From c5bb34b10f497c960c818c46ed7a3fa2567d0d3c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= Date: Wed, 5 Jun 2024 12:29:37 +0100 Subject: [PATCH] Fix crash in event loop (CVE-2024-4418), event loop stall, GSource leak & udev leak MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Daniel P. Berrangé --- ...ev-reference-leak-with-invalid-flags.patch | 41 ++++++++ ...f-GSource-in-use-for-interrupting-ma.patch | 49 +++++++++ ...rary-GSource-is-removed-from-client-.patch | 99 +++++++++++++++++++ ...-race-in-waking-up-client-event-loop.patch | 76 ++++++++++++++ libvirt.spec | 12 ++- 5 files changed, 276 insertions(+), 1 deletion(-) create mode 100644 0001-interface-fix-udev-reference-leak-with-invalid-flags.patch create mode 100644 0001-rpc-avoid-leak-of-GSource-in-use-for-interrupting-ma.patch create mode 100644 0001-rpc-ensure-temporary-GSource-is-removed-from-client-.patch create mode 100644 0001-rpc-fix-race-in-waking-up-client-event-loop.patch diff --git a/0001-interface-fix-udev-reference-leak-with-invalid-flags.patch b/0001-interface-fix-udev-reference-leak-with-invalid-flags.patch new file mode 100644 index 0000000..cdc4a2b --- /dev/null +++ b/0001-interface-fix-udev-reference-leak-with-invalid-flags.patch @@ -0,0 +1,41 @@ +From 3499354e12a1c1832bf4030693a64e03ceb79d05 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= +Date: Wed, 5 Jun 2024 11:16:21 +0100 +Subject: [PATCH] interface: fix udev reference leak with invalid flags +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +The udevInterfaceGetXMLDesc method takes a reference on the udev +driver as its first action. If the virCheckFlags() condition +fails, however, this reference is never released. + +Reviewed-by: Peter Krempa +Signed-off-by: Daniel P. Berrangé +--- + src/interface/interface_backend_udev.c | 4 +++- + 1 file changed, 3 insertions(+), 1 deletion(-) + +diff --git a/src/interface/interface_backend_udev.c b/src/interface/interface_backend_udev.c +index fdf11a8318..e1a50389c9 100644 +--- a/src/interface/interface_backend_udev.c ++++ b/src/interface/interface_backend_udev.c +@@ -1027,12 +1027,14 @@ static char * + udevInterfaceGetXMLDesc(virInterfacePtr ifinfo, + unsigned int flags) + { +- struct udev *udev = udev_ref(driver->udev); ++ struct udev *udev = NULL; + g_autoptr(virInterfaceDef) ifacedef = NULL; + char *xmlstr = NULL; + + virCheckFlags(VIR_INTERFACE_XML_INACTIVE, NULL); + ++ udev = udev_ref(driver->udev); ++ + /* Recursively build up the interface XML based on the requested + * interface name + */ +-- +2.45.1 + diff --git a/0001-rpc-avoid-leak-of-GSource-in-use-for-interrupting-ma.patch b/0001-rpc-avoid-leak-of-GSource-in-use-for-interrupting-ma.patch new file mode 100644 index 0000000..afa729d --- /dev/null +++ b/0001-rpc-avoid-leak-of-GSource-in-use-for-interrupting-ma.patch @@ -0,0 +1,49 @@ +From 98f1cf88fa7e0f992d93f376418fbfb3996a9690 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= +Date: Fri, 17 May 2024 14:55:24 +0100 +Subject: [PATCH] rpc: avoid leak of GSource in use for interrupting main loop +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +We never release the reference on the GSource created for +interrupting the main loop, nor do we remove it from the +main context if our thread is woken up prior to the wakeup +callback firing. + +This can result in a leak of GSource objects, along with an +ever growing list of GSources attached to the main context, +which will gradually slow down execution of the loop, as +several operations are O(N) for the number of attached GSource +objects. + +Reviewed-by: Michal Privoznik +Signed-off-by: Daniel P. Berrangé +--- + src/rpc/virnetclient.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c +index 147b0d661a..6d424eb599 100644 +--- a/src/rpc/virnetclient.c ++++ b/src/rpc/virnetclient.c +@@ -1946,7 +1946,7 @@ static int virNetClientIO(virNetClient *client, + /* Check to see if another thread is dispatching */ + if (client->haveTheBuck) { + /* Force other thread to wakeup from poll */ +- GSource *wakeup = g_idle_source_new(); ++ g_autoptr(GSource) wakeup = g_idle_source_new(); + g_source_set_callback(wakeup, virNetClientIOWakeup, client->eventLoop, NULL); + g_source_attach(wakeup, client->eventCtx); + +@@ -1968,6 +1968,7 @@ static int virNetClientIO(virNetClient *client, + return -1; + } + ++ g_source_destroy(wakeup); + VIR_DEBUG("Woken up from sleep head=%p call=%p", + client->waitDispatch, thiscall); + /* Three reasons we can be woken up +-- +2.45.1 + diff --git a/0001-rpc-ensure-temporary-GSource-is-removed-from-client-.patch b/0001-rpc-ensure-temporary-GSource-is-removed-from-client-.patch new file mode 100644 index 0000000..f9c5b8b --- /dev/null +++ b/0001-rpc-ensure-temporary-GSource-is-removed-from-client-.patch @@ -0,0 +1,99 @@ +From 8074d64dc2eca846d6a61efe1a9b7428a0ce1dd1 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= +Date: Tue, 30 Apr 2024 11:51:15 +0100 +Subject: [PATCH] rpc: ensure temporary GSource is removed from client event + loop +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +Users are seeing periodic segfaults from libvirt client apps, +especially thread heavy ones like virt-manager. A typical +stack trace would end up in the virNetClientIOEventFD method, +with illegal access to stale stack data. eg + +==238721==ERROR: AddressSanitizer: stack-use-after-return on address 0x75cd18709788 at pc 0x75cd3111f907 bp 0x75cd181ff550 sp 0x75cd181ff548 +WRITE of size 4 at 0x75cd18709788 thread T11 + #0 0x75cd3111f906 in virNetClientIOEventFD /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:1634:15 + #1 0x75cd3210d198 (/usr/lib/libglib-2.0.so.0+0x5a198) (BuildId: 0a2311dfbbc6c215dc36f4b6bdd2b4b6fbae55a2) + #2 0x75cd3216c3be (/usr/lib/libglib-2.0.so.0+0xb93be) (BuildId: 0a2311dfbbc6c215dc36f4b6bdd2b4b6fbae55a2) + #3 0x75cd3210ddc6 in g_main_loop_run (/usr/lib/libglib-2.0.so.0+0x5adc6) (BuildId: 0a2311dfbbc6c215dc36f4b6bdd2b4b6fbae55a2) + #4 0x75cd3111a47c in virNetClientIOEventLoop /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:1722:9 + #5 0x75cd3111a47c in virNetClientIO /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:2002:10 + #6 0x75cd3111a47c in virNetClientSendInternal /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:2170:11 + #7 0x75cd311198a8 in virNetClientSendWithReply /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclient.c:2198:11 + #8 0x75cd31111653 in virNetClientProgramCall /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/rpc/virnetclientprogram.c:318:9 + #9 0x75cd31241c8f in callFull /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/remote/remote_driver.c:6054:10 + #10 0x75cd31241c8f in call /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/remote/remote_driver.c:6076:12 + #11 0x75cd31241c8f in remoteNetworkGetXMLDesc /usr/src/debug/libvirt/libvirt-10.2.0/build/src/remote/remote_client_bodies.h:5959:9 + #12 0x75cd31410ff7 in virNetworkGetXMLDesc /usr/src/debug/libvirt/libvirt-10.2.0/build/../src/libvirt-network.c:952:15 + +The root cause is a bad assumption in the virNetClientIOEventLoop +method. This method is run by whichever thread currently owns the +buck, and is responsible for handling I/O. Inside a for(;;) loop, +this method creates a temporary GSource, adds it to the event loop +and runs g_main_loop_run(). When I/O is ready, the GSource callback +(virNetClientIOEventFD) will fire and call g_main_loop_quit(), and +return G_SOURCE_REMOVE which results in the temporary GSource being +destroyed. A g_autoptr() will then remove the last reference. + +What was overlooked, is that a second thread can come along and +while it can't enter virNetClientIOEventLoop, it will register an +idle source that uses virNetClientIOWakeup to interrupt the +original thread's 'g_main_loop_run' call. When this happens the +virNetClientIOEventFD callback never runs, and so the temporary +GSource is not destroyed. The g_autoptr() will remove a reference, +but by virtue of still being attached to the event context, there +is an extra reference held causing GSource to be leaked. The +next time 'g_main_loop_run' is called, the original GSource will +trigger its callback, and access data that was allocated on the +stack by the previous thread, and likely SEGV. + +To solve this, the thread calling 'g_main_loop_run' must call +g_source_destroy, immediately upon return, to guarantee that +the temporary GSource is removed. + +CVE-2024-4418 +Reviewed-by: Ján Tomko +Reported-by: Martin Shirokov +Tested-by: Martin Shirokov +Signed-off-by: Daniel P. Berrangé +--- + src/rpc/virnetclient.c | 14 +++++++++++++- + 1 file changed, 13 insertions(+), 1 deletion(-) + +diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c +index 68098b1c8d..147b0d661a 100644 +--- a/src/rpc/virnetclient.c ++++ b/src/rpc/virnetclient.c +@@ -1657,7 +1657,7 @@ static int virNetClientIOEventLoop(virNetClient *client, + #endif /* !WIN32 */ + int timeout = -1; + virNetMessage *msg = NULL; +- g_autoptr(GSource) G_GNUC_UNUSED source = NULL; ++ g_autoptr(GSource) source = NULL; + GIOCondition ev = 0; + struct virNetClientIOEventData data = { + .client = client, +@@ -1721,6 +1721,18 @@ static int virNetClientIOEventLoop(virNetClient *client, + + g_main_loop_run(client->eventLoop); + ++ /* ++ * If virNetClientIOEventFD ran, this GSource will already be ++ * destroyed due to G_SOURCE_REMOVE. It is harmless to re-destroy ++ * it, since we still own a reference. ++ * ++ * If virNetClientIOWakeup ran, it will have interrupted the ++ * g_main_loop_run call, before virNetClientIOEventFD could ++ * run, and thus the GSource is still registered, and we need ++ * to destroy it since it is referencing stack memory for 'data' ++ */ ++ g_source_destroy(source); ++ + #ifndef WIN32 + ignore_value(pthread_sigmask(SIG_SETMASK, &oldmask, NULL)); + #endif /* !WIN32 */ +-- +2.45.1 + diff --git a/0001-rpc-fix-race-in-waking-up-client-event-loop.patch b/0001-rpc-fix-race-in-waking-up-client-event-loop.patch new file mode 100644 index 0000000..0eff797 --- /dev/null +++ b/0001-rpc-fix-race-in-waking-up-client-event-loop.patch @@ -0,0 +1,76 @@ +From 7cb03e6a28e465c49f0cabe8fe2e7d21edb5aadf Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= +Date: Mon, 18 Dec 2023 12:17:18 +0000 +Subject: [PATCH] rpc: fix race in waking up client event loop +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +The first thread to issue a client RPC request will own the event +loop execution, sitting in the virNetClientIOEventLoop function. + +It releases the client lock while running: + + virNetClientUnlock() + g_main_loop_run() + virNetClientLock() + +If a second thread arrives with an RPC request, it will queue it +for the first thread to process. To inform the first thread that +there's a new request it calls g_main_loop_quit() to break it out +of the main loop. + +This works if the first thread is in g_main_loop_run() at that +time. There is a small window of opportunity, however, where +the first thread has released the client lock, but not yet got +into g_main_loop_run(). If that happens, the wakeup from the +second thread is lost. + +This patch deals with that by changing the way the wakeup is +performed. Instead of directly calling g_main_loop_quit(), the +second thread creates an idle source to run the quit function +from within the first thread. This guarantees that the first +thread will see the wakeup. + +Tested by: Fima Shevrin +Reviewed-by: Michal Privoznik +Reviewed-by: Denis V. Lunev +Signed-off-by: Daniel P. Berrangé +--- + src/rpc/virnetclient.c | 13 ++++++++++++- + 1 file changed, 12 insertions(+), 1 deletion(-) + +diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c +index 4ab8af68c5..68098b1c8d 100644 +--- a/src/rpc/virnetclient.c ++++ b/src/rpc/virnetclient.c +@@ -1848,6 +1848,15 @@ static void virNetClientIOUpdateCallback(virNetClient *client, + } + + ++static gboolean virNetClientIOWakeup(gpointer opaque) ++{ ++ GMainLoop *loop = opaque; ++ ++ g_main_loop_quit(loop); ++ ++ return G_SOURCE_REMOVE; ++} ++ + /* + * This function sends a message to remote server and awaits a reply + * +@@ -1925,7 +1934,9 @@ static int virNetClientIO(virNetClient *client, + /* Check to see if another thread is dispatching */ + if (client->haveTheBuck) { + /* Force other thread to wakeup from poll */ +- g_main_loop_quit(client->eventLoop); ++ GSource *wakeup = g_idle_source_new(); ++ g_source_set_callback(wakeup, virNetClientIOWakeup, client->eventLoop, NULL); ++ g_source_attach(wakeup, client->eventCtx); + + /* If we are non-blocking, detach the thread and keep the call in the + * queue. */ +-- +2.43.0 + diff --git a/libvirt.spec b/libvirt.spec index 6d319ac..e708ede 100644 --- a/libvirt.spec +++ b/libvirt.spec @@ -229,7 +229,7 @@ Summary: Library providing a simple virtualization API Name: libvirt Version: 9.7.0 -Release: 3%{?dist} +Release: 4%{?dist} License: GPL-2.0-or-later AND LGPL-2.1-only AND LGPL-2.1-or-later AND OFL-1.1 URL: https://libvirt.org/ @@ -246,6 +246,10 @@ Patch1: libvirt-regression-input-default-bus.patch Patch2: 0001-interface-fix-udev_device_get_sysattr_value-return-v.patch # CVE listing interfaces with missized array Patch3: 0001-Fix-off-by-one-error-in-udevListInterfacesByStatus.patch +Patch4: 0001-rpc-ensure-temporary-GSource-is-removed-from-client-.patch +Patch5: 0001-rpc-fix-race-in-waking-up-client-event-loop.patch +Patch6: 0001-rpc-avoid-leak-of-GSource-in-use-for-interrupting-ma.patch +Patch7: 0001-interface-fix-udev-reference-leak-with-invalid-flags.patch Requires: libvirt-daemon = %{version}-%{release} Requires: libvirt-daemon-config-network = %{version}-%{release} @@ -2479,6 +2483,12 @@ exit 0 %changelog +* Wed Jun 5 2024 Daniel P. Berrangé - 9.7.0-4 +- Fix crash in event loop (CVE-2024-4418) +- Fix I/O stall when multiple threads issue RPC calls +- Fix leak of GSource handle +- Fix leak of udev reference (rhbz #2266017) + * Tue Mar 12 2024 Daniel P. Berrangé - 9.7.0-3 - Fix crash listing interfaces with missing link status attribute (rhbz #2266014) - Fix crash listing interfaces with missized array (CVE-2024-1441)