Compare commits

..

12 Commits

Author SHA1 Message Date
Mark McLoughlin 21a9f1ec41 * Thu Oct 29 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-14
- Fix xen driver recounting (#531429)
- Fix crash on virsh error (#531429)
- Fix segfault where XML parsing fails in qemu disk hotplug
- Fix segfault where interface target device name is ommitted (#523418)
2009-10-29 10:40:45 +00:00
Mark McLoughlin 39bacac57d * Thu Oct 29 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-14
- Make libvirt-devel require libvirt-client, not libvirt
2009-10-29 10:14:00 +00:00
Mark McLoughlin 7d9775ba12 * Mon Oct 19 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-13
- Misc fixes to qemu machine types handling
- A couple of XML formatting fixes
2009-10-19 10:14:07 +00:00
Mark McLoughlin 3cfccddffa Add the second patch for #523158 2009-10-13 15:43:51 +00:00
Mark McLoughlin 1bc3776fdb * Tue Oct 13 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-12
- Fix restore of qemu guest using raw save format (#523158)
2009-10-13 15:34:19 +00:00
Mark McLoughlin 3a44160f46 * Fri Oct 9 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-11
- Fix libvirtd memory leak during error reply sending (#528162)
- Add several PCI hot-unplug typo fixes from upstream
2009-10-09 14:55:28 +00:00
Mark McLoughlin 5e8ea6c64c * Tue Oct 6 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-10
- Create /var/log/libvirt/{lxc,uml} dirs for logrotate
- Make libvirt-python dependon on libvirt-client
- Sync misc minor changes from upstream spec
2009-10-06 12:42:16 +00:00
Mark McLoughlin 00ce651fb8 * Tue Oct 6 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-9
- Change logrotate config to weekly (#526769)
2009-10-06 09:43:52 +00:00
Mark McLoughlin 3c684a55ed - Re-label qcow2 backing files (#497131) 2009-10-01 15:17:32 +00:00
Mark McLoughlin b1ea570e48 * Thu Oct 1 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-8
- Disable sound backend, even when selinux is disabled (#524499)
2009-10-01 08:35:16 +00:00
Mark McLoughlin 6ccf4c1a0c * Wed Sep 30 2009 Mark McLoughlin <markmc@redhat.com> - 0.7.1-7
- Fix USB device passthrough (#522683)
2009-09-30 17:57:50 +00:00
Jesse Keating dbaa6786af Initialize branch F-12 for libvirt 2009-09-29 05:24:04 +00:00
41 changed files with 5520 additions and 4514 deletions
-31
View File
@@ -1,31 +0,0 @@
[suppress_function]
symbol_version_regexp = LIBVIRT_PRIVATE.*
soname_regexp = libvirt\\.so.*
[suppress_function]
symbol_version_regexp = LIBVIRT_ADMIN_PRIVATE.*
soname_regexp = libvirt-admin\\.so.*
[suppress_variable]
symbol_version_regexp = LIBVIRT_PRIVATE.*
soname_regexp = libvirt\\.so.*
[suppress_variable]
symbol_version_regexp = LIBVIRT_ADMIN_PRIVATE.*
soname_regexp = libvirt-admin\\.so.*
[suppress_function]
symbol_version_regexp = .*
soname_regexp = libvirt_storage_.*\\.so.*
[suppress_variable]
symbol_version_regexp = .*
soname_regexp = libvirt_storage_.*\\.so.*
[suppress_function]
symbol_version_regexp = .*
soname_regexp = libvirt_driver_.*\\.so.*
[suppress_variable]
symbol_version_regexp = .*
soname_regexp = libvirt_driver_.*\\.so.*
+13
View File
@@ -0,0 +1,13 @@
.build*.log
*.rpm
i686
x86_64
libvirt-*.tar.gz
libvirt-0.6.0.tar.gz
libvirt-0.6.1.tar.gz
libvirt-0.6.2.tar.gz
libvirt-0.6.3.tar.gz
libvirt-0.6.4.tar.gz
libvirt-0.6.5.tar.gz
libvirt-0.7.0.tar.gz
libvirt-0.7.1.tar.gz
-5
View File
@@ -1,5 +0,0 @@
.build*.log
*.rpm
i686
x86_64
libvirt-*.tar.xz
@@ -1,29 +0,0 @@
From: Martin Kletzander <mkletzan@redhat.com>
Date: Fri, 16 Aug 2024 13:56:51 +0200
Subject: [PATCH] virarptable: Properly calculate rtattr length
Content-type: text/plain
Use convenience macro which does almost the same thing we were doing,
but also pads out the payload length to a multiple of NLMSG_ALIGNTO (4)
bytes.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
---
src/util/virarptable.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/util/virarptable.c b/src/util/virarptable.c
index 299dddd664..d8e41c5a86 100644
--- a/src/util/virarptable.c
+++ b/src/util/virarptable.c
@@ -102,8 +102,7 @@ virArpTableGet(void)
return table;
VIR_WARNINGS_NO_CAST_ALIGN
- parse_rtattr(tb, NDA_MAX, NDA_RTA(r),
- nh->nlmsg_len - NLMSG_LENGTH(sizeof(*r)));
+ parse_rtattr(tb, NDA_MAX, NDA_RTA(r), NLMSG_PAYLOAD(nh, sizeof(*r)));
VIR_WARNINGS_RESET
if (tb[NDA_DST] == NULL || tb[NDA_LLADDR] == NULL)
@@ -1,34 +0,0 @@
From: Martin Kletzander <mkletzan@redhat.com>
Date: Fri, 16 Aug 2024 13:59:15 +0200
Subject: [PATCH] virarptable: Fix check for message length
Content-type: text/plain
The previous check was all wrong since it calculated the how long would
the netlink message be if the netlink header was the payload and then
subtracted that from the whole message length, a variable that was not
used later in the code. This check can fail if there are no additional
payloads, struct rtattr in particular, which we are parsing later,
however the RTA_OK macro would've caught that anyway.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
---
src/util/virarptable.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/util/virarptable.c b/src/util/virarptable.c
index d8e41c5a86..45ee76766f 100644
--- a/src/util/virarptable.c
+++ b/src/util/virarptable.c
@@ -81,10 +81,9 @@ virArpTableGet(void)
for (; NLMSG_OK(nh, msglen); nh = NLMSG_NEXT(nh, msglen)) {
VIR_WARNINGS_RESET
struct ndmsg *r = NLMSG_DATA(nh);
- int len = nh->nlmsg_len;
void *addr;
- if ((len -= NLMSG_LENGTH(sizeof(*nh))) < 0) {
+ if (nh->nlmsg_len < NLMSG_SPACE(sizeof(*r))) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("wrong nlmsg len"));
goto cleanup;
@@ -1,46 +0,0 @@
From: Martin Kletzander <mkletzan@redhat.com>
Date: Fri, 16 Aug 2024 14:02:48 +0200
Subject: [PATCH] virarptable: End parsing earlier in case of NLMSG_DONE
Content-type: text/plain
Check for the last multipart message right as the first thing. The
presumption probably was that the last message might still contain a
payload we want to parse. However that cannot be true since that would
have to be a type RTM_NEWNEIGH. This was not caught because older
kernels were note sending NLMSG_DONE and probably relied on the fact
that the parsing just stops after all the messages are walked through,
which the NLMSG_OK macro successfully did.
Resolves: https://issues.redhat.com/browse/RHEL-52449
Resolves: https://bugzilla.redhat.com/2302245
Fixes: a176d67cdfaf5b8237a7e3a80d8be0e6bdf2d8fd
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
---
src/util/virarptable.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/util/virarptable.c b/src/util/virarptable.c
index 45ee76766f..20d11f97b0 100644
--- a/src/util/virarptable.c
+++ b/src/util/virarptable.c
@@ -83,6 +83,9 @@ virArpTableGet(void)
struct ndmsg *r = NLMSG_DATA(nh);
void *addr;
+ if (nh->nlmsg_type == NLMSG_DONE)
+ break;
+
if (nh->nlmsg_len < NLMSG_SPACE(sizeof(*r))) {
virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
_("wrong nlmsg len"));
@@ -97,9 +100,6 @@ virArpTableGet(void)
(!(r->ndm_state == NUD_STALE || r->ndm_state == NUD_REACHABLE)))
continue;
- if (nh->nlmsg_type == NLMSG_DONE)
- return table;
-
VIR_WARNINGS_NO_CAST_ALIGN
parse_rtattr(tb, NDA_MAX, NDA_RTA(r), NLMSG_PAYLOAD(nh, sizeof(*r)));
VIR_WARNINGS_RESET
@@ -1,315 +0,0 @@
From 807e2670f2704c41f0a1dca81a5d2f2f9336137c Mon Sep 17 00:00:00 2001
From: Laine Stump <laine@redhat.com>
Date: Mon, 25 Nov 2024 22:24:44 -0500
Subject: [PATCH 4/9] util: use a single flags arg for virNetDevBandwidthSet(),
not multiple bools
Having two bools in the arg list is on the borderline of being
confusing to anyone trying to read the code, but we're about to add a
3rd. This patch replaces the two bools with a single flags argument
which will instead have one or more bits from virNetDevBandwidthFlags
set.
Signed-off-by: Laine Stump <laine@redhat.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
---
src/lxc/lxc_driver.c | 8 ++++++--
src/lxc/lxc_process.c | 8 ++++++--
src/network/bridge_driver.c | 10 ++++++++--
src/qemu/qemu_command.c | 11 ++++++++---
src/qemu/qemu_driver.c | 29 ++++++++++++++-------------
src/qemu/qemu_hotplug.c | 22 +++++++++++++++------
src/util/virnetdevbandwidth.c | 36 ++++++++++++++++++++--------------
src/util/virnetdevbandwidth.h | 9 +++++++--
tests/virnetdevbandwidthtest.c | 8 +++++++-
9 files changed, 94 insertions(+), 47 deletions(-)
diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c
index 534e257f30..b693980dbb 100644
--- a/src/lxc/lxc_driver.c
+++ b/src/lxc/lxc_driver.c
@@ -3570,8 +3570,12 @@ lxcDomainAttachDeviceNetLive(virLXCDriver *driver,
actualBandwidth = virDomainNetGetActualBandwidth(net);
if (actualBandwidth) {
if (virNetDevSupportsBandwidth(actualType)) {
- if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false,
- !virDomainNetTypeSharesHostView(net)) < 0)
+ unsigned int flags = 0;
+
+ if (!virDomainNetTypeSharesHostView(net))
+ flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
+
+ if (virNetDevBandwidthSet(net->ifname, actualBandwidth, flags) < 0)
goto cleanup;
} else {
VIR_WARN("setting bandwidth on interfaces of "
diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c
index f5eb5383ec..0e689fbb70 100644
--- a/src/lxc/lxc_process.c
+++ b/src/lxc/lxc_process.c
@@ -605,8 +605,12 @@ virLXCProcessSetupInterfaces(virLXCDriver *driver,
actualBandwidth = virDomainNetGetActualBandwidth(net);
if (actualBandwidth) {
if (virNetDevSupportsBandwidth(type)) {
- if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false,
- !virDomainNetTypeSharesHostView(net)) < 0)
+ unsigned int flags = 0;
+
+ if (!virDomainNetTypeSharesHostView(net))
+ flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
+
+ if (virNetDevBandwidthSet(net->ifname, actualBandwidth, flags) < 0)
goto cleanup;
} else {
VIR_WARN("setting bandwidth on interfaces of "
diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index 32572c755f..1c53636450 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -2058,8 +2058,11 @@ networkStartNetworkVirtual(virNetworkDriverState *driver,
}
}
- if (virNetDevBandwidthSet(def->bridge, def->bandwidth, true, true) < 0)
+ if (virNetDevBandwidthSet(def->bridge, def->bandwidth,
+ VIR_NETDEV_BANDWIDTH_SET_HIERARCHICAL_CLASS
+ | VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED) < 0) {
goto error;
+ }
return 0;
@@ -2141,8 +2144,11 @@ networkStartNetworkBridge(virNetworkObj *obj)
* type BRIDGE, is started. On failure, undo anything you've done,
* and return -1. On success return 0.
*/
- if (virNetDevBandwidthSet(def->bridge, def->bandwidth, true, true) < 0)
+ if (virNetDevBandwidthSet(def->bridge, def->bandwidth,
+ VIR_NETDEV_BANDWIDTH_SET_HIERARCHICAL_CLASS
+ | VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED) < 0) {
goto error;
+ }
if (networkStartHandleMACTableManagerMode(obj) < 0)
goto error;
diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c
index f15e6bda1e..b4815e5e71 100644
--- a/src/qemu/qemu_command.c
+++ b/src/qemu/qemu_command.c
@@ -8840,9 +8840,14 @@ qemuBuildInterfaceCommandLine(virQEMUDriver *driver,
def->uuid,
!virDomainNetTypeSharesHostView(net)) < 0)
goto cleanup;
- } else if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false,
- !virDomainNetTypeSharesHostView(net)) < 0) {
- goto cleanup;
+ } else {
+ unsigned int flags = 0;
+
+ if (!virDomainNetTypeSharesHostView(net))
+ flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
+
+ if (virNetDevBandwidthSet(net->ifname, actualBandwidth, flags) < 0)
+ goto cleanup;
}
} else {
VIR_WARN("setting bandwidth on interfaces of "
diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c
index 736602333e..14929616e5 100644
--- a/src/qemu/qemu_driver.c
+++ b/src/qemu/qemu_driver.c
@@ -9941,21 +9941,22 @@ qemuDomainSetInterfaceParameters(virDomainPtr dom,
virErrorRestore(&orig_err);
goto endjob;
}
- } else if (virNetDevBandwidthSet(net->ifname, newBandwidth, false,
- !virDomainNetTypeSharesHostView(net)) < 0) {
- virErrorPtr orig_err;
-
- virErrorPreserveLast(&orig_err);
- ignore_value(virNetDevBandwidthSet(net->ifname,
- net->bandwidth,
- false,
- !virDomainNetTypeSharesHostView(net)));
- if (net->bandwidth) {
- ignore_value(virDomainNetBandwidthUpdate(net,
- net->bandwidth));
+ } else {
+ unsigned int bwflags = 0;
+
+ if (!virDomainNetTypeSharesHostView(net))
+ bwflags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
+
+ if (virNetDevBandwidthSet(net->ifname, newBandwidth, bwflags) < 0) {
+ virErrorPtr orig_err;
+
+ virErrorPreserveLast(&orig_err);
+ ignore_value(virNetDevBandwidthSet(net->ifname, net->bandwidth, bwflags));
+ if (net->bandwidth)
+ ignore_value(virDomainNetBandwidthUpdate(net, net->bandwidth));
+ virErrorRestore(&orig_err);
+ goto endjob;
}
- virErrorRestore(&orig_err);
- goto endjob;
}
/* If the old bandwidth was cleared out, restore qdisc. */
diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c
index 7cb1800504..d5e7e99359 100644
--- a/src/qemu/qemu_hotplug.c
+++ b/src/qemu/qemu_hotplug.c
@@ -1279,9 +1279,14 @@ qemuDomainAttachNetDevice(virQEMUDriver *driver,
vm->def->uuid,
!virDomainNetTypeSharesHostView(net)) < 0)
goto cleanup;
- } else if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false,
- !virDomainNetTypeSharesHostView(net)) < 0) {
- goto cleanup;
+ } else {
+ int flags = 0;
+
+ if (!virDomainNetTypeSharesHostView(net))
+ flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
+
+ if (virNetDevBandwidthSet(net->ifname, actualBandwidth, flags) < 0)
+ goto cleanup;
}
} else {
VIR_WARN("setting bandwidth on interfaces of "
@@ -4082,9 +4087,14 @@ qemuDomainChangeNet(virQEMUDriver *driver,
vm->def->uuid,
!virDomainNetTypeSharesHostView(newdev)) < 0)
goto cleanup;
- } else if (virNetDevBandwidthSet(newdev->ifname, newb, false,
- !virDomainNetTypeSharesHostView(newdev)) < 0) {
- goto cleanup;
+ } else {
+ int flags = 0;
+
+ if (!virDomainNetTypeSharesHostView(newdev))
+ flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
+
+ if (virNetDevBandwidthSet(newdev->ifname, newb, flags) < 0)
+ goto cleanup;
}
} else {
if (virDomainInterfaceClearQoS(vm->def, olddev) < 0)
diff --git a/src/util/virnetdevbandwidth.c b/src/util/virnetdevbandwidth.c
index 2b58c58d3e..1baad849c6 100644
--- a/src/util/virnetdevbandwidth.c
+++ b/src/util/virnetdevbandwidth.c
@@ -173,30 +173,35 @@ virNetDevBandwidthManipulateFilter(const char *ifname,
* virNetDevBandwidthSet:
* @ifname: on which interface
* @bandwidth: rates to set (may be NULL)
- * @hierarchical_class: whether to create hierarchical class
- * @swapped: true if IN/OUT should be set contrariwise
+ * @flags: bits indicating certain optional actions
*
+
* This function enables QoS on specified interface
* and set given traffic limits for both, incoming
- * and outgoing traffic. Any previous setting get
- * overwritten. If @hierarchical_class is TRUE, create
- * hierarchical class. It is used to guarantee minimal
- * throughput ('floor' attribute in NIC).
+ * and outgoing traffic.
+ *
+ * @flags bits and their meanings:
+ *
+ * VIR_NETDEV_BANDWIDTH_SET_HIERARCHICAL_CLASS
+ * whether to create a hierarchical class
+ * A hiearchical class structure is used to implement a minimal
+ * throughput guarantee ('floor' attribute in NIC).
*
- * If @swapped is set, the IN part of @bandwidth is set on
- * @ifname's TX, and vice versa. If it is not set, IN is set on
- * RX and OUT on TX. This is because for some types of interfaces
- * domain and the host live on the same side of the interface (so
- * domain's RX/TX is host's RX/TX), and for some it's swapped
- * (domain's RX/TX is hosts's TX/RX).
+ * VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED
+ * set if IN/OUT should be set backwards from what's indicated in
+ * the bandwidth, i.e. the IN part of @bandwidth is set on
+ * @ifname's TX, and the OUT part of @bandwidth is set on
+ * @ifname's RX. This is needed because for some types of
+ * interfaces the domain and the host live on the same side of the
+ * interface (so domain's RX/TX is host's RX/TX), and for some
+ * it's swapped (domain's RX/TX is hosts's TX/RX).
*
* Return 0 on success, -1 otherwise.
*/
int
virNetDevBandwidthSet(const char *ifname,
const virNetDevBandwidth *bandwidth,
- bool hierarchical_class,
- bool swapped)
+ unsigned int flags)
{
int ret = -1;
virNetDevBandwidthRate *rx = NULL; /* From domain POV */
@@ -205,6 +210,7 @@ virNetDevBandwidthSet(const char *ifname,
char *average = NULL;
char *peak = NULL;
char *burst = NULL;
+ bool hierarchical_class = flags & VIR_NETDEV_BANDWIDTH_SET_HIERARCHICAL_CLASS;
if (!bandwidth) {
/* nothing to be enabled */
@@ -224,7 +230,7 @@ virNetDevBandwidthSet(const char *ifname,
return -1;
}
- if (swapped) {
+ if (flags & VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED) {
rx = bandwidth->out;
tx = bandwidth->in;
} else {
diff --git a/src/util/virnetdevbandwidth.h b/src/util/virnetdevbandwidth.h
index 6d268fb119..80dc654486 100644
--- a/src/util/virnetdevbandwidth.h
+++ b/src/util/virnetdevbandwidth.h
@@ -39,11 +39,16 @@ void virNetDevBandwidthFree(virNetDevBandwidth *def);
G_DEFINE_AUTOPTR_CLEANUP_FUNC(virNetDevBandwidth, virNetDevBandwidthFree);
+typedef enum {
+ VIR_NETDEV_BANDWIDTH_SET_HIERARCHICAL_CLASS = (1 << 0),
+ VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED = (1 << 1),
+} virNetDevBandwidthSetFlags;
+
int virNetDevBandwidthSet(const char *ifname,
const virNetDevBandwidth *bandwidth,
- bool hierarchical_class,
- bool swapped)
+ unsigned int flags)
G_GNUC_WARN_UNUSED_RESULT;
+
int virNetDevBandwidthClear(const char *ifname);
int virNetDevBandwidthCopy(virNetDevBandwidth **dest,
const virNetDevBandwidth *src)
diff --git a/tests/virnetdevbandwidthtest.c b/tests/virnetdevbandwidthtest.c
index f7c38faa2e..6529ff4026 100644
--- a/tests/virnetdevbandwidthtest.c
+++ b/tests/virnetdevbandwidthtest.c
@@ -82,8 +82,14 @@ testVirNetDevBandwidthSet(const void *data)
if (virNetDevOpenvswitchInterfaceSetQos(iface, band, info->uuid, true) < 0)
return -1;
} else {
+ unsigned int flags = VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
+
+ if (info->hierarchical_class)
+ flags |= VIR_NETDEV_BANDWIDTH_SET_HIERARCHICAL_CLASS;
+
exp_cmd = info->exp_cmd_tc;
- if (virNetDevBandwidthSet(iface, band, info->hierarchical_class, true) < 0)
+
+ if (virNetDevBandwidthSet(iface, band, flags) < 0)
return -1;
}
--
2.47.1
@@ -1,185 +0,0 @@
From 490f58382dca2a415a5f16b6133f298d853bb379 Mon Sep 17 00:00:00 2001
From: Laine Stump <laine@redhat.com>
Date: Mon, 25 Nov 2024 22:24:45 -0500
Subject: [PATCH 5/9] util: make it optional to clear existing tc
qdiscs/filters in virNetDevBandwidthSet()
virNetDevBandwidthSet() always clears all existing qdiscs and their
subordinate filters before adding all the new qdiscs/filters. This is
normally exactly what we want, but there is one case (the network
driver) where the Qdisc added by virNetDevBandwidthSet() may already
be in use by the nftables backend (which will add a rule to fix the
checksum of dhcp packets); in that case, we *don't* want
virNetDevBandwidthSet() to clear out the qdisc that was already added
for nftables, and none of the bandwidth filters have been added yet,
so there already aren't any "old" filters that need to be removed
either - it is safe to just skip virNetDevBandwidthClear() in this
case.
To allow the network driver to set bandwidth without first clearing
it, this patch adds the flag VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL to the
virNetDevBandwidthSetFlags enum, and recognizes it in
virNetDevBandwidthSet() - if the flag is set, then
virNetDevBandwidth() will call virNetDevBandwidthClear() just as it
always has. But if the flag isn't set it *won't* call
virNetDevBandwidthClear().
As suggested above, VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL is set for all
calls to virNetdevBandwidthSet() except for two places in the network
driver.
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
---
src/lxc/lxc_driver.c | 2 +-
src/lxc/lxc_process.c | 2 +-
src/qemu/qemu_command.c | 2 +-
src/qemu/qemu_driver.c | 2 +-
src/qemu/qemu_hotplug.c | 4 ++--
src/util/virnetdevbandwidth.c | 21 ++++++++++++++++++++-
src/util/virnetdevbandwidth.h | 1 +
tests/virnetdevbandwidthtest.c | 3 ++-
8 files changed, 29 insertions(+), 8 deletions(-)
diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c
index b693980dbb..81581c74df 100644
--- a/src/lxc/lxc_driver.c
+++ b/src/lxc/lxc_driver.c
@@ -3570,7 +3570,7 @@ lxcDomainAttachDeviceNetLive(virLXCDriver *driver,
actualBandwidth = virDomainNetGetActualBandwidth(net);
if (actualBandwidth) {
if (virNetDevSupportsBandwidth(actualType)) {
- unsigned int flags = 0;
+ unsigned int flags = VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL;
if (!virDomainNetTypeSharesHostView(net))
flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c
index 0e689fbb70..081ce03a57 100644
--- a/src/lxc/lxc_process.c
+++ b/src/lxc/lxc_process.c
@@ -605,7 +605,7 @@ virLXCProcessSetupInterfaces(virLXCDriver *driver,
actualBandwidth = virDomainNetGetActualBandwidth(net);
if (actualBandwidth) {
if (virNetDevSupportsBandwidth(type)) {
- unsigned int flags = 0;
+ unsigned int flags = VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL;
if (!virDomainNetTypeSharesHostView(net))
flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c
index b4815e5e71..ed54fd4c5b 100644
--- a/src/qemu/qemu_command.c
+++ b/src/qemu/qemu_command.c
@@ -8841,7 +8841,7 @@ qemuBuildInterfaceCommandLine(virQEMUDriver *driver,
!virDomainNetTypeSharesHostView(net)) < 0)
goto cleanup;
} else {
- unsigned int flags = 0;
+ unsigned int flags = VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL;
if (!virDomainNetTypeSharesHostView(net))
flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c
index 14929616e5..9549065b1f 100644
--- a/src/qemu/qemu_driver.c
+++ b/src/qemu/qemu_driver.c
@@ -9942,7 +9942,7 @@ qemuDomainSetInterfaceParameters(virDomainPtr dom,
goto endjob;
}
} else {
- unsigned int bwflags = 0;
+ unsigned int bwflags = VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL;
if (!virDomainNetTypeSharesHostView(net))
bwflags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c
index d5e7e99359..ceda4119cd 100644
--- a/src/qemu/qemu_hotplug.c
+++ b/src/qemu/qemu_hotplug.c
@@ -1280,7 +1280,7 @@ qemuDomainAttachNetDevice(virQEMUDriver *driver,
!virDomainNetTypeSharesHostView(net)) < 0)
goto cleanup;
} else {
- int flags = 0;
+ int flags = VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL;
if (!virDomainNetTypeSharesHostView(net))
flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
@@ -4088,7 +4088,7 @@ qemuDomainChangeNet(virQEMUDriver *driver,
!virDomainNetTypeSharesHostView(newdev)) < 0)
goto cleanup;
} else {
- int flags = 0;
+ int flags = VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL;
if (!virDomainNetTypeSharesHostView(newdev))
flags |= VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
diff --git a/src/util/virnetdevbandwidth.c b/src/util/virnetdevbandwidth.c
index 1baad849c6..9c48844c5d 100644
--- a/src/util/virnetdevbandwidth.c
+++ b/src/util/virnetdevbandwidth.c
@@ -196,6 +196,21 @@ virNetDevBandwidthManipulateFilter(const char *ifname,
* interface (so domain's RX/TX is host's RX/TX), and for some
* it's swapped (domain's RX/TX is hosts's TX/RX).
*
+ * VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL
+ * If VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL is set, then the root
+ * qdisc is deleted before adding any new qdisc/class/filter,
+ * which causes any pre-existing filters to also be deleted. If
+ * not set, then it's assumed that there are no existing rules (or
+ * that those already there need to be kept). The caller should
+ * set this flag for an existing interface that is having its
+ * bandwidth settings modified, but can leave it unset if the
+ * interface was newly created and this is the first time
+ * bandwidth has been set, but someone else might have already
+ * added the qdisc (e.g. this is the case when the network driver
+ * is setting bandwidth for a virtual network bridge device - the
+ * nftables backend may have already added qdisc handle 1:0 and a
+ * filter, and we don't want to delete them)
+ *
* Return 0 on success, -1 otherwise.
*/
int
@@ -238,7 +253,11 @@ virNetDevBandwidthSet(const char *ifname,
tx = bandwidth->out;
}
- virNetDevBandwidthClear(ifname);
+ /* Only if the caller requests, clear everything including root
+ * qdisc and all filters before adding everything.
+ */
+ if (flags & VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL)
+ virNetDevBandwidthClear(ifname);
if (tx && tx->average) {
average = g_strdup_printf("%llukbps", tx->average);
diff --git a/src/util/virnetdevbandwidth.h b/src/util/virnetdevbandwidth.h
index 80dc654486..744aa4c826 100644
--- a/src/util/virnetdevbandwidth.h
+++ b/src/util/virnetdevbandwidth.h
@@ -42,6 +42,7 @@ G_DEFINE_AUTOPTR_CLEANUP_FUNC(virNetDevBandwidth, virNetDevBandwidthFree);
typedef enum {
VIR_NETDEV_BANDWIDTH_SET_HIERARCHICAL_CLASS = (1 << 0),
VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED = (1 << 1),
+ VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL = (1 << 2),
} virNetDevBandwidthSetFlags;
int virNetDevBandwidthSet(const char *ifname,
diff --git a/tests/virnetdevbandwidthtest.c b/tests/virnetdevbandwidthtest.c
index 6529ff4026..6d5c847ad7 100644
--- a/tests/virnetdevbandwidthtest.c
+++ b/tests/virnetdevbandwidthtest.c
@@ -82,7 +82,8 @@ testVirNetDevBandwidthSet(const void *data)
if (virNetDevOpenvswitchInterfaceSetQos(iface, band, info->uuid, true) < 0)
return -1;
} else {
- unsigned int flags = VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED;
+ unsigned int flags = VIR_NETDEV_BANDWIDTH_SET_DIR_SWAPPED |
+ VIR_NETDEV_BANDWIDTH_SET_CLEAR_ALL;
if (info->hierarchical_class)
flags |= VIR_NETDEV_BANDWIDTH_SET_HIERARCHICAL_CLASS;
--
2.47.1
@@ -1,98 +0,0 @@
From faebbbbfa3b1bd4120852b3f416c8073ab82d5c5 Mon Sep 17 00:00:00 2001
From: Laine Stump <laine@redhat.com>
Date: Mon, 25 Nov 2024 22:24:46 -0500
Subject: [PATCH 6/9] util: put the command that adds a tx filter qdisc into a
separate function
virNetDevBandwidthSet() adds a queue discipline (qdisc) for each
interface that it will need to add tc transmit filters to, and the
filters are then attached to the qdisc.
There are other circumstances where some other function will need to
add tc transmit filters to an interface (in particular an upcoming
patch to the network driver nftables backend that will use a tc tx
filter to fix the checksum of dhcp packets), so that function will
also need a qdisc for the tx filter. To assure both always use exactly
the same qdisc, this patch puts the command that adds the tx filter
qdisc into a separate helper function that can (and will) be called
from either place
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
---
src/libvirt_private.syms | 1 +
src/util/virnetdevbandwidth.c | 30 +++++++++++++++++++++++++-----
src/util/virnetdevbandwidth.h | 3 +++
3 files changed, 29 insertions(+), 5 deletions(-)
diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms
index d15d6a6a9d..0211cee967 100644
--- a/src/libvirt_private.syms
+++ b/src/libvirt_private.syms
@@ -2859,6 +2859,7 @@ virNetDevVFInterfaceStats;
# util/virnetdevbandwidth.h
+virNetDevBandWidthAddTxFilterParentQdisc;
virNetDevBandwidthClear;
virNetDevBandwidthCopy;
virNetDevBandwidthEqual;
diff --git a/src/util/virnetdevbandwidth.c b/src/util/virnetdevbandwidth.c
index 9c48844c5d..90eebe6576 100644
--- a/src/util/virnetdevbandwidth.c
+++ b/src/util/virnetdevbandwidth.c
@@ -266,11 +266,7 @@ virNetDevBandwidthSet(const char *ifname,
if (tx->burst)
burst = g_strdup_printf("%llukb", tx->burst);
- cmd = virCommandNew(TC);
- virCommandAddArgList(cmd, "qdisc", "add", "dev", ifname, "root",
- "handle", "1:", "htb", "default",
- hierarchical_class ? "2" : "1", NULL);
- if (virCommandRun(cmd, NULL) < 0)
+ if (virNetDevBandWidthAddTxFilterParentQdisc(ifname, hierarchical_class) < 0)
goto cleanup;
/* If we are creating a hierarchical class, all non guaranteed traffic
@@ -794,3 +790,27 @@ virNetDevBandwidthSetRootQDisc(const char *ifname,
return 0;
}
+
+/**
+ * virNetDevBandwidthAddTxFilterParentQdisc:
+ * @ifname: name of interface that needs a qdisc to attach tx filters to
+ * @hierarchical_class: true if hierarchical classes will be used on this interface
+ *
+ * Add a root Qdisc (Queueing Discipline) for attaching Tx filters to
+ * @ifname.
+ *
+ * returns 0 on success, -1 on failure
+ */
+int
+virNetDevBandWidthAddTxFilterParentQdisc(const char *ifname,
+ bool hierarchical_class)
+{
+ g_autoptr(virCommand) cmd = NULL;
+
+ cmd = virCommandNew(TC);
+ virCommandAddArgList(cmd, "qdisc", "add", "dev", ifname, "root",
+ "handle", "1:", "htb", "default",
+ hierarchical_class ? "2" : "1", NULL);
+
+ return virCommandRun(cmd, NULL);
+}
diff --git a/src/util/virnetdevbandwidth.h b/src/util/virnetdevbandwidth.h
index 744aa4c826..65c1500637 100644
--- a/src/util/virnetdevbandwidth.h
+++ b/src/util/virnetdevbandwidth.h
@@ -84,3 +84,6 @@ int virNetDevBandwidthUpdateFilter(const char *ifname,
int virNetDevBandwidthSetRootQDisc(const char *ifname,
const char *qdisc)
G_NO_INLINE;
+
+int virNetDevBandWidthAddTxFilterParentQdisc(const char *ifname,
+ bool hierarchical_class);
--
2.47.1
@@ -1,107 +0,0 @@
From 73c0fb19ce5b816ee81ede691252855c75391c9a Mon Sep 17 00:00:00 2001
From: Laine Stump <laine@redhat.com>
Date: Mon, 25 Nov 2024 22:24:47 -0500
Subject: [PATCH 7/9] util: don't re-add the qdisc used for tx filters if it
already exists
There will soon be two separate users of tc on virtual networks, and
both will use the "qdisc root handle 1: htb" to add tx filters. One or the
other could get the first chance to add the qdisc, and then if at a
later time the other decides to use it, we need to prevent the 2nd
user from attempting to re-add the qdisc (because that just generates
an error).
We do this by running "tc qdisc show dev $bridge handle 1:" then
checking if the output of that command contains both "qdisc" and " 1:
".[*] If it does then the qdisc has already been added. If not then we
need to add it now.
[*]As of this writing, the output more exactly starts with "qdisc
htb 1: root", but our comparison is made purposefully generous to
increase the chances that it will continue to work properly if tc
modifies the format of its output.
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
---
src/util/virnetdevbandwidth.c | 35 ++++++++++++++++++++++++++++------
tests/virnetdevbandwidthtest.c | 3 +++
2 files changed, 32 insertions(+), 6 deletions(-)
diff --git a/src/util/virnetdevbandwidth.c b/src/util/virnetdevbandwidth.c
index 90eebe6576..5c6a65528c 100644
--- a/src/util/virnetdevbandwidth.c
+++ b/src/util/virnetdevbandwidth.c
@@ -805,12 +805,35 @@ int
virNetDevBandWidthAddTxFilterParentQdisc(const char *ifname,
bool hierarchical_class)
{
- g_autoptr(virCommand) cmd = NULL;
+ g_autoptr(virCommand) testCmd = NULL;
+ g_autofree char *testResult = NULL;
- cmd = virCommandNew(TC);
- virCommandAddArgList(cmd, "qdisc", "add", "dev", ifname, "root",
- "handle", "1:", "htb", "default",
- hierarchical_class ? "2" : "1", NULL);
+ /* first check it the qdisc with handle 1: was already added for
+ * this interface by someone else
+ */
+ testCmd = virCommandNew(TC);
+ virCommandAddArgList(testCmd, "qdisc", "show", "dev", ifname,
+ "handle", "1:", NULL);
+ virCommandSetOutputBuffer(testCmd, &testResult);
- return virCommandRun(cmd, NULL);
+ if (virCommandRun(testCmd, NULL) < 0)
+ return -1;
+
+ /* output will be something like: "qdisc htb 1: root refcnt ..."
+ * if the qdisc was already added. We just search for "qdisc" and
+ * " 1: " anywhere in the output to allow for tc changing its
+ * output format.
+ */
+ if (!(testResult && strstr(testResult, "qdisc") && strstr(testResult, " 1: "))) {
+ /* didn't find qdisc in output, so we need to add one */
+ g_autoptr(virCommand) addCmd = virCommandNew(TC);
+
+ virCommandAddArgList(addCmd, "qdisc", "add", "dev", ifname, "root",
+ "handle", "1:", "htb", "default",
+ hierarchical_class ? "2" : "1", NULL);
+
+ return virCommandRun(addCmd, NULL);
+ }
+
+ return 0;
}
diff --git a/tests/virnetdevbandwidthtest.c b/tests/virnetdevbandwidthtest.c
index 6d5c847ad7..31aa7f469d 100644
--- a/tests/virnetdevbandwidthtest.c
+++ b/tests/virnetdevbandwidthtest.c
@@ -147,6 +147,7 @@ mymain(void)
"</bandwidth>",
TC " qdisc del dev eth0 root\n"
TC " qdisc del dev eth0 ingress\n"
+ TC " qdisc show dev eth0 handle 1:\n"
TC " qdisc add dev eth0 root handle 1: htb default 1\n"
TC " class add dev eth0 parent 1: classid 1:1 htb rate 1024kbps quantum 87\n"
TC " qdisc add dev eth0 parent 1:1 handle 2: sfq perturb 10\n"
@@ -177,6 +178,7 @@ mymain(void)
"</bandwidth>",
TC " qdisc del dev eth0 root\n"
TC " qdisc del dev eth0 ingress\n"
+ TC " qdisc show dev eth0 handle 1:\n"
TC " qdisc add dev eth0 root handle 1: htb default 1\n"
TC " class add dev eth0 parent 1: classid 1:1 htb rate 1kbps ceil 2kbps burst 4kb quantum 1\n"
TC " qdisc add dev eth0 parent 1:1 handle 2: sfq perturb 10\n"
@@ -199,6 +201,7 @@ mymain(void)
"</bandwidth>",
TC " qdisc del dev eth0 root\n"
TC " qdisc del dev eth0 ingress\n"
+ TC " qdisc show dev eth0 handle 1:\n"
TC " qdisc add dev eth0 root handle 1: htb default 1\n"
TC " class add dev eth0 parent 1: classid 1:1 htb rate 4294967295kbps quantum 366503875\n"
TC " qdisc add dev eth0 parent 1:1 handle 2: sfq perturb 10\n"
--
2.47.1
@@ -1,171 +0,0 @@
From dac9cb9030ac03d18f59884864a0a253e3c9f8f1 Mon Sep 17 00:00:00 2001
From: Laine Stump <laine@redhat.com>
Date: Mon, 25 Nov 2024 22:24:48 -0500
Subject: [PATCH 8/9] util: add new "tc" layer for virFirewallCmd objects
If the layer of a virFirewallCmd is "tc", then the "tc" utility will
be executed using the arguments that had been added to the
virFirewallCmd
tc layer doesn't support auto-rollback command creation (any rollback
needs to be added manually with virFirewallAddRollbackCmd()), and also
tc layer isn't supported by the iptables backend (it would have been
straightforward to add, but the iptables backend doesn't need it, and
I didn't want to take the chance of causing a regression in that
code for no good reason).
Signed-off-by: Laine Stump <laine@redhat.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
---
src/network/network_nftables.c | 1 +
src/util/virfirewall.c | 66 +++++++++++++++++++++-------------
src/util/virfirewall.h | 1 +
src/util/virfirewalld.c | 1 +
4 files changed, 44 insertions(+), 25 deletions(-)
diff --git a/src/network/network_nftables.c b/src/network/network_nftables.c
index 268d1f12ca..cc184105c3 100644
--- a/src/network/network_nftables.c
+++ b/src/network/network_nftables.c
@@ -73,6 +73,7 @@ VIR_ENUM_IMPL(nftablesLayer,
"",
"ip",
"ip6",
+ "",
);
diff --git a/src/util/virfirewall.c b/src/util/virfirewall.c
index 811b787ecc..9389bcf541 100644
--- a/src/util/virfirewall.c
+++ b/src/util/virfirewall.c
@@ -47,6 +47,7 @@ VIR_ENUM_IMPL(virFirewallLayer,
"ethernet",
"ipv4",
"ipv6",
+ "tc",
);
typedef struct _virFirewallGroup virFirewallGroup;
@@ -57,6 +58,7 @@ VIR_ENUM_IMPL(virFirewallLayerCommand,
EBTABLES,
IPTABLES,
IP6TABLES,
+ TC,
);
struct _virFirewallCmd {
@@ -591,6 +593,7 @@ virFirewallCmdIptablesApply(virFirewall *firewall,
case VIR_FIREWALL_LAYER_IPV6:
virCommandAddArg(cmd, "-w");
break;
+ case VIR_FIREWALL_LAYER_TC:
case VIR_FIREWALL_LAYER_LAST:
break;
}
@@ -672,39 +675,52 @@ virFirewallCmdNftablesApply(virFirewall *firewall G_GNUC_UNUSED,
size_t i;
int status;
- cmd = virCommandNew(NFT);
+ if (fwCmd->layer == VIR_FIREWALL_LAYER_TC) {
- if ((virFirewallTransactionGetFlags(firewall) & VIR_FIREWALL_TRANSACTION_AUTO_ROLLBACK) &&
- fwCmd->argsLen > 1) {
- /* skip any leading options to get to command verb */
- for (i = 0; i < fwCmd->argsLen - 1; i++) {
- if (fwCmd->args[i][0] != '-')
- break;
- }
+ /* for VIR_FIREWALL_LAYER_TC, we run the 'tc' (traffic control) command with
+ * the supplied args.
+ */
+ cmd = virCommandNew(TC);
- if (i + 1 < fwCmd->argsLen &&
- VIR_NFTABLES_ARG_IS_CREATE(fwCmd->args[i])) {
+ /* NB: RAW commands don't support auto-rollback command creation */
- cmdIdx = i;
- objectType = fwCmd->args[i + 1];
+ } else {
- /* we currently only handle auto-rollback for rules,
- * chains, and tables, and those all can be "rolled
- * back" by a delete command using the handle that is
- * returned when "-ae" is added to the add/insert
- * command.
- */
- if (STREQ_NULLABLE(objectType, "rule") ||
- STREQ_NULLABLE(objectType, "chain") ||
- STREQ_NULLABLE(objectType, "table")) {
+ cmd = virCommandNew(NFT);
- needRollback = true;
- /* this option to nft instructs it to add the
- * "handle" of the created object to stdout
+ if ((virFirewallTransactionGetFlags(firewall) & VIR_FIREWALL_TRANSACTION_AUTO_ROLLBACK) &&
+ fwCmd->argsLen > 1) {
+ /* skip any leading options to get to command verb */
+ for (i = 0; i < fwCmd->argsLen - 1; i++) {
+ if (fwCmd->args[i][0] != '-')
+ break;
+ }
+
+ if (i + 1 < fwCmd->argsLen &&
+ VIR_NFTABLES_ARG_IS_CREATE(fwCmd->args[i])) {
+
+ cmdIdx = i;
+ objectType = fwCmd->args[i + 1];
+
+ /* we currently only handle auto-rollback for rules,
+ * chains, and tables, and those all can be "rolled
+ * back" by a delete command using the handle that is
+ * returned when "-ae" is added to the add/insert
+ * command.
*/
- virCommandAddArg(cmd, "-ae");
+ if (STREQ_NULLABLE(objectType, "rule") ||
+ STREQ_NULLABLE(objectType, "chain") ||
+ STREQ_NULLABLE(objectType, "table")) {
+
+ needRollback = true;
+ /* this option to nft instructs it to add the
+ * "handle" of the created object to stdout
+ */
+ virCommandAddArg(cmd, "-ae");
+ }
}
}
+
}
for (i = 0; i < fwCmd->argsLen; i++)
diff --git a/src/util/virfirewall.h b/src/util/virfirewall.h
index bce51259d2..d42e60884b 100644
--- a/src/util/virfirewall.h
+++ b/src/util/virfirewall.h
@@ -39,6 +39,7 @@ typedef enum {
VIR_FIREWALL_LAYER_ETHERNET,
VIR_FIREWALL_LAYER_IPV4,
VIR_FIREWALL_LAYER_IPV6,
+ VIR_FIREWALL_LAYER_TC,
VIR_FIREWALL_LAYER_LAST,
} virFirewallLayer;
diff --git a/src/util/virfirewalld.c b/src/util/virfirewalld.c
index 827e201dbb..124523c420 100644
--- a/src/util/virfirewalld.c
+++ b/src/util/virfirewalld.c
@@ -43,6 +43,7 @@ VIR_LOG_INIT("util.firewalld");
VIR_ENUM_DECL(virFirewallLayerFirewallD);
VIR_ENUM_IMPL(virFirewallLayerFirewallD,
VIR_FIREWALL_LAYER_LAST,
+ "",
"eb",
"ipv4",
"ipv6",
--
2.47.1
@@ -1,687 +0,0 @@
From b1e2318a0d609fcdff04fcf88953ea87cdd02b95 Mon Sep 17 00:00:00 2001
From: Laine Stump <laine@redhat.com>
Date: Mon, 25 Nov 2024 22:24:49 -0500
Subject: [PATCH 9/9] network: add tc filter rule to nftables backend to fix
checksum of DHCP responses
Please see the commit log for commit v10.9.0-rc1-1-g42ab0148dd for the
history and explanation of the problem that this patch is fixing.
A shorter explanation is that when a guest is connected to a libvirt
virtual network using a virtio-net adapter with in-kernel "vhost-net"
packet processing enabled, it will fail to acquire an IP address from
a DHCP seever running on the host.
In commit v10.9.0-rc1-1-g42ab0148dd we tried fixing this by *zeroing
out* the checksums of these packets with an nftables rule (nftables
can't recompute the checksum, but it can set it to 0) . This
*appeared* to work initially, but it turned out that zeroing the
checksum ends up breaking dhcp packets on *non* virtio/vhost-net guest
interfaces. That attempt was reverted in commit v10.9.0-rc2.
Fortunately, there is an existing way to recompute the checksum of a
packet as it leaves an interface - the "tc" (traffic control) utility
that libvirt already uses for bandwidth management. This patch uses a
tc filter rule to match dhcp response packets on the bridge and
recompute their checksum.
The filter rule must be attached to a tc qdisc, which may also have a
filter attached for bandwidth management (in the <bandwidth> element
of the network config). Not only must we add the qdisc only once
(which was already handled by the patch two prior to this one), but
also the filter rule for checksum fixing and the filter rule for
bandwidth management must be different priorities so they don't clash;
this is solved by adding the checksum-fix filter with "priority 2",
while the bandwidth management filter remains "priority 1" (both will
always be evaluated anyway, it's just a matter of which is evaluated
first).
So far this method has worked with every different guest we could
throw at it, including several that failed with the previous method.
Fixes: b89c4991daa0ee9371f10937fab3b03c5ffdabc6
Reported-by: Rich Jones <rjones@redhat.com>
Reported-by: Andrea Bolognani <abologna@redhat.com>
Fix-Suggested-by: Eric Garver <egarver@redhat.com>
Fix-Suggested-by: Phil Sutter <psutter@redhat.com>
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
---
src/network/network_nftables.c | 68 +++++++++++++++++++
.../forward-dev-linux.nftables | 40 +++++++++++
.../isolated-linux.nftables | 40 +++++++++++
.../nat-default-linux.nftables | 40 +++++++++++
.../nat-ipv6-linux.nftables | 40 +++++++++++
.../nat-ipv6-masquerade-linux.nftables | 40 +++++++++++
.../nat-many-ips-linux.nftables | 40 +++++++++++
.../nat-no-dhcp-linux.nftables | 40 +++++++++++
.../nat-port-range-ipv6-linux.nftables | 40 +++++++++++
.../nat-port-range-linux.nftables | 40 +++++++++++
.../nat-tftp-linux.nftables | 40 +++++++++++
.../route-default-linux.nftables | 40 +++++++++++
12 files changed, 508 insertions(+)
diff --git a/src/network/network_nftables.c b/src/network/network_nftables.c
index cc184105c3..748edb0273 100644
--- a/src/network/network_nftables.c
+++ b/src/network/network_nftables.c
@@ -29,6 +29,7 @@
#include "internal.h"
#include "virfirewalld.h"
+#include "vircommand.h"
#include "virerror.h"
#include "virlog.h"
#include "virhash.h"
@@ -924,6 +925,67 @@ nftablesAddIPSpecificFirewallRules(virFirewall *fw,
}
+/**
+ * nftablesAddUdpChecksumFixWithTC:
+ *
+ * Add a tc filter rule to @ifname (the bridge device of this network)
+ * that will recompute the checksum of udp packets output from @iface with
+ * destination port @port.
+ *
+ * Normally the checksum should be filled by some part of the basic
+ * network stack, but there are cases (e.g. DHCP response packets sent
+ * from virtualization host to a QEMU guest when the guest NIC uses
+ * vhost-net packet processing) when the host (sender) thinks that
+ * packet checksums will be computed elsewhere (and so leaves a
+ * partially computed checksum in the packet header) while the guest
+ * (receiver) thinks that the checksum has already been fully
+ * computed; in the meantime none of the code in between has actually
+ * finished computing the checksum.
+ *
+ * An example of this is DHCP response packets from host to guest. If
+ * the checksum of each of these packets isn't properly computed, then
+ * many guests (e.g. FreeBSD) will drop them with reason BAD CHECKSUM;
+ * this tc filter rule will fix the ip and udp checksums, and the
+ * FreeBSD dhcp client will happily accept the packet.
+ *
+ * (NB: if you're wondering how the tc qdisc and filter are removed
+ * when the network is destroyed, the answer is that the kernel
+ * automatically (and properly) removes them for us, so we don't need
+ * to worry about keeping track/deleting as we do with nftables rules)
+ */
+static int
+nftablesAddUdpChecksumFixWithTC(virFirewall *fw,
+ const char *iface,
+ int port)
+{
+ g_autofree char *portstr = g_strdup_printf("%d", port);
+
+ /* this will add the qdisc (that the filter below is attached to)
+ * unless it already exists
+ */
+ if (virNetDevBandWidthAddTxFilterParentQdisc(iface, true) < 0)
+ return -1;
+
+ /* add a filter to catch all udp packets with dst "port" and
+ * recompute their checksum
+ */
+ virFirewallAddCmd(fw, VIR_FIREWALL_LAYER_TC,
+ "filter", "add", "dev", iface,
+ "prio", "2", "protocol", "ip", "parent", "1:",
+ "u32", "match", "ip", "dport", portstr, "ffff",
+ "action", "csum", "ip", "and", "udp",
+ NULL);
+
+ virFirewallAddRollbackCmd(fw, VIR_FIREWALL_LAYER_TC,
+ "filter", "del", "dev", iface,
+ "prio", "2", "protocol", "ip", "parent", "1:",
+ "u32", "match", "ip", "dport", portstr, "ffff",
+ "action", "csum", "ip", "and", "udp",
+ NULL);
+ return 0;
+}
+
+
/* nftablesAddFirewallrules:
*
* @def - the network that needs an nftables firewall added
@@ -944,6 +1006,12 @@ nftablesAddFirewallRules(virNetworkDef *def, virFirewall **fwRemoval)
virFirewallStartTransaction(fw, VIR_FIREWALL_TRANSACTION_AUTO_ROLLBACK);
+ /* add the tc filter rule needed to fixup the checksum of dhcp
+ * response packets going from host to guest.
+ */
+ if (nftablesAddUdpChecksumFixWithTC(fw, def->bridge, 68) < 0)
+ return -1;
+
nftablesAddGeneralFirewallRules(fw, def);
for (i = 0;
diff --git a/tests/networkxml2firewalldata/forward-dev-linux.nftables b/tests/networkxml2firewalldata/forward-dev-linux.nftables
index 8badb74beb..6772383b37 100644
--- a/tests/networkxml2firewalldata/forward-dev-linux.nftables
+++ b/tests/networkxml2firewalldata/forward-dev-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/isolated-linux.nftables b/tests/networkxml2firewalldata/isolated-linux.nftables
index d1b4dac178..546a18b75a 100644
--- a/tests/networkxml2firewalldata/isolated-linux.nftables
+++ b/tests/networkxml2firewalldata/isolated-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/nat-default-linux.nftables b/tests/networkxml2firewalldata/nat-default-linux.nftables
index 28508292f9..08623c1381 100644
--- a/tests/networkxml2firewalldata/nat-default-linux.nftables
+++ b/tests/networkxml2firewalldata/nat-default-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/nat-ipv6-linux.nftables b/tests/networkxml2firewalldata/nat-ipv6-linux.nftables
index d8a9ba706d..3fd6b94eef 100644
--- a/tests/networkxml2firewalldata/nat-ipv6-linux.nftables
+++ b/tests/networkxml2firewalldata/nat-ipv6-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/nat-ipv6-masquerade-linux.nftables b/tests/networkxml2firewalldata/nat-ipv6-masquerade-linux.nftables
index a7f09cda59..2811e098d1 100644
--- a/tests/networkxml2firewalldata/nat-ipv6-masquerade-linux.nftables
+++ b/tests/networkxml2firewalldata/nat-ipv6-masquerade-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/nat-many-ips-linux.nftables b/tests/networkxml2firewalldata/nat-many-ips-linux.nftables
index b826fe6134..5409d5b552 100644
--- a/tests/networkxml2firewalldata/nat-many-ips-linux.nftables
+++ b/tests/networkxml2firewalldata/nat-many-ips-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/nat-no-dhcp-linux.nftables b/tests/networkxml2firewalldata/nat-no-dhcp-linux.nftables
index d8a9ba706d..3fd6b94eef 100644
--- a/tests/networkxml2firewalldata/nat-no-dhcp-linux.nftables
+++ b/tests/networkxml2firewalldata/nat-no-dhcp-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/nat-port-range-ipv6-linux.nftables b/tests/networkxml2firewalldata/nat-port-range-ipv6-linux.nftables
index ceaed6fa40..d74417cdb3 100644
--- a/tests/networkxml2firewalldata/nat-port-range-ipv6-linux.nftables
+++ b/tests/networkxml2firewalldata/nat-port-range-ipv6-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/nat-port-range-linux.nftables b/tests/networkxml2firewalldata/nat-port-range-linux.nftables
index 1dc37a26ec..b55bb287a9 100644
--- a/tests/networkxml2firewalldata/nat-port-range-linux.nftables
+++ b/tests/networkxml2firewalldata/nat-port-range-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/nat-tftp-linux.nftables b/tests/networkxml2firewalldata/nat-tftp-linux.nftables
index 28508292f9..08623c1381 100644
--- a/tests/networkxml2firewalldata/nat-tftp-linux.nftables
+++ b/tests/networkxml2firewalldata/nat-tftp-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
diff --git a/tests/networkxml2firewalldata/route-default-linux.nftables b/tests/networkxml2firewalldata/route-default-linux.nftables
index 282c9542a5..76d6902517 100644
--- a/tests/networkxml2firewalldata/route-default-linux.nftables
+++ b/tests/networkxml2firewalldata/route-default-linux.nftables
@@ -1,3 +1,43 @@
+tc \
+qdisc \
+show \
+dev \
+virbr0 \
+handle \
+1:
+tc \
+qdisc \
+add \
+dev \
+virbr0 \
+root \
+handle \
+1: \
+htb \
+default \
+2
+tc \
+filter \
+add \
+dev \
+virbr0 \
+prio \
+2 \
+protocol \
+ip \
+parent \
+1: \
+u32 \
+match \
+ip \
+dport \
+68 \
+ffff \
+action \
+csum \
+ip \
+and \
+udp
nft \
-ae insert \
rule \
--
2.47.1
@@ -1,51 +0,0 @@
From 114c0ec656e879ab4d67919914bb24cf5993106d Mon Sep 17 00:00:00 2001
Message-ID: <114c0ec656e879ab4d67919914bb24cf5993106d.1734201785.git.crobinso@redhat.com>
From: Laine Stump <laine@redhat.com>
Date: Mon, 2 Sep 2024 16:13:08 -0400
Subject: [PATCH] network: permit <forward mode='open'/> when a network has no
IP address
Content-type: text/plain
The whole point of <forward mode='open'/> is to supress libvirt from
adding any firewall rules for a network, and someone might want to
create a network with no IP address (i.e. they don't want the guests
to have connectivity to the host via this interface) and no firewall
rules (they don't want any, or they want to add their own). So there's
no reason to fail when a network has <forward mode='open'/> and also
has no IP address.
Kind-of-Resolves: https://gitlab.com/libvirt/libvirt/-/issues/588
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
Signed-off-by: Cole Robinson <crobinso@redhat.com>
---
src/conf/network_conf.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c
index 5cf419acf1..320e1b089a 100644
--- a/src/conf/network_conf.c
+++ b/src/conf/network_conf.c
@@ -1789,7 +1789,6 @@ virNetworkDefParseXML(xmlXPathContextPtr ctxt,
case VIR_NETWORK_FORWARD_ROUTE:
case VIR_NETWORK_FORWARD_NAT:
- case VIR_NETWORK_FORWARD_OPEN:
/* It's pointless to specify L3 forwarding without specifying
* the network we're on.
*/
@@ -1806,8 +1805,10 @@ virNetworkDefParseXML(xmlXPathContextPtr ctxt,
def->name);
return NULL;
}
+ break;
- if (def->forward.type == VIR_NETWORK_FORWARD_OPEN && def->forward.nifs) {
+ case VIR_NETWORK_FORWARD_OPEN:
+ if (def->forward.nifs) {
/* an open network by definition can't place any restrictions
* on what traffic is allowed or where it goes, so specifying
* a forwarding device is nonsensical.
--
2.47.1
@@ -1,64 +0,0 @@
From d51179fa82448f4720f1645f0b7100df80508cc4 Mon Sep 17 00:00:00 2001
From: Pavel Hrdina <phrdina@redhat.com>
Date: Thu, 9 Jan 2025 16:23:44 +0100
Subject: [PATCH] qemu: snapshot: delete disk image only if parent snapshot is
external
Content-type: text/plain
When we are deleting external snapshot that is not active we only need
to delete overlay disk image of the parent snapshot. This works
correctly even if parent snapshot is external and active as it will have
another overlay created when user reverted to that snapshot.
In case the parent snapshot is internal there are no overlay disk images
created as everything is stored internally within the disk image. In
this case we would delete the actual disk image storing internal
snapshots and most likely the original disk image as well resulting in
data loss once the VM is shutoff.
Fixes: https://gitlab.com/libvirt/libvirt/-/issues/734
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
---
src/qemu/qemu_snapshot.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/src/qemu/qemu_snapshot.c b/src/qemu/qemu_snapshot.c
index 18b2e478f6..80cd54bf33 100644
--- a/src/qemu/qemu_snapshot.c
+++ b/src/qemu/qemu_snapshot.c
@@ -3144,6 +3144,8 @@ qemuSnapshotDeleteExternalPrepareData(virDomainObj *vm,
return -1;
}
+ data->parentSnap = qemuSnapshotFindParentSnapForDisk(snap, data->snapDisk);
+
if (data->merge) {
virStorageSource *snapDiskSrc = NULL;
@@ -3185,8 +3187,6 @@ qemuSnapshotDeleteExternalPrepareData(virDomainObj *vm,
qemuSnapshotGetDisksWithBackingStore(vm, snap, data);
}
- data->parentSnap = qemuSnapshotFindParentSnapForDisk(snap, data->snapDisk);
-
if (data->parentSnap && !virDomainSnapshotIsExternal(data->parentSnap)) {
virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s",
_("deleting external snapshot that has internal snapshot as parent not supported"));
@@ -3642,10 +3642,12 @@ qemuSnapshotDiscardExternal(virDomainObj *vm,
if (!data->job)
goto error;
} else {
- if (virStorageSourceInit(data->parentDomDisk->src) < 0 ||
- virStorageSourceUnlink(data->parentDomDisk->src) < 0) {
- VIR_WARN("Failed to remove snapshot image '%s'",
- data->snapDisk->name);
+ if (data->parentSnap && virDomainSnapshotIsExternal(data->parentSnap)) {
+ if (virStorageSourceInit(data->parentDomDisk->src) < 0 ||
+ virStorageSourceUnlink(data->parentDomDisk->src) < 0) {
+ VIR_WARN("Failed to remove snapshot image '%s'",
+ data->snapDisk->name);
+ }
}
}
}
+21
View File
@@ -0,0 +1,21 @@
# Makefile for source rpm: libvirt
# $Id$
NAME := libvirt
SPECFILE = $(firstword $(wildcard *.spec))
define find-makefile-common
for d in common ../common ../../common ; do if [ -f $$d/Makefile.common ] ; then if [ -f $$d/CVS/Root -a -w $$/Makefile.common ] ; then cd $$d ; cvs -Q update ; fi ; echo "$$d/Makefile.common" ; break ; fi ; done
endef
MAKEFILE_COMMON := $(shell $(find-makefile-common))
ifeq ($(MAKEFILE_COMMON),)
# attempt a checkout
define checkout-makefile-common
test -f CVS/Root && { cvs -Q -d $$(cat CVS/Root) checkout common && echo "common/Makefile.common" ; } || { echo "ERROR: I can't figure out how to checkout the 'common' module." ; exit -1 ; } >&2
endef
MAKEFILE_COMMON := $(shell $(checkout-makefile-common))
endif
include $(MAKEFILE_COMMON)
+1
View File
@@ -0,0 +1 @@
F-12
@@ -0,0 +1,34 @@
From bcd4180124afa20580d720912e2179b3a2f9295a Mon Sep 17 00:00:00 2001
From: Daniel Veillard <veillard@redhat.com>
Date: Mon, 5 Oct 2009 17:03:14 +0200
Subject: [PATCH] 526769 change logrotate config default to weekly
* daemon/libvirtd.logrotate.in: change to weekly rotation of logs,
keep a month worth of data and also extend to cover LXC and UML
domain logs
(cherry picked from commit 529325bbdd050af89bda5a5c1a01b5553c49a57e)
Fedora-patch: libvirt-change-logrotate-config-to-weekly.patch
---
qemud/libvirtd.logrotate.in | 6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/qemud/libvirtd.logrotate.in b/qemud/libvirtd.logrotate.in
index 9b42630..093651c 100644
--- a/qemud/libvirtd.logrotate.in
+++ b/qemud/libvirtd.logrotate.in
@@ -1,7 +1,7 @@
-@localstatedir@/log/libvirt/qemu/*.log {
- daily
+@localstatedir@/log/libvirt/qemu/*.log @localstatedir@/log/libvirt/uml/*.log @localstatedir@/log/libvirt/lxc/*.log {
+ weekly
missingok
- rotate 7
+ rotate 4
compress
delaycompress
copytruncate
--
1.6.2.5
+77
View File
@@ -0,0 +1,77 @@
From 4721ceb9b85daabe53804627473b06ced821c695 Mon Sep 17 00:00:00 2001
From: Daniel P. Berrange <berrange@redhat.com>
Date: Mon, 14 Sep 2009 11:23:20 +0100
Subject: [PATCH] Allow control over QEMU audio backend
When using VNC for graphics + keyboard + mouse, we shouldn't
then use the host OS for audio. Audio should go back over
VNC.
When using SDL for graphics, we should use the host OS for
audio since that's where the display is. We need to allow
certain QEMU env variables to be passed through to guest
too to allow choice of QEMU audio backend.
* qemud/libvirtd.sysconf: Mention QEMU/SDL audio env vars
* src/qemu_conf.c: Passthrough QEMU/SDL audio env for SDL display,
disable host audio for VNC display
(cherry picked from commit b08e6d38ae7a0ed70300d7d82107f83fddb60f44)
Fedora-patch: libvirt-disable-audio-backend.patch
---
qemud/libvirtd.sysconf | 8 ++++++++
src/qemu_conf.c | 14 ++++++++++++++
2 files changed, 22 insertions(+), 0 deletions(-)
diff --git a/qemud/libvirtd.sysconf b/qemud/libvirtd.sysconf
index fe4596a..28080a0 100644
--- a/qemud/libvirtd.sysconf
+++ b/qemud/libvirtd.sysconf
@@ -7,3 +7,11 @@
# Override Kerberos service keytab for SASL/GSSAPI
#KRB5_KTNAME=/etc/libvirt/krb5.tab
+
+# Override the QEMU/SDL default audio driver probing when
+# starting virtual machines using SDL graphics
+#
+# NB these have no effect for VMs using VNC
+#QEMU_AUDIO_DRV=sdl
+#
+#SDL_AUDIODRIVER=pulse
diff --git a/src/qemu_conf.c b/src/qemu_conf.c
index f92bcef..0dd0624 100644
--- a/src/qemu_conf.c
+++ b/src/qemu_conf.c
@@ -2109,6 +2109,13 @@ int qemudBuildCommandLine(virConnectPtr conn,
ADD_ARG_LIT("-k");
ADD_ARG_LIT(def->graphics[0]->data.vnc.keymap);
}
+
+ /* QEMU implements a VNC extension for providing audio, so we
+ * set the audio backend to none, to prevent it opening the
+ * host OS audio devices since that causes security issues
+ * and is non-sensical when using VNC.
+ */
+ ADD_ENV_LIT("QEMU_AUDIO_DRV=none");
} else if ((def->ngraphics == 1) &&
def->graphics[0]->type == VIR_DOMAIN_GRAPHICS_TYPE_SDL) {
char *xauth = NULL;
@@ -2131,6 +2138,13 @@ int qemudBuildCommandLine(virConnectPtr conn,
ADD_ENV(display);
if (def->graphics[0]->data.sdl.fullscreen)
ADD_ARG_LIT("-full-screen");
+
+ /* If using SDL for video, then we should just let it
+ * use QEMU's host audio drivers, possibly SDL too
+ * User can set these two before starting libvirtd
+ */
+ ADD_ENV_COPY("QEMU_AUDIO_DRV");
+ ADD_ENV_COPY("SDL_AUDIODRIVER");
}
if (def->nvideos) {
--
1.6.2.5
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,31 @@
From a5fa9f63fcffbf70465386672f24edac439866b9 Mon Sep 17 00:00:00 2001
From: Daniel P. Berrange <berrange@redhat.com>
Date: Thu, 24 Sep 2009 15:42:25 +0100
Subject: [PATCH] Fix crash in device hotplug cleanup code
* src/qemu/qemu_driver.c: Fix crash in scenario where XML
parsing of hotplugged device failed & thus 'dev' is NULL
(cherry picked from commit 879cd8cc2ba00f795913f296556e05f25afa7877)
Fedora-patch: libvirt-fix-crash-on-device-hotplug-parse-error.patch
---
src/qemu_driver.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/qemu_driver.c b/src/qemu_driver.c
index 0ce403c..c956258 100644
--- a/src/qemu_driver.c
+++ b/src/qemu_driver.c
@@ -5912,7 +5912,7 @@ cleanup:
if (cgroup)
virCgroupFree(&cgroup);
- if (ret < 0) {
+ if (ret < 0 && dev != NULL) {
if (qemuDomainSetDeviceOwnership(dom->conn, driver, dev, 1) < 0)
VIR_WARN0("Fail to restore disk device ownership");
virDomainDeviceDefFree(dev);
--
1.6.2.5
@@ -0,0 +1,33 @@
From 7bc1491deba6338e514504d1b68fe097e7f2bf19 Mon Sep 17 00:00:00 2001
From: Daniel Veillard <veillard@redhat.com>
Date: Thu, 1 Oct 2009 11:54:38 +0200
Subject: [PATCH] Avoid a libvirtd crash on broken input 523418
* src/conf/domain_conf.c: a simple typo in an XML domain file could lead
to a crash, because we called STRPREFIX() on the looked up value without
checking it was non-null.
(cherry picked from commit 79d233b5ca62f86ab22d271d1f08ec20060eee88)
Fedora-patch: libvirt-fix-crash-on-missing-iface-target-dev.patch
---
src/domain_conf.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/src/domain_conf.c b/src/domain_conf.c
index c424c67..476cdd7 100644
--- a/src/domain_conf.c
+++ b/src/domain_conf.c
@@ -1031,7 +1031,8 @@ virDomainNetDefParseXML(virConnectPtr conn,
} else if ((ifname == NULL) &&
xmlStrEqual(cur->name, BAD_CAST "target")) {
ifname = virXMLPropString(cur, "dev");
- if (STRPREFIX((const char*)ifname, "vnet")) {
+ if ((ifname != NULL) &&
+ (STRPREFIX((const char*)ifname, "vnet"))) {
/* An auto-generated target name, blank it out */
VIR_FREE(ifname);
}
--
1.6.2.5
+29
View File
@@ -0,0 +1,29 @@
From 58c38896a67c170063401d8091bae7dca8842923 Mon Sep 17 00:00:00 2001
From: Jiri Denemark <jdenemar@redhat.com>
Date: Wed, 23 Sep 2009 18:46:23 +0200
Subject: [PATCH] Fix a typo in virDiskHasValidPciAddr()
(cherry-picked from commit 3620e3cdcfe56cc4475b5ef1a0a893757240b795)
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Fedora-patch: libvirt-fix-device-detach-typo1.patch
---
src/domain_conf.h | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/domain_conf.h b/src/domain_conf.h
index 09368d9..d494e54 100644
--- a/src/domain_conf.h
+++ b/src/domain_conf.h
@@ -125,7 +125,7 @@ struct _virDomainDiskDef {
static inline int
virDiskHasValidPciAddr(virDomainDiskDefPtr def)
{
- return def->pci_addr.domain || def->pci_addr.domain || def->pci_addr.slot;
+ return def->pci_addr.domain || def->pci_addr.bus || def->pci_addr.slot;
}
--
1.6.2.5
+30
View File
@@ -0,0 +1,30 @@
From 81e967c716ce8c085be8baad9169f7772452d187 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 24 Sep 2009 08:55:55 +0100
Subject: [PATCH] Fix a typo in virNetHasValidPciAddr() too
* src/domain_conf.h: check domain/bus/slot, not domain/domain/slot
(cherry-picked from commit 6bfffce91635bb08de601747e94ed1182c0f47eb)
Fedora-patch: libvirt-fix-device-detach-typo2.patch
---
src/domain_conf.h | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/domain_conf.h b/src/domain_conf.h
index d494e54..7c918a7 100644
--- a/src/domain_conf.h
+++ b/src/domain_conf.h
@@ -207,7 +207,7 @@ struct _virDomainNetDef {
static inline int
virNetHasValidPciAddr(virDomainNetDefPtr def)
{
- return def->pci_addr.domain || def->pci_addr.domain || def->pci_addr.slot;
+ return def->pci_addr.domain || def->pci_addr.bus || def->pci_addr.slot;
}
enum virDomainChrSrcType {
--
1.6.2.5
+30
View File
@@ -0,0 +1,30 @@
From 3a64779ec5a89d38d64e07bca2b11b19e1882d7a Mon Sep 17 00:00:00 2001
From: Charles Duffy <charles@dyfis.net>
Date: Thu, 24 Sep 2009 09:00:24 +0100
Subject: [PATCH] Fix unitialized variable in qemudDomainDetachHostPciDevice()
* src/qemu_driver.c: initialize detach var
(cherry-picked from commit 580ad29288751234bee47ac9f6c04dac1dc529ea)
Fedora-patch: libvirt-fix-device-detach-typo3.patch
---
src/qemu_driver.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/qemu_driver.c b/src/qemu_driver.c
index 7c7b985..550a59c 100644
--- a/src/qemu_driver.c
+++ b/src/qemu_driver.c
@@ -6126,7 +6126,7 @@ static int qemudDomainDetachHostPciDevice(virConnectPtr conn,
virDomainObjPtr vm,
virDomainDeviceDefPtr dev)
{
- virDomainHostdevDefPtr detach;
+ virDomainHostdevDefPtr detach = NULL;
char *cmd, *reply;
int i, ret;
pciDevice *pci;
--
1.6.2.5
@@ -0,0 +1,32 @@
From 7692e1e19487c28454b1e5f6488d5574c70883f2 Mon Sep 17 00:00:00 2001
From: Chris Lalancette <clalance@redhat.com>
Date: Mon, 21 Sep 2009 14:53:31 +0200
Subject: [PATCH] Don't do virSetConnError when virDrvSupportsFeature is successful.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Fedora-patch: libvirt-fix-drv-supports-feature-bogus-error.patch
---
src/libvirt.c | 7 +++++--
1 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/libvirt.c b/src/libvirt.c
index 4a11688..fa59dc7 100644
--- a/src/libvirt.c
+++ b/src/libvirt.c
@@ -1349,8 +1349,11 @@ virDrvSupportsFeature (virConnectPtr conn, int feature)
}
ret = VIR_DRV_SUPPORTS_FEATURE (conn->driver, conn, feature);
- /* Copy to connection error object for back compatability */
- virSetConnError(conn);
+
+ if (ret < 0)
+ /* Copy to connection error object for back compatability */
+ virSetConnError(conn);
+
return ret;
}
--
1.6.2.5
@@ -0,0 +1,79 @@
From 71de8d92f20a9a9ee76d4d5df77ff477f1b7d441 Mon Sep 17 00:00:00 2001
From: Matthias Bolte <matthias.bolte@googlemail.com>
Date: Wed, 30 Sep 2009 02:17:27 +0200
Subject: [PATCH] Fix memory leaks in libvirtd's message processing
Commit 47cab734995fa9521b1df05d37e9978eedd8d3a2 changed the way how
qemud_client_message objects were reused. Before this commit
remoteDispatchClientRequest() reused the received message for normal responses
and to report non-fatal errors. If a fatal error occurred qemudWorker() frees
the message. After this commit non-fatal errors are reported by
remoteSerializeReplyError() using a new qemud_client_message object and the
original message leaks.
To fix this leak the original message has to be freed if
remoteSerializeReplyError() succeeds. If remoteSerializeReplyError()
fails the original message is freed in qemudWorker().
* daemon/dispatch.c: free qemud_client_message objects that will not be reused
and would leak otherwise, also free the allocated qemud_client_message object
in remoteSerializeError() if an error occurs
(cherry-picked from commit c6f1459eb998619ab21a92d9bb87341f26978181)
Fedora-patch: libvirt-fix-libvirtd-leak-in-error-reply.patch
---
qemud/dispatch.c | 15 +++++++++++++--
1 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/qemud/dispatch.c b/qemud/dispatch.c
index a60f2f4..ddb3215 100644
--- a/qemud/dispatch.c
+++ b/qemud/dispatch.c
@@ -191,6 +191,7 @@ remoteSerializeError(struct qemud_client *client,
xdr_error:
xdr_destroy(&xdr);
+ VIR_FREE(msg);
fatal_error:
xdr_free((xdrproc_t)xdr_remote_error, (char *)rerr);
return -1;
@@ -336,6 +337,7 @@ remoteDispatchClientRequest (struct qemud_server *server,
struct qemud_client *client,
struct qemud_client_message *msg)
{
+ int ret;
remote_error rerr;
memset(&rerr, 0, sizeof rerr);
@@ -364,7 +366,12 @@ remoteDispatchClientRequest (struct qemud_server *server,
}
error:
- return remoteSerializeReplyError(client, &rerr, &msg->hdr);
+ ret = remoteSerializeReplyError(client, &rerr, &msg->hdr);
+
+ if (ret >= 0)
+ VIR_FREE(msg);
+
+ return ret;
}
@@ -521,8 +528,12 @@ remoteDispatchClientCall (struct qemud_server *server,
rpc_error:
/* Semi-bad stuff happened, we can still try to send back
* an RPC error message to client */
- return remoteSerializeReplyError(client, &rerr, &msg->hdr);
+ rv = remoteSerializeReplyError(client, &rerr, &msg->hdr);
+
+ if (rv >= 0)
+ VIR_FREE(msg);
+ return rv;
xdr_error:
/* Seriously bad stuff happened, so we'll kill off this client
--
1.6.2.5
@@ -0,0 +1,46 @@
From ba585ed6cff624c6c0f1f9801382fd6846466ee0 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 17 Sep 2009 15:31:08 +0100
Subject: [PATCH] Fix net/disk hot-unplug segfault
When we hot-unplug the last device, we're currently double-freeing
the device definition.
Reported by Michal Nowak here:
https://bugzilla.redhat.com/523953
* src/qemu_driver.c: fix double free
(cherry-picked from commit 8881ae1bf8783006777429403cc543c33187175d)
Fedora-patch: libvirt-fix-net-hotunplug-double-free.patch
---
src/qemu_driver.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/qemu_driver.c b/src/qemu_driver.c
index a65334f..de31581 100644
--- a/src/qemu_driver.c
+++ b/src/qemu_driver.c
@@ -5998,7 +5998,7 @@ try_command:
/* ignore, harmless */
}
} else {
- VIR_FREE(vm->def->disks[0]);
+ VIR_FREE(vm->def->disks);
vm->def->ndisks = 0;
}
virDomainDiskDefFree(detach);
@@ -6100,7 +6100,7 @@ qemudDomainDetachNetDevice(virConnectPtr conn,
/* ignore, harmless */
}
} else {
- VIR_FREE(vm->def->nets[0]);
+ VIR_FREE(vm->def->nets);
vm->def->nnets = 0;
}
virDomainNetDefFree(detach);
--
1.6.2.5
@@ -0,0 +1,50 @@
From 17831d20051f8de8f1f7d661e8a23f4fe67c2153 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 17 Sep 2009 15:32:45 +0100
Subject: [PATCH] Fix leak in PCI hostdev hot-unplug
* src/qemu_driver.c: sync the hostdev hot-unplug code with the disk/net
code.
(cherry-picked from commit a70da51ff76ed860bfc0cdee2e1d556da997c557)
Fedora-patch: libvirt-fix-pci-hostdev-hotunplug-leak.patch
---
src/qemu_driver.c | 20 +++++++++++++-------
1 files changed, 13 insertions(+), 7 deletions(-)
diff --git a/src/qemu_driver.c b/src/qemu_driver.c
index de31581..2ddcdc0 100644
--- a/src/qemu_driver.c
+++ b/src/qemu_driver.c
@@ -6206,14 +6206,20 @@ static int qemudDomainDetachHostPciDevice(virConnectPtr conn,
pciFreeDevice(conn, pci);
}
- if (i != --vm->def->nhostdevs)
- memmove(&vm->def->hostdevs[i],
- &vm->def->hostdevs[i+1],
- sizeof(*vm->def->hostdevs) * (vm->def->nhostdevs-i));
- if (VIR_REALLOC_N(vm->def->hostdevs, vm->def->nhostdevs) < 0) {
- virReportOOMError(conn);
- ret = -1;
+ if (vm->def->nhostdevs > 1) {
+ memmove(vm->def->hostdevs + i,
+ vm->def->hostdevs + i + 1,
+ sizeof(*vm->def->hostdevs) *
+ (vm->def->nhostdevs - (i + 1)));
+ vm->def->nhostdevs--;
+ if (VIR_REALLOC_N(vm->def->hostdevs, vm->def->nhostdevs) < 0) {
+ /* ignore, harmless */
+ }
+ } else {
+ VIR_FREE(vm->def->hostdevs);
+ vm->def->nhostdevs = 0;
}
+ virDomainHostdevDefFree(detach);
return ret;
}
--
1.6.2.5
+53
View File
@@ -0,0 +1,53 @@
From f1be5a4714e194a84840343e0937fe62463a18dc Mon Sep 17 00:00:00 2001
From: Charles Duffy <Charles_Duffy@dell.com>
Date: Fri, 18 Sep 2009 11:32:35 -0500
Subject: [PATCH] Prevent attempt to call cat -c during virDomainSave to raw
Fedora-patch: libvirt-fix-qemu-raw-format-save.patch
---
src/qemu_driver.c | 28 ++++++++++++++++++----------
1 files changed, 18 insertions(+), 10 deletions(-)
diff --git a/src/qemu_driver.c b/src/qemu_driver.c
index 2ddcdc0..7c7b985 100644
--- a/src/qemu_driver.c
+++ b/src/qemu_driver.c
@@ -3905,17 +3905,25 @@ static int qemudDomainSave(virDomainPtr dom,
goto cleanup;
}
- const char *prog = qemudSaveCompressionTypeToString(header.compressed);
- if (prog == NULL) {
- qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR,
- _("Invalid compress format %d"), header.compressed);
- goto cleanup;
- }
+ {
+ const char *prog = qemudSaveCompressionTypeToString(header.compressed);
+ const char *args;
- if (STREQ (prog, "raw"))
- prog = "cat";
- internalret = virAsprintf(&command, "migrate \"exec:"
- "%s -c >> '%s' 2>/dev/null\"", prog, safe_path);
+ if (prog == NULL) {
+ qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR,
+ _("Invalid compress format %d"), header.compressed);
+ goto cleanup;
+ }
+
+ if (STREQ (prog, "raw")) {
+ prog = "cat";
+ args = "";
+ } else {
+ args = "-c";
+ }
+ internalret = virAsprintf(&command, "migrate \"exec:"
+ "%s %s >> '%s' 2>/dev/null\"", prog, args, safe_path);
+ }
if (internalret < 0) {
virReportOOMError(dom->conn);
--
1.6.2.5
+38
View File
@@ -0,0 +1,38 @@
From 0b846a30468a6b4586407f020ccde7bb51afaf98 Mon Sep 17 00:00:00 2001
From: Daniel P. Berrange <berrange@redhat.com>
Date: Mon, 12 Oct 2009 20:03:50 +0100
Subject: [PATCH] Fix QEMU restore from file in raw format
The logic for running the decompression programs was broken in
commit f238709304f9f6c57204cdd943e542cbae38fa5f, so that for
non-raw formats the decompression program was never run, and
for raw formats, it tried to exec an argv[] with initial NULL
in the program name.
* src/qemu/qemu_driver.c: Fix logic in runing decompression program
(cherry picked from commit 74b379cbd5ba9f472a3a2d5710e497966b1a3a37)
Fedora-patch: libvirt-fix-qemu-restore-from-raw1.patch
---
src/qemu_driver.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/src/qemu_driver.c b/src/qemu_driver.c
index 550a59c..0ce403c 100644
--- a/src/qemu_driver.c
+++ b/src/qemu_driver.c
@@ -4541,9 +4541,8 @@ static int qemudDomainRestore(virConnectPtr conn,
goto cleanup;
}
- if (header.compressed != QEMUD_SAVE_FORMAT_RAW)
+ if (header.compressed != QEMUD_SAVE_FORMAT_RAW) {
intermediate_argv[0] = prog;
- else {
intermediatefd = fd;
fd = -1;
if (virExec(conn, intermediate_argv, NULL, NULL,
--
1.6.2.5
+120
View File
@@ -0,0 +1,120 @@
From 57d7cc602d14c6b50e2826e427a5de124e479f95 Mon Sep 17 00:00:00 2001
From: Daniel P. Berrange <berrange@redhat.com>
Date: Mon, 12 Oct 2009 20:32:33 +0100
Subject: [PATCH] Fix virFileReadLimFD/virFileReadAll to handle EINTR
The fread_file_lim() function uses fread() but never handles
EINTR results, causing unexpected failures when reading QEMU
help arg info. It was unneccessarily using FILE * instead
of plain UNIX file handles, which prevented use of saferead()
* src/util/util.c: Switch fread_file_lim over to use saferead
instead of fread, remove FILE * use, and rename
(cherry picked from commit 11a36d956cb8a5e439e535bff3e0cfce50a64bca)
Fedora-patch: libvirt-fix-qemu-restore-from-raw2.patch
---
src/util.c | 45 ++++++++++++---------------------------------
1 files changed, 12 insertions(+), 33 deletions(-)
diff --git a/src/util.c b/src/util.c
index 1878e33..7bc3a66 100644
--- a/src/util.c
+++ b/src/util.c
@@ -887,7 +887,7 @@ virExec(virConnectPtr conn,
number of bytes. If the length of the input is <= max_len, and
upon error while reading that data, it works just like fread_file. */
static char *
-fread_file_lim (FILE *stream, size_t max_len, size_t *length)
+saferead_lim (int fd, size_t max_len, size_t *length)
{
char *buf = NULL;
size_t alloc = 0;
@@ -895,8 +895,8 @@ fread_file_lim (FILE *stream, size_t max_len, size_t *length)
int save_errno;
for (;;) {
- size_t count;
- size_t requested;
+ int count;
+ int requested;
if (size + BUFSIZ + 1 > alloc) {
alloc += alloc / 2;
@@ -912,12 +912,12 @@ fread_file_lim (FILE *stream, size_t max_len, size_t *length)
/* Ensure that (size + requested <= max_len); */
requested = MIN (size < max_len ? max_len - size : 0,
alloc - size - 1);
- count = fread (buf + size, 1, requested, stream);
+ count = saferead (fd, buf + size, requested);
size += count;
if (count != requested || requested == 0) {
save_errno = errno;
- if (ferror (stream))
+ if (count < 0)
break;
buf[size] = '\0';
*length = size;
@@ -930,12 +930,12 @@ fread_file_lim (FILE *stream, size_t max_len, size_t *length)
return NULL;
}
-/* A wrapper around fread_file_lim that maps a failure due to
+/* A wrapper around saferead_lim that maps a failure due to
exceeding the maximum size limitation to EOVERFLOW. */
-static int virFileReadLimFP(FILE *fp, int maxlen, char **buf)
+int virFileReadLimFD(int fd, int maxlen, char **buf)
{
size_t len;
- char *s = fread_file_lim (fp, maxlen+1, &len);
+ char *s = saferead_lim (fd, maxlen+1, &len);
if (s == NULL)
return -1;
if (len > maxlen || (int)len != len) {
@@ -949,37 +949,16 @@ static int virFileReadLimFP(FILE *fp, int maxlen, char **buf)
return len;
}
-/* Like virFileReadLimFP, but use a file descriptor rather than a FILE*. */
-int virFileReadLimFD(int fd_arg, int maxlen, char **buf)
-{
- int fd = dup (fd_arg);
- if (fd >= 0) {
- FILE *fp = fdopen (fd, "r");
- if (fp) {
- int len = virFileReadLimFP (fp, maxlen, buf);
- int saved_errno = errno;
- fclose (fp);
- errno = saved_errno;
- return len;
- } else {
- int saved_errno = errno;
- close (fd);
- errno = saved_errno;
- }
- }
- return -1;
-}
-
int virFileReadAll(const char *path, int maxlen, char **buf)
{
- FILE *fh = fopen(path, "r");
- if (fh == NULL) {
+ int fd = open(path, O_RDONLY);
+ if (fd < 0) {
virReportSystemError(NULL, errno, _("Failed to open file '%s'"), path);
return -1;
}
- int len = virFileReadLimFP (fh, maxlen, buf);
- fclose(fh);
+ int len = virFileReadLimFD(fd, maxlen, buf);
+ close(fd);
if (len < 0) {
virReportSystemError(NULL, errno, _("Failed to read file '%s'"), path);
return -1;
--
1.6.2.5
+41
View File
@@ -0,0 +1,41 @@
From 6b12148864cf6a1d22a2cf4e0e9c48e9946331cb Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Wed, 30 Sep 2009 18:37:03 +0100
Subject: [PATCH] Fix USB device re-labelling
A simple misplaced break out of a switch results in:
libvir: error : Failed to open file '/sys/bus/pci/devices/0000:00:54c./vendor': No such file or directory
libvir: error : Failed to open file '/sys/bus/pci/devices/0000:00:54c./device': No such file or directory
libvir: error : this function is not supported by the hypervisor: Failed to read product/vendor ID for 0000:00:54c.
when trying to passthrough a USB host device to qemu.
* src/security_selinux.c: fix a switch/break thinko
Fedora-patch: libvirt-fix-usb-device-passthrough.patch
---
src/security_selinux.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/src/security_selinux.c b/src/security_selinux.c
index bc295b1..b4dc153 100644
--- a/src/security_selinux.c
+++ b/src/security_selinux.c
@@ -464,12 +464,11 @@ SELinuxSetSecurityHostdevLabel(virConnectPtr conn,
ret = usbDeviceFileIterate(conn, usb, SELinuxSetSecurityUSBLabel, vm);
usbFreeDevice(conn, usb);
-
- break;
} else {
/* XXX deal with product/vendor better */
ret = 0;
}
+ break;
}
case VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI: {
--
1.6.2.5
+103
View File
@@ -0,0 +1,103 @@
From cab81502320d97dac4c5c12e7496f30896709c49 Mon Sep 17 00:00:00 2001
From: Matthias Bolte <matthias.bolte@googlemail.com>
Date: Tue, 22 Sep 2009 15:12:48 +0200
Subject: [PATCH] Fix xen driver refcounting.
The commit cb51aa48a777ddae6997faa9f28350cb62655ffd "Fix up connection
reference counting." changed the driver closing and virConnectPtr
unref-logic in virConnectClose().
Before this commit virConnectClose() closed all drivers of the given
virConnectPtr and virUnrefConnect()'ed it afterwards. After this
commit the driver-closing is done in virUnrefConnect() if and only if
the ref-count of the virConnectPtr dropped to zero.
This change in execution order leads to a virConnectPtr leak, at least
for connections to Xen.
The relevant call sequences:
virConnectOpen() -> xenUnifiedOpen() ...
... xenInotifyOpen() -> virConnectRef(conn)
... xenStoreOpen() -> xenStoreAddWatch() -> conn->refs++
virConnectClose() -> xenUnifiedClose() ...
... xenInotifyClose() -> virUnrefConnect(conn)
... xenStoreClose() -> xenStoreRemoveWatch() -> virUnrefConnect(conn)
Before the commit this additional virConnectRef/virUnrefConnect calls
where no problem, because virConnectClose() closed the drivers
explicitly and the additional refs added by the Xen subdrivers were
removed properly. After the commit this additional refs result in a
virConnectPtr leak (including a leak of the hypercall file handle;
that's how I noticed this problem), because now the drivers are only
close if and only if the ref-count drops to zero, but this cannot
happen anymore, because the additional refs from the Xen subdrivers
would only be removed if the drivers get closed, but that doesn't
happen because the ref-count cannot drop to zero.
The fix for this problem is simple: remove the
virConnectRef/virUnrefConnect calls from the Xen subdrivers (see
attached patch). Maybe someone could explain why the Xen Inotify and
Xen Store driver do this extra ref-counting, but none of the other Xen
subdrivers. It seems unnecessary to me and can be removed without
problems.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
(cherry picked from commit 6ed7374c5a6c6a2b1b1801d7d041dc7f09748592)
Fedora-patch: libvirt-fix-xen-driver-refcounting.patch
---
src/xen_inotify.c | 2 --
src/xs_internal.c | 3 ---
2 files changed, 0 insertions(+), 5 deletions(-)
diff --git a/src/xen_inotify.c b/src/xen_inotify.c
index e312b9e..ecaefaf 100644
--- a/src/xen_inotify.c
+++ b/src/xen_inotify.c
@@ -463,7 +463,6 @@ xenInotifyOpen(virConnectPtr conn ATTRIBUTE_UNUSED,
DEBUG0("Failed to add inotify handle, disabling events");
}
- virConnectRef(conn);
return 0;
}
@@ -486,7 +485,6 @@ xenInotifyClose(virConnectPtr conn)
if (priv->inotifyWatch != -1)
virEventRemoveHandle(priv->inotifyWatch);
close(priv->inotifyFD);
- virUnrefConnect(conn);
return 0;
}
diff --git a/src/xs_internal.c b/src/xs_internal.c
index 1f54b1f..a18dcad 100644
--- a/src/xs_internal.c
+++ b/src/xs_internal.c
@@ -1139,8 +1139,6 @@ int xenStoreAddWatch(virConnectPtr conn,
list->watches[n] = watch;
list->count++;
- conn->refs++;
-
return xs_watch(priv->xshandle, watch->path, watch->token);
}
@@ -1190,7 +1188,6 @@ int xenStoreRemoveWatch(virConnectPtr conn,
; /* Failure to reduce memory allocation isn't fatal */
}
list->count--;
- virUnrefConnect(conn);
return 0;
}
}
--
1.6.2.5
@@ -0,0 +1,54 @@
From b7e3ac4f23befe67518b57e34691c301820a436c Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Tue, 6 Oct 2009 12:33:17 +0100
Subject: [PATCH] Create /var/log/libvirt/{lxc,uml} dirs
Otherwise logrotate barfs:
error: error accessing /var/log/libvirt/uml: No such file or directory
error: libvirtd:1 glob failed for /var/log/libvirt/uml/*.log
error: found error in /var/log/libvirt/qemu/*.log /var/log/libvirt/uml/*.log /var/log/libvirt/lxc/*.log , skipping
* qemud/Makefile.am: always create /var/log/libvirt/{lxc,uml} when
installing the logrotate conf; not ideal, but easier than making
the logrotate conf depend on which drivers are enabled
Fedora-patch: libvirt-logrotate-create-lxc-uml-dirs.patch
---
qemud/Makefile.am | 6 ++++--
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/qemud/Makefile.am b/qemud/Makefile.am
index 3d143da..a7f4bdf 100644
--- a/qemud/Makefile.am
+++ b/qemud/Makefile.am
@@ -176,7 +176,7 @@ install-data-local: install-init install-data-sasl install-data-polkit \
test -e $(DESTDIR)$(sysconfdir)/libvirt/qemu/networks/autostart/default.xml || \
ln -s ../default.xml \
$(DESTDIR)$(sysconfdir)/libvirt/qemu/networks/autostart/default.xml
- mkdir -p $(DESTDIR)$(localstatedir)/log/libvirt/qemu
+ mkdir -p $(DESTDIR)$(localstatedir)/log/libvirt
mkdir -p $(DESTDIR)$(localstatedir)/run/libvirt
mkdir -p $(DESTDIR)$(localstatedir)/lib/libvirt
@@ -184,7 +184,7 @@ uninstall-local:: uninstall-init uninstall-data-sasl install-data-polkit
rm -f $(DESTDIR)$(sysconfdir)/libvirt/qemu/networks/autostart/default.xml
rm -f $(DESTDIR)$(sysconfdir)/$(default_xml_dest)
rmdir $(DESTDIR)$(sysconfdir)/libvirt/qemu/networks/autostart || :
- rmdir $(DESTDIR)$(localstatedir)/log/libvirt/qemu || :
+ rmdir $(DESTDIR)$(localstatedir)/log/libvirt || :
rmdir $(DESTDIR)$(localstatedir)/run/libvirt || :
rmdir $(DESTDIR)$(localstatedir)/lib/libvirt || :
@@ -240,6 +240,8 @@ libvirtd.logrotate: libvirtd.logrotate.in
install-logrotate: libvirtd.logrotate
mkdir -p $(DESTDIR)$(localstatedir)/log/libvirt/qemu/
+ mkdir -p $(DESTDIR)$(localstatedir)/log/libvirt/lxc/
+ mkdir -p $(DESTDIR)$(localstatedir)/log/libvirt/uml/
mkdir -p $(DESTDIR)$(sysconfdir)/logrotate.d/
$(INSTALL_DATA) $< $(DESTDIR)$(sysconfdir)/logrotate.d/libvirtd
--
1.6.2.5
@@ -0,0 +1,32 @@
From ba3bc9b22a21b8e9e110166c98be70e2ad6469cb Mon Sep 17 00:00:00 2001
From: Cole Robinson <crobinso@redhat.com>
Date: Tue, 13 Oct 2009 11:31:27 -0400
Subject: [PATCH] network: Fix printing XML 'delay' attribute
When specifying bridge delay via network XML define, we were looking for
the 'delay' attribute, but would dump the value as 'forwardDelay'. Have
the output match the expected input (and schema).
(cherry picked from commit 3b13aa3db37bf5a692bccfa015a01999043e797b)
Fedora-patch: libvirt-network-delay-attribute-formatting.patch
---
src/network_conf.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/network_conf.c b/src/network_conf.c
index 3764bb4..f75c457 100644
--- a/src/network_conf.c
+++ b/src/network_conf.c
@@ -587,7 +587,7 @@ char *virNetworkDefFormat(virConnectPtr conn,
virBufferAddLit(&buf, " <bridge");
if (def->bridge)
virBufferEscapeString(&buf, " name='%s'", def->bridge);
- virBufferVSprintf(&buf, " stp='%s' forwardDelay='%ld' />\n",
+ virBufferVSprintf(&buf, " stp='%s' delay='%ld' />\n",
def->stp ? "on" : "off",
def->delay);
--
1.6.2.5
+177
View File
@@ -0,0 +1,177 @@
From d7722ed1cb04aa8a7b9fbf880882841867b69ab0 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Mon, 12 Oct 2009 10:52:13 +0100
Subject: [PATCH] Take domain type into account when looking up default machine
If one has e.g.
<guest>
<os_type>hvm</os_type>
<arch name='x86_64'>
<wordsize>64</wordsize>
<emulator>/usr/bin/qemu-system-x86_64</emulator>
<machine>pc-0.11</machine>
<machine canonical='pc-0.11'>pc</machine>
<machine>pc-0.10</machine>
<machine>isapc</machine>
<domain type='qemu'>
</domain>
<domain type='kvm'>
<emulator>/usr/bin/kvm</emulator>
<machine>pc</machine>
<machine>isapc</machine>
</domain>
</arch>
</guest>
and start a guest with:
<domain type='kvm'>
...
<os>
<type arch='x86_64'>hvm</type>
...
</os>
</domain>
then the default machine type should be 'pc' and not 'pc-0.11'
Issue was reported by Anton Protopopov.
* src/capabilities.[ch]: pass the domain type to
virCapabilitiesDefaultGuestArch() and use it to look up the default
machine type from a specific guest domain if needed.
* src/conf/domain_conf.c, src/xen/xm_internal.c: update
* tests/qemuxml2argvdata/qemuxml2argv-machine-aliases2.xml: update
the domain type to 'kvm' and remove the machine type to check
that the default gets looked up correctly
(cherry picked from commit 73c901a8075c09203545fc81164c1e5f11c67c89)
Fedora-patch: libvirt-qemu-machine-type-fixes1.patch
---
src/capabilities.c | 31 ++++++++++++++++---
src/capabilities.h | 3 +-
src/domain_conf.c | 3 +-
src/xm_internal.c | 3 +-
.../qemuxml2argv-machine-aliases2.xml | 4 +-
5 files changed, 34 insertions(+), 10 deletions(-)
diff --git a/src/capabilities.c b/src/capabilities.c
index 38fe7fc..6ebddf5 100644
--- a/src/capabilities.c
+++ b/src/capabilities.c
@@ -549,22 +549,43 @@ virCapabilitiesDefaultGuestArch(virCapsPtr caps,
* @caps: capabilities to query
* @ostype: OS type to search for
* @arch: architecture to search for
+ * @domain: domain type to search for
*
* Returns the first machine variant associated with
- * the requested operating system type and architecture
+ * the requested operating system type, architecture
+ * and domain type
*/
extern const char *
virCapabilitiesDefaultGuestMachine(virCapsPtr caps,
const char *ostype,
- const char *arch)
+ const char *arch,
+ const char *domain)
{
int i;
+
for (i = 0 ; i < caps->nguests ; i++) {
- if (STREQ(caps->guests[i]->ostype, ostype) &&
- STREQ(caps->guests[i]->arch.name, arch) &&
- caps->guests[i]->arch.defaultInfo.nmachines)
+ virCapsGuestPtr guest = caps->guests[i];
+ int j;
+
+ if (!STREQ(guest->ostype, ostype) || !STREQ(guest->arch.name, arch))
+ continue;
+
+ for (j = 0; j < guest->arch.ndomains; j++) {
+ virCapsGuestDomainPtr dom= guest->arch.domains[j];
+
+ if (!STREQ(dom->type, domain))
+ continue;
+
+ if (!dom->info.nmachines)
+ break;
+
+ return dom->info.machines[0]->name;
+ }
+
+ if (guest->arch.defaultInfo.nmachines)
return caps->guests[i]->arch.defaultInfo.machines[0]->name;
}
+
return NULL;
}
diff --git a/src/capabilities.h b/src/capabilities.h
index b958d95..2f24605 100644
--- a/src/capabilities.h
+++ b/src/capabilities.h
@@ -207,7 +207,8 @@ virCapabilitiesDefaultGuestArch(virCapsPtr caps,
extern const char *
virCapabilitiesDefaultGuestMachine(virCapsPtr caps,
const char *ostype,
- const char *arch);
+ const char *arch,
+ const char *domain);
extern const char *
virCapabilitiesDefaultGuestEmulator(virCapsPtr caps,
const char *ostype,
diff --git a/src/domain_conf.c b/src/domain_conf.c
index 5ae0775..c424c67 100644
--- a/src/domain_conf.c
+++ b/src/domain_conf.c
@@ -2664,7 +2664,8 @@ static virDomainDefPtr virDomainDefParseXML(virConnectPtr conn,
if (!def->os.machine) {
const char *defaultMachine = virCapabilitiesDefaultGuestMachine(caps,
def->os.type,
- def->os.arch);
+ def->os.arch,
+ virDomainVirtTypeToString(def->virtType));
if (defaultMachine != NULL) {
if (!(def->os.machine = strdup(defaultMachine))) {
virReportOOMError(conn);
diff --git a/src/xm_internal.c b/src/xm_internal.c
index de3aca9..6d351d4 100644
--- a/src/xm_internal.c
+++ b/src/xm_internal.c
@@ -720,7 +720,8 @@ xenXMDomainConfigParse(virConnectPtr conn, virConfPtr conf) {
defaultMachine = virCapabilitiesDefaultGuestMachine(priv->caps,
def->os.type,
- def->os.arch);
+ def->os.arch,
+ virDomainVirtTypeToString(def->virtType));
if (defaultMachine != NULL) {
if (!(def->os.machine = strdup(defaultMachine)))
goto no_memory;
diff --git a/tests/qemuxml2argvdata/qemuxml2argv-machine-aliases2.xml b/tests/qemuxml2argvdata/qemuxml2argv-machine-aliases2.xml
index 6f62243..a2c6254 100644
--- a/tests/qemuxml2argvdata/qemuxml2argv-machine-aliases2.xml
+++ b/tests/qemuxml2argvdata/qemuxml2argv-machine-aliases2.xml
@@ -1,11 +1,11 @@
-<domain type='qemu'>
+<domain type='kvm'>
<name>QEMUGuest1</name>
<uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid>
<memory>219200</memory>
<currentMemory>219200</currentMemory>
<vcpu>1</vcpu>
<os>
- <type arch='x86_64' machine='pc'>hvm</type>
+ <type arch='x86_64'>hvm</type>
<boot dev='hd'/>
</os>
<clock offset='utc'/>
--
1.6.2.5
+42
View File
@@ -0,0 +1,42 @@
From bb64cc4cbe5d3c69057f63be2c1acaca72038e5a Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 15 Oct 2009 12:09:17 +0100
Subject: [PATCH] Don't copy old machines from a domain which has none
If the the qemu and kvm binaries are the same, we don't include machine
types in the kvm domain info.
However, the code which refreshes the machine types info from the
previous capabilities structure first looks at the kvm domain's info,
finds it matches and then copies the empty machine types list over
for the top-level qemu domain.
That doesn't make sense, we shouldn't copy an empty machin types list.
* src/qemu/qemu_conf.c: qemudGetOldMachinesFromInfo(): don't copy an
empty machine types list.
(cherry picked from commit 2210f8a3a8e2774ca4fb8b42e21899e5b85ca913)
Fedora-patch: libvirt-qemu-machine-type-fixes2.patch
---
src/qemu_conf.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/src/qemu_conf.c b/src/qemu_conf.c
index 0dd0624..34a7fe1 100644
--- a/src/qemu_conf.c
+++ b/src/qemu_conf.c
@@ -505,6 +505,9 @@ qemudGetOldMachinesFromInfo(virCapsGuestDomainInfoPtr info,
virCapsGuestMachinePtr *list;
int i;
+ if (!info->nmachines)
+ return 0;
+
if (!info->emulator || !STREQ(emulator, info->emulator))
return 0;
--
1.6.2.5
@@ -0,0 +1,30 @@
From a44bce591a8d746a4a00c8609cb0111c76271cab Mon Sep 17 00:00:00 2001
From: Cole Robinson <crobinso@redhat.com>
Date: Thu, 8 Oct 2009 18:05:36 -0400
Subject: [PATCH] storage: Fix generating iscsi 'auth' xml
We were missing a closing tag, so the XML wasn't proper.
(cherry picked from commit 826cbac4591fd5929b497299a90d3a65226b2825)
Fedora-patch: libvirt-storage-iscsi-auth-xml-formatting.patch
---
src/storage_conf.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/storage_conf.c b/src/storage_conf.c
index 788de15..1633aac 100644
--- a/src/storage_conf.c
+++ b/src/storage_conf.c
@@ -799,7 +799,7 @@ virStoragePoolSourceFormat(virConnectPtr conn,
if (src->authType == VIR_STORAGE_POOL_AUTH_CHAP)
- virBufferVSprintf(buf," <auth type='chap' login='%s' passwd='%s'>\n",
+ virBufferVSprintf(buf," <auth type='chap' login='%s' passwd='%s'/>\n",
src->auth.chap.login,
src->auth.chap.passwd);
virBufferAddLit(buf," </source>\n");
--
1.6.2.5
File diff suppressed because it is too large Load Diff
+1020 -2690
View File
File diff suppressed because it is too large Load Diff
+1 -1
View File
@@ -1 +1 @@
SHA512 (libvirt-10.6.0.tar.xz) = edec79e89669d5e9a46be35e0d6334a6ed3bbf32426679549bd998bde24cba52b0378843f41a3abb5d781ad53e2a6a54619a0bad3f168c11fb41736cc6af6568
f1cd360a5da38b847e166c6482141940 libvirt-0.7.1.tar.gz