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> --- 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 | 11 ++++++++--- tests/virnetdevbandwidthtest.c | 8 +++++++- 9 files changed, 95 insertions(+), 48 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 0e31e5e4b9..a64cfef1a0 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 083ab83ec6..7eba7a2c46 100644 --- a/src/lxc/lxc_process.c +++ b/src/lxc/lxc_process.c @@ -609,8 +609,12 @@ 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 d408f17de7..e700a614a9 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -2109,8 +2109,11 @@ 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; @@ -2190,8 +2193,11 @@ 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 f4430275dc..3d0a5dd460 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -8694,9 +8694,14 @@ 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 5b9c55f704..425d583e99 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -9938,21 +9938,22 @@ 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 55512476e4..b06ae61a44 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1331,9 +1331,14 @@ 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 " @@ -4181,9 +4186,14 @@ 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..414d6c97eb 100644 --- a/src/util/virnetdevbandwidth.h +++ b/src/util/virnetdevbandwidth.h @@ -35,15 +35,20 @@ struct _virNetDevBandwidth { virNetDevBandwidthRate *out; }; -void virNetDevBandwidthFree(virNetDevBandwidth *def); + 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..3a87d876c8 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 { + 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; } -- 2.47.0

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> --- 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 5b9b44ef96..90a7c6fa01 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -2874,6 +2874,7 @@ 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 69c204389c..080d509d6e 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); -- 2.47.0

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> --- 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 f8b5ab665d..b3605bd40e 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..754bc18162 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 0a886780ad..21a9e02061 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", -- 2.47.0

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> --- 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 b3605bd40e..5d716264bf 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 \ -- 2.47.0