- Devel - libvirt List Archives

[libvirt] Reboot with SR-IOV devices confuses IOMMU (bug #1294677)
by Jakub Kicinski 04 Jan '16

04 Jan '16

Hi! I hit an issue with PCI passthrough - after I reboot the VM IOMMU mappings are incorrect and devices will access invalid memory. The issue is quite easy to reproduce, I'm using NICs (e.g. Intel 40 Gig Ethernet NICs hit the issue reliably, I wasn't able to get Mellanox NICs to work reliably due to other bugs) but you could probably just pass through any PCI device which does a bit of DMA. With NICs I do the following to reproduce: - configure the NIC for SR-IOV passthrough [1]; - create two standard VMs; - configure VMs with 4GB current allocation and 15GB maximum allocation of memory (my machines have 32 or 64GB total); - pass a VF to each machine; Note1: the current/maximum allocation of memory seem to play a role here. I'm not 100% sure, however, if it causes the bug or just makes it more likely to be triggered. Note2: I leave <on_reboot>restart</on_reboot> so that VMs can reboot. I was able to reproduce easily on 3 distinct machines: dual CPU Haswell E, single CPU Haswell E, single Sandy Bridge EP. With the VMs created above do the following: (1) boot; (2) configure VF interfaces; (3) run ping -c30 to confirm they can communicate; (4) run iperf -P4 -t30 between the machines; (5) reboot; (6) goto 2; First time (fresh after boot) ping and iperf should work fine. After first reboot, there should already be communication problems. From traffic inspection with tcpdump it appears that VFs receive zeroed packets. Only some of the packets are zeroed so depending on your luck the communication may work for a while or just have limited throughput. Usually it breaks down when ARP or important TCP segment gets placed in area that device reads as zero. Reboot will not fix this condition, shutdown/start will. I reproduced this with fully up-to-date Ubuntu 14.04 (both host and VM). I also tried kernel from linux-next.git (4ef7675344d68) and qemu from tip of their git (38a762f) and bug is still there. libvirt in Ubuntu 14.04 is at 1.2.2. As already said - I reproduce this easily within a minute on a range of machines. If someone looks into this and have problems hitting the bug - please let me know. Kuba [1] For Intel NICs supported by i40e: http://www.intel.es/content/www/es/es/embedded/products/networking/xl710-sr…

1 1

[libvirt] [PATCH] maint: update to latest gnulib
by Eric Blake 04 Jan '16

04 Jan '16

Required for the copyright year bump to keep 'make syntax-check' happy, and also pulls in several portability fixes. * .gnulib: Update to latest. * bootstrap: Resync from upstream. * gnulib/local/m4/ssize_t.m4.diff: Regenerate. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Pushing under the build-breaker rule. .gnulib | 2 +- bootstrap | 2 +- gnulib/local/m4/ssize_t.m4.diff | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/.gnulib b/.gnulib index f39477d..6cc32c6 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit f39477dba778e99392948dd3dd19ec0d46aee932 +Subproject commit 6cc32c63e80bc1a30c521b2f07f2b54909b59892 diff --git a/bootstrap b/bootstrap index e72894c..294c0bc 100755 --- a/bootstrap +++ b/bootstrap @@ -4,7 +4,7 @@ scriptversion=2014-12-08.12; # UTC # Bootstrap this package from checked-out sources. -# Copyright (C) 2003-2015 Free Software Foundation, Inc. +# Copyright (C) 2003-2016 Free Software Foundation, Inc. # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by diff --git a/gnulib/local/m4/ssize_t.m4.diff b/gnulib/local/m4/ssize_t.m4.diff index 12cff12..c4863b9 100644 --- a/gnulib/local/m4/ssize_t.m4.diff +++ b/gnulib/local/m4/ssize_t.m4.diff @@ -5,7 +5,7 @@ index 209d64c..5ea72a1 100644 @@ -1,4 +1,4 @@ -# ssize_t.m4 serial 5 (gettext-0.18.2) +# ssize_t.m4 serial 6 (gettext-0.18.2) - dnl Copyright (C) 2001-2003, 2006, 2010-2015 Free Software Foundation, Inc. + dnl Copyright (C) 2001-2003, 2006, 2010-2016 Free Software Foundation, Inc. dnl This file is free software; the Free Software Foundation dnl gives unlimited permission to copy and/or distribute it, @@ -17,7 +17,21 @@ AC_DEFUN([gt_TYPE_SSIZE_T], -- 2.4.3

1 1

[libvirt] [PATCH] tests: Add newlines with VIR_TEST_REGENERATE_OUTPUT
by Cole Robinson 04 Jan '16

04 Jan '16

Since test files are formatted predictably nowadays, we can make VIR_TEST_REGENERATE_OUTPUT handle most cases for us with a simple replacement. test-wrap-argv.pl is still canon, but this bit makes it easier to confirm test output changes during active development. --- tests/testutils.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/tests/testutils.c b/tests/testutils.c index 857e819..6487a62 100644 --- a/tests/testutils.c +++ b/tests/testutils.c @@ -598,6 +598,7 @@ virtTestCompareToFile(const char *strcontent, int ret = -1; char *filecontent = NULL; char *fixedcontent = NULL; + char *regencontent = NULL; bool regenerate = !!virTestGetFlag("VIR_TEST_REGENERATE_OUTPUT"); if (virtTestLoadFile(filename, &filecontent) < 0 && !regenerate) @@ -613,7 +614,10 @@ virtTestCompareToFile(const char *strcontent, if (STRNEQ_NULLABLE(fixedcontent ? fixedcontent : strcontent, filecontent)) { if (regenerate) { - if (virFileWriteStr(filename, strcontent, 0666) < 0) + if (!(regencontent = virStringReplace(strcontent, " -", " \\\n-"))) + goto failure; + + if (virFileWriteStr(filename, regencontent, 0666) < 0) goto failure; goto out; } @@ -626,6 +630,7 @@ virtTestCompareToFile(const char *strcontent, failure: VIR_FREE(fixedcontent); VIR_FREE(filecontent); + VIR_FREE(regencontent); return ret; } -- 2.5.0

1 0

[libvirt] [PATCH] qemu: Fix return value of qemuDomainGetBlockJobInfo
by Michal Privoznik 04 Jan '16

04 Jan '16

While reviewing 1b43885d1784640 I've noticed a virReportError() followed by a goto endjob; without setting the correct return value. Problem is, if block job is so fast that it's bandwidth does not fit into ulong, an error is reported. However, by that time @ret is already set to 1 which means success. Since the scenario can be hardly considered successful, we should return a value meaning error. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_driver.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 304165c..1161aa0 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -16546,6 +16546,7 @@ qemuDomainGetBlockJobInfo(virDomainPtr dom, virReportError(VIR_ERR_OVERFLOW, _("bandwidth %llu cannot be represented in result"), rawInfo.bandwidth); + ret = -1; goto endjob; } -- 2.4.10

2 1

[libvirt] [PATCH] qemu: Fix all callers of qemuMonitorGetBlockJobInfo()
by Andrea Bolognani 04 Jan '16

04 Jan '16

Commit 1b43885 modified one of the callers of this function to take into account the possible return value of 0 when the block job can't be found. This commit finishes the job by updating the remaining caller. --- src/qemu/qemu_driver.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 304165c..c4573d9 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -16150,14 +16150,13 @@ qemuDomainBlockPivot(virQEMUDriverPtr driver, rc = qemuMonitorGetBlockJobInfo(priv->mon, disk->info.alias, &info); if (qemuDomainObjExitMonitor(driver, vm) < 0) goto cleanup; - if (rc < 0) + if (rc <= 0) goto cleanup; - if (rc == 1 && - (info.ready == 1 || - (info.ready == -1 && - info.end == info.cur && - (info.type == VIR_DOMAIN_BLOCK_JOB_TYPE_COPY || - info.type == VIR_DOMAIN_BLOCK_JOB_TYPE_COMMIT)))) + if (info.ready == 1 || + (info.ready == -1 && + info.end == info.cur && + (info.type == VIR_DOMAIN_BLOCK_JOB_TYPE_COPY || + info.type == VIR_DOMAIN_BLOCK_JOB_TYPE_COMMIT))) disk->mirrorState = VIR_DOMAIN_DISK_MIRROR_STATE_READY; } -- 2.5.0

3 3

[libvirt] [PATCH] rbd: Return VIR_STORAGE_FILE_RAW as format for RBD volumes
by Wido den Hollander 04 Jan '16

04 Jan '16

This used to return 'unkown' and that was not correct. A vol-dumpxml now returns: <volume type='network'> <name>image3</name> <key>libvirt/image3</key> <source> </source> <capacity unit='bytes'>10737418240</capacity> <allocation unit='bytes'>10737418240</allocation> <target> <path>libvirt/image3</path> <format type='raw'/> </target> </volume> The RBD driver will now error out if a different format than RAW is provided when creating a volume. Signed-off-by: Wido den Hollander <wido(a)widodh.nl> --- src/conf/storage_conf.c | 3 ++- src/storage/storage_backend_rbd.c | 13 +++++++++++++ 2 files changed, 15 insertions(+), 1 deletion(-) diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index 9b8abea..7c81bab 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -220,7 +220,8 @@ static virStoragePoolTypeInfo poolTypeInfo[] = { }, .volOptions = { .defaultFormat = VIR_STORAGE_FILE_RAW, - .formatToString = virStoragePoolFormatDiskTypeToString, + .formatFromString = virStorageVolumeFormatFromString, + .formatToString = virStorageFileFormatTypeToString, } }, {.poolType = VIR_STORAGE_POOL_SHEEPDOG, diff --git a/src/storage/storage_backend_rbd.c b/src/storage/storage_backend_rbd.c index cdbfdee..80684eb 100644 --- a/src/storage/storage_backend_rbd.c +++ b/src/storage/storage_backend_rbd.c @@ -306,6 +306,7 @@ static int volStorageBackendRBDRefreshVolInfo(virStorageVolDefPtr vol, vol->target.capacity = info.size; vol->target.allocation = info.obj_size * info.num_objs; vol->type = VIR_STORAGE_VOL_NETWORK; + vol->target.format = VIR_STORAGE_FILE_RAW; VIR_FREE(vol->target.path); if (virAsprintf(&vol->target.path, "%s/%s", @@ -557,6 +558,12 @@ virStorageBackendRBDCreateVol(virConnectPtr conn ATTRIBUTE_UNUSED, { vol->type = VIR_STORAGE_VOL_NETWORK; + if (vol->target.format != VIR_STORAGE_FILE_RAW) { + virReportError(VIR_ERR_NO_SUPPORT, "%s", + _("only RAW volumes are supported by this storage pool")); + return -VIR_ERR_NO_SUPPORT; + } + VIR_FREE(vol->target.path); if (virAsprintf(&vol->target.path, "%s/%s", pool->def->source.name, @@ -603,6 +610,12 @@ virStorageBackendRBDBuildVol(virConnectPtr conn, goto cleanup; } + if (vol->target.format != VIR_STORAGE_FILE_RAW) { + virReportError(VIR_ERR_NO_SUPPORT, "%s", + _("only RAW volumes are supported by this storage pool")); + goto cleanup; + } + if (virStorageBackendRBDOpenRADOSConn(&ptr, conn, &pool->def->source) < 0) goto cleanup; -- 1.9.1

2 1

[libvirt] [PATCH 0/3] Various Valgrind fixes
by Michael Chapman 04 Jan '16

04 Jan '16

Valgrind reported a couple of memory leaks and jumps conditional on uninitialized values. Happy New Year! Michael Chapman (3): qemu: do not copy out non-existent block job info qemu: do not leak NBD disk data in migration cookie storage: do not leak storage pool XML filename src/qemu/qemu_driver.c | 4 ++-- src/qemu/qemu_migration.c | 5 +++-- src/storage/storage_driver.c | 10 +++++----- 3 files changed, 10 insertions(+), 9 deletions(-) -- 2.4.3

3 5

[libvirt] [libvirt-glib][PATCH] libvirt-glib-1.0.pc.in: Correctly mark variable
by Michal Privoznik 04 Jan '16

04 Jan '16

In the pkg-config file for libvirt-glib we have a typo: Libs.private: @LIBVIRT_LIBS @GLIB2_LIBS@ Noticed the missing '@' after LIBVIRT_LIBS? Well, I just did. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- libvirt-glib-1.0.pc.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libvirt-glib-1.0.pc.in b/libvirt-glib-1.0.pc.in index 1e8956d..d00be9f 100644 --- a/libvirt-glib-1.0.pc.in +++ b/libvirt-glib-1.0.pc.in @@ -8,5 +8,5 @@ Version: @VERSION@ Description: libvirt glib library Requires: glib-2.0 Libs: -L${libdir} -lvirt-glib-1.0 -Libs.private: @LIBVIRT_LIBS @GLIB2_LIBS@ +Libs.private: @LIBVIRT_LIBS@ @GLIB2_LIBS@ Cflags: -I${includedir}/libvirt-glib-1.0 @LIBVIRT_CFLAGS@ @GLIB2_CFLAGS@ -- 2.4.10

3 3

[libvirt] Non-contiguous NUMA cell numbers
by Andrea Bolognani 04 Jan '16

04 Jan '16

Hi all, libvirt currently doesn't allow you to configure a guest with something like <cpu> <topology sockets='2' cores='5' threads='8'/> <numa> <cell id='0' cpus='0-39' memory='1048576' unit='KiB'/> <cell id='16' cpus='40-79' memory='1048576' unit='KiB'/> </numa> </cpu> with the following error: XML error: Exactly one 'cell' element per guest NUMA cell allowed, non-contiguous ranges or ranges not starting from 0 are not allowed The error message is very specific about not allowing gaps in NUMA cell numbering; however, on the very same host, I have $ numactl --hard | head -1 available: 4 nodes (0-1,16-17) so gaps in numbering are definitely possible, at least on ppc64. Should we consider relaxing this requirement and allow arbitrarily-numbered NUMA cells? Is there a specific reason why it was forbidden in the first place? Cheers. -- Andrea Bolognani Software Engineer - Virtualization Team

4 4

[libvirt] [PATCH v8 0/5] Partial posting of updates to Add quorum to libvirt
by John Ferlan 04 Jan '16

04 Jan '16

I took a few cycles to reformat the initial 5 patches of the add quorum support posted as a v7 here: http://www.redhat.com/archives/libvir-list/2015-December/msg00119.html These patches are easily separable from the rest of the series. Rather than see things linger for months on end, trying to make small leaps. Changes in the series are to reorder things slightly, follow some of the review comments from previous series, and a few other things along the way. I've looked at this way too long to - hopefully I haven't missed anything. Most I forget... One I do remember is the virStorageSourceCopy changes in patch 5 where I believe the "Get" and "Set" wanted to use the 'i' index rather than having the "Get" use the 'i' and the "Set" always use '0'. Matthias Gatto (5): virstoragefile: Add helper to get storage source backingStore virstoragefile: Add helper to set storage source backingStore virstoragefile: Use helper to get storage source backing store virstoragefile: Use helper to set storage source backing store virstoragefile: Convert storage source backingStore into backingStores src/conf/domain_conf.c | 25 +++++---- src/conf/storage_conf.c | 29 ++++++----- src/libvirt_private.syms | 2 + src/qemu/qemu_cgroup.c | 6 ++- src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_driver.c | 35 ++++++++----- src/qemu/qemu_monitor_json.c | 4 +- src/security/security_dac.c | 3 +- src/security/security_selinux.c | 5 +- src/security/virt-aa-helper.c | 2 +- src/storage/storage_backend.c | 34 +++++++----- src/storage/storage_backend_fs.c | 46 +++++++++------- src/storage/storage_backend_gluster.c | 11 ++-- src/storage/storage_backend_logical.c | 15 ++++-- src/storage/storage_driver.c | 3 +- src/util/virstoragefile.c | 98 ++++++++++++++++++++++++++++++----- src/util/virstoragefile.h | 10 +++- tests/virstoragetest.c | 18 +++---- 18 files changed, 243 insertions(+), 105 deletions(-) -- 2.5.0

2 6