup; else { -- 1.6.6.638.g2bc54

2 3

[libvirt] [PATCH] remove useless comparisons in lxc_driver, vbox_tmpl.c
by Jim Meyering 18 Jan '10

18 Jan '10

Here are two change sets. They remove useless comparisons of the style - if (def->nets[i]->mac) { where the mac member is declared as an array. >From d22b04d2684d5a16f797d8df0c2ef036b5b1b2ce Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 18 Jan 2010 09:49:57 +0100 Subject: [PATCH 1/2] lxc_driver: remove useless comparison * src/lxc/lxc_driver.c (lxcSetupInterfaces): Remove always-true array-is-non-NULL test. git grep 'mac\[.*\];'|grep -F .h src/conf/domain_conf.h: unsigned char mac[VIR_MAC_BUFLEN]; --- src/lxc/lxc_driver.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 86606c7..94ec874 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -834,7 +834,7 @@ static int lxcSetupInterfaces(virConnectPtr conn, goto error_exit; } - if (def->nets[i]->mac) { + { char macaddr[VIR_MAC_STRING_BUFLEN]; virFormatMacAddr(def->nets[i]->mac, macaddr); if (0 != (rc = setMacAddr(containerVeth, macaddr))) { -- 1.6.6.638.g2bc54 >From c751c0cf3352b08d674bb59efd987f106fc1fffb Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 18 Jan 2010 09:58:57 +0100 Subject: [PATCH 2/2] vbox_tmpl.c: remove useless array-is-non-NULL comparisons * src/vbox/vbox_tmpl.c (vboxStorageVolDelete): Remove always-true array-is-non-NULL test. git grep 'key\[.*\];'|grep -F .h src/datatypes.h: char key[PATH_MAX]; (vboxStorageVolGetInfo): Likewise. (vboxStorageVolGetXMLDesc): Likewise. (vboxStorageVolGetPath): Likewise. (vboxDomainDefineXML): Likewise. (but now with "mac[]") --- src/vbox/vbox_tmpl.c | 16 ++-------------- 1 files changed, 2 insertions(+), 14 deletions(-) diff --git a/src/vbox/vbox_tmpl.c b/src/vbox/vbox_tmpl.c index 07696c0..e40c848 100644 --- a/src/vbox/vbox_tmpl.c +++ b/src/vbox/vbox_tmpl.c @@ -4084,9 +4084,7 @@ static virDomainPtr vboxDomainDefineXML(virConnectPtr conn, const char *xml) { } VBOX_UTF8_TO_UTF16(macaddrvbox, &MACAddress); - if (def->nets[i]->mac) { - adapter->vtbl->SetMACAddress(adapter, MACAddress); - } + adapter->vtbl->SetMACAddress(adapter, MACAddress); VBOX_UTF16_FREE(MACAddress); } } @@ -6645,9 +6643,6 @@ static int vboxStorageVolDelete(virStorageVolPtr vol, int i = 0; int j = 0; - if (!vol->key) - return ret; - vboxUtf8toIID(vol->conn, vol->key, &hddIID); if (!hddIID) return ret; @@ -6774,8 +6769,7 @@ static int vboxStorageVolGetInfo(virStorageVolPtr vol, virStorageVolInfoPtr info vboxIID *hddIID = NULL; nsresult rc; - if ( !vol->key - || !info) + if (!info) return ret; vboxUtf8toIID(vol->conn, vol->key, &hddIID); @@ -6824,9 +6818,6 @@ static char *vboxStorageVolGetXMLDesc(virStorageVolPtr vol, unsigned int flags A int defOk = 0; nsresult rc; - if (!vol->key) - return ret; - memset(&pool, 0, sizeof(pool)); memset(&def, 0, sizeof(def)); @@ -6919,9 +6910,6 @@ static char *vboxStorageVolGetPath(virStorageVolPtr vol) { vboxIID *hddIID = NULL; nsresult rc; - if (!vol->key) - return ret; - vboxUtf8toIID(vol->conn, vol->key, &hddIID); if (!hddIID) return ret; -- 1.6.6.638.g2bc54

2 1

[libvirt] [PATCH] secret_driver.c: remove dead cleanup code
by Jim Meyering 18 Jan '10

18 Jan '10

Here's another example of appropriate use of assert. The while loop removed below is currently guaranteed never to be executed, since "list" is always NULL at that point. However, you can argue that simply removing the loop is a little risky: what if someone changes things to "goto cleanup" before "list" reaches the final NULL? That's why I added the assertion: to catch the potential (albeit unlikely) future coding error. It would be a shame to include "real" run-time code to detect a situation like that that will probably never arise. And if you do, technically you'd have to document that the function would return some sort of error code if a developer changes its guts in a way that violates this invariant. That would be nonsensical. >From 6baa2b0b8f9963fa60ff0946afcefd57586e2720 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 18 Jan 2010 10:46:57 +0100 Subject: [PATCH] secret_driver.c: remove dead cleanup code * src/secret/secret_driver.c (loadSecrets): Remove dead code after "cleanup:". At that point, "list" is known to be NULL. Instead, add an assertion. Include <assert.h>. --- src/secret/secret_driver.c | 10 +++------- 1 files changed, 3 insertions(+), 7 deletions(-) diff --git a/src/secret/secret_driver.c b/src/secret/secret_driver.c index 1eef468..b4b4f8a 100644 --- a/src/secret/secret_driver.c +++ b/src/secret/secret_driver.c @@ -1,7 +1,7 @@ /* * secret_driver.c: local driver for secret manipulation API * - * Copyright (C) 2009 Red Hat, Inc. + * Copyright (C) 2009-2010 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -28,6 +28,7 @@ #include <string.h> #include <sys/stat.h> #include <unistd.h> +#include <assert.h> #include "internal.h" #include "base64.h" @@ -518,12 +519,7 @@ loadSecrets(virConnectPtr conn, virSecretDriverStatePtr driver, ret = 0; cleanup: - while (list != NULL) { - virSecretEntryPtr s; - - s = listUnlink(&list); - secretFree(s); - } + assert (list == NULL); if (dir != NULL) closedir(dir); return ret; -- 1.6.6.638.g2bc54

2 1

[libvirt] [PATCH] Use closest CPU model when decoding from CPUID
by Jiri Denemark 18 Jan '10

18 Jan '10

Current implementation of x86Decode() used for CPUID -> model+features translation does not always select the closest CPU model. When walking through all models from cpu_map.xml the function considers a new candidate as a better choice than a previously selected candidate only if the new one is a superset of the old one. In case the new candidate is closer to host CPU but lacks some feature comparing to the old candidate, the function does not choose well. This patch changes the algorithm so that the closest model is always selected. That is, the model which requires the lowest number of additional features to describe host CPU. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/cpu/cpu_x86.c | 122 ++++++++++++++++++++++++++++++++++------------------- 1 files changed, 78 insertions(+), 44 deletions(-) diff --git a/src/cpu/cpu_x86.c b/src/cpu/cpu_x86.c index 298b632..dae7c90 100644 --- a/src/cpu/cpu_x86.c +++ b/src/cpu/cpu_x86.c @@ -238,6 +238,55 @@ x86DataFromModel(const struct x86_model *model) } +static virCPUDefPtr +x86DataToCPU(const union cpuData *data, + const struct x86_model *model, + const struct x86_map *map) +{ + virCPUDefPtr cpu; + union cpuData *tmp = NULL; + struct cpuX86cpuid *cpuid; + const struct x86_feature *feature; + int i; + + if (VIR_ALLOC(cpu) < 0 || + (cpu->model = strdup(model->name)) == NULL || + (tmp = x86DataCopy(data)) == NULL) + goto no_memory; + + for (i = 0; i < model->ncpuid; i++) { + x86cpuidClearBits(x86DataCpuid(tmp, model->cpuid[i].function), + model->cpuid + i); + } + + feature = map->features; + while (feature != NULL) { + for (i = 0; i < feature->ncpuid; i++) { + if ((cpuid = x86DataCpuid(tmp, feature->cpuid[i].function)) + && x86cpuidMatchMasked(cpuid, feature->cpuid + i)) { + x86cpuidClearBits(cpuid, feature->cpuid + i); + if (virCPUDefAddFeature(NULL, cpu, feature->name, + VIR_CPU_FEATURE_REQUIRE) < 0) + goto error; + } + } + + feature = feature->next; + } + +cleanup: + x86DataFree(tmp); + return cpu; + +no_memory: + virReportOOMError(NULL); +error: + virCPUDefFree(cpu); + cpu = NULL; + goto cleanup; +} + + static void x86FeatureFree(struct x86_feature *feature) { @@ -896,10 +945,9 @@ x86Decode(virCPUDefPtr cpu, { int ret = -1; struct x86_map *map; - const struct x86_feature *feature; - const struct x86_model *model = NULL; const struct x86_model *candidate; - union cpuData *tmp = NULL; + virCPUDefPtr cpuCandidate; + virCPUDefPtr cpuModel = NULL; struct cpuX86cpuid *cpuid; int i; @@ -908,6 +956,8 @@ x86Decode(virCPUDefPtr cpu, candidate = map->models; while (candidate != NULL) { + bool allowed = (models == NULL); + for (i = 0; i < candidate->ncpuid; i++) { cpuid = x86DataCpuid(data, candidate->cpuid[i].function); if (cpuid == NULL @@ -915,65 +965,49 @@ x86Decode(virCPUDefPtr cpu, goto next; } - if (model == NULL - || x86ModelCompare(model, candidate) == SUBSET) { - bool found = false; - for (i = 0; i < nmodels; i++) { - if (STREQ(models[i], candidate->name)) { - found = true; - break; - } + for (i = 0; i < nmodels; i++) { + if (STREQ(models[i], candidate->name)) { + allowed = true; + break; } + } - if (nmodels > 0 && !found) { - VIR_DEBUG("CPU model %s not allowed by hypervisor; ignoring", - candidate->name); - } - else - model = candidate; + if (!allowed) { + VIR_DEBUG("CPU model %s not allowed by hypervisor; ignoring", + candidate->name); + goto next; } + if (!(cpuCandidate = x86DataToCPU(data, candidate, map))) + goto out; + + if (cpuModel == NULL + || cpuModel->nfeatures > cpuCandidate->nfeatures) { + virCPUDefFree(cpuModel); + cpuModel = cpuCandidate; + } else + virCPUDefFree(cpuCandidate); + next: candidate = candidate->next; } - if (model == NULL) { + if (cpuModel == NULL) { virCPUReportError(NULL, VIR_ERR_INTERNAL_ERROR, "%s", _("Cannot find suitable CPU model for given data")); goto out; } - if ((cpu->model = strdup(model->name)) == NULL - || (tmp = x86DataCopy(data)) == NULL) { - virReportOOMError(NULL); - goto out; - } - - for (i = 0; i < model->ncpuid; i++) { - x86cpuidClearBits(x86DataCpuid(tmp, model->cpuid[i].function), - model->cpuid + i); - } - - feature = map->features; - while (feature != NULL) { - for (i = 0; i < feature->ncpuid; i++) { - if ((cpuid = x86DataCpuid(tmp, feature->cpuid[i].function)) - && x86cpuidMatchMasked(cpuid, feature->cpuid + i)) { - x86cpuidClearBits(cpuid, feature->cpuid + i); - if (virCPUDefAddFeature(NULL, cpu, feature->name, - VIR_CPU_FEATURE_REQUIRE) < 0) - goto out; - } - } - - feature = feature->next; - } + cpu->model = cpuModel->model; + cpu->nfeatures = cpuModel->nfeatures; + cpu->features = cpuModel->features; + VIR_FREE(cpuModel); ret = 0; out: - x86DataFree(tmp); x86MapFree(map); + virCPUDefFree(cpuModel); return ret; } -- 1.6.6

3 2

[libvirt] [PATCH 1/2] Standardise ./configure --help options reporting.
by Diego Elio Pettenò 18 Jan '10

18 Jan '10

Always use AC_HELP_STRING for the help text for options at ./configure, so that the output is properly aligned. Use proper quadrigraphs for outputting the brackets. Always use autoconf-style [default=$foo] output to state the default, both where it was stated before and where it wasn't. This time, include Matthias Bolte notes regarding defaults, and removing PFX specification from phyp. --- configure.in | 100 +++++++++++++++++++++++++++++----------------------------- 1 files changed, 50 insertions(+), 50 deletions(-) diff --git a/configure.in b/configure.in index 012a96f..447d57e 100644 --- a/configure.in +++ b/configure.in @@ -205,33 +205,33 @@ fi dnl Allow to build without Xen, QEMU/KVM, test or remote driver AC_ARG_WITH([xen], -[ --with-xen add XEN support (on)],[],[with_xen=yes]) + AC_HELP_STRING([--with-xen], [add XEN support @<:@default=yes@:>@]),[],[with_xen=yes]) AC_ARG_WITH([xen-inotify], -[ --with-xen-inotify add XEN inotify support (on)],[],[with_xen_inotify=check]) + AC_HELP_STRING([--with-xen-inotify], [add XEN inotify support @<:@default=check@:>@]),[],[with_xen_inotify=check]) AC_ARG_WITH([qemu], -[ --with-qemu add QEMU/KVM support (on)],[],[with_qemu=yes]) + AC_HELP_STRING([--with-qemu], [add QEMU/KVM support @<:@default=yes@:>@]),[],[with_qemu=yes]) AC_ARG_WITH([uml], -[ --with-uml add UML support (on)],[],[with_uml=check]) + AC_HELP_STRING([--with-uml], [add UML support @<:@default=check@:>@]),[],[with_uml=check]) AC_ARG_WITH([openvz], -[ --with-openvz add OpenVZ support (on)],[],[with_openvz=yes]) + AC_HELP_STRING([--with-openvz], [add OpenVZ support @<:@default=yes@:>@]),[],[with_openvz=yes]) AC_ARG_WITH([libssh2], -[ --with-libssh2=[PFX] libssh2 location],[],[with_libssh2=yes]) + AC_HELP_STRING([--with-libssh2=@<:@PFX@:>@], [libssh2 location @<:@default=/usr/local/lib@:>@]),[],[with_libssh2=yes]) AC_ARG_WITH([phyp], -[ --with-phyp=[PFX] add PHYP support (on)],[],[with_phyp=check]) + AC_HELP_STRING([--with-phyp], [add PHYP support @<:@default=check@:>@]),[],[with_phyp=check]) AC_ARG_WITH([vbox], -[ --with-vbox add VirtualBox support (on)],[],[with_vbox=yes]) + AC_HELP_STRING([--with-vbox], [add VirtualBox support @<:@default=yes@:>@]),[],[with_vbox=yes]) AC_ARG_WITH([lxc], -[ --with-lxc add Linux Container support (on)],[],[with_lxc=check]) + AC_HELP_STRING([--with-lxc], [add Linux Container support @<:@default=check@:>@]),[],[with_lxc=check]) AC_ARG_WITH([one], -[ --with-one add ONE support (on)],[],[with_one=check]) + AC_HELP_STRING([--with-one], [add ONE support @<:@default=check@:>@]),[],[with_one=check]) AC_ARG_WITH([esx], -[ --with-esx add ESX support (on)],[],[with_esx=check]) + AC_HELP_STRING([--with-esx], [add ESX support @<:@default=check@:>@]),[],[with_esx=check]) AC_ARG_WITH([test], -[ --with-test add test driver support (on)],[],[with_test=yes]) + AC_HELP_STRING([--with-test], [add test driver support @<:@default=yes@:>@]),[],[with_test=yes]) AC_ARG_WITH([remote], -[ --with-remote add remote driver support (on)],[],[with_remote=yes]) + AC_HELP_STRING([--with-remote], [add remote driver support @<:@default=yes@:>@]),[],[with_remote=yes]) AC_ARG_WITH([libvirtd], -[ --with-libvirtd add libvirtd support (on)],[],[with_libvirtd=yes]) + AC_HELP_STRING([--with-libvirtd], [add libvirtd support @<:@default=yes@:>@]),[],[with_libvirtd=yes]) dnl dnl specific tests to setup DV devel environments with debug etc ... @@ -245,8 +245,8 @@ AC_SUBST([STATIC_BINARIES]) dnl --enable-debug=(yes|no) AC_ARG_ENABLE([debug], - [AC_HELP_STRING([--enable-debug=no/yes], - [enable debugging output])],[],[enable_debug=yes]) + [AC_HELP_STRING([--enable-debug=@<:@no|yes@:>@], + [enable debugging output @<:@default=yes@:>@])],[],[enable_debug=yes]) AM_CONDITIONAL([ENABLE_DEBUG], test x"$enable_debug" = x"yes") if test x"$enable_debug" = x"yes"; then AC_DEFINE([ENABLE_DEBUG], [], [whether debugging is enabled]) @@ -254,7 +254,7 @@ fi AC_MSG_CHECKING([where to write libvirtd PID file]) -AC_ARG_WITH([remote-pid-file], [AC_HELP_STRING([--with-remote-pid-file=[pidfile|none]], [PID file for libvirtd])]) +AC_ARG_WITH([remote-pid-file], [AC_HELP_STRING([--with-remote-pid-file=@<:@pidfile|none@:>@], [PID file for libvirtd])]) if test "x$with_remote_pid_file" == "x" ; then REMOTE_PID_FILE="$localstatedir/run/libvirtd.pid" elif test "x$with_remote_pid_file" == "xnone" ; then @@ -270,8 +270,8 @@ dnl init script flavor dnl AC_MSG_CHECKING([for init script flavor]) AC_ARG_WITH([init-script], - [AC_HELP_STRING([--with-init-script=[redhat|auto|none]], - [Style of init script to install (defaults to auto)])]) + [AC_HELP_STRING([--with-init-script=@<:@redhat|auto|none@:>@], + [Style of init script to install @<:@default=auto@:>@])]) if test "x$with_init_script" = "x" -o "x$with_init_script" = "xauto"; then if test -f /etc/redhat-release ; then with_init_script=redhat @@ -284,8 +284,8 @@ AC_MSG_RESULT($with_init_script) dnl RHEL-5 has a peculiar version of Xen, which requires some special casing AC_ARG_WITH([rhel5-api], - [AC_HELP_STRING([--with-rhel5-api=[ARG]], - [build for the RHEL-5 API [default=no]])]) + [AC_HELP_STRING([--with-rhel5-api=@<:@ARG@:>@], + [build for the RHEL-5 API @<:@default=no@:>@])]) if test x"$with_rhel5_api" = x"yes"; then AC_DEFINE([WITH_RHEL5_API], [1], [whether building for the RHEL-5 API]) fi @@ -509,7 +509,7 @@ LIBXML_CFLAGS="" LIBXML_LIBS="" LIBXML_FOUND="no" -AC_ARG_WITH([libxml], [ --with-libxml=[PFX] libxml2 location]) +AC_ARG_WITH([libxml], AC_HELP_STRING([--with-libxml=@<:@PFX@:>@], [libxml2 location])) if test "x$with_libxml" = "xno" ; then AC_MSG_CHECKING(for libxml2 libraries >= $LIBXML_REQUIRED) AC_MSG_ERROR([libxml2 >= $LIBXML_REQUIRED is required for libvirt]) @@ -596,7 +596,7 @@ LIBS="$old_libs" dnl Cyrus SASL AC_ARG_WITH([sasl], - [ --with-sasl use cyrus SASL for authentication], + AC_HELP_STRING([--with-sasl], [use cyrus SASL for authentication @<:@default=check@:>@]), [], [with_sasl=check]) @@ -651,7 +651,7 @@ AC_SUBST([SASL_LIBS]) dnl YAJL JSON library http://lloyd.github.com/yajl/ AC_ARG_WITH([yajl], - [ --with-yajl use YAJL for JSON parsing/formatting], + AC_HELP_STRING([--with-yajl], [use YAJL for JSON parsing/formatting @<:@default=check@:>@]), [], [with_yajl=check]) @@ -704,7 +704,7 @@ POLKIT_CFLAGS= POLKIT_LIBS= PKCHECK_PATH= AC_ARG_WITH([polkit], - [ --with-polkit use PolicyKit for UNIX socket access checks], + AC_HELP_STRING([--with-polkit], [use PolicyKit for UNIX socket access checks @<:@default=check@:>@]), [], [with_polkit=check]) @@ -762,7 +762,7 @@ AC_SUBST([POLKIT_LIBS]) dnl Avahi library AC_ARG_WITH([avahi], - [ --with-avahi use avahi to advertise remote daemon], + AC_HELP_STRING([--with-avahi], [use avahi to advertise remote daemon @<:@default=check@:>@]), [], [with_avahi=check]) @@ -789,7 +789,7 @@ AC_SUBST([AVAHI_LIBS]) dnl SELinux AC_ARG_WITH([selinux], - [ --with-selinux use SELinux to manage security], + AC_HELP_STRING([--with-selinux], [use SELinux to manage security @<:@default=check@:>@]), [], [with_selinux=check]) @@ -824,7 +824,7 @@ AC_SUBST([SELINUX_LIBS]) AC_ARG_WITH([secdriver-selinux], - [ --with-secdriver-selinux use SELinux security driver], + AC_HELP_STRING([--with-secdriver-selinux], [use SELinux security driver @<:@default=check@:>@]), [], [with_secdriver_selinux=check]) @@ -862,7 +862,7 @@ AM_CONDITIONAL([WITH_SECDRIVER_SELINUX], [test "$with_secdriver_selinux" != "no" dnl AppArmor AC_ARG_WITH([apparmor], - [ --with-apparmor use AppArmor to manage security], + AC_HELP_STRING([--with-apparmor], [use AppArmor to manage security @<:@default=check@:>@]), [], [with_apparmor=check]) @@ -901,7 +901,7 @@ AC_SUBST([APPARMOR_LIBS]) AC_ARG_WITH([secdriver-apparmor], - [ --with-secdriver-apparmor use AppArmor security driver], + AC_HELP_STRING([--with-secdriver-apparmor], [use AppArmor security driver @<:@default=check@:>@]), [], [with_secdriver_apparmor=check]) @@ -940,7 +940,7 @@ AM_CONDITIONAL([WITH_SECDRIVER_APPARMOR], [test "$with_secdriver_apparmor" != "n dnl NUMA lib AC_ARG_WITH([numactl], - [ --with-numactl use numactl for host topology info], + AC_HELP_STRING([--with-numactl], [use numactl for host topology info @<:@default=check@:>@]), [], [with_numactl=check]) @@ -1075,7 +1075,7 @@ AM_CONDITIONAL([WITH_PHYP],[test "$with_phyp" = "yes"]) dnl libcap-ng AC_ARG_WITH([capng], - [ --with-capng use libcap-ng to reduce libvirtd privileges], + AC_HELP_STRING([--with-capng], [use libcap-ng to reduce libvirtd privileges @<:@default=check@:>@]), [], [with_capng=check]) @@ -1158,7 +1158,7 @@ AC_SUBST([VIRSH_LIBS]) AC_ARG_WITH([network], -[ --with-network with virtual network driver (on)],[],[with_network=yes]) + AC_HELP_STRING([--with-network], [with virtual network driver @<:@default=yes@:>@]),[],[with_network=yes]) if test "$with_libvirtd" = "no" ; then with_network=no fi @@ -1176,7 +1176,7 @@ AM_CONDITIONAL([WITH_BRIDGE], [test "$with_bridge" = "yes"]) dnl netcf library AC_ARG_WITH([netcf], -[ --with-netcf libnetcf support to configure physical host network interfaces], + AC_HELP_STRING([--with-netcf], [libnetcf support to configure physical host network interfaces @<:@default=check@:>@]), [], [with_netcf=check]) NETCF_CFLAGS= @@ -1212,17 +1212,17 @@ AM_CONDITIONAL([WITH_SECRETS], [test "$with_secrets" = "yes"]) AC_ARG_WITH([storage-fs], -[ --with-storage-fs with FileSystem backend for the storage driver (on)],[],[with_storage_fs=check]) + AC_HELP_STRING([--with-storage-fs], [with FileSystem backend for the storage driver @<:@default=check@:>@]),[],[with_storage_fs=check]) AC_ARG_WITH([storage-lvm], -[ --with-storage-lvm with LVM backend for the storage driver (on)],[],[with_storage_lvm=check]) + AC_HELP_STRING([--with-storage-lvm], [with LVM backend for the storage driver @<:@default=check@:>@]),[],[with_storage_lvm=check]) AC_ARG_WITH([storage-iscsi], -[ --with-storage-iscsi with iSCSI backend for the storage driver (on)],[],[with_storage_iscsi=check]) + AC_HELP_STRING([--with-storage-iscsi], [with iSCSI backend for the storage driver @<:@default=check@:>@]),[],[with_storage_iscsi=check]) AC_ARG_WITH([storage-scsi], -[ --with-storage-scsi with SCSI backend for the storage driver (on)],[],[with_storage_scsi=check]) + AC_HELP_STRING([--with-storage-scsi], [with SCSI backend for the storage driver @<:@default=check@:>@]),[],[with_storage_scsi=check]) AC_ARG_WITH([storage-mpath], -[ --with-storage-mpath with mpath backend for the storage driver (on)],[],[with_storage_mpath=check]) + AC_HELP_STRING([--with-storage-mpath], [with mpath backend for the storage driver @<:@default=check@:>@]),[],[with_storage_mpath=check]) AC_ARG_WITH([storage-disk], -[ --with-storage-disk with GPartd Disk backend for the storage driver (on)],[],[with_storage_disk=check]) + AC_HELP_STRING([--with-storage-disk], [with GPartd Disk backend for the storage driver @<:@default=check@:>@]),[],[with_storage_disk=check]) with_storage_dir=yes if test "$with_libvirtd" = "no"; then @@ -1464,7 +1464,7 @@ dnl check for python dnl AC_ARG_WITH([python], -[ --with-python Build python bindings (on)],[],[with_python=yes]) + AC_HELP_STRING([--with-python], [Build python bindings @<:@default=yes@:>@]),[],[with_python=yes]) if test "$enable_shared:$with_python" = no:yes; then AC_MSG_WARN([Disabling shared libraries is incompatible with building Python extensions.]) @@ -1561,7 +1561,7 @@ AC_MSG_RESULT($RUNNING_XEND) AM_CONDITIONAL([ENABLE_XEN_TESTS], [test "$RUNNING_XEN" != "no" -a "$RUNNING_XEND" != "no"]) AC_ARG_ENABLE([test-coverage], -[ --enable-test-coverage turn on code coverage instrumentation], + AC_HELP_STRING([--enable-test-coverage], [turn on code coverage instrumentation @<:@default=no@:>@]), [case "${enableval}" in yes|no) ;; *) AC_MSG_ERROR([bad value ${enableval} for test-coverage option]) ;; @@ -1578,7 +1578,7 @@ if test "${enable_coverage}" = yes; then fi AC_ARG_ENABLE([test-oom], -[ --enable-test-oom memory allocation failure checking], + AC_HELP_STRING([--enable-test-oom], [memory allocation failure checking @<:@default=no@:>@]), [case "${enableval}" in yes|no) ;; *) AC_MSG_ERROR([bad value ${enableval} for test-oom option]) ;; @@ -1598,7 +1598,7 @@ fi AC_ARG_ENABLE([test-locking], -[ --enable-test-locking thread locking tests using CIL], + AC_HELP_STRING([--enable-test-locking], [thread locking tests using CIL @<:@default=no@:>@]), [case "${enableval}" in yes|no) ;; *) AC_MSG_ERROR([bad value ${enableval} for test-locking option]) ;; @@ -1615,7 +1615,7 @@ AM_CONDITIONAL([WITH_CIL],[test "$enable_locking" = "yes"]) dnl Enable building the proxy? AC_ARG_WITH([xen-proxy], -[ --with-xen-proxy add XEN setuid proxy support (on)],[],[with_xen_proxy=auto]) + AC_HELP_STRING([--with-xen-proxy], [add XEN setuid proxy support @<:@default=auto@:>@]),[],[with_xen_proxy=auto]) AC_MSG_CHECKING([if Xen setuid proxy is needed]) if test "$with_xen_proxy" = "auto"; then @@ -1675,7 +1675,7 @@ AM_CONDITIONAL([WITH_WIN_ICON], [test "$WINDRES" != "no"]) dnl Driver-Modules library AC_ARG_WITH([driver-modules], - [ --with-driver-modules build drivers as loadable modules], + AC_HELP_STRING([--with-driver-modules], [build drivers as loadable modules @<:@default=no@:>@]), [], [with_driver_modules=no]) @@ -1714,7 +1714,7 @@ dnl HAL library check for host device enumeration HAL_CFLAGS= HAL_LIBS= AC_ARG_WITH([hal], - [ --with-hal use HAL for host device enumeration], + AC_HELP_STRING([--with-hal], [use HAL for host device enumeration @<:@default=check@:>@]), [], [with_hal=check]) @@ -1757,7 +1757,7 @@ UDEV_LIBS= PCIACCESS_CFLAGS= PCIACCESS_LIBS= AC_ARG_WITH([udev], - [ --with-udev use libudev for host device enumeration], + AC_HELP_STRING([--with-udev], [use libudev for host device enumeration @<:@default=check@:>@]), [], [with_udev=check]) @@ -1809,11 +1809,11 @@ AM_CONDITIONAL([WITH_LINUX], [test `uname -s` = "Linux"]) AC_ARG_WITH([qemu-user], - [ --with-qemu-user username to run QEMU system instance as], + AC_HELP_STRING([--with-qemu-user], [username to run QEMU system instance as @<:@default=root@:>@]), [QEMU_USER=${withval}], [QEMU_USER=root]) AC_ARG_WITH([qemu-group], - [ --with-qemu-group groupname to run QEMU system instance as], + AC_HELP_STRING([--with-qemu-group], [groupname to run QEMU system instance as @<:@default=root@:>@]), [QEMU_GROUP=${withval}], [QEMU_GROUP=root]) AC_DEFINE_UNQUOTED([QEMU_USER], ["$QEMU_USER"], [QEMU user account]) -- 1.6.6

3 5

[libvirt] [PATCH 0/4] A bit of configure clear-up
by Diego Elio Pettenò 18 Jan '10

18 Jan '10

Since on the whole it seems like the configure could see some cleanup, as it seem to use lots of older constructs and repeated code, I tried doing some basic cleaning. My changes also require autoconf-2.63… it could be something older as well, but 2.63 is new enough I'd say. If nobody has reasons for 2.59 compatibility to stick around, it would probably improve a lot. HTH! -- Diego

4 14

[libvirt] [PATCH] qemu: Use log output for pty assignment if 'info chardev' is unavailable
by Matthias Bolte 18 Jan '10

18 Jan '10

qemudFindCharDevicePTYsMonitor reports an error if 'info chardev' didn't provide information for a requested device, even if the log output parsing had found the pty path for that device. This makes pty assignment fail for older QEMU/KVM versions. For example KVM 72 on Debian doesn't support 'info chardev', so qemuMonitorTextGetPtyPaths cannot parse any useful information and the hash for device-id-to-pty-path mapping stays empty. Make qemudFindCharDevicePTYsMonitor report an error only if the log output parsing and the 'info chardev' parsing failed to provide the pty path. --- src/qemu/qemu_driver.c | 14 +++++++++++--- 1 files changed, 11 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index d13119f..783d9fc 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1428,9 +1428,17 @@ qemudFindCharDevicePTYsMonitor(virConnectPtr conn, \ const char *path = (const char *) virHashLookup(paths, id); \ if (path == NULL) { \ - qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR,\ - _("no assigned pty for device %s"), id); \ - return -1; \ + if (chr->data.file.path == NULL) { \ + /* neither the log output nor 'info chardev' had a */ \ + /* pty path for this chardev, report an error */ \ + qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, \ + _("no assigned pty for device %s"), id); \ + return -1; \ + } else { \ + /* 'info chardev' had no pty path for this chardev, */\ + /* but the log output had, so we're fine */ \ + continue; \ + } \ } \ \ VIR_FREE(chr->data.file.path); \ -- 1.6.3.3

2 2

[libvirt] [PATCH] esx: Add VNC support
by Matthias Bolte 18 Jan '10

18 Jan '10

* src/conf/domain_conf.c: add defaults for the video device * src/esx/esx_vmx.[ch]: add VNC support to the VMX handling * tests/vmx2xmltest.c, tests/xml2vmxtest.c: add tests for the VNC support --- src/conf/domain_conf.c | 5 + src/esx/esx_vmx.c | 146 +++++++++++++++++++++++++++- src/esx/esx_vmx.h | 6 + tests/vmx2xmldata/vmx2xml-graphics-vnc.vmx | 6 + tests/vmx2xmldata/vmx2xml-graphics-vnc.xml | 17 +++ tests/vmx2xmltest.c | 4 +- tests/xml2vmxdata/xml2vmx-graphics-vnc.vmx | 11 ++ tests/xml2vmxdata/xml2vmx-graphics-vnc.xml | 11 ++ tests/xml2vmxtest.c | 2 + 9 files changed, 203 insertions(+), 5 deletions(-) create mode 100644 tests/vmx2xmldata/vmx2xml-graphics-vnc.vmx create mode 100644 tests/vmx2xmldata/vmx2xml-graphics-vnc.xml create mode 100644 tests/xml2vmxdata/xml2vmx-graphics-vnc.vmx create mode 100644 tests/xml2vmxdata/xml2vmx-graphics-vnc.xml diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 126e7e1..5c427fb 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -2490,6 +2490,8 @@ virDomainVideoDefaultRAM(virDomainDefPtr def, case VIR_DOMAIN_VIDEO_TYPE_VMVGA: if (def->virtType == VIR_DOMAIN_VIRT_VBOX) return 8 * 1024; + else if (def->virtType == VIR_DOMAIN_VIRT_VMWARE) + return 4 * 1024; else return 9 * 1024; break; @@ -2523,6 +2525,9 @@ virDomainVideoDefaultType(virDomainDefPtr def) case VIR_DOMAIN_VIRT_VBOX: return VIR_DOMAIN_VIDEO_TYPE_VBOX; + case VIR_DOMAIN_VIRT_VMWARE: + return VIR_DOMAIN_VIDEO_TYPE_VMVGA; + default: return -1; } diff --git a/src/esx/esx_vmx.c b/src/esx/esx_vmx.c index 8e016df..777a60a 100644 --- a/src/esx/esx_vmx.c +++ b/src/esx/esx_vmx.c @@ -731,7 +731,7 @@ esxVMX_ParseConfig(esxVI_Context *ctx, const char *vmx, char *guestOS = NULL; int controller; int port; - int present; + int present; // boolean char *scsi_virtualDev = NULL; int id; @@ -967,7 +967,20 @@ esxVMX_ParseConfig(esxVI_Context *ctx, const char *vmx, def->localtime*/ /* def:graphics */ - /* FIXME */ + if (VIR_ALLOC_N(def->graphics, 1) < 0) { + virReportOOMError(NULL); + goto failure; + } + + def->ngraphics = 0; + + if (esxVMX_ParseVNC(conf, &def->graphics[def->ngraphics]) < 0) { + goto failure; + } + + if (def->graphics[def->ngraphics] != NULL) { + ++def->ngraphics; + } /* def:disks: 4 * 15 scsi + 2 * 2 ide + 2 floppy = 66 */ if (VIR_ALLOC_N(def->disks, 66) < 0) { @@ -1155,6 +1168,70 @@ esxVMX_ParseConfig(esxVI_Context *ctx, const char *vmx, int +esxVMX_ParseVNC(virConfPtr conf, virDomainGraphicsDefPtr *def) +{ + int enabled = 0; // boolean + long long port = 0; + + if (def == NULL || *def != NULL) { + ESX_ERROR(VIR_ERR_INTERNAL_ERROR, "Invalid argument"); + return -1; + } + + if (esxUtil_GetConfigBoolean(conf, "RemoteDisplay.vnc.enabled", + &enabled, 0, 1) < 0) { + return -1; + } + + if (! enabled) { + return 0; + } + + if (VIR_ALLOC(*def) < 0) { + virReportOOMError(NULL); + goto failure; + } + + (*def)->type = VIR_DOMAIN_GRAPHICS_TYPE_VNC; + + if (esxUtil_GetConfigLong(conf, "RemoteDisplay.vnc.port", + &port, -1, 1) < 0 || + esxUtil_GetConfigString(conf, "RemoteDisplay.vnc.ip", + &(*def)->data.vnc.listenAddr, 1) < 0 || + esxUtil_GetConfigString(conf, "RemoteDisplay.vnc.keymap", + &(*def)->data.vnc.keymap, 1) < 0 || + esxUtil_GetConfigString(conf, "RemoteDisplay.vnc.password", + &(*def)->data.vnc.passwd, 1) < 0) { + goto failure; + } + + if (port < 0) { + VIR_WARN0("VNC is enabled but VMX entry 'RemoteDisplay.vnc.port' " + "is missing, the VNC port is unknown"); + + (*def)->data.vnc.port = 0; + (*def)->data.vnc.autoport = 1; + } else { + if (port < 5900 || port > 5964) { + VIR_WARN("VNC port %lld it out of [5900..5964] range", port); + } + + (*def)->data.vnc.port = port; + (*def)->data.vnc.autoport = 0; + } + + return 0; + + failure: + virDomainGraphicsDefFree(*def); + *def = NULL; + + return -1; +} + + + +int esxVMX_ParseSCSIController(virConfPtr conf, int controller, int *present, char **virtualDev) { @@ -2195,7 +2272,7 @@ esxVMX_FormatConfig(esxVI_Context *ctx, virDomainDefPtr def, (int)(def->memory / 1024)); } - /* vmx:numvcpus -> def:vcpus */ + /* def:vcpus -> vmx:numvcpus */ if (def->vcpus <= 0 || (def->vcpus % 2 != 0 && def->vcpus != 1)) { ESX_ERROR(VIR_ERR_INTERNAL_ERROR, "Expecting domain XML entry 'vcpu' to be an unsigned " @@ -2241,6 +2318,24 @@ esxVMX_FormatConfig(esxVI_Context *ctx, virDomainDefPtr def, virBufferAddLit(&buffer, "\"\n"); } + /* def:graphics */ + for (i = 0; i < def->ngraphics; ++i) { + switch (def->graphics[i]->type) { + case VIR_DOMAIN_GRAPHICS_TYPE_VNC: + if (esxVMX_FormatVNC(def->graphics[i], &buffer) < 0) { + goto failure; + } + + break; + + default: + ESX_ERROR(VIR_ERR_INTERNAL_ERROR, + "Unsupported graphics type '%s'", + virDomainGraphicsTypeToString(def->graphics[i]->type)); + goto failure; + } + } + /* def:disks */ int scsi_present[4] = { 0, 0, 0, 0 }; char *scsi_virtualDev[4] = { NULL, NULL, NULL, NULL }; @@ -2285,7 +2380,7 @@ esxVMX_FormatConfig(esxVI_Context *ctx, virDomainDefPtr def, default: ESX_ERROR(VIR_ERR_INTERNAL_ERROR, - "Unsuppotred disk device type '%s'", + "Unsupported disk device type '%s'", virDomainDiskDeviceTypeToString(def->disks[i]->device)); goto failure; } @@ -2341,6 +2436,49 @@ esxVMX_FormatConfig(esxVI_Context *ctx, virDomainDefPtr def, int +esxVMX_FormatVNC(virDomainGraphicsDefPtr def, virBufferPtr buffer) +{ + if (def->type != VIR_DOMAIN_GRAPHICS_TYPE_VNC) { + ESX_ERROR(VIR_ERR_INTERNAL_ERROR, "Invalid argument"); + return -1; + } + + virBufferVSprintf(buffer, "RemoteDisplay.vnc.enabled = \"true\"\n"); + + if (def->data.vnc.autoport) { + VIR_WARN0("VNC autoport is enabled, but the automatically assigned " + "VNC port cannot be read back"); + } else { + if (def->data.vnc.port < 5900 || def->data.vnc.port > 5964) { + VIR_WARN("VNC port %d it out of [5900..5964] range", + def->data.vnc.port); + } + + virBufferVSprintf(buffer, "RemoteDisplay.vnc.port = \"%d\"\n", + def->data.vnc.port); + } + + if (def->data.vnc.listenAddr != NULL) { + virBufferVSprintf(buffer, "RemoteDisplay.vnc.ip = \"%s\"\n", + def->data.vnc.listenAddr); + } + + if (def->data.vnc.keymap != NULL) { + virBufferVSprintf(buffer, "RemoteDisplay.vnc.keymap = \"%s\"\n", + def->data.vnc.keymap); + } + + if (def->data.vnc.passwd != NULL) { + virBufferVSprintf(buffer, "RemoteDisplay.vnc.password = \"%s\"\n", + def->data.vnc.passwd); + } + + return 0; +} + + + +int esxVMX_FormatHardDisk(esxVI_Context *ctx, virDomainDiskDefPtr def, virBufferPtr buffer) { diff --git a/src/esx/esx_vmx.h b/src/esx/esx_vmx.h index 325005e..6193409 100644 --- a/src/esx/esx_vmx.h +++ b/src/esx/esx_vmx.h @@ -61,6 +61,9 @@ esxVMX_ParseConfig(esxVI_Context *ctx, const char *vmx, esxVI_APIVersion apiVersion); int +esxVMX_ParseVNC(virConfPtr conf, virDomainGraphicsDefPtr *def); + +int esxVMX_ParseSCSIController(virConfPtr conf, int controller, int *present, char **virtualDev); @@ -96,6 +99,9 @@ esxVMX_FormatConfig(esxVI_Context *ctx, virDomainDefPtr def, esxVI_APIVersion apiVersion); int +esxVMX_FormatVNC(virDomainGraphicsDefPtr def, virBufferPtr buffer); + +int esxVMX_FormatHardDisk(esxVI_Context *ctx, virDomainDiskDefPtr def, virBufferPtr buffer); diff --git a/tests/vmx2xmldata/vmx2xml-graphics-vnc.vmx b/tests/vmx2xmldata/vmx2xml-graphics-vnc.vmx new file mode 100644 index 0000000..bd3e55f --- /dev/null +++ b/tests/vmx2xmldata/vmx2xml-graphics-vnc.vmx @@ -0,0 +1,6 @@ +config.version = "8" +virtualHW.version = "4" +RemoteDisplay.vnc.enabled = "true" +RemoteDisplay.vnc.port = "5903" +RemoteDisplay.vnc.keymap = "de" +RemoteDisplay.vnc.password = "password" diff --git a/tests/vmx2xmldata/vmx2xml-graphics-vnc.xml b/tests/vmx2xmldata/vmx2xml-graphics-vnc.xml new file mode 100644 index 0000000..159324d --- /dev/null +++ b/tests/vmx2xmldata/vmx2xml-graphics-vnc.xml @@ -0,0 +1,17 @@ +<domain type='vmware'> + <uuid>00000000-0000-0000-0000-000000000000</uuid> + <memory>32768</memory> + <currentMemory>32768</currentMemory> + <vcpu>1</vcpu> + <os> + <type arch='i686'>hvm</type> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <input type='mouse' bus='ps2'/> + <graphics type='vnc' port='5903' autoport='no' keymap='de' passwd='password'/> + </devices> +</domain> diff --git a/tests/vmx2xmltest.c b/tests/vmx2xmltest.c index b83f23f..8ea0b12 100644 --- a/tests/vmx2xmltest.c +++ b/tests/vmx2xmltest.c @@ -42,7 +42,7 @@ testCompareFiles(const char *vmx, const char *xml, esxVI_APIVersion apiVersion) goto failure; } - formatted = virDomainDefFormat(NULL, def, 0); + formatted = virDomainDefFormat(NULL, def, VIR_DOMAIN_XML_SECURE); if (formatted == NULL) { goto failure; @@ -120,6 +120,8 @@ mymain(int argc, char **argv) DO_TEST("minimal", "minimal", esxVI_APIVersion_25); DO_TEST("minimal-64bit", "minimal-64bit", esxVI_APIVersion_25); + DO_TEST("graphics-vnc", "graphics-vnc", esxVI_APIVersion_25); + DO_TEST("scsi-buslogic", "scsi-buslogic", esxVI_APIVersion_25); DO_TEST("scsi-writethrough", "scsi-writethrough", esxVI_APIVersion_25); diff --git a/tests/xml2vmxdata/xml2vmx-graphics-vnc.vmx b/tests/xml2vmxdata/xml2vmx-graphics-vnc.vmx new file mode 100644 index 0000000..b4966b5 --- /dev/null +++ b/tests/xml2vmxdata/xml2vmx-graphics-vnc.vmx @@ -0,0 +1,11 @@ +config.version = "8" +virtualHW.version = "4" +guestOS = "other" +uuid.bios = "56 4d 9b ef ac d9 b4 e0-c8 f0 ae a8 b9 10 35 15" +displayName = "minimal" +memsize = "4" +numvcpus = "1" +RemoteDisplay.vnc.enabled = "true" +RemoteDisplay.vnc.port = "5903" +RemoteDisplay.vnc.keymap = "de" +RemoteDisplay.vnc.password = "password" diff --git a/tests/xml2vmxdata/xml2vmx-graphics-vnc.xml b/tests/xml2vmxdata/xml2vmx-graphics-vnc.xml new file mode 100644 index 0000000..400dcdd --- /dev/null +++ b/tests/xml2vmxdata/xml2vmx-graphics-vnc.xml @@ -0,0 +1,11 @@ +<domain type='vmware'> + <name>minimal</name> + <uuid>564d9bef-acd9-b4e0-c8f0-aea8b9103515</uuid> + <memory>4096</memory> + <os> + <type>hvm</type> + </os> + <devices> + <graphics type='vnc' port='5903' autoport='no' keymap='de' passwd='password'/> + </devices> +</domain> diff --git a/tests/xml2vmxtest.c b/tests/xml2vmxtest.c index dfaca02..530dd2a 100644 --- a/tests/xml2vmxtest.c +++ b/tests/xml2vmxtest.c @@ -173,6 +173,8 @@ mymain(int argc, char **argv) DO_TEST("minimal", "minimal", esxVI_APIVersion_25); DO_TEST("minimal-64bit", "minimal-64bit", esxVI_APIVersion_25); + DO_TEST("graphics-vnc", "graphics-vnc", esxVI_APIVersion_25); + DO_TEST("scsi-buslogic", "scsi-buslogic", esxVI_APIVersion_25); DO_TEST("scsi-writethrough", "scsi-writethrough", esxVI_APIVersion_25); -- 1.6.3.3

2 2

[libvirt] [PATCH] esx: Make the domain part of the hostname optional
by Matthias Bolte 18 Jan '10

18 Jan '10

Before the driver assumed that there is always a domain part. That's not true. Now the domain part is handled as optional. --- src/esx/esx_driver.c | 17 ++++++++++------- 1 files changed, 10 insertions(+), 7 deletions(-) diff --git a/src/esx/esx_driver.c b/src/esx/esx_driver.c index 4ed4533..aeb3d07 100644 --- a/src/esx/esx_driver.c +++ b/src/esx/esx_driver.c @@ -773,14 +773,17 @@ esxGetHostname(virConnectPtr conn) } if (domainName == NULL || strlen(domainName) < 1) { - ESX_ERROR(VIR_ERR_INTERNAL_ERROR, - "Missing or empty 'domainName' property"); - goto failure; - } + complete = strdup(hostName); - if (virAsprintf(&complete, "%s.%s", hostName, domainName) < 0) { - virReportOOMError(NULL); - goto failure; + if (complete == NULL) { + virReportOOMError(NULL); + goto failure; + } + } else { + if (virAsprintf(&complete, "%s.%s", hostName, domainName) < 0) { + virReportOOMError(NULL); + goto failure; + } } cleanup: -- 1.6.3.3

2 2

[libvirt] [PATCH] esx: Add stubs for secondary driver types
by Matthias Bolte 18 Jan '10

18 Jan '10

This stops libvirt from probing for a libvirtd on the ESX server and sets the base for the implementation of the secondary drivers. --- src/Makefile.am | 5 ++ src/esx/esx_device_monitor.c | 93 +++++++++++++++++++++++++++++++ src/esx/esx_device_monitor.h | 29 ++++++++++ src/esx/esx_driver.c | 28 +++++----- src/esx/esx_driver.h | 20 +++++++- src/esx/esx_interface_driver.c | 96 ++++++++++++++++++++++++++++++++ src/esx/esx_interface_driver.h | 29 ++++++++++ src/esx/esx_network_driver.c | 101 ++++++++++++++++++++++++++++++++++ src/esx/esx_network_driver.h | 29 ++++++++++ src/esx/esx_secret_driver.c | 91 +++++++++++++++++++++++++++++++ src/esx/esx_secret_driver.h | 28 ++++++++++ src/esx/esx_storage_driver.c | 117 ++++++++++++++++++++++++++++++++++++++++ src/esx/esx_storage_driver.h | 29 ++++++++++ 13 files changed, 680 insertions(+), 15 deletions(-) create mode 100644 src/esx/esx_device_monitor.c create mode 100644 src/esx/esx_device_monitor.h create mode 100644 src/esx/esx_interface_driver.c create mode 100644 src/esx/esx_interface_driver.h create mode 100644 src/esx/esx_network_driver.c create mode 100644 src/esx/esx_network_driver.h create mode 100644 src/esx/esx_secret_driver.c create mode 100644 src/esx/esx_secret_driver.h create mode 100644 src/esx/esx_storage_driver.c create mode 100644 src/esx/esx_storage_driver.h diff --git a/src/Makefile.am b/src/Makefile.am index 324030b..7d426ae 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -214,6 +214,11 @@ ONE_DRIVER_SOURCES = \ ESX_DRIVER_SOURCES = \ esx/esx_driver.c esx/esx_driver.h \ + esx/esx_interface_driver.c esx/esx_interface_driver.h \ + esx/esx_network_driver.c esx/esx_network_driver.h \ + esx/esx_storage_driver.c esx/esx_storage_driver.h \ + esx/esx_device_monitor.c esx/esx_device_monitor.h \ + esx/esx_secret_driver.c esx/esx_secret_driver.h \ esx/esx_util.c esx/esx_util.h \ esx/esx_vi.c esx/esx_vi.h \ esx/esx_vi_methods.c esx/esx_vi_methods.h \ diff --git a/src/esx/esx_device_monitor.c b/src/esx/esx_device_monitor.c new file mode 100644 index 0000000..c813adf --- /dev/null +++ b/src/esx/esx_device_monitor.c @@ -0,0 +1,93 @@ + +/* + * esx_device_monitor.c: device monitor methods for managing VMware ESX + * host devices + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <config.h> + +#include "internal.h" +#include "virterror_internal.h" +#include "util.h" +#include "memory.h" +#include "logging.h" +#include "uuid.h" +#include "esx_driver.h" +#include "esx_device_monitor.h" +#include "esx_vi.h" +#include "esx_vi_methods.h" +#include "esx_util.h" + +#define VIR_FROM_THIS VIR_FROM_ESX + +#define ESX_ERROR(conn, code, fmt...) \ + virReportErrorHelper(conn, VIR_FROM_ESX, code, __FILE__, __FUNCTION__, \ + __LINE__, fmt) + + + +static virDrvOpenStatus +esxDeviceOpen(virConnectPtr conn, + virConnectAuthPtr auth ATTRIBUTE_UNUSED, + int flags ATTRIBUTE_UNUSED) +{ + if (STRNEQ(conn->driver->name, "ESX")) { + return VIR_DRV_OPEN_DECLINED; + } + + conn->devMonPrivateData = conn->privateData; + + return VIR_DRV_OPEN_SUCCESS; +} + + + +static int +esxDeviceClose(virConnectPtr conn) +{ + conn->devMonPrivateData = NULL; + + return 0; +} + + + +static virDeviceMonitor esxDeviceMonitor = { + "ESX", /* name */ + esxDeviceOpen, /* open */ + esxDeviceClose, /* close */ + NULL, /* numOfDevices */ + NULL, /* listDevices */ + NULL, /* deviceLookupByName */ + NULL, /* deviceDumpXML */ + NULL, /* deviceGetParent */ + NULL, /* deviceNumOfCaps */ + NULL, /* deviceListCaps */ + NULL, /* deviceCreateXML */ + NULL, /* deviceDestroy */ +}; + + + +int +esxDeviceRegister(void) +{ + return virRegisterDeviceMonitor(&esxDeviceMonitor); +} diff --git a/src/esx/esx_device_monitor.h b/src/esx/esx_device_monitor.h new file mode 100644 index 0000000..e72a5b6 --- /dev/null +++ b/src/esx/esx_device_monitor.h @@ -0,0 +1,29 @@ + +/* + * esx_device_monitor.h: device monitor methods for managing VMware ESX + * host devices + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#ifndef __ESX_DEVICE_MONITOR_H__ +#define __ESX_DEVICE_MONITOR_H__ + +int esxDeviceRegister(void); + +#endif /* __ESX_DEVICE_MONITOR_H__ */ diff --git a/src/esx/esx_driver.c b/src/esx/esx_driver.c index ddda66e..5aa3237 100644 --- a/src/esx/esx_driver.c +++ b/src/esx/esx_driver.c @@ -2,7 +2,7 @@ /* * esx_driver.c: core driver methods for managing VMware ESX hosts * - * Copyright (C) 2009 Matthias Bolte <matthias.bolte(a)googlemail.com> + * Copyright (C) 2009, 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> * Copyright (C) 2009 Maximilian Wilhelm <max(a)rfc2324.org> * * This library is free software; you can redistribute it and/or @@ -33,6 +33,11 @@ #include "logging.h" #include "uuid.h" #include "esx_driver.h" +#include "esx_interface_driver.h" +#include "esx_network_driver.h" +#include "esx_storage_driver.h" +#include "esx_device_monitor.h" +#include "esx_secret_driver.h" #include "esx_vi.h" #include "esx_vi_methods.h" #include "esx_util.h" @@ -46,18 +51,6 @@ static int esxDomainGetMaxVcpus(virDomainPtr domain); -typedef struct _esxPrivate { - esxVI_Context *host; - esxVI_Context *vCenter; - virCapsPtr caps; - char *transport; - int32_t maxVcpus; - esxVI_Boolean supportsVMotion; - esxVI_Boolean supportsLongMode; /* aka x86_64 */ - esxVI_Boolean autoAnswer; - int32_t usedCpuTimeCounterId; -} esxPrivate; - static esxVI_Boolean @@ -3479,7 +3472,14 @@ static virDriver esxDriver = { int esxRegister(void) { - virRegisterDriver(&esxDriver); + if (virRegisterDriver(&esxDriver) < 0 || + esxInterfaceRegister() < 0 || + esxNetworkRegister() < 0 || + esxStorageRegister() < 0 || + esxDeviceRegister() < 0 || + esxSecretRegister() < 0) { + return -1; + } return 0; } diff --git a/src/esx/esx_driver.h b/src/esx/esx_driver.h index 85f4b32..18e821b 100644 --- a/src/esx/esx_driver.h +++ b/src/esx/esx_driver.h @@ -2,7 +2,7 @@ /* * esx_driver.h: core driver methods for managing VMware ESX hosts * - * Copyright (C) 2009 Matthias Bolte <matthias.bolte(a)googlemail.com> + * Copyright (C) 2009, 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> * Copyright (C) 2009 Maximilian Wilhelm <max(a)rfc2324.org> * * This library is free software; you can redistribute it and/or @@ -24,6 +24,24 @@ #ifndef __ESX_DRIVER_H__ #define __ESX_DRIVER_H__ +#include "internal.h" +#include "capabilities.h" +#include "esx_vi.h" + +typedef struct _esxPrivate { + esxVI_Context *host; + esxVI_Context *vCenter; + virCapsPtr caps; + char *transport; + int32_t maxVcpus; + esxVI_Boolean supportsVMotion; + esxVI_Boolean supportsLongMode; /* aka x86_64 */ + esxVI_Boolean autoAnswer; + int32_t usedCpuTimeCounterId; +} esxPrivate; + + + int esxRegister(void); #endif /* __ESX_DRIVER_H__ */ diff --git a/src/esx/esx_interface_driver.c b/src/esx/esx_interface_driver.c new file mode 100644 index 0000000..7664e4a --- /dev/null +++ b/src/esx/esx_interface_driver.c @@ -0,0 +1,96 @@ + +/* + * esx_interface_driver.h: interface driver methods for managing VMware ESX + * host interfaces + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <config.h> + +#include "internal.h" +#include "virterror_internal.h" +#include "util.h" +#include "memory.h" +#include "logging.h" +#include "uuid.h" +#include "esx_driver.h" +#include "esx_interface_driver.h" +#include "esx_vi.h" +#include "esx_vi_methods.h" +#include "esx_util.h" + +#define VIR_FROM_THIS VIR_FROM_ESX + +#define ESX_ERROR(conn, code, fmt...) \ + virReportErrorHelper(conn, VIR_FROM_ESX, code, __FILE__, __FUNCTION__, \ + __LINE__, fmt) + + + +static virDrvOpenStatus +esxInterfaceOpen(virConnectPtr conn, + virConnectAuthPtr auth ATTRIBUTE_UNUSED, + int flags ATTRIBUTE_UNUSED) +{ + if (STRNEQ(conn->driver->name, "ESX")) { + return VIR_DRV_OPEN_DECLINED; + } + + conn->interfacePrivateData = conn->privateData; + + return VIR_DRV_OPEN_SUCCESS; +} + + + +static int +esxInterfaceClose(virConnectPtr conn) +{ + conn->interfacePrivateData = NULL; + + return 0; +} + + + +static virInterfaceDriver esxInterfaceDriver = { + "ESX", /* name */ + esxInterfaceOpen, /* open */ + esxInterfaceClose, /* close */ + NULL, /* numOfInterfaces */ + NULL, /* listInterfaces */ + NULL, /* numOfDefinedInterfaces */ + NULL, /* listDefinedInterfaces */ + NULL, /* interfaceLookupByName */ + NULL, /* interfaceLookupByMACString */ + NULL, /* interfaceGetXMLDesc */ + NULL, /* interfaceDefineXML */ + NULL, /* interfaceUndefine */ + NULL, /* interfaceCreate */ + NULL, /* interfaceDestroy */ + NULL, /* interfaceIsActive */ +}; + + + +int +esxInterfaceRegister(void) +{ + return virRegisterInterfaceDriver(&esxInterfaceDriver); +} diff --git a/src/esx/esx_interface_driver.h b/src/esx/esx_interface_driver.h new file mode 100644 index 0000000..d30641a --- /dev/null +++ b/src/esx/esx_interface_driver.h @@ -0,0 +1,29 @@ + +/* + * esx_interface_driver.h: interface driver methods for managing VMware ESX + * host interfaces + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#ifndef __ESX_INTERFACE_DRIVER_H__ +#define __ESX_INTERFACE_DRIVER_H__ + +int esxInterfaceRegister(void); + +#endif /* __ESX_INTERFACE_DRIVER_H__ */ diff --git a/src/esx/esx_network_driver.c b/src/esx/esx_network_driver.c new file mode 100644 index 0000000..9532352 --- /dev/null +++ b/src/esx/esx_network_driver.c @@ -0,0 +1,101 @@ + +/* + * esx_network_driver.c: network driver methods for managing VMware ESX + * host networks + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <config.h> + +#include "internal.h" +#include "virterror_internal.h" +#include "util.h" +#include "memory.h" +#include "logging.h" +#include "uuid.h" +#include "esx_driver.h" +#include "esx_network_driver.h" +#include "esx_vi.h" +#include "esx_vi_methods.h" +#include "esx_util.h" + +#define VIR_FROM_THIS VIR_FROM_ESX + +#define ESX_ERROR(conn, code, fmt...) \ + virReportErrorHelper(conn, VIR_FROM_ESX, code, __FILE__, __FUNCTION__, \ + __LINE__, fmt) + + + +static virDrvOpenStatus +esxNetworkOpen(virConnectPtr conn, + virConnectAuthPtr auth ATTRIBUTE_UNUSED, + int flags ATTRIBUTE_UNUSED) +{ + if (STRNEQ(conn->driver->name, "ESX")) { + return VIR_DRV_OPEN_DECLINED; + } + + conn->networkPrivateData = conn->privateData; + + return VIR_DRV_OPEN_SUCCESS; +} + + + +static int +esxNetworkClose(virConnectPtr conn) +{ + conn->networkPrivateData = NULL; + + return 0; +} + + + +static virNetworkDriver esxNetworkDriver = { + "ESX", /* name */ + esxNetworkOpen, /* open */ + esxNetworkClose, /* close */ + NULL, /* numOfNetworks */ + NULL, /* listNetworks */ + NULL, /* numOfDefinedNetworks */ + NULL, /* listDefinedNetworks */ + NULL, /* networkLookupByUUID */ + NULL, /* networkLookupByName */ + NULL, /* networkCreateXML */ + NULL, /* networkDefineXML */ + NULL, /* networkUndefine */ + NULL, /* networkCreate */ + NULL, /* networkDestroy */ + NULL, /* networkDumpXML */ + NULL, /* networkGetBridgeName */ + NULL, /* networkGetAutostart */ + NULL, /* networkSetAutostart */ + NULL, /* networkIsActive */ + NULL, /* networkIsPersistent */ +}; + + + +int +esxNetworkRegister(void) +{ + return virRegisterNetworkDriver(&esxNetworkDriver); +} diff --git a/src/esx/esx_network_driver.h b/src/esx/esx_network_driver.h new file mode 100644 index 0000000..f117e03 --- /dev/null +++ b/src/esx/esx_network_driver.h @@ -0,0 +1,29 @@ + +/* + * esx_network_driver.h: network driver methods for managing VMware ESX + * host networks + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#ifndef __ESX_NETWORK_DRIVER_H__ +#define __ESX_NETWORK_DRIVER_H__ + +int esxNetworkRegister(void); + +#endif /* __ESX_NETWORK_DRIVER_H__ */ diff --git a/src/esx/esx_secret_driver.c b/src/esx/esx_secret_driver.c new file mode 100644 index 0000000..e451b13 --- /dev/null +++ b/src/esx/esx_secret_driver.c @@ -0,0 +1,91 @@ + +/* + * esx_secret_driver.c: secret driver methods for VMware ESX secret manipulation + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <config.h> + +#include "internal.h" +#include "virterror_internal.h" +#include "util.h" +#include "memory.h" +#include "logging.h" +#include "uuid.h" +#include "esx_driver.h" +#include "esx_secret_driver.h" +#include "esx_vi.h" +#include "esx_vi_methods.h" +#include "esx_util.h" + +#define VIR_FROM_THIS VIR_FROM_ESX + +#define ESX_ERROR(conn, code, fmt...) \ + virReportErrorHelper(conn, VIR_FROM_ESX, code, __FILE__, __FUNCTION__, \ + __LINE__, fmt) + + + +static virDrvOpenStatus +esxSecretOpen(virConnectPtr conn, virConnectAuthPtr auth ATTRIBUTE_UNUSED, + int flags ATTRIBUTE_UNUSED) +{ + if (STRNEQ(conn->driver->name, "ESX")) { + return VIR_DRV_OPEN_DECLINED; + } + + conn->secretPrivateData = conn->privateData; + + return VIR_DRV_OPEN_SUCCESS; +} + + + +static int +esxSecretClose(virConnectPtr conn) +{ + conn->secretPrivateData = NULL; + + return 0; +} + + + +static virSecretDriver esxSecretDriver = { + "ESX", /* name */ + esxSecretOpen, /* open */ + esxSecretClose, /* close */ + NULL, /* numOfSecrets */ + NULL, /* listSecrets */ + NULL, /* lookupByUUID */ + NULL, /* lookupByUsage */ + NULL, /* defineXML */ + NULL, /* getXMLDesc */ + NULL, /* setValue */ + NULL, /* getValue */ + NULL, /* undefine */ +}; + + + +int +esxSecretRegister(void) +{ + return virRegisterSecretDriver(&esxSecretDriver); +} diff --git a/src/esx/esx_secret_driver.h b/src/esx/esx_secret_driver.h new file mode 100644 index 0000000..c0d4a48 --- /dev/null +++ b/src/esx/esx_secret_driver.h @@ -0,0 +1,28 @@ + +/* + * esx_secret_driver.h: secret driver methods for VMware ESX secret manipulation + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#ifndef __ESX_SECRET_DRIVER_H__ +#define __ESX_SECRET_DRIVER_H__ + +int esxSecretRegister(void); + +#endif /* __ESX_SECRET_DRIVER_H__ */ diff --git a/src/esx/esx_storage_driver.c b/src/esx/esx_storage_driver.c new file mode 100644 index 0000000..7e9b98d --- /dev/null +++ b/src/esx/esx_storage_driver.c @@ -0,0 +1,117 @@ + +/* + * esx_storage_driver.c: storage driver methods for managing VMware ESX + * host storage + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <config.h> + +#include "internal.h" +#include "virterror_internal.h" +#include "util.h" +#include "memory.h" +#include "logging.h" +#include "uuid.h" +#include "esx_driver.h" +#include "esx_storage_driver.h" +#include "esx_vi.h" +#include "esx_vi_methods.h" +#include "esx_util.h" + +#define VIR_FROM_THIS VIR_FROM_ESX + +#define ESX_ERROR(conn, code, fmt...) \ + virReportErrorHelper(conn, VIR_FROM_ESX, code, __FILE__, __FUNCTION__, \ + __LINE__, fmt) + + + +static virDrvOpenStatus +esxStorageOpen(virConnectPtr conn, + virConnectAuthPtr auth ATTRIBUTE_UNUSED, + int flags ATTRIBUTE_UNUSED) +{ + if (STRNEQ(conn->driver->name, "ESX")) { + return VIR_DRV_OPEN_DECLINED; + } + + conn->storagePrivateData = conn->privateData; + + return VIR_DRV_OPEN_SUCCESS; +} + + + +static int +esxStorageClose(virConnectPtr conn) +{ + conn->storagePrivateData = NULL; + + return 0; +} + + + +static virStorageDriver esxStorageDriver = { + "ESX", /* name */ + esxStorageOpen, /* open */ + esxStorageClose, /* close */ + NULL, /* numOfPools */ + NULL, /* listPools */ + NULL, /* numOfDefinedPools */ + NULL, /* listDefinedPools */ + NULL, /* findPoolSources */ + NULL, /* poolLookupByName */ + NULL, /* poolLookupByUUID */ + NULL, /* poolLookupByVolume */ + NULL, /* poolCreateXML */ + NULL, /* poolDefineXML */ + NULL, /* poolBuild */ + NULL, /* poolUndefine */ + NULL, /* poolCreate */ + NULL, /* poolDestroy */ + NULL, /* poolDelete */ + NULL, /* poolRefresh */ + NULL, /* poolGetInfo */ + NULL, /* poolGetXMLDesc */ + NULL, /* poolGetAutostart */ + NULL, /* poolSetAutostart */ + NULL, /* poolNumOfVolumes */ + NULL, /* poolListVolumes */ + NULL, /* volLookupByName */ + NULL, /* volLookupByKey */ + NULL, /* volLookupByPath */ + NULL, /* volCreateXML */ + NULL, /* volCreateXMLFrom */ + NULL, /* volDelete */ + NULL, /* volGetInfo */ + NULL, /* volGetXMLDesc */ + NULL, /* volGetPath */ + NULL, /* poolIsActive */ + NULL, /* poolIsPersistent */ +}; + + + +int +esxStorageRegister(void) +{ + return virRegisterStorageDriver(&esxStorageDriver); +} diff --git a/src/esx/esx_storage_driver.h b/src/esx/esx_storage_driver.h new file mode 100644 index 0000000..91a5d03 --- /dev/null +++ b/src/esx/esx_storage_driver.h @@ -0,0 +1,29 @@ + +/* + * esx_storage_driver.h: storage driver methods for managing VMware ESX + * host storage + * + * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * + * 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, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#ifndef __ESX_STORAGE_DRIVER_H__ +#define __ESX_STORAGE_DRIVER_H__ + +int esxStorageRegister(void); + +#endif /* __ESX_STORAGE_DRIVER_H__ */ -- 1.6.3.3

2 2

[libvirt] LVM storage pool with existing mirrored volumes
by Ben Roberts 17 Jan '10

17 Jan '10

Hi, I tried to setup a lvm-backed storage pool, reusing an existing volume group which already contains a handful of mirrored logical volumes. For reference, I am using lvm 2.02.56-r3, qemu-kvm 0.12.1.2 and libvirt 0.7.5 on an up-to-date ~amd64 gentoo system. The first issue I encountered is when trying to activate the storage pool, the process fails with "internal error lvs command failed". This appears to be because the output of lvs is slightly different when using mirrored lvs, and is breaking the parser: /sbin/lvs --separator , --noheadings --units b --unbuffered --nosuffix --options lv_name,origin,uuid,devices,seg_size,vg_extent_size bigpool vms,,0PzrRS-8Ljs-SUq0-ywFH-BI3d-RI0V-aibupu,vms_mimage_0(0),vms_mimage_1(0),107374182400,4194304 backups,,Snk2NL-oJe3-GpTL-Qihs-ZjjG-xPyj-ufmwCp,backups_mimage_0(0),backups_mimage_1(0),214748364800,4194304 The device column of output would normally contain something like "/dev/sda(1234)", but here contains "vms_mimage_0(0),vms_mimage_1(0). This is causing the parser to misinterpret the output, and it believes the lv name to be suffixed with a trailing comma, eg "vms,". When trying to access "/dev/bigpool/vms,", the command errors with File or Directory not Found. When adding the --all parameter to lvs, the hidden mirroring lvs are visible, and these have the expected structure: vms,,0PzrRS-8Ljs-SUq0-ywFH-BI3d-RI0V-aibupu,vms_mimage_0(0),vms_mimage_1(0),107374182400,4194304 [vms_mlog],,vnOzg8-ynKc-kM8F-RKdM-xXC0-PJlG-qcgmFh,/dev/sda2(0),4194304,4194304 [vms_mimage_0],,2pXjHt-aARa-FOO7-ZL6N-ahpu-ebdr-P6Qxj3,/dev/sdb1(0),107374182400,4194304 [vms_mimage_1],,ITd8CT-JpKi-Y2a6-a4jc-2yDa-9VmG-gWgRd8,/dev/sdc1(51200),107374182400,4194304 Presumably, without the --all parameter and listing the internal mirroring volumes, libvirt will miscalculate sizes of LVs in the VG. This bug appears to prevent the use of any volume group which contains one or more mirrored logical volumes as a backend storage pool. -- Ben Roberts Computer Science MEng, Part IV Electronics and Computer Science, University of Southampton

3 2

[libvirt] [PATCH] Man page for LXC BZ 528709
by David Jorm 17 Jan '10

17 Jan '10

--- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -114,6 +114,10 @@ allow to connect locally as root to the daemon supervising QEmu and KVM domains allow to connect locally as a normal user to his own set of QEmu and KVM domains +=item lxc:/// + +connect to a local linux container + =back For remote access see the documentation page on how to make URIs.

2 1

[libvirt] [PATCH] Man page BZ 548485
by David Jorm 17 Jan '10

17 Jan '10

--- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -342,7 +342,15 @@ severed upon restore, as TCP timeouts may have expired. =item B<schedinfo> optional I<--weight> B<number> optional I<--cap> B<number> I<domain-id> -Allows to show (and set) the domain scheduler parameters. +Allows you to show (and set) the domain scheduler parameters. The parameters available for each hypervisor are: + +LXC, QEMU/KVM (posix scheduler): cpu_shares + +Xen (credit scheduler): weight, cap + +ESX (allocation scheduler): reservation, limit, shares + +B<Note>: The cpu_shares parameter has a valid value range of 0-262144. B<Note>: The weight and cap parameters are defined only for the XEN_CREDIT scheduler and are now I<DEPRECATED>.

2 1

[libvirt] [PATCH] Minor fixes for API extension doc
by Jim Fehlig 17 Jan '10

17 Jan '10

Update the API Extensions doc to reflect new source directory layout. --- docs/api_extension.html.in | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/api_extension.html.in b/docs/api_extension.html.in index 59d3414..de6eedc 100644 --- a/docs/api_extension.html.in +++ b/docs/api_extension.html.in @@ -178,7 +178,7 @@ involves making two additions to: - <code>qemud/remote_protocol.x</code> + <code>src/remote/remote_protocol.x</code> First, create two new structs for each new function that you're adding @@ -198,7 +198,7 @@ Once these changes are in place, it's necessary to run 'make rpcgen' - in the qemud directory to create the .c and .h files required by the + in the src directory to create the .c and .h files required by the remote protocol code. This must be done on a Linux host using the GLibC rpcgen program. Other rpcgen versions may generate code which results in bogus compile time warnings @@ -213,7 +213,7 @@ the rpcgen generated .h files. The remote method calls go in: - <code>src/remote_internal.c</code> + <code>src/remote/remote_internal.c</code> Each remote method invocation does the following: @@ -243,7 +243,7 @@ The server side dispatchers are implemented in: - <code>qemud/remote.c</code> + <code>daemon/remote.c</code> Again, this step uses the .h files generated by make rpcgen. -- 1.6.0.2

3 2