Compare commits

...

61 Commits

Author SHA1 Message Date
Daniel P. Berrangé 5ff9ee11bb Rebuild for wireshark soname change rhbz#2031322
Now with missing patches

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2021-12-13 15:33:14 +00:00
Daniel P. Berrangé a5abe2ec98 Rebuild for wireshark soname change rhbz#2031322
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2021-12-13 15:21:21 +00:00
Cole Robinson 3e09e1c917 libvirt-7.0.0-7
More CAP_SETPCAP warning fixes (bz #1924218)
Handle unknown firmware.json errors
CVE-2021-3667: Fix deadlock on virStoragePoolLookupByTargetPath failure (bz #1986113)
2021-07-27 14:18:27 -04:00
Cole Robinson aa95d82b58 libvirt-7.0.0-6
CVE-2021-3631 libvirt: insecure sVirt label generation (bz #1977760)
2021-07-02 16:32:25 -04:00
Cole Robinson 476c3411a3 libvirt-7.0.0-5
Crash in udev driver populate_vendor (bz #1966851)
Fix CAP_SETPCAP syslog warning (bz #1924218)
2021-06-29 09:51:22 -04:00
Cole Robinson bbfa59c8da Increase meson test timeout to fix builds on s390x copr
Signed-off-by: Cole Robinson <crobinso@redhat.com>
2021-02-03 14:23:33 -05:00
Laine Stump b97ad8956c Disable netcf backend for interface driver
Signed-off-by: Laine Stump <laine@redhat.com>
2021-02-02 12:47:35 -05:00
Fedora Release Engineering f9639a0012 - Rebuilt for https://fedoraproject.org/wiki/Fedora_34_Mass_Rebuild
Signed-off-by: Fedora Release Engineering <releng@fedoraproject.org>
2021-01-26 18:16:33 +00:00
Daniel P. Berrangé c71e60c76c Rebase to 7.0.0 release
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2021-01-15 17:44:42 +00:00
Daniel P. Berrangé 9764121247 Cull 2019 vintage changelogs
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2021-01-05 18:12:21 +00:00
Cole Robinson a18ed3c492 Rebuild for wireshark soname bump
Signed-off-by: Cole Robinson <crobinso@redhat.com>
2020-12-04 10:25:53 -05:00
Richard W.M. Jones 9e2eeb32e3 Build libvirt-daemon-kvm for riscv64. 2020-12-04 11:38:09 +00:00
Cole Robinson 8bf91f5d22 libvirt-6.10.0-1
Update to version 6.10.0
2020-12-01 10:16:26 -05:00
Daniel P. Berrangé 91d8eff569 Re-apply reverted fix for disabling glusterfs, curl, openswman and libiscsi
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-11-04 09:13:29 +00:00
Cole Robinson a761ea338f libvirt-6.9.0-1
Update to version 6.9.0
2020-11-03 11:22:45 -05:00
Daniel P. Berrangé 7eef88add4 Really fix disabling glusterfs, curl, openwsman and libiscsi
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-11-02 12:50:07 +00:00
Daniel P. Berrangé 0c88a64234 Fix name of meson option for disabling glusterfs
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-11-02 11:09:26 +00:00
Daniel P. Berrangé c2f3be1f4d Fix stat() mocking for new glibc
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-10-30 15:36:08 +00:00
Daniel P. Berrangé 18b1c9d77e Fix disablement of libssh/libssh2 & remove obsolete conditions
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-10-28 14:52:45 +00:00
Cole Robinson 4528832026 Fix glib errors in virsystemdtest
Signed-off-by: Cole Robinson <crobinso@redhat.com>
2020-10-02 16:34:07 -04:00
Cole Robinson 037cd36ba2 libvirt-6.8.0-1
Update to version 6.8.0
2020-10-01 13:08:10 -04:00
Daniel P. Berrangé b94fb6a393 Fix QEMU start when KVM is not loaded (rhbz#1875327)
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-09-04 14:05:21 +01:00
Cole Robinson b0c05ac2ae libvirt-6.7.0-1
Update to version 6.7.0
2020-09-01 10:36:57 -04:00
Merlin Mathesius 15c305caf3 Fix so no platforms depend upon ceph for 32-bit architectures
Include upstream patches to fix GCC 10 warnings

Signed-off-by: Merlin Mathesius <mmathesi@redhat.com>
2020-08-21 11:28:09 -05:00
Daniel P. Berrangé bc1b3e59a7 Fix creation of pools on non-btrfs (rhbz#1870197)
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-08-21 11:47:16 +01:00
Daniel P. Berrangé 3c4492c415 Add a reminder to re-enable LTO in next release
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-08-07 17:30:14 +01:00
Cole Robinson fecc0295a2 libvirt-6.6.0-1
Update to version 6.6.0
2020-08-04 10:35:23 -04:00
Daniel P. Berrangé 325dc837c6 Disable LTO to avoid LD_PRELOAD breakage
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-08-03 15:31:10 +01:00
Fedora Release Engineering 5a45cb7f0d - Rebuilt for https://fedoraproject.org/wiki/Fedora_33_Mass_Rebuild
Signed-off-by: Fedora Release Engineering <releng@fedoraproject.org>
2020-07-28 05:51:41 +00:00
Cole Robinson 21269530e8 libvirt-6.5.0-1
Update to version 6.5.0
2020-07-03 10:01:57 -04:00
Cole Robinson fc18e983e7 libvirt-6.4.0-1
Update to version 6.4.0
2020-06-02 13:43:43 -04:00
Cole Robinson 4a7cc743a9 libvirt-6.3.0-1
Update to version 6.3.0
2020-05-05 19:33:44 -04:00
Cole Robinson 4398397311 libvirt-6.2.0-1
Update to version 6.2.0
2020-04-02 16:10:01 -04:00
Felipe Borges 401f9087bc libvirt-6.1.0-2.fc32
Check for disk type correctly in virDomainDiskTranslateSourcePool

Signed-off-by: Felipe Borges <feborges@redhat.com>
2020-03-24 11:34:46 +01:00
Cole Robinson dd3698988e libvirt-6.1.0-1.fc33
Update to version 6.1.0
2020-03-04 08:25:20 -05:00
Cole Robinson 3df49002c2 libvirt-6.0.0-3.fc33
Rebuild for libiscsi soname bump
2020-02-25 14:47:54 -05:00
Fedora Release Engineering 1bf0aedb9f - Rebuilt for https://fedoraproject.org/wiki/Fedora_32_Mass_Rebuild
Signed-off-by: Fedora Release Engineering <releng@fedoraproject.org>
2020-01-29 10:43:22 +00:00
Cole Robinson 79accd5a6e libvirt-6.0.0-1.fc32
Update to version 6.0.0
2020-01-15 11:17:10 -05:00
Adam Williamson 422d548fce Rebuild for new xen-libs 2019-12-19 09:02:02 -08:00
Daniel P. Berrangé 20ec6e955c Add abigail suppressions list
Stop abigail complaining about libvirt private symbols or any of the
loadable driver modules.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-12-03 18:01:54 +00:00
Cole Robinson 406a235cbc Fix test suite
Signed-off-by: Cole Robinson <crobinso@redhat.com>
2019-12-03 10:22:18 -05:00
Cole Robinson 4d15e56a38 libvirt-5.10.0-1.fc32
Update to version 5.10.0
2019-12-03 09:54:02 -05:00
Cole Robinson 8693111790 libvirt-5.9.0-1.fc32
Update to version 5.9.0
2019-11-11 13:36:07 -05:00
Cole Robinson 8461a521f2 libvirt-5.8.0-1.fc32
Update to version 5.8.0
2019-10-07 16:35:23 -04:00
Cole Robinson 511e2c5124 libvirt-5.7.0-3.fc32
Fix VM startup when legacy cgroups are defined (bz #1612383)
2019-09-26 15:47:05 -04:00
Daniel P. Berrangé 749dc3d3cb Fix systemd socket activation with TLS socket 2019-09-20 10:14:22 +01:00
Cole Robinson 73c9e38f2c Remove old Makefile
Signed-off-by: Cole Robinson <crobinso@redhat.com>
2019-09-17 10:54:06 -04:00
Cole Robinson afe8184d3f libvirt-5.7.0-1.fc32
Update to version 5.7.0
2019-09-03 12:35:57 -04:00
Cole Robinson 7d77bf76aa libvirt-5.6.0-1.fc31
Update to version 5.6.0
2019-08-06 13:23:41 -04:00
Fedora Release Engineering 7f0f1b831e - Rebuilt for https://fedoraproject.org/wiki/Fedora_31_Mass_Rebuild
Signed-off-by: Fedora Release Engineering <releng@fedoraproject.org>
2019-07-25 14:40:37 +00:00
Cole Robinson d964be0097 libvirt-5.5.0-1.fc31
Rebased to version 5.5.0
2019-07-03 11:25:28 -04:00
Cole Robinson a2479f539e libvirt-5.4.0-2.fc31
CVE-2019-10161: arbitrary file read/exec via virDomainSaveImageGetXMLDesc API (bz #1722463, bz #1720115)
CVE-2019-10166: virDomainManagedSaveDefineXML API exposed to readonly clients (bz #1722462, bz #1720114)
CVE-2019-10167: arbitrary command execution via virConnectGetDomainCapabilities API (bz #1722464, bz #1720117)
CVE-2019-10168: arbitrary command execution via virConnectBaselineHypervisorCPU and virConnectCompareHypervisorCPU APIs (bz #1722466, bz #1720118)
2019-06-20 12:30:57 -04:00
Daniel P. Berrangé c2c89ec6a3 Update to 5.4.0 release
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-06-12 11:29:13 +02:00
Daniel P. Berrangé d61e2404ba Fix systemd socket permissions (CVE-2019-10132)
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-05-21 16:58:15 +01:00
Daniel P. Berrangé a2be167dfe Define md-clear CPUID bit
CVE-2018-12126, CVE-2018-12127, CVE-2018-12130, CVE-2019-11091

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-05-14 19:17:13 +01:00
Daniel P. Berrangé 15ca09e1da Update to 5.3.0 release
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-05-07 14:50:34 +01:00
Cole Robinson 2cc07e6366 libvirt-5.2.0-2.fc31
Rebuild for xen 4.12 soname bump
2019-04-08 10:24:27 -04:00
Daniel P. Berrangé 4dd4fe78a4 Remove patch chunk touching Makefile.am
We don't need to run make dist from a build and the filename is broken
and would require re-running autotools anyway.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-04-03 13:59:48 +01:00
Daniel P. Berrangé 0bc22fb6f7 Remove patching on mock-noinline.pl which is missing in dist
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-04-03 11:47:23 +01:00
Daniel P. Berrangé 76097a6961 Add fix for mocking tests on 32-bit
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-04-03 11:35:55 +01:00
Daniel P. Berrangé 98cbf39fd3 Update to 5.2.0 release
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2019-04-03 11:06:44 +01:00
18 changed files with 1028 additions and 1011 deletions
+31
View File
@@ -0,0 +1,31 @@
[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.*
@@ -0,0 +1,56 @@
From: Laine Stump <laine@redhat.com>
Date: Thu, 21 Jan 2021 16:01:06 -0500
Subject: [PATCH] build: support explicitly disabling netcf
placing "-Dnetcf=disabled" on the meson commandline was ignored,
meaning that even with that option the build would get WITH_NETCF if
the netcf-devel package was found - the only way to disable it was to
uninstall netcf-devel.
This patch adds the small bit of logic to check the netcf meson
commandline option (in addition to whether netcf-devel is installed)
before defining WITH_NETCF.
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Neal Gompa <ngompa13@gmail.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
(cherry picked from commit 06169a115d46d8870a96d293c2faf6ea87e71020)
---
meson.build | 10 ++++++----
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/meson.build b/meson.build
index b5164f68ed..e9d6d9f82e 100644
--- a/meson.build
+++ b/meson.build
@@ -1155,8 +1155,10 @@ libm_dep = cc.find_library('m', required : false)
netcf_version = '0.1.8'
netcf_dep = dependency('netcf', version: '>=' + netcf_version, required: get_option('netcf'))
-if netcf_dep.found()
- conf.set('WITH_NETCF', 1)
+if not get_option('netcf').disabled()
+ if netcf_dep.found()
+ conf.set('WITH_NETCF', 1)
+ endif
endif
have_gnu_gettext_tools = false
@@ -1550,7 +1552,7 @@ elif get_option('driver_hyperv').enabled()
error('openwsman is required for the Hyper-V driver')
endif
-if not get_option('driver_interface').disabled() and conf.has('WITH_LIBVIRTD') and (udev_dep.found() or netcf_dep.found())
+if not get_option('driver_interface').disabled() and conf.has('WITH_LIBVIRTD') and (udev_dep.found() or conf.has('WITH_NETCF'))
conf.set('WITH_INTERFACE', 1)
elif get_option('driver_interface').enabled()
error('Requested the Interface driver without netcf or udev and libvirtd support')
@@ -2362,7 +2364,7 @@ libs_summary = {
'libssh': libssh_dep.found(),
'libssh2': libssh2_dep.found(),
'libutil': libutil_dep.found(),
- 'netcf': netcf_dep.found(),
+ 'netcf': conf.has('WITH_NETCF'),
'NLS': have_gnu_gettext_tools,
'numactl': numactl_dep.found(),
'openwsman': openwsman_dep.found(),
@@ -1,149 +0,0 @@
From 092320f10b47bd6aca1f29278fcdc6b0efaf636a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Mon, 18 Mar 2019 10:58:48 +0000
Subject: [PATCH 1/5] storage: split off code for calling rbd_list
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
The rbd_list method has a quite unpleasant signature returning an
array of strings in a single buffer instead of an array. It is
being deprecated in favour of rbd_list2. To maintain clarity of
code when supporting both APIs in parallel, split the rbd_list
code out into a separate method.
In splitting this we now honour the rbd_list failures.
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 28c8403ed07896d6d7e06d7726ed904027206719)
---
src/storage/storage_backend_rbd.c | 83 +++++++++++++++++++++----------
1 file changed, 58 insertions(+), 25 deletions(-)
diff --git a/src/storage/storage_backend_rbd.c b/src/storage/storage_backend_rbd.c
index 2b7af1db23..0865163756 100644
--- a/src/storage/storage_backend_rbd.c
+++ b/src/storage/storage_backend_rbd.c
@@ -565,19 +565,68 @@ volStorageBackendRBDRefreshVolInfo(virStorageVolDefPtr vol,
return ret;
}
+
+static char **
+virStorageBackendRBDGetVolNames(virStorageBackendRBDStatePtr ptr)
+{
+ char **names = NULL;
+ size_t nnames = 0;
+ int rc;
+ size_t max_size = 1024;
+ VIR_AUTOFREE(char *) namebuf = NULL;
+ const char *name;
+
+ while (true) {
+ if (VIR_ALLOC_N(namebuf, max_size) < 0)
+ goto error;
+
+ rc = rbd_list(ptr->ioctx, namebuf, &max_size);
+ if (rc >= 0)
+ break;
+ if (rc != -ERANGE) {
+ virReportSystemError(-rc, "%s", _("Unable to list RBD images"));
+ goto error;
+ }
+ VIR_FREE(namebuf);
+ }
+
+ for (name = namebuf; name < namebuf + max_size;) {
+ VIR_AUTOFREE(char *) namedup = NULL;
+
+ if (STREQ(name, ""))
+ break;
+
+ if (VIR_STRDUP(namedup, name) < 0)
+ goto error;
+
+ if (VIR_APPEND_ELEMENT(names, nnames, namedup) < 0)
+ goto error;
+
+ name += strlen(name) + 1;
+ }
+
+ if (VIR_EXPAND_N(names, nnames, 1) < 0)
+ goto error;
+
+ return names;
+
+ error:
+ virStringListFreeCount(names, nnames);
+ return NULL;
+}
+
+
static int
virStorageBackendRBDRefreshPool(virStoragePoolObjPtr pool)
{
- size_t max_size = 1024;
int ret = -1;
- int len = -1;
int r = 0;
- char *name;
virStoragePoolDefPtr def = virStoragePoolObjGetDef(pool);
virStorageBackendRBDStatePtr ptr = NULL;
struct rados_cluster_stat_t clusterstat;
struct rados_pool_stat_t poolstat;
- VIR_AUTOFREE(char *) names = NULL;
+ char **names = NULL;
+ size_t i;
if (!(ptr = virStorageBackendRBDNewState(pool)))
goto cleanup;
@@ -602,33 +651,16 @@ virStorageBackendRBDRefreshPool(virStoragePoolObjPtr pool)
def->source.name, clusterstat.kb, clusterstat.kb_avail,
poolstat.num_bytes);
- while (true) {
- if (VIR_ALLOC_N(names, max_size) < 0)
- goto cleanup;
-
- len = rbd_list(ptr->ioctx, names, &max_size);
- if (len >= 0)
- break;
- if (len != -ERANGE) {
- VIR_WARN("%s", "A problem occurred while listing RBD images");
- goto cleanup;
- }
- VIR_FREE(names);
- }
+ if (!(names = virStorageBackendRBDGetVolNames(ptr)))
+ goto cleanup;
- for (name = names; name < names + max_size;) {
+ for (i = 0; names[i] != NULL; i++) {
VIR_AUTOPTR(virStorageVolDef) vol = NULL;
- if (STREQ(name, ""))
- break;
-
if (VIR_ALLOC(vol) < 0)
goto cleanup;
- if (VIR_STRDUP(vol->name, name) < 0)
- goto cleanup;
-
- name += strlen(name) + 1;
+ VIR_STEAL_PTR(vol->name, names[i]);
r = volStorageBackendRBDRefreshVolInfo(vol, pool, ptr);
@@ -661,6 +693,7 @@ virStorageBackendRBDRefreshPool(virStoragePoolObjPtr pool)
ret = 0;
cleanup:
+ virStringListFree(names);
virStorageBackendRBDFreeState(&ptr);
return ret;
}
--
2.20.1
@@ -0,0 +1,72 @@
From: wangjian <wangjian161@huawei.com>
Date: Fri, 26 Mar 2021 11:21:16 +0800
Subject: [PATCH] node_device_udev: Serialize access to pci_get_strings)_
Since the functions provided by libpciaccess are not thread-safe,
when the udev-event and nodedev-init threads of libvirt call the
pci_get_strings function provided by libpaciaccess at the same
time the following can happen:
nodedev-init thread:
nodeStateInitializeEnumerate ->
udevEnumerateDevices->
udevProcessDeviceListEntry ->
udevAddOneDevice ->
udevGetDeviceDetails->
udevProcessPCI ->
udevTranslatePCIIds ->
pci_get_strings -> (libpciaccess)
find_device_name ->
populate_vendor ->
d = realloc( vend->devices, (vend->num_devices + 1), * sizeof( struct pci_device_leaf ) );
vend->num_devices++;
udev-event thread:
udevEventHandleThread ->
udevHandleOneDevice ->
udevAddOneDevice->
udevGetDeviceDetails->
udevProcessPCI ->
udevTranslatePCIIds ->
pci_get_strings -> (libpciaccess)
find_device_name ->
populate_vendor ->
d = realloc( vend->devices, (vend->num_devices + 1), * sizeof( struct pci_device_leaf ) );
vend->num_devices++;
Signed-off-by: WangJian <wangjian161@huawei.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
(cherry picked from commit 59788a5caea5f292c86e07a31ee2b853d68db87e)
---
src/node_device/node_device_udev.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c
index 55a2731681..6f0defe908 100644
--- a/src/node_device/node_device_udev.c
+++ b/src/node_device/node_device_udev.c
@@ -328,6 +328,7 @@ udevGenerateDeviceName(struct udev_device *device,
return 0;
}
+static virMutex pciaccessMutex = VIR_MUTEX_INITIALIZER;
static int
udevTranslatePCIIds(unsigned int vendor,
@@ -346,12 +347,14 @@ udevTranslatePCIIds(unsigned int vendor,
m.device_class_mask = 0;
m.match_data = 0;
- /* pci_get_strings returns void */
+ /* pci_get_strings returns void and unfortunately is not thread safe. */
+ virMutexLock(&pciaccessMutex);
pci_get_strings(&m,
&device_name,
&vendor_name,
NULL,
NULL);
+ virMutexUnlock(&pciaccessMutex);
*vendor_string = g_strdup(vendor_name);
*product_string = g_strdup(device_name);
@@ -1,96 +0,0 @@
From e8ec2592202387cca8e45cf15bd55ed5a952f3e3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Mon, 18 Mar 2019 11:11:38 +0000
Subject: [PATCH 2/5] storage: add support for new rbd_list2 method
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
The rbd_list method has been deprecated in Ceph >= 14.0.0
in favour of the new rbd_list2 method which populates an
array of structs.
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 3aa190f2a43a632b542a6ba751a6c3ab4d51f1dd)
---
m4/virt-storage-rbd.m4 | 1 +
src/storage/storage_backend_rbd.c | 43 +++++++++++++++++++++++++++++++
2 files changed, 44 insertions(+)
diff --git a/m4/virt-storage-rbd.m4 b/m4/virt-storage-rbd.m4
index 17e2115309..f3d9d04908 100644
--- a/m4/virt-storage-rbd.m4
+++ b/m4/virt-storage-rbd.m4
@@ -33,6 +33,7 @@ AC_DEFUN([LIBVIRT_STORAGE_CHECK_RBD], [
old_LIBS="$LIBS"
LIBS="$LIBS $LIBRBD_LIBS"
AC_CHECK_FUNCS([rbd_get_features],[],[LIBRBD_FOUND=no])
+ AC_CHECK_FUNCS([rbd_list2])
LIBS="$old_LIBS"
fi
diff --git a/src/storage/storage_backend_rbd.c b/src/storage/storage_backend_rbd.c
index 0865163756..bfc3419f9c 100644
--- a/src/storage/storage_backend_rbd.c
+++ b/src/storage/storage_backend_rbd.c
@@ -566,6 +566,48 @@ volStorageBackendRBDRefreshVolInfo(virStorageVolDefPtr vol,
}
+#ifdef HAVE_RBD_LIST2
+static char **
+virStorageBackendRBDGetVolNames(virStorageBackendRBDStatePtr ptr)
+{
+ char **names = NULL;
+ size_t nnames = 0;
+ int rc;
+ rbd_image_spec_t *images = NULL;
+ size_t nimages = 16;
+ size_t i;
+
+ while (true) {
+ if (VIR_ALLOC_N(images, nimages) < 0)
+ goto error;
+
+ rc = rbd_list2(ptr->ioctx, images, &nimages);
+ if (rc >= 0)
+ break;
+ if (rc != -ERANGE) {
+ virReportSystemError(-rc, "%s", _("Unable to list RBD images"));
+ goto error;
+ }
+ }
+
+ if (VIR_ALLOC_N(names, nimages + 1) < 0)
+ goto error;
+ nnames = nimages;
+
+ for (i = 0; i < nimages; i++)
+ VIR_STEAL_PTR(names[i], images->name);
+
+ return names;
+
+ error:
+ virStringListFreeCount(names, nnames);
+ rbd_image_spec_list_cleanup(images, nimages);
+ VIR_FREE(images);
+ return NULL;
+}
+
+#else /* ! HAVE_RBD_LIST2 */
+
static char **
virStorageBackendRBDGetVolNames(virStorageBackendRBDStatePtr ptr)
{
@@ -614,6 +656,7 @@ virStorageBackendRBDGetVolNames(virStorageBackendRBDStatePtr ptr)
virStringListFreeCount(names, nnames);
return NULL;
}
+#endif /* ! HAVE_RBD_LIST2 */
static int
--
2.20.1
@@ -1,137 +0,0 @@
From b990740b12117eaaf2797141a53a30b41f07c791 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Mon, 18 Mar 2019 17:31:21 +0000
Subject: [PATCH 3/5] network: improve error report when firewall chain
creation fails
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
During startup we create some top level chains in which all
virtual network firewall rules will be placed. The upfront
creation is done to avoid slowing down creation of individual
virtual networks by checking for chain existance every time.
There are some factors which can cause this upfront creation
to fail and while a message will get into the libvirtd log
this won't be seen by users who later try to start a virtual
network. Instead they'll just get a message saying that the
libvirt top level chain does not exist. This message is
accurate, but unhelpful for solving the root cause.
This patch thus saves any error during daemon startup and
reports it when trying to create a virtual network later.
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 9f4e35dc73ec9e940aa61bc7c140c2b800218ef3)
---
src/network/bridge_driver.c | 3 +--
src/network/bridge_driver_linux.c | 31 +++++++++++++++++++++-------
src/network/bridge_driver_nop.c | 3 +--
src/network/bridge_driver_platform.h | 2 +-
4 files changed, 27 insertions(+), 12 deletions(-)
diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index b3ca5b8a15..1da60f0a21 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -2108,8 +2108,7 @@ static void
networkReloadFirewallRules(virNetworkDriverStatePtr driver, bool startup)
{
VIR_INFO("Reloading iptables rules");
- if (networkPreReloadFirewallRules(startup) < 0)
- return;
+ networkPreReloadFirewallRules(startup);
virNetworkObjListForEach(driver->networks,
networkReloadFirewallRulesHelper,
NULL);
diff --git a/src/network/bridge_driver_linux.c b/src/network/bridge_driver_linux.c
index b10d0a6c4d..c899f4b6d0 100644
--- a/src/network/bridge_driver_linux.c
+++ b/src/network/bridge_driver_linux.c
@@ -35,11 +35,25 @@ VIR_LOG_INIT("network.bridge_driver_linux");
#define PROC_NET_ROUTE "/proc/net/route"
-int networkPreReloadFirewallRules(bool startup)
+static virErrorPtr errInit;
+
+void networkPreReloadFirewallRules(bool startup)
{
- int ret = iptablesSetupPrivateChains();
- if (ret < 0)
- return -1;
+ int rc;
+
+ /* We create global rules upfront as we don't want
+ * the perf hit of conditionally figuring out whether
+ * to create them each time a network is started.
+ *
+ * Any errors here are saved to be reported at time
+ * of starting the network though as that makes them
+ * more likely to be seen by a human
+ */
+ rc = iptablesSetupPrivateChains();
+ if (rc < 0) {
+ errInit = virSaveLastError();
+ virResetLastError();
+ }
/*
* If this is initial startup, and we just created the
@@ -54,10 +68,8 @@ int networkPreReloadFirewallRules(bool startup)
* rules will be present. Thus we can safely just tell it
* to always delete from the builin chain
*/
- if (startup && ret == 1)
+ if (startup && rc == 1)
iptablesSetDeletePrivate(false);
-
- return 0;
}
@@ -671,6 +683,11 @@ int networkAddFirewallRules(virNetworkDefPtr def)
virFirewallPtr fw = NULL;
int ret = -1;
+ if (errInit) {
+ virSetError(errInit);
+ return -1;
+ }
+
if (def->bridgeZone) {
/* if a firewalld zone has been specified, fail/log an error
diff --git a/src/network/bridge_driver_nop.c b/src/network/bridge_driver_nop.c
index a0e57012f9..ea9db338cb 100644
--- a/src/network/bridge_driver_nop.c
+++ b/src/network/bridge_driver_nop.c
@@ -19,9 +19,8 @@
#include <config.h>
-int networkPreReloadFirewallRules(bool startup ATTRIBUTE_UNUSED)
+void networkPreReloadFirewallRules(bool startup ATTRIBUTE_UNUSED)
{
- return 0;
}
diff --git a/src/network/bridge_driver_platform.h b/src/network/bridge_driver_platform.h
index baeb22bc3e..95fd64bdc7 100644
--- a/src/network/bridge_driver_platform.h
+++ b/src/network/bridge_driver_platform.h
@@ -58,7 +58,7 @@ struct _virNetworkDriverState {
typedef struct _virNetworkDriverState virNetworkDriverState;
typedef virNetworkDriverState *virNetworkDriverStatePtr;
-int networkPreReloadFirewallRules(bool startup);
+void networkPreReloadFirewallRules(bool startup);
void networkPostReloadFirewallRules(bool startup);
int networkCheckRouteCollision(virNetworkDefPtr def);
--
2.20.1
@@ -0,0 +1,53 @@
From: Michal Privoznik <mprivozn@redhat.com>
Date: Thu, 24 Jun 2021 16:58:09 +0200
Subject: [PATCH] virSetUIDGIDWithCaps: Don't drop CAP_SETPCAP right away
There are few cases where we execute a virCommand with all caps
cleared (virCommandClearCaps()). For instance
dnsmasqCapsRefreshInternal() does just that. This means, that
after fork() and before exec() the virSetUIDGIDWithCaps() is
called. But since the caller did not want to change anything,
just drop capabilities, these are the values of arguments:
virSetUIDGIDWithCaps (uid=-1, gid=-1, groups=0x0, ngroups=0,
capBits=0, clearExistingCaps=true)
This means that indeed all capabilities will be dropped,
including CAP_SETPCAP. But this capability controls whether
capabilities can be set, IOW whether capng_apply() succeeds.
There are two calls of capng_apply() in the function. The
CAP_SETPCAP is dropped after the first call and thus the other
call (capng_apply(CAPNG_SELECT_BOUNDS);) fails.
The solution is to keep the capability for as long as needed
(just like CAP_SETGID and CAP_SETUID) and drop it only at the
very end (just like CAP_SETGID and CAP_SETUID).
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1949388
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
(cherry picked from commit 438b50dda8a863fdc988e9ab612f097cc1626e8a)
---
src/util/virutil.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/src/util/virutil.c b/src/util/virutil.c
index a0cd0f1bcd..7ae23a7061 100644
--- a/src/util/virutil.c
+++ b/src/util/virutil.c
@@ -1202,12 +1202,10 @@ virSetUIDGIDWithCaps(uid_t uid, gid_t gid, gid_t *groups, int ngroups,
}
# ifdef PR_CAPBSET_DROP
/* If newer kernel, we need also need setpcap to change the bounding set */
- if ((capBits || need_setgid || need_setuid) &&
- !capng_have_capability(CAPNG_EFFECTIVE, CAP_SETPCAP)) {
+ if (!capng_have_capability(CAPNG_EFFECTIVE, CAP_SETPCAP)) {
need_setpcap = true;
- }
- if (need_setpcap)
capng_update(CAPNG_ADD, CAPNG_EFFECTIVE|CAPNG_PERMITTED, CAP_SETPCAP);
+ }
# endif
/* Tell system we want to keep caps across uid change */
@@ -1,153 +0,0 @@
From 095c45036615a84c7150ea801d6932bdde1d5b49 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Mon, 18 Mar 2019 16:49:32 +0000
Subject: [PATCH 4/5] network: split setup of ipv4 and ipv6 top level chains
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
During startup libvirtd creates top level chains for both ipv4
and ipv6 protocols. If this fails for any reason then startup
of virtual networks is blocked.
The default virtual network, however, only requires use of ipv4
and some servers have ipv6 disabled so it is expected that ipv6
chain creation will fail. There could equally be servers with
no ipv4, only ipv6.
This patch thus makes error reporting a little more fine grained
so that it works more sensibly when either ipv4 or ipv6 is
disabled on the server. Only the protocols that are actually
used by the virtual network have errors reported.
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 686803a1a2e1e0641916b1c9e2c7e3910fe598d4)
---
src/network/bridge_driver_linux.c | 34 +++++++++++++++++++++++++------
src/util/viriptables.c | 14 ++++---------
src/util/viriptables.h | 2 +-
3 files changed, 33 insertions(+), 17 deletions(-)
diff --git a/src/network/bridge_driver_linux.c b/src/network/bridge_driver_linux.c
index c899f4b6d0..50fc197134 100644
--- a/src/network/bridge_driver_linux.c
+++ b/src/network/bridge_driver_linux.c
@@ -35,10 +35,12 @@ VIR_LOG_INIT("network.bridge_driver_linux");
#define PROC_NET_ROUTE "/proc/net/route"
-static virErrorPtr errInit;
+static virErrorPtr errInitV4;
+static virErrorPtr errInitV6;
void networkPreReloadFirewallRules(bool startup)
{
+ bool created = false;
int rc;
/* We create global rules upfront as we don't want
@@ -49,11 +51,21 @@ void networkPreReloadFirewallRules(bool startup)
* of starting the network though as that makes them
* more likely to be seen by a human
*/
- rc = iptablesSetupPrivateChains();
+ rc = iptablesSetupPrivateChains(VIR_FIREWALL_LAYER_IPV4);
if (rc < 0) {
- errInit = virSaveLastError();
+ errInitV4 = virSaveLastError();
virResetLastError();
}
+ if (rc)
+ created = true;
+
+ rc = iptablesSetupPrivateChains(VIR_FIREWALL_LAYER_IPV6);
+ if (rc < 0) {
+ errInitV6 = virSaveLastError();
+ virResetLastError();
+ }
+ if (rc)
+ created = true;
/*
* If this is initial startup, and we just created the
@@ -68,7 +80,7 @@ void networkPreReloadFirewallRules(bool startup)
* rules will be present. Thus we can safely just tell it
* to always delete from the builin chain
*/
- if (startup && rc == 1)
+ if (startup && created)
iptablesSetDeletePrivate(false);
}
@@ -683,8 +695,18 @@ int networkAddFirewallRules(virNetworkDefPtr def)
virFirewallPtr fw = NULL;
int ret = -1;
- if (errInit) {
- virSetError(errInit);
+ if (errInitV4 &&
+ (virNetworkDefGetIPByIndex(def, AF_INET, 0) ||
+ virNetworkDefGetRouteByIndex(def, AF_INET, 0))) {
+ virSetError(errInitV4);
+ return -1;
+ }
+
+ if (errInitV6 &&
+ (virNetworkDefGetIPByIndex(def, AF_INET6, 0) ||
+ virNetworkDefGetRouteByIndex(def, AF_INET6, 0) ||
+ def->ipv6nogw)) {
+ virSetError(errInitV6);
return -1;
}
diff --git a/src/util/viriptables.c b/src/util/viriptables.c
index d67b640a3b..0e3c0ad73a 100644
--- a/src/util/viriptables.c
+++ b/src/util/viriptables.c
@@ -127,7 +127,7 @@ iptablesPrivateChainCreate(virFirewallPtr fw,
int
-iptablesSetupPrivateChains(void)
+iptablesSetupPrivateChains(virFirewallLayer layer)
{
virFirewallPtr fw = NULL;
int ret = -1;
@@ -143,17 +143,11 @@ iptablesSetupPrivateChains(void)
};
bool changed = false;
iptablesGlobalChainData data[] = {
- { VIR_FIREWALL_LAYER_IPV4, "filter",
+ { layer, "filter",
filter_chains, ARRAY_CARDINALITY(filter_chains), &changed },
- { VIR_FIREWALL_LAYER_IPV4, "nat",
+ { layer, "nat",
natmangle_chains, ARRAY_CARDINALITY(natmangle_chains), &changed },
- { VIR_FIREWALL_LAYER_IPV4, "mangle",
- natmangle_chains, ARRAY_CARDINALITY(natmangle_chains), &changed },
- { VIR_FIREWALL_LAYER_IPV6, "filter",
- filter_chains, ARRAY_CARDINALITY(filter_chains), &changed },
- { VIR_FIREWALL_LAYER_IPV6, "nat",
- natmangle_chains, ARRAY_CARDINALITY(natmangle_chains), &changed },
- { VIR_FIREWALL_LAYER_IPV6, "mangle",
+ { layer, "mangle",
natmangle_chains, ARRAY_CARDINALITY(natmangle_chains), &changed },
};
size_t i;
diff --git a/src/util/viriptables.h b/src/util/viriptables.h
index 903f390f89..e680407ec8 100644
--- a/src/util/viriptables.h
+++ b/src/util/viriptables.h
@@ -24,7 +24,7 @@
# include "virsocketaddr.h"
# include "virfirewall.h"
-int iptablesSetupPrivateChains (void);
+int iptablesSetupPrivateChains (virFirewallLayer layer);
void iptablesSetDeletePrivate (bool pvt);
--
2.20.1
@@ -0,0 +1,51 @@
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Mon, 28 Jun 2021 13:09:04 +0100
Subject: [PATCH] security: fix SELinux label generation logic
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
A process can access a file if the set of MCS categories
for the file is equal-to *or* a subset-of, the set of
MCS categories for the process.
If there are two VMs:
a) svirt_t:s0:c117
b) svirt_t:s0:c117,c720
Then VM (b) is able to access files labelled for VM (a).
IOW, we must discard case where the categories are equal
because that is a subset of many other valid category pairs.
Fixes: https://gitlab.com/libvirt/libvirt/-/issues/153
CVE-2021-3631
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 15073504dbb624d3f6c911e85557019d3620fdb2)
---
src/security/security_selinux.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/src/security/security_selinux.c b/src/security/security_selinux.c
index 2fc6ef2616..61a871ec3d 100644
--- a/src/security/security_selinux.c
+++ b/src/security/security_selinux.c
@@ -389,7 +389,15 @@ virSecuritySELinuxMCSFind(virSecurityManagerPtr mgr,
VIR_DEBUG("Try cat %s:c%d,c%d", sens, c1 + catMin, c2 + catMin);
if (c1 == c2) {
- mcs = g_strdup_printf("%s:c%d", sens, catMin + c1);
+ /*
+ * A process can access a file if the set of MCS categories
+ * for the file is equal-to *or* a subset-of, the set of
+ * MCS categories for the process.
+ *
+ * IOW, we must discard case where the categories are equal
+ * because that is a subset of other category pairs.
+ */
+ continue;
} else {
if (c1 > c2) {
int t = c1;
@@ -1,50 +0,0 @@
From 3e02ee9b5da7fc7197aaa6d57563349a7670b8a1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= <berrange@redhat.com>
Date: Wed, 13 Mar 2019 16:21:15 +0000
Subject: [PATCH 5/5] network: avoid trying to create global firewall rules if
unprivileged
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
The unprivileged libvirtd does not have permission to create firewall
rules, or bridge devices, or do anything to the host network in
general. Historically we still activate the network driver though and
let the network start API call fail.
The startup code path which reloads firewall rules on active networks
would thus effectively be a no-op when unprivileged as it is impossible
for there to be any active networks
With the change to use a global set of firewall chains, however, we now
have code that is run unconditionally.
Ideally we would not register the network driver at all when
unprivileged, but the entanglement with the virt drivers currently makes
that impractical. As a temporary hack, we just make the firewall reload
into a no-op.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 5d010c3df6152cf5fb00f1f67d22151241f4a8a2)
---
src/network/bridge_driver.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index 1da60f0a21..0e1d5efd8e 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -2108,6 +2108,10 @@ static void
networkReloadFirewallRules(virNetworkDriverStatePtr driver, bool startup)
{
VIR_INFO("Reloading iptables rules");
+ /* Ideally we'd not even register the driver when unprivilegd
+ * but until we untangle the virt driver that's not viable */
+ if (!driver->privileged)
+ return;
networkPreReloadFirewallRules(startup);
virNetworkObjListForEach(driver->networks,
networkReloadFirewallRulesHelper,
--
2.20.1
@@ -0,0 +1,43 @@
From: Michal Privoznik <mprivozn@redhat.com>
Date: Thu, 22 Jul 2021 14:26:00 +0200
Subject: [PATCH] virSetUIDGIDWithCaps: Set bounding capabilities only with
CAP_SETPCAP
In one of my previous patches I've tried to postpone dropping
CAP_SETPCAP until the very end because it's needed for
capng_apply(). What I did not realize back then was that we might
not have the capability to begin with. Because of unknown reasons
capng_apply() pollutes logs only for CAPNG_SELECT_BOUNDS and not
for CAPNG_SELECT_CAPS.
Reproducer is really simple: run libvirtd as a regular user.
During its initialization, libvirtd will spawn some binaries
(dnsmasq, qemu-*, etc.) and while doing so it will try to drop
capabilities.
Anyway, let's call capng_apply(CAPNG_SELECT_BOUNDS) only if we
have the CAP_SETPCAP (which is tracked in need_setpcap variable).
Fixes: 438b50dda8a863fdc988e9ab612f097cc1626e8a
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1924218
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Cole Robinson <crobinso@redhat.com>
(cherry picked from commit a2476f37a7789eb9315b77bb451f4754ef4ef15b)
---
src/util/virutil.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/util/virutil.c b/src/util/virutil.c
index 7ae23a7061..333f99e91d 100644
--- a/src/util/virutil.c
+++ b/src/util/virutil.c
@@ -1269,7 +1269,8 @@ virSetUIDGIDWithCaps(uid_t uid, gid_t gid, gid_t *groups, int ngroups,
* do this if we failed to get the capability above, so ignore the
* return value.
*/
- capng_apply(CAPNG_SELECT_BOUNDS);
+ if (!need_setpcap)
+ capng_apply(CAPNG_SELECT_BOUNDS);
/* Drop the caps that allow setuid/gid (unless they were requested) */
if (need_setgid)
@@ -0,0 +1,57 @@
From: Pavel Hrdina <phrdina@redhat.com>
Date: Mon, 10 May 2021 15:07:09 +0200
Subject: [PATCH] qemu_firmware: don't error out for unknown firmware features
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
When QEMU introduces new firmware features libvirt will fail until we
list that feature in our code as well which doesn't sound right.
We should simply ignore the new feature until we add a proper support
for it.
Reported-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 61d95a1073833ec4323c1ef28e71e913c55aa7b9)
---
src/qemu/qemu_firmware.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/src/qemu/qemu_firmware.c b/src/qemu/qemu_firmware.c
index 639cff7459..e602de22e3 100644
--- a/src/qemu/qemu_firmware.c
+++ b/src/qemu/qemu_firmware.c
@@ -573,6 +573,7 @@ qemuFirmwareFeatureParse(const char *path,
virJSONValuePtr featuresJSON;
g_autoptr(qemuFirmwareFeature) features = NULL;
size_t nfeatures;
+ size_t nparsed = 0;
size_t i;
if (!(featuresJSON = virJSONValueObjectGetArray(doc, "features"))) {
@@ -592,17 +593,16 @@ qemuFirmwareFeatureParse(const char *path,
int tmp;
if ((tmp = qemuFirmwareFeatureTypeFromString(tmpStr)) <= 0) {
- virReportError(VIR_ERR_INTERNAL_ERROR,
- _("unknown feature %s"),
- tmpStr);
- return -1;
+ VIR_DEBUG("ignoring unknown QEMU firmware feature '%s'", tmpStr);
+ continue;
}
- features[i] = tmp;
+ features[nparsed] = tmp;
+ nparsed++;
}
fw->features = g_steal_pointer(&features);
- fw->nfeatures = nfeatures;
+ fw->nfeatures = nparsed;
return 0;
}
@@ -0,0 +1,33 @@
From: Peter Krempa <pkrempa@redhat.com>
Date: Wed, 21 Jul 2021 11:22:25 +0200
Subject: [PATCH] storage_driver: Unlock object on ACL fail in
storagePoolLookupByTargetPath
'virStoragePoolObjListSearch' returns a locked and refed object, thus we
must release it on ACL permission failure.
Fixes: 7aa0e8c0cb8
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1984318
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
(cherry picked from commit 447f69dec47e1b0bd15ecd7cd49a9fd3b050fb87)
---
src/storage/storage_driver.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/storage/storage_driver.c b/src/storage/storage_driver.c
index 16bc53aa46..2787c1671b 100644
--- a/src/storage/storage_driver.c
+++ b/src/storage/storage_driver.c
@@ -1739,8 +1739,10 @@ storagePoolLookupByTargetPath(virConnectPtr conn,
storagePoolLookupByTargetPathCallback,
cleanpath))) {
def = virStoragePoolObjGetDef(obj);
- if (virStoragePoolLookupByTargetPathEnsureACL(conn, def) < 0)
+ if (virStoragePoolLookupByTargetPathEnsureACL(conn, def) < 0) {
+ virStoragePoolObjEndAPI(&obj);
return NULL;
+ }
pool = virGetStoragePool(conn, def->name, def->uuid, NULL, NULL);
virStoragePoolObjEndAPI(&obj);
@@ -0,0 +1,81 @@
From 7e299ba649b1288d529c7595c0e6060c9ae0ff2a Mon Sep 17 00:00:00 2001
From: Michal Privoznik <mprivozn@redhat.com>
Date: Mon, 29 Nov 2021 09:57:49 +0100
Subject: [PATCH 1/2] wireshark: Switch to tvb_bytes_to_str()
When the dissector sees a byte sequence that is either an opaque
data (xdr_opaque) or a byte sequence (xdr_bytes) it formats the
bytes as a hex numbers using our own implementation. But
wireshark already provides a function for it: tvb_bytes_to_str().
NB, the reason why it returns a const string is so that callers
don't try to free it - the string is allocated using an allocator
which will decide when to free it.
The wireshark formatter was introduced in wireshark commit of
v1.99.2~479 and thus is present in the version we require at
least (2.6.0).
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
---
tools/wireshark/src/packet-libvirt.c | 30 ++++++++--------------------
1 file changed, 8 insertions(+), 22 deletions(-)
diff --git a/tools/wireshark/src/packet-libvirt.c b/tools/wireshark/src/packet-libvirt.c
index f43919b05d..cb922b8070 100644
--- a/tools/wireshark/src/packet-libvirt.c
+++ b/tools/wireshark/src/packet-libvirt.c
@@ -158,24 +158,6 @@ dissect_xdr_string(tvbuff_t *tvb, proto_tree *tree, XDR *xdrs, int hf,
}
}
-static const gchar *
-format_xdr_bytes(guint8 *bytes, guint32 length)
-{
- gchar *buf;
- guint32 i;
-
- if (length == 0)
- return "";
- buf = wmem_alloc(wmem_packet_scope(), length*2 + 1);
- for (i = 0; i < length; i++) {
- /* We know that buf has enough size to contain
- 2 * length + '\0' characters. */
- g_snprintf(buf, 2*(length - i) + 1, "%02x", bytes[i]);
- buf += 2;
- }
- return buf - length*2;
-}
-
static gboolean
dissect_xdr_opaque(tvbuff_t *tvb, proto_tree *tree, XDR *xdrs, int hf,
guint32 size)
@@ -187,8 +169,10 @@ dissect_xdr_opaque(tvbuff_t *tvb, proto_tree *tree, XDR *xdrs, int hf,
val = g_malloc(size);
start = xdr_getpos(xdrs);
if ((rc = xdr_opaque(xdrs, (caddr_t)val, size))) {
- proto_tree_add_bytes_format_value(tree, hf, tvb, start, xdr_getpos(xdrs) - start,
- NULL, "%s", format_xdr_bytes(val, size));
+ gint len = xdr_getpos(xdrs) - start;
+ const char *s = tvb_bytes_to_str(wmem_packet_scope(), tvb, start, len);
+
+ proto_tree_add_bytes_format_value(tree, hf, tvb, start, len, NULL, "%s", s);
} else {
proto_tree_add_item(tree, hf_libvirt_unknown, tvb, start, -1, ENC_NA);
}
@@ -207,8 +191,10 @@ dissect_xdr_bytes(tvbuff_t *tvb, proto_tree *tree, XDR *xdrs, int hf,
start = xdr_getpos(xdrs);
if (xdr_bytes(xdrs, (char **)&val, &length, maxlen)) {
- proto_tree_add_bytes_format_value(tree, hf, tvb, start, xdr_getpos(xdrs) - start,
- NULL, "%s", format_xdr_bytes(val, length));
+ gint len = xdr_getpos(xdrs) - start;
+ const char *s = tvb_bytes_to_str(wmem_packet_scope(), tvb, start, len);
+
+ proto_tree_add_bytes_format_value(tree, hf, tvb, start, len, NULL, "%s", s);
/* Seems I can't call xdr_free() for this case.
It will raises SEGV by referencing out of bounds call stack */
free(val);
--
2.33.1
@@ -0,0 +1,33 @@
From 010613cfd8dae6d85602a84c5c95b2d441e1b3d1 Mon Sep 17 00:00:00 2001
From: Michal Privoznik <mprivozn@redhat.com>
Date: Mon, 29 Nov 2021 10:20:05 +0100
Subject: [PATCH 2/2] wireshark: Drop needless comment in dissect_xdr_bytes()
In the dissect_xdr_bytes() there's a comment that the string
allocated by xdr_bytes() can't be freed using xdr_free(). Well,
that is expected because xdr_bytes() used plain calloc() AND the
string is not an XDR struct but plain 'char *' type. Passing it
to xdr_free() must result in weird things happening.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
---
tools/wireshark/src/packet-libvirt.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/tools/wireshark/src/packet-libvirt.c b/tools/wireshark/src/packet-libvirt.c
index cb922b8070..eeacbcdf0e 100644
--- a/tools/wireshark/src/packet-libvirt.c
+++ b/tools/wireshark/src/packet-libvirt.c
@@ -195,8 +195,6 @@ dissect_xdr_bytes(tvbuff_t *tvb, proto_tree *tree, XDR *xdrs, int hf,
const char *s = tvb_bytes_to_str(wmem_packet_scope(), tvb, start, len);
proto_tree_add_bytes_format_value(tree, hf, tvb, start, len, NULL, "%s", s);
- /* Seems I can't call xdr_free() for this case.
- It will raises SEGV by referencing out of bounds call stack */
free(val);
return TRUE;
} else {
--
2.33.1
-21
View File
@@ -1,21 +0,0 @@
# 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 $$d/Makefile.common ] ; then cd $$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)
+517 -404
View File
File diff suppressed because it is too large Load Diff
+1 -1
View File
@@ -1 +1 @@
SHA512 (libvirt-5.1.0.tar.xz) = ca64d7be683614bdeb20a8865655fe80f911cf13c00aed2334db3a2e4131e1dd6fe5e9663a24e6f82161ad5aa53f1a2637cd21730eed46e4764b7eebced94f3f
SHA512 (libvirt-7.0.0.tar.xz) = dd6db5ec4971cf4c6059795fd81d5a3a889b10740e34c3c92271eda1c683c99df2c8f923398065d8a7c4f987a20eb1da617d5297ba8ea5a31f154412af50c343