- Devel - libvirt List Archives

[libvirt] RFC: Modelling timers/ clocks / tick policies in libvirt
by Daniel P. Berrange 05 Mar '10

05 Mar '10

One of the things we've not dealt with in libvirt yet is how to model the various evil hacks most virt products have for dealing with timers in guests. This email tries to outlines the problems & way each virt system has dealt with them. Finally it suggests how to manage this in libvirt domain XML. Comments please :-) Virtual machine timer management in libvirt =========================================== On PC hardware there are a number of terrible timers / clock sources available to operating systems * PIT Timer with periodic interrupts * RTC Time of Day clock, continuous running Timer with periodic interrupts * Local APIC Timer Timer with periodic interrupts * ACPI Timer Timer with periodic interrupts * TSC Read via rdtsc instruction. No interrupts Unreliable on some hardware. eg changes frequency. Not synced between cores Different HZ across hosts * HPET Multiple timers with periodic interrupts Can replace PIT/RTC timers They all generally suck in real hardware, and this gets worse in virtual machines. Many different approaches to making them suck less in VMWare, Xen & KVM, but there are some reasonably common concepts.... Virtual timekeeping problems ---------------------------- Three primary problems / areas to deal with: * Time of day clock (RTC) - Initialized to UTC/Localtime/Timezone/UTC+offset - Two modes of operation: 1. Guest wallclock: only runs when guest is executing. ie stopped across save/restore, etc 2. Host wallclock: runs continuously with host wall time. * Interrupt timers - Ticks can not always be delivered on time Policies to deal with "missed" ticks: 1. Deliver at normal rate without catchup 2. Deliver at higher rate to catch up 3. Merge into 1 tick & deliver asap 4. Discard all missed ticks * TSC - rdtsc instruction can be exposed to guests in two ways 1. Trap + emulate (slow, but more reliable) 2. Native (fast, but possibly unreliable) Optionally also expose a 'rdtscp' instruction Possiblly set a fixed HZ independant of host. VMWare timekeeping ------------------ * All timers run in "apparant time" ie track guest wallclock * Missed tick policy is to deliver at higher rate to catchup * TSC can be switched between native/emulate (virtual_rdtsc=TRUE|FALSE) * TSC can have hardcoded HZ in emulate mode (apparantHZ=VALUE) * RTC time of day is synced to host at startup (rtc.diffFromUTC or rtc.startTime) * VMWare tools reset guest TOD if it gets out of sync Xen timekeeping --------------- * Virtual platform timer (VPT) used as source for other timers * VPT has 4 modes 0: delay_for_missed_ticks Missed ticks are delivered when next scheduled, at the normal rate. RTC runs in guest wallclock, so is delayed. No catchup is attempted 1: no_delay_for_missed_ticks Missed ticks are delivered when next scheduled, at the normal rate. RTC runs in host wallclock, so is not delayed. 2: no_missed_ticks_pending Missed ticks are discarded & next tick is delivered normally. RTC runs in host wallclock. 3: one_missed_tick_pending Missed interrupts are collapsed into a single late tick. RTC run in host wallclock. * HPET Optionally enabled * TSC. Can run in 4 modes - auto: emulate if host TSC is unstable. native with invariant TSC - native: always native regardless of host TSC stability - emulate: trap + emulate regardless of host TSC invariant - pvrdtsc: native, requiring invariant TSC. Also exposes rdtscp instruction KVM timekeeping --------------- * PIT: can be in kernel, or userspace (userspace deprecated for KVM) Default tick policies differ for both impls - Userspace: Default: missed ticks are delivered when next scheduled at normal rate -tdf flag enable tick reinjection to catchup - Kernel: Default: Missed ticks are delivered at higher rate to catch up -no-kvm-pit-reinjection to disable tick reinjection catchup * RTC TOD clock can run in host or guest wallclock (clock=host|guest) Default: missed ticks are delivered when next scheduled at normal rate -rtc-td-hack or -clock driftfix=slew: missed ticks are delivered at a higher rate to catchup * TSC - Always runs native. * HPET - Optionally enabled/disabled Mapping in libvirt XML ---------------------- Currently supports setting Time of Day clock via <clock offset="utc"/> Always sync to UTC <clock offset="localtime"/> Always sync to host timezone <clock offset="timezone" timezone='Europe/Paris'/> Sync to arbitrary timezone <clock offset="variable" adjustment='123456'/> Sync to UTC + arbitrary offset Proposal to model all timers policies as sub-elements of this <clock/> In general we wil allow zero or more <timer/> elements following the syntax: <timer name='platform|pit|rtc|hpet|tsc' wallclock='host|guest' tickpolicy='none|catchup|merge|discard' frequency='123' mode='auto|native|emulate|paravirt' present='yes|no' /> Meaning of 'name': Names map to regular PC timers / clocks. 'Platform' refers to the (optional) master virtual clock source that may be used to drive policy of "other" clocks (eg used in Xen, which clocks are controlled by the platform clock is to be undefined because it has varied in Xen over time). Meaning of 'tickpolicy': none: continue to deliver at normal rate (ie ticks are delayed) catchup: deliver at higher rate to catchup merge: ticks merged into 1 single tick discard: all missed ticks are discarded Meaning of 'wallclock': Only valid for name='rtc' or 'platform' host: RTC wallclock always tracks host time guest: RTC wallclock always tracks host time Meaning of 'frequency': Set a fixed frequency in HZ. NB: Only relevant for TSC. All other timers are fixed (PIT, RTC), or fully guest controlled frequency (HPET) Meaning of 'mode': Control how the clock is exposed to guest. auto: native if safe, otherwise emulate native: always native emulate: always emulate paravirt: native + paravirtualize NB: Only relevant for TSC. All other timers are always emulated. Meaing of 'present': Used to override default set of timers visible to the guest. eg to enable or disable the HPET Mapping to VMWare ----------------- eg with guest config showing diffFromUTC='123456' apparentHZ='123456' virtual_rdtsc=False libvirt XML gets: <clock mode='variable' adjustment='123456'> <timer name='tsc' frequency='123456' mode='native'/> </clock> Mapping to Xen -------------- eg with guest config showing timer_mode=3 hpet=1 tsc_mode=2 localtime=1 <clock mode='localtime'> <timer name='platform' tickpolicy='merge' wallclock='host'/> <timer name='hpet'/> <timer name='tsc' mode='native'/> </clock> Mapping to KVM -------------- eg with guest ARGV showing -no-kvm-pit-reinjection -clock base=localtime,clock=guest,driftfix=slew -no-hpet <clock mode='localtime'> <timer name='rtc' tickpolicy='catchup' wallclock='guest'/> <timer name='pit' tickpolicy='none'/> <timer name='hpet' present='no'/> </clock> Further reading --------------- VMWare has the best doc: http://www.vmware.com/pdf/vmware_timekeeping.pdf Xen: Docs on 'tsc_mode' at $SOURCETREE/docs/misc/tscmode.txt Docs for 'timer_mode' in the source code only: xen/include/public/hvm/params.h KVM: No docs at all. Guess from -help descriptions, reading source code & asking clever people about it :-) -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://deltacloud.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

1 0

[libvirt] [PATCH 0/3] Misc fixes for LXC driver
by Daniel P. Berrange 05 Mar '10

05 Mar '10

This series provides some small fixes to the LXC driver which bring it closer to being able to run a full Fedora OS with regular init (upstart) startup process

2 6

[libvirt] Get IP of ghost from remote with libvirt
by angelo alfano 05 Mar '10

05 Mar '10

Hello everyone, sorry for my English, I need to create a script (preferably in Python) that runs on a host machine, I return the IP address of a Ghost running, providing its hostname. In this VM is assigned the IP in DHCP with a network with this configuration: $ virsh net-dumpxml network1 Connecting to uri: qemu:///system <network> <name>network1</name> <uuid>6050d8d4-4030-fdd5-ddfa-041c89bbdfa7</uuid> <forward mode='nat'/> <bridge name='virbr0' stp='on' forwardDelay='0' /> <ip address='192.168.100.1' netmask='255.255.255.0'> <dhcp> <range start='192.168.100.2' end='192.168.100.254' /> </dhcp> </ip> </network> While the configuration of one of the VM is: $ virsh dumpxml test1 Connecting to uri: qemu:///system <domain type='kvm' id='1'> <name>test1</name> <uuid>cf71c7d3-0d05-d51c-492e-f5ad2be3ec0e</uuid> <memory>524288</memory> <currentMemory>524288</currentMemory> <vcpu>1</vcpu> <os> <type arch='x86_64' machine='pc-0.11'>hvm</type> <boot dev='hd'/> </os> <features> <acpi/> <apic/> <pae/> </features> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <devices> <emulator>/usr/bin/kvm</emulator> <disk type='file' device='cdrom'> <target dev='hdc' bus='ide'/> <readonly/> </disk> <disk type='file' device='disk'> <source file='/var/lib/libvirt/images/test1.img'/> <target dev='vda' bus='virtio'/> </disk> <interface type='network'> <mac address='54:52:00:4e:59:0d'/> <source network='network1'/> <target dev='vnet0'/> <model type='virtio'/> </interface> <serial type='pty'> <source path='/dev/pts/1'/> <target port='0'/> </serial> <console type='pty' tty='/dev/pts/1'> <source path='/dev/pts/1'/> <target port='0'/> </console> <input type='mouse' bus='ps2'/> <graphics type='vnc' port='5900' autoport='yes' keymap='it'/> <sound model='es1370'/> <video> <model type='cirrus' vram='9216' heads='1'/> </video> </devices> <seclabel type='dynamic' model='apparmor'> <label>libvirt-cf71c7d3-0d05-d51c-492e-f5ad2be3ec0e</label> <imagelabel>libvirt-cf71c7d3-0d05-d51c-492e-f5ad2be3ec0e</imagelabel> </seclabel> </domain> Give me a hand? Thanks -- Angelo

1 0

[libvirt] Assigning a static IP to the Linux Container through libvirt.
by Kumar L Srikanth-B22348 05 Mar '10

05 Mar '10

Hi, I want to assign a static IP 172.16.1.20 to one of the containers created using libvirt. For that, I defined a network 'mynetwork' using net-define. The XML format for that network is shown below: <network> <name>mynetwork</name> <uuid>f76c44dd-bcef-6b45-4a50-00d575a369ad</uuid> <bridge name='virbr1' stp='on' delay='0' /> <ip address='172.16.1.20' netmask='255.255.255.0'> </ip> </network> I started the network 'mynetwork' using net-start. virsh # net-list --all Name State Autostart ----------------------------------------- default active yes mynetwork active no I created a domain 'vm1_fedora' with libvirt XML: <domain type='lxc' id='1'> <name>vm1_fedora</name> <memory>500000</memory> <os> <type>exe</type> <init>/bin/bash</init> </os> <vcpu>1</vcpu> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>destroy</on_crash> <devices> <emulator>/usr/libexec/libvirt_lxc</emulator> <filesystem type='mount'> <source dir='/root/lxc/fedora12'/> <target dir='/'/> </filesystem> <filesystem type='mount'> <source dir='/root/lxc/fedora12/dev'/> <target dir='/dev'/> </filesystem> <interface type='network'> <source network='mynetwork'/> </interface> <console type='pty' /> </devices> </domain> But, when I start that domain and see for 'ifconfig', I am not able to see the IP address 172.16.1.20 there. [root@localhost /]# ifconfig eth0 Link encap:Ethernet HWaddr 52:54:00:53:4D:BF inet6 addr: fe80::5054:ff:fe53:4dbf/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:18 errors:0 dropped:0 overruns:0 frame:0 TX packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:840 (840.0 b) TX bytes:384 (384.0 b) lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) can you please let me know, how can I assign a static IP to the Linux Container? Regards, Srikanth.

2 2

[libvirt] Assigning a Static IP to the Domain.
by Kumar L Srikanth-B22348 05 Mar '10

05 Mar '10

Hi, I am using libvirt version 0.7.6, to create a Linux Container. With the domain XML format, I am able to create a Domain and able to access. But, I am struck at assigning an IP address to the Domain. I want to assign a static IP to eth0 of the Domain and ale to ping that from the host machine. For that, I created a bridge br0 [which can be linked to the interface eth0 of the Domain created by libvirt] with the IP 172.16.1.10. I want to assign an IP 172.16.1.16 to the Domain. Can any one please guide me? Regards, Srikanth.

1 0

[libvirt] [PATCH] Change default for storage uid/gid from getuid()/getgid() to -1/-1
by Laine Stump 05 Mar '10

05 Mar '10

This allows the config to have a setting that means "leave it alone", eg when building a pool where the directory already exists the user may want the current uid/gid of the directory left intact. This actually gets us back to older behavior - before recent changes to the pool building code, we weren't as insistent about honoring the uid/gid settings in the XML, and virt-manager was taking advantage of this behavior. As a side benefit, removing calls to getuid/getgid from the XML parsing functions also seems like a good idea. And having a default that is different from a common/useful value (0 == root) is a good thing in general, as it removes ambiguity from decisions (at least one place in the code was checking for (perms.uid == 0) to see if a special uid was requested). Note that this will only affect newly created pools and volumes. Due to the way that the XML is parsed, then formatted for newly created volumes, all existing pools/volumes already have an explicit uid and gid set. src/conf/storage_conf.c: Remove calls to setuid/setgid for default values of uid/gid, and set them to -1 instead src/storage/storage_backend.c: src/storage/storage_backend_fs.c: Make account for the new default values of perms.uid and perms.gid. --- src/conf/storage_conf.c | 8 +++--- src/storage/storage_backend.c | 25 +++++++++++++---------- src/storage/storage_backend_fs.c | 39 +++++++++++++++++++++++++++---------- 3 files changed, 46 insertions(+), 26 deletions(-) diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index 19a1db9..f59f109 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -539,8 +539,8 @@ virStorageDefParsePerms(xmlXPathContextPtr ctxt, if (node == NULL) { /* Set default values if there is not <permissions> element */ perms->mode = defaultmode; - perms->uid = getuid(); - perms->gid = getgid(); + perms->uid = -1; + perms->gid = -1; perms->label = NULL; return 0; } @@ -564,7 +564,7 @@ virStorageDefParsePerms(xmlXPathContextPtr ctxt, } if (virXPathNode("./owner", ctxt) == NULL) { - perms->uid = getuid(); + perms->uid = -1; } else { if (virXPathLong("number(./owner)", ctxt, &v) < 0) { virStorageReportError(VIR_ERR_XML_ERROR, @@ -575,7 +575,7 @@ virStorageDefParsePerms(xmlXPathContextPtr ctxt, } if (virXPathNode("./group", ctxt) == NULL) { - perms->gid = getgid(); + perms->gid = -1; } else { if (virXPathLong("number(./group)", ctxt, &v) < 0) { virStorageReportError(VIR_ERR_XML_ERROR, diff --git a/src/storage/storage_backend.c b/src/storage/storage_backend.c index 3742493..ee6a32e 100644 --- a/src/storage/storage_backend.c +++ b/src/storage/storage_backend.c @@ -241,11 +241,10 @@ virStorageBackendCreateBlockFrom(virConnectPtr conn ATTRIBUTE_UNUSED, uid = (vol->target.perms.uid != st.st_uid) ? vol->target.perms.uid : -1; gid = (vol->target.perms.gid != st.st_gid) ? vol->target.perms.gid : -1; if (((uid != -1) || (gid != -1)) - && (fchown(fd, vol->target.perms.uid, vol->target.perms.gid) < 0)) { + && (fchown(fd, uid, gid) < 0)) { virReportSystemError(errno, _("cannot chown '%s' to (%u, %u)"), - vol->target.path, vol->target.perms.uid, - vol->target.perms.gid); + vol->target.path, uid, gid); goto cleanup; } if (fchmod(fd, vol->target.perms.mode) < 0) { @@ -356,10 +355,12 @@ virStorageBackendCreateRaw(virConnectPtr conn ATTRIBUTE_UNUSED, goto cleanup; } + uid_t uid = (vol->target.perms.uid == -1) ? getuid() : vol->target.perms.uid; + gid_t gid = (vol->target.perms.gid == -1) ? getgid() : vol->target.perms.gid; + if ((createstat = virFileOperation(vol->target.path, O_RDWR | O_CREAT | O_EXCL | O_DSYNC, - vol->target.perms.mode, - vol->target.perms.uid, vol->target.perms.gid, + vol->target.perms.mode, uid, gid, createRawFileOpHook, &hdata, VIR_FILE_OP_FORCE_PERMS | (pool->def->type == VIR_STORAGE_POOL_NETFS @@ -491,14 +492,14 @@ cleanup: static int virStorageBuildSetUIDHook(void *data) { virStorageVolDefPtr vol = data; - if ((vol->target.perms.gid != 0) + if ((vol->target.perms.gid != -1) && (setgid(vol->target.perms.gid) != 0)) { virReportSystemError(errno, _("Cannot set gid to %u before creating %s"), vol->target.perms.gid, vol->target.path); return -1; } - if ((vol->target.perms.uid != 0) + if ((vol->target.perms.uid != -1) && (setuid(vol->target.perms.uid) != 0)) { virReportSystemError(errno, _("Cannot set uid to %u before creating %s"), @@ -517,8 +518,11 @@ static int virStorageBackendCreateExecCommand(virStoragePoolObjPtr pool, int filecreated = 0; if ((pool->def->type == VIR_STORAGE_POOL_NETFS) - && (getuid() == 0) - && ((vol->target.perms.uid != 0) || (vol->target.perms.gid != 0))) { + && (((getuid() == 0) + && (vol->target.perms.uid != -1) + && (vol->target.perms.uid != 0)) + || ((vol->target.perms.gid != -1) + && (vol->target.perms.gid != getgid())))) { if (virRunWithHook(cmdargv, virStorageBuildSetUIDHook, vol, NULL) == 0) { /* command was successfully run, check if the file was created */ @@ -547,8 +551,7 @@ static int virStorageBackendCreateExecCommand(virStoragePoolObjPtr pool, && (chown(vol->target.path, uid, gid) < 0)) { virReportSystemError(errno, _("cannot chown %s to (%u, %u)"), - vol->target.path, vol->target.perms.uid, - vol->target.perms.gid); + vol->target.path, uid, gid); return -1; } if (chmod(vol->target.path, vol->target.perms.mode) < 0) { diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index 8279d78..dfd417f 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -529,16 +529,28 @@ virStorageBackendFileSystemBuild(virConnectPtr conn ATTRIBUTE_UNUSED, /* Now create the final dir in the path with the uid/gid/mode * requested in the config. If the dir already exists, just set * the perms. */ - if ((err = virDirCreate(pool->def->target.path, - pool->def->target.perms.mode, - pool->def->target.perms.uid, - pool->def->target.perms.gid, - VIR_DIR_CREATE_FORCE_PERMS | VIR_DIR_CREATE_ALLOW_EXIST | - (pool->def->type == VIR_STORAGE_POOL_NETFS - ? VIR_DIR_CREATE_AS_UID : 0)) != 0)) { - virReportSystemError(err, _("cannot create path '%s'"), - pool->def->target.path); - goto error; + + struct stat st; + + if ((stat(pool->def->target.path, &st) < 0) + || (pool->def->target.perms.uid != -1)) { + + uid_t uid = (pool->def->target.perms.uid == -1) + ? getuid() : pool->def->target.perms.uid; + gid_t gid = (pool->def->target.perms.gid == -1) + ? getgid() : pool->def->target.perms.gid; + + if ((err = virDirCreate(pool->def->target.path, + pool->def->target.perms.mode, + uid, gid, + VIR_DIR_CREATE_FORCE_PERMS | + VIR_DIR_CREATE_ALLOW_EXIST | + (pool->def->type == VIR_STORAGE_POOL_NETFS + ? VIR_DIR_CREATE_AS_UID : 0)) != 0)) { + virReportSystemError(err, _("cannot create path '%s'"), + pool->def->target.path); + goto error; + } } ret = 0; error: @@ -777,8 +789,13 @@ static int createFileDir(virConnectPtr conn ATTRIBUTE_UNUSED, return -1; } + uid_t uid = (vol->target.perms.uid == -1) + ? getuid() : vol->target.perms.uid; + gid_t gid = (vol->target.perms.gid == -1) + ? getgid() : vol->target.perms.gid; + if ((err = virDirCreate(vol->target.path, vol->target.perms.mode, - vol->target.perms.uid, vol->target.perms.gid, + uid, gid, VIR_DIR_CREATE_FORCE_PERMS | (pool->def->type == VIR_STORAGE_POOL_NETFS ? VIR_DIR_CREATE_AS_UID : 0))) != 0) { -- 1.6.6.1

3 4

[libvirt] [PATCH 0/9] Storage volume wiping API
by David Allan 04 Mar '10

04 Mar '10

Here's the revised volume wiping API patchset (thanks to Dan for the improved name). I'm sending a replacement patchset since the name change made for a very large and messy incremental. This set contains everybody's feedback, which was all very helpful. Many thanks. Dave David Allan (9): Add public API for volume wiping Define the internal driver API for vol wiping Add vol wiping to ESX storage driver struct Implement the public API for vol wiping Define wire protocol format for vol wiping Implement RPC client for vol wiping Implement the remote dispatch bits of vol wiping Simplified version of volume wiping based on feedback from the list. Virsh support for vol wiping daemon/remote.c | 32 +++++ daemon/remote_dispatch_args.h | 1 + daemon/remote_dispatch_prototypes.h | 8 ++ daemon/remote_dispatch_table.h | 5 + include/libvirt/libvirt.h.in | 2 + src/driver.h | 5 + src/esx/esx_storage_driver.c | 1 + src/libvirt.c | 47 ++++++++ src/libvirt_public.syms | 1 + src/remote/remote_driver.c | 27 ++++ src/remote/remote_protocol.c | 11 ++ src/remote/remote_protocol.h | 9 ++ src/remote/remote_protocol.x | 8 +- src/storage/storage_driver.c | 224 +++++++++++++++++++++++++++++++++++ tools/virsh.c | 42 +++++++ 15 files changed, 422 insertions(+), 1 deletions(-)

1 9

[libvirt] Libvirt Auth
by Marco Mornati 04 Mar '10

04 Mar '10

Hello guys, I've created a plugin for Hudson (CI) to manage virtmachines using libvirt java (start vm for a build and stop just after). I've just a problem in authentication, in fact I cannot send uname and password using code, libvirt ask me for the credentials on CLI. someone could explain me How I can use authentican or where I can find an usage guide or example? Thanks a lot Bye Marco

3 8

[libvirt] FW: Libvirt and openvz?
by Thomas Graves 04 Mar '10

04 Mar '10

I just recently started using openvz and I'm trying to use libvirt/virsh with it. I followed the directions on page: http://libvirt.org/drvopenvz.html But I get an error trying to run define with the example config file: virsh define virt.conf error: Failed to define domain from virt.conf error: unknown OS type exe sudo virsh -c openvz:/// list error: could not connect to openvz:/// error: failed to connect to the hypervisor I'm using libvirt 0.7.2 with latest rhel5.4 openvz. Does anyone have more example configuration files or other setup instructions? Does Libvirt allow you to specify other resource limits and controls that openvz has - like vzctl set --devnodes device:rw or vzctl set kmemsize? Thanks, Tom

1 1

[libvirt] [PATCH] fix rawhide/F13 --no-add-needed/dlopen-related build failures
by Jim Meyering 04 Mar '10

04 Mar '10

Diego, I saw no follow-up here, http://thread.gmane.org/gmane.comp.emulators.libvirt/21398/focus=21671 so I've fixed your two patches to address Eric's review feedback, and confirmed that they solve the problem on Rawhide. >From 55c182d89cd6e5417eafe9f4bdf2e0ef7f8721f5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Diego=20Elio=20Petten=C3=B2?= <flameeyes(a)gmail.com> Date: Thu, 4 Mar 2010 16:45:02 +0100 Subject: [PATCH 1/2] build: avoid dlopen-related link failure on rawhide/F13 Instead of using AC_CHECK_LIB and hardcoding -ldl, search for the library needed to get dlopen() and then use the cached value. --- configure.ac | 7 +++++-- 1 files changed, 5 insertions(+), 2 deletions(-) diff --git a/configure.ac b/configure.ac index 682b8b5..d5d62ea 100644 --- a/configure.ac +++ b/configure.ac @@ -1700,7 +1700,7 @@ if test "x$with_driver_modules" = "xyes" ; then old_libs="$LIBS" fail=0 AC_CHECK_HEADER([dlfcn.h],[],[fail=1]) - AC_CHECK_LIB([dl], [dlopen],[],[fail=1]) + AC_SEARCH_LIBS([dlopen], [dl], [], [fail=1]) test $fail = 1 && AC_MSG_ERROR([You must have dlfcn.h / dlopen() support to build driver modules]) @@ -1709,7 +1709,10 @@ if test "x$with_driver_modules" = "xyes" ; then fi if test "$with_driver_modules" = "yes"; then DRIVER_MODULE_CFLAGS="-export-dynamic" - DRIVER_MODULE_LIBS="-ldl" + case $ac_cv_search_dlopen in + no*) DRIVER_MODULE_LIBS= ;; + *) DRIVER_MODULE_LIBS=$ac_cv_search_dlopen ;; + esac AC_DEFINE_UNQUOTED([WITH_DRIVER_MODULES], 1, [whether to build drivers as modules]) fi AM_CONDITIONAL([WITH_DRIVER_MODULES], [test "$with_driver_modules" != "no"]) -- 1.7.0.1.475.gaf89b >From ca7a9f299b0bd415f8434b3c9e6c70528d2f85cc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Diego=20Elio=20Petten=C3=B2?= <flameeyes(a)gmail.com> Date: Thu, 4 Mar 2010 16:48:10 +0100 Subject: [PATCH 2/2] build: vbox: avoid build failure when linking with --no-add-needed With the recent changes to the linking defaults in Fedora 13 (namely enabling --no-add-needed behaviour by default), we have to pass the dlopen()-providing libraries directly at the link of the module; use the same AC_SEARCH_LIBS function as used before to look for it and add it to the Makefile. --- configure.ac | 6 ++++++ src/Makefile.am | 1 + 2 files changed, 7 insertions(+), 0 deletions(-) diff --git a/configure.ac b/configure.ac index d5d62ea..923d7b8 100644 --- a/configure.ac +++ b/configure.ac @@ -303,6 +303,12 @@ fi AM_CONDITIONAL([WITH_OPENVZ], [test "$with_openvz" = "yes"]) if test "x$with_vbox" = "xyes"; then + AC_SEARCH_LIBS([dlopen], [dl], [], [AC_MSG_ERROR([Unable to find dlopen()])]) + case $ac_cv_search_dlopen in + no*) DLOPEN_LIBS= ;; + *) DLOPEN_LIBS=$ac_cv_search_dlopen ;; + esac + AC_SUBST([DLOPEN_LIBS]) AC_DEFINE_UNQUOTED([WITH_VBOX], 1, [whether VirtualBox driver is enabled]) fi AM_CONDITIONAL([WITH_VBOX], [test "$with_vbox" = "yes"]) diff --git a/src/Makefile.am b/src/Makefile.am index 2051e5d..67f8b6d 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -463,6 +463,7 @@ libvirt_driver_vbox_la_CFLAGS = \ if WITH_DRIVER_MODULES libvirt_driver_vbox_la_LDFLAGS = -module -avoid-version endif +libvirt_driver_vbox_la_LIBADD = $(DLOPEN_LIBS) libvirt_driver_vbox_la_SOURCES = $(VBOX_DRIVER_SOURCES) endif -- 1.7.0.1.475.gaf89b

2 1