[libvirt] [PATCH] qemu: Introduce qemuDomainDelChardevTLSObjects
Let's make a comment deletion helper similar to the Add helper that can be called after the ExitMonitor. The modify qemuDomainRemoveChrDevice and qemuDomainRemoveRNGDevice to call the helper instead of inlining the copy and pasted code. Signed-off-by: John Ferlan <jferlan@redhat.com> --- An offshoot of the redirdev hot unplug patches from Chen Hanxiao. Let's create a helper that can remove the TLS chardev objects for the specific devices that may have used them. src/qemu/qemu_hotplug.c | 107 +++++++++++++++++++++++------------------------- 1 file changed, 52 insertions(+), 55 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 7de04c85a..85f47bee2 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1562,6 +1562,47 @@ qemuDomainAddChardevTLSObjects(virConnectPtr conn, } +static int +qemuDomainDelChardevTLSObjects(virQEMUDriverPtr driver, + virDomainObjPtr vm, + const char *inAlias) +{ + int ret = -1; + virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); + qemuDomainObjPrivatePtr priv = vm->privateData; + char *tlsAlias = NULL; + char *secAlias = NULL; + + if (!(tlsAlias = qemuAliasTLSObjFromSrcAlias(inAlias))) + goto cleanup; + + /* Best shot at this as the secinfo is destroyed after process launch + * and this path does not recreate it. Thus, if the config has the + * secret UUID and we have a serial TCP chardev, then formulate a + * secAlias which we'll attempt to destroy. */ + if (cfg->chardevTLSx509secretUUID && + !(secAlias = qemuDomainGetSecretAESAlias(inAlias, false))) + goto cleanup; + + qemuDomainObjEnterMonitor(driver, vm); + + ignore_value(qemuMonitorDelObject(priv->mon, tlsAlias)); + if (secAlias) + ignore_value(qemuMonitorDelObject(priv->mon, secAlias)); + + if (qemuDomainObjExitMonitor(driver, vm) < 0) + goto cleanup; + + ret = 0; + + cleanup: + VIR_FREE(tlsAlias); + VIR_FREE(secAlias); + virObjectUnref(cfg); + return ret; +} + + int qemuDomainAttachRedirdevDevice(virConnectPtr conn, virQEMUDriverPtr driver, virDomainObjPtr vm, @@ -4120,10 +4161,7 @@ qemuDomainRemoveChrDevice(virQEMUDriverPtr driver, virDomainChrDefPtr chr) { virObjectEventPtr event; - virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); char *charAlias = NULL; - char *tlsAlias = NULL; - char *secAlias = NULL; qemuDomainObjPrivatePtr priv = vm->privateData; int ret = -1; int rc; @@ -4134,34 +4172,18 @@ qemuDomainRemoveChrDevice(virQEMUDriverPtr driver, if (!(charAlias = qemuAliasChardevFromDevAlias(chr->info.alias))) goto cleanup; - if (chr->source->type == VIR_DOMAIN_CHR_TYPE_TCP && - chr->source->data.tcp.haveTLS == VIR_TRISTATE_BOOL_YES) { - - if (!(tlsAlias = qemuAliasTLSObjFromSrcAlias(charAlias))) - goto cleanup; - - /* Best shot at this as the secinfo is destroyed after process launch - * and this path does not recreate it. Thus, if the config has the - * secret UUID and we have a serial TCP chardev, then formulate a - * secAlias which we'll attempt to destroy. */ - if (cfg->chardevTLSx509secretUUID && - !(secAlias = qemuDomainGetSecretAESAlias(charAlias, false))) - goto cleanup; - } - qemuDomainObjEnterMonitor(driver, vm); rc = qemuMonitorDetachCharDev(priv->mon, charAlias); - if (rc == 0) { - if (tlsAlias) - ignore_value(qemuMonitorDelObject(priv->mon, tlsAlias)); - if (secAlias) - ignore_value(qemuMonitorDelObject(priv->mon, secAlias)); - } - if (qemuDomainObjExitMonitor(driver, vm) < 0) goto cleanup; + if (chr->source->type == VIR_DOMAIN_CHR_TYPE_TCP && + chr->source->data.tcp.haveTLS == VIR_TRISTATE_BOOL_YES && + rc == 0 && + qemuDomainDelChardevTLSObjects(driver, vm, charAlias) < 0) + goto cleanup; + virDomainAuditChardev(vm, chr, NULL, "detach", rc == 0); if (rc < 0) @@ -4185,9 +4207,6 @@ qemuDomainRemoveChrDevice(virQEMUDriverPtr driver, cleanup: VIR_FREE(charAlias); - VIR_FREE(tlsAlias); - VIR_FREE(secAlias); - virObjectUnref(cfg); return ret; } @@ -4198,11 +4217,8 @@ qemuDomainRemoveRNGDevice(virQEMUDriverPtr driver, virDomainRNGDefPtr rng) { virObjectEventPtr event; - virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); char *charAlias = NULL; char *objAlias = NULL; - char *tlsAlias = NULL; - char *secAlias = NULL; qemuDomainObjPrivatePtr priv = vm->privateData; ssize_t idx; int ret = -1; @@ -4218,34 +4234,18 @@ qemuDomainRemoveRNGDevice(virQEMUDriverPtr driver, if (!(charAlias = qemuAliasChardevFromDevAlias(rng->info.alias))) goto cleanup; - if (rng->backend == VIR_DOMAIN_RNG_BACKEND_EGD) { - if (!(tlsAlias = qemuAliasTLSObjFromSrcAlias(charAlias))) - goto cleanup; - - /* Best shot at this as the secinfo is destroyed after process launch - * and this path does not recreate it. Thus, if the config has the - * secret UUID and we have a serial TCP chardev, then formulate a - * secAlias which we'll attempt to destroy. */ - if (cfg->chardevTLSx509secretUUID && - !(secAlias = qemuDomainGetSecretAESAlias(charAlias, false))) - goto cleanup; - } - qemuDomainObjEnterMonitor(driver, vm); rc = qemuMonitorDelObject(priv->mon, objAlias); - if (rc == 0 && rng->backend == VIR_DOMAIN_RNG_BACKEND_EGD) { - ignore_value(qemuMonitorDetachCharDev(priv->mon, charAlias)); - if (tlsAlias) - ignore_value(qemuMonitorDelObject(priv->mon, tlsAlias)); - if (secAlias) - ignore_value(qemuMonitorDelObject(priv->mon, secAlias)); - } - if (qemuDomainObjExitMonitor(driver, vm) < 0) goto cleanup; + if (rng->backend == VIR_DOMAIN_RNG_BACKEND_EGD && + rc == 0 && + qemuDomainDelChardevTLSObjects(driver, vm, charAlias) < 0) + goto cleanup; + virDomainAuditRNG(vm, rng, NULL, "detach", rc == 0); if (rc < 0) @@ -4269,9 +4269,6 @@ qemuDomainRemoveRNGDevice(virQEMUDriverPtr driver, cleanup: VIR_FREE(charAlias); VIR_FREE(objAlias); - VIR_FREE(tlsAlias); - VIR_FREE(secAlias); - virObjectUnref(cfg); return ret; } -- 2.13.6
Similar to qemuDomainAddChardevTLSObjects let's move the chardev source must be TCP and it has the @haveTLS flag set checks before trying to delete the TLS objects. For the Chr device this represents no change; however, for RNG device this is an additionaly check that was missed in commit id '68808516'. Before adding the objects, TCP and haveTLS are checked. Signed-off-by: John Ferlan <jferlan@redhat.com> --- Thanks for the review... Of course after re-reading my own patch "in the morning" I realized I could/should have taken the next step to check the chardev source for TCP and haveTLS being set similar to how the Add code does things. So that's what this patch does. src/qemu/qemu_hotplug.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 85f47bee2..b79807300 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1565,6 +1565,7 @@ qemuDomainAddChardevTLSObjects(virConnectPtr conn, static int qemuDomainDelChardevTLSObjects(virQEMUDriverPtr driver, virDomainObjPtr vm, + virDomainChrSourceDefPtr dev, const char *inAlias) { int ret = -1; @@ -1573,6 +1574,12 @@ qemuDomainDelChardevTLSObjects(virQEMUDriverPtr driver, char *tlsAlias = NULL; char *secAlias = NULL; + if (dev->type != VIR_DOMAIN_CHR_TYPE_TCP || + dev->data.tcp.haveTLS != VIR_TRISTATE_BOOL_YES) { + ret = 0; + goto cleanup; + } + if (!(tlsAlias = qemuAliasTLSObjFromSrcAlias(inAlias))) goto cleanup; @@ -4178,10 +4185,8 @@ qemuDomainRemoveChrDevice(virQEMUDriverPtr driver, if (qemuDomainObjExitMonitor(driver, vm) < 0) goto cleanup; - if (chr->source->type == VIR_DOMAIN_CHR_TYPE_TCP && - chr->source->data.tcp.haveTLS == VIR_TRISTATE_BOOL_YES && - rc == 0 && - qemuDomainDelChardevTLSObjects(driver, vm, charAlias) < 0) + if (rc == 0 && + qemuDomainDelChardevTLSObjects(driver, vm, chr->source, charAlias) < 0) goto cleanup; virDomainAuditChardev(vm, chr, NULL, "detach", rc == 0); @@ -4243,7 +4248,8 @@ qemuDomainRemoveRNGDevice(virQEMUDriverPtr driver, if (rng->backend == VIR_DOMAIN_RNG_BACKEND_EGD && rc == 0 && - qemuDomainDelChardevTLSObjects(driver, vm, charAlias) < 0) + qemuDomainDelChardevTLSObjects(driver, vm, rng->source.chardev, + charAlias) < 0) goto cleanup; virDomainAuditRNG(vm, rng, NULL, "detach", rc == 0); -- 2.13.6
At 2017-12-20 19:49:41, "John Ferlan" <jferlan@redhat.com> wrote:
Similar to qemuDomainAddChardevTLSObjects let's move the chardev source must be TCP and it has the @haveTLS flag set checks before trying to delete the TLS objects.
For the Chr device this represents no change; however, for RNG device this is an additionaly check that was missed in commit id '68808516'. Before adding the objects, TCP and haveTLS are checked.
Signed-off-by: John Ferlan <jferlan@redhat.com> ---
Thanks for the review... Of course after re-reading my own patch "in the morning" I realized I could/should have taken the next step to check the chardev source for TCP and haveTLS being set similar to how the Add code does things. So that's what this patch does.
Ahh, more helpful helper. Reviewed-by: Chen Hanxiao <chenhanxiao@gmail.com> Regards, - Chen
At 2017-12-20 08:01:37, "John Ferlan" <jferlan@redhat.com> wrote:
Let's make a comment deletion helper similar to the Add helper that can be called after the ExitMonitor.
The modify qemuDomainRemoveChrDevice and qemuDomainRemoveRNGDevice to call the helper instead of inlining the copy and pasted code.
Signed-off-by: John Ferlan <jferlan@redhat.com> ---
An offshoot of the redirdev hot unplug patches from Chen Hanxiao. Let's create a helper that can remove the TLS chardev objects for the specific devices that may have used them.
Reviewed-by: Chen Hanxiao <chenhanxiao@gmail.com> Regards, - Chen
participants (2)
-
Chen Hanxiao -
John Ferlan