[libvirt] [PATCH] qemu: hotplug: Don't access srcPriv when it's not allocated
The private data of a virStorageSource which is backing an iSCSI hostdev may be NULL if no authentication is present. The code handling the hotplug would attempt to extract the authentication info stored in 'secinfo' without checking if it is allocated which resulted in a crash. Here we opt the easy way to check if srcPriv is not NULL so that we don't duplicate all the logic which selects whether the disk source has a secret. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1597550 Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/qemu/qemu_hotplug.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index fcd8eb0ffa..075f2fb72e 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -2240,7 +2240,8 @@ qemuDomainAttachHostSCSIDevice(virQEMUDriverPtr driver, if (scsisrc->protocol == VIR_DOMAIN_HOSTDEV_SCSI_PROTOCOL_TYPE_ISCSI) { qemuDomainStorageSourcePrivatePtr srcPriv = QEMU_DOMAIN_STORAGE_SOURCE_PRIVATE(scsisrc->u.iscsi.src); - secinfo = srcPriv->secinfo; + if (srcPriv) + secinfo = srcPriv->secinfo; } if (secinfo && secinfo->type == VIR_DOMAIN_SECRET_INFO_TYPE_AES) { -- 2.16.2
On 07/03/2018 08:05 AM, Peter Krempa wrote:
The private data of a virStorageSource which is backing an iSCSI hostdev may be NULL if no authentication is present. The code handling the hotplug would attempt to extract the authentication info stored in 'secinfo' without checking if it is allocated which resulted in a crash.
Here we opt the easy way to check if srcPriv is not NULL so that we don't duplicate all the logic which selects whether the disk source has a secret.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1597550
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/qemu/qemu_hotplug.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
Reviewed-by: John Ferlan <jferlan@redhat.com> John Feels like we've done this before... 97202988 <sigh> and 6050affb7f
On Tue, Jul 03, 2018 at 09:34:56 -0400, John Ferlan wrote:
On 07/03/2018 08:05 AM, Peter Krempa wrote:
The private data of a virStorageSource which is backing an iSCSI hostdev may be NULL if no authentication is present. The code handling the hotplug would attempt to extract the authentication info stored in 'secinfo' without checking if it is allocated which resulted in a crash.
Here we opt the easy way to check if srcPriv is not NULL so that we don't duplicate all the logic which selects whether the disk source has a secret.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1597550
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/qemu/qemu_hotplug.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
Reviewed-by: John Ferlan <jferlan@redhat.com>
John
Feels like we've done this before...
97202988 <sigh> and 6050affb7f
Specifically the former commit was not good enough to cover this case despite touching this code.
On Tue, Jul 03, 2018 at 04:22 PM +0200, Peter Krempa <pkrempa@redhat.com> wrote:
On Tue, Jul 03, 2018 at 09:34:56 -0400, John Ferlan wrote:
On 07/03/2018 08:05 AM, Peter Krempa wrote:
The private data of a virStorageSource which is backing an iSCSI hostdev may be NULL if no authentication is present. The code handling the hotplug would attempt to extract the authentication info stored in 'secinfo' without checking if it is allocated which resulted in a crash.
Here we opt the easy way to check if srcPriv is not NULL so that we don't duplicate all the logic which selects whether the disk source has a secret.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1597550
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/qemu/qemu_hotplug.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
Reviewed-by: John Ferlan <jferlan@redhat.com>
John
Feels like we've done this before...
97202988 <sigh> and 6050affb7f
Specifically the former commit was not good enough to cover this case despite touching this code.
Would it make sense to always allocate the qemuDomainStorageSourcePrivate struct? We’d be wasting some memory, but at the same time the code would be easier to read and less prone to error. e.g. qemuDomainDeviceDiskDefPostParseRestoreSecAlias would be simpler. For the privateData of virConnect, virDomainVsockDef, virDisk, virDomainVcpu… it’s implemented this way. Anyway: Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com> and thanks for fixing it!
-- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
-- Beste Grüße / Kind regards Marc Hartmayer IBM Deutschland Research & Development GmbH Vorsitzende des Aufsichtsrats: Martina Koederitz Geschäftsführung: Dirk Wittkopp Sitz der Gesellschaft: Böblingen Registergericht: Amtsgericht Stuttgart, HRB 243294
On Tue, Jul 03, 2018 at 17:05:31 +0200, Marc Hartmayer wrote:
On Tue, Jul 03, 2018 at 04:22 PM +0200, Peter Krempa <pkrempa@redhat.com> wrote:
On Tue, Jul 03, 2018 at 09:34:56 -0400, John Ferlan wrote:
Feels like we've done this before...
97202988 <sigh> and 6050affb7f
Specifically the former commit was not good enough to cover this case despite touching this code.
Would it make sense to always allocate the qemuDomainStorageSourcePrivate struct? We’d be wasting some memory, but at the same time the code would be easier to read and less prone to error. e.g. qemuDomainDeviceDiskDefPostParseRestoreSecAlias would be simpler.
For the privateData of virConnect, virDomainVsockDef, virDisk, virDomainVcpu… it’s implemented this way.
Anyway: Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com> and thanks for fixing it!
virStorageSource is not allocated using a helper so this would require refactoring all allocations and passing in of the private data structure which contains the callback to allocate the private data. I was thinking about it but did not find fixing it compelling enough to do it myself.
participants (3)
-
John Ferlan -
Marc Hartmayer -
Peter Krempa