Skip to contentSkip to conversation

migration: Use QAPI_CLONE_MEMBERS in migrate_params_test_apply

PATCH In review Fabiano Rosas 4 participants 8 messages Apr 14 – Apr 24

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 migration: Use QAPI_CLONE_MEMBERS in migrate_params_test_apply

Use QAPI_CLONE_MEMBERS instead of making an assignment. The QAPI method makes the handling of the TLS strings more intuitive because it clones them as well.

This also fixes a segfault when a NULL TLS option is accessed as part of a validation check for another option (e.g. in the zero-copy + multifd compression case). Details follow:

Currently, after copying s->parameters to the temporary MigrationParameters object before migrate_params_check(), the references in temporary object to the TLS options are dropped, either because:

a) the user set a new option, in which case that's fine as s->parameters still holds the reference to the old memory or,

b) the user did not set a new option, in which case keeping the references in the temporary object would later cause them to be freed along with it, leading to double-free when s->parameters is also freed later on.

In this second case, it was overlooked that the TLS options can be accessed already during migrate_params_check() as part of validation of another option. Those pointers should not have been cleared.

Using QAPI_CLONE_MEMBERS fixes the issue because the temporary object is not stealing a reference from s->parameters anymore.

---
NOTE#1: For the release we could have a simpler fix just checking for NULL, but that would allow two unsupported configurations to be
accepted: zero-copy with either multifd compression or TLS.

NOTE#2: CI is red due to pre-existing failure in functional tests with "exec:socat" migration.

---
 migration/options.c | 26 ++++++++++++--------------
 1 file changed, 12 insertions(+), 14 deletions(-)
migration/options.c +12 -14
index 7556fbc06b..68441f0276 100644
@@ -1279,9 +1279,9 @@ bool migrate_params_check(MigrationParameters *params, Error **errp)
 static void migrate_params_test_apply(MigrationParameters *params,
                                       MigrationParameters *dest)
 {
-    *dest = migrate_get_current()->parameters;
+    MigrationState *s = migrate_get_current();
 
-    /* TODO use QAPI_CLONE() instead of duplicating it inline */
+    QAPI_CLONE_MEMBERS(MigrationParameters, dest, &s->parameters);
 
     if (params->has_throttle_trigger_threshold) {
         dest->throttle_trigger_threshold = params->throttle_trigger_threshold;
@@ -1300,24 +1300,18 @@ static void migrate_params_test_apply(MigrationParameters *params,
     }
 
     if (params->tls_creds) {
+        qapi_free_StrOrNull(dest->tls_creds);
         dest->tls_creds = QAPI_CLONE(StrOrNull, params->tls_creds);
-    } else {
-        /* clear the reference, it's owned by s->parameters */
-        dest->tls_creds = NULL;
     }
 
     if (params->tls_hostname) {
+        qapi_free_StrOrNull(dest->tls_hostname);
         dest->tls_hostname = QAPI_CLONE(StrOrNull, params->tls_hostname);
-    } else {
-        /* clear the reference, it's owned by s->parameters */
-        dest->tls_hostname = NULL;
     }
 
     if (params->tls_authz) {
+        qapi_free_StrOrNull(dest->tls_authz);
         dest->tls_authz = QAPI_CLONE(StrOrNull, params->tls_authz);
-    } else {
-        /* clear the reference, it's owned by s->parameters */
-        dest->tls_authz = NULL;
     }
 
     if (params->has_max_bandwidth) {
@@ -1374,8 +1368,9 @@ static void migrate_params_test_apply(MigrationParameters *params,
     }
 
     if (params->has_block_bitmap_mapping) {
-        dest->has_block_bitmap_mapping = true;
-        dest->block_bitmap_mapping = params->block_bitmap_mapping;
+        qapi_free_BitmapMigrationNodeAliasList(dest->block_bitmap_mapping);
+        dest->block_bitmap_mapping = QAPI_CLONE(BitmapMigrationNodeAliasList,
+                                                params->block_bitmap_mapping);
     }
 
     if (params->has_x_vcpu_dirty_limit_period) {
@@ -1399,7 +1394,8 @@ static void migrate_params_test_apply(MigrationParameters *params,
     }
 
     if (params->has_cpr_exec_command) {
-        dest->cpr_exec_command = params->cpr_exec_command;
+        qapi_free_strList(dest->cpr_exec_command);
+        dest->cpr_exec_command = QAPI_CLONE(strList, params->cpr_exec_command);
     }
 }
 
@@ -1555,4 +1551,6 @@ void qmp_migrate_set_parameters(MigrationParameters *params, Error **errp)
     }
 
     migrate_tls_opts_free(&tmp);
+    qapi_free_BitmapMigrationNodeAliasList(tmp.block_bitmap_mapping);
+    qapi_free_strList(tmp.cpr_exec_command);
 }
Signature · 1 line
2.51.0
Tested-by on v1 1/1
#1 lore raw
Quoted 33 lines · Fabiano Rosas

Use QAPI_CLONE_MEMBERS instead of making an assignment. The QAPI method makes the handling of the TLS strings more intuitive because it clones them as well.

This also fixes a segfault when a NULL TLS option is accessed as part of a validation check for another option (e.g. in the zero-copy + multifd compression case). Details follow:

Currently, after copying s->parameters to the temporary MigrationParameters object before migrate_params_check(), the references in temporary object to the TLS options are dropped, either because:

a) the user set a new option, in which case that's fine as

    s->parameters still holds the reference to the old memory or,

b) the user did not set a new option, in which case keeping the references in the temporary object would later cause them to be freed along with it, leading to double-free when s->parameters is also freed later on.

In this second case, it was overlooked that the TLS options can be accessed already during migrate_params_check() as part of validation of another option. Those pointers should not have been cleared.

Using QAPI_CLONE_MEMBERS fixes the issue because the temporary object is not stealing a reference from s->parameters anymore.

Quoted 8 lines

---
NOTE#1: For the release we could have a simpler fix just checking for NULL, but that would allow two unsupported configurations to be
accepted: zero-copy with either multifd compression or TLS.

NOTE#2: CI is red due to pre-existing failure in functional tests with "exec:socat" migration.
---

Thanks for the quick fix Fabiano.

The patch seems to fix the issue, so:

Thanks,
Maciej

#2 lore raw
In reply to Fabiano Rosas
Quoted 33 lines · Fabiano Rosas

Use QAPI_CLONE_MEMBERS instead of making an assignment. The QAPI method makes the handling of the TLS strings more intuitive because it clones them as well.

This also fixes a segfault when a NULL TLS option is accessed as part of a validation check for another option (e.g. in the zero-copy + multifd compression case). Details follow:

Currently, after copying s->parameters to the temporary MigrationParameters object before migrate_params_check(), the references in temporary object to the TLS options are dropped, either because:

a) the user set a new option, in which case that's fine as s->parameters still holds the reference to the old memory or,

b) the user did not set a new option, in which case keeping the references in the temporary object would later cause them to be freed along with it, leading to double-free when s->parameters is also freed later on.

In this second case, it was overlooked that the TLS options can be accessed already during migrate_params_check() as part of validation of another option. Those pointers should not have been cleared.

Using QAPI_CLONE_MEMBERS fixes the issue because the temporary object is not stealing a reference from s->parameters anymore.

Quoted 5 lines

---
NOTE#1: For the release we could have a simpler fix just checking for NULL, but that would allow two unsupported configurations to be
accepted: zero-copy with either multifd compression or TLS.

Peter Xu just pointed out on IRC that it's actually only the zero-copy + TLS configuration that will be allowed with the simpler fix. I think it's best we go that route instead and hold this patch until after the release.

#3 lore raw
Quoted 42 lines · Fabiano Rosas
Quoted 33 lines · Fabiano Rosas

Use QAPI_CLONE_MEMBERS instead of making an assignment. The QAPI method makes the handling of the TLS strings more intuitive because it clones them as well.

This also fixes a segfault when a NULL TLS option is accessed as part of a validation check for another option (e.g. in the zero-copy + multifd compression case). Details follow:

Currently, after copying s->parameters to the temporary MigrationParameters object before migrate_params_check(), the references in temporary object to the TLS options are dropped, either because:

a) the user set a new option, in which case that's fine as s->parameters still holds the reference to the old memory or,

b) the user did not set a new option, in which case keeping the references in the temporary object would later cause them to be freed along with it, leading to double-free when s->parameters is also freed later on.

In this second case, it was overlooked that the TLS options can be accessed already during migrate_params_check() as part of validation of another option. Those pointers should not have been cleared.

Using QAPI_CLONE_MEMBERS fixes the issue because the temporary object is not stealing a reference from s->parameters anymore.

Quoted 5 lines

---
NOTE#1: For the release we could have a simpler fix just checking for NULL, but that would allow two unsupported configurations to be
accepted: zero-copy with either multifd compression or TLS.

Quoted 4 lines

Peter Xu just pointed out on IRC that it's actually only the zero-copy + TLS configuration that will be allowed with the simpler fix. I think it's best we go that route instead and hold this patch until after the release.

rc4 has already been tagged and I really don't want to put anything more into 11.0 unless it is a super high priority release-blocker kind of a bug. The fact that this is a fix for a commit that's been in git for pretty much the whole of the 11.0 cycle without anybody noticing earlier suggests it probably isn't that important ?

If we want to put it in then the commit message needs to clearly state the severity (e.g. "everybody trying to use migration hits a segfault") that makes it release-critical.

Otherwise we should put this into 11.1 and with the usual cc:stable so it is backported to the 11.0 stable series.

thanks
-- PMM

#4 lore raw
Quoted 54 lines · Peter Maydell
Quoted 42 lines · Fabiano Rosas
Quoted 33 lines · Fabiano Rosas

Use QAPI_CLONE_MEMBERS instead of making an assignment. The QAPI method makes the handling of the TLS strings more intuitive because it clones them as well.

This also fixes a segfault when a NULL TLS option is accessed as part of a validation check for another option (e.g. in the zero-copy + multifd compression case). Details follow:

Currently, after copying s->parameters to the temporary MigrationParameters object before migrate_params_check(), the references in temporary object to the TLS options are dropped, either because:

a) the user set a new option, in which case that's fine as s->parameters still holds the reference to the old memory or,

b) the user did not set a new option, in which case keeping the references in the temporary object would later cause them to be freed along with it, leading to double-free when s->parameters is also freed later on.

In this second case, it was overlooked that the TLS options can be accessed already during migrate_params_check() as part of validation of another option. Those pointers should not have been cleared.

Using QAPI_CLONE_MEMBERS fixes the issue because the temporary object is not stealing a reference from s->parameters anymore.

Quoted 5 lines

---
NOTE#1: For the release we could have a simpler fix just checking for NULL, but that would allow two unsupported configurations to be
accepted: zero-copy with either multifd compression or TLS.

Quoted 4 lines

Peter Xu just pointed out on IRC that it's actually only the zero-copy + TLS configuration that will be allowed with the simpler fix. I think it's best we go that route instead and hold this patch until after the release.

rc4 has already been tagged and I really don't want to put anything more into 11.0 unless it is a super high priority release-blocker kind of a bug. The fact that this is a fix for a commit that's been in git for pretty much the whole of the 11.0 cycle without anybody noticing earlier suggests it probably isn't that important ?

Quoted 7 lines

If we want to put it in then the commit message needs to clearly state the severity (e.g. "everybody trying to use migration hits a segfault") that makes it release-critical.

Otherwise we should put this into 11.1 and with the usual cc:stable so it is backported to the 11.0 stable series.

Makes sense.

@Peter Xu, do you agree with leaving this to stable? Given that it needs multifd and zero-copy-send to be both set, I'd say it doesn't have that large of a user base.

@Maciej, what's the impact for you?

#5 lore raw
Quoted 64 lines · Fabiano Rosas
Quoted 54 lines · Peter Maydell
Quoted 42 lines · Fabiano Rosas
Quoted 33 lines · Fabiano Rosas

Use QAPI_CLONE_MEMBERS instead of making an assignment. The QAPI method makes the handling of the TLS strings more intuitive because it clones them as well.

This also fixes a segfault when a NULL TLS option is accessed as part of a validation check for another option (e.g. in the zero-copy + multifd compression case). Details follow:

Currently, after copying s->parameters to the temporary MigrationParameters object before migrate_params_check(), the references in temporary object to the TLS options are dropped, either because:

a) the user set a new option, in which case that's fine as

    s->parameters still holds the reference to the old memory or,

b) the user did not set a new option, in which case keeping the references in the temporary object would later cause them to be freed along with it, leading to double-free when s->parameters is also freed later on.

In this second case, it was overlooked that the TLS options can be accessed already during migrate_params_check() as part of validation of another option. Those pointers should not have been cleared.

Using QAPI_CLONE_MEMBERS fixes the issue because the temporary object is not stealing a reference from s->parameters anymore.

Quoted 5 lines

---
NOTE#1: For the release we could have a simpler fix just checking for NULL, but that would allow two unsupported configurations to be
accepted: zero-copy with either multifd compression or TLS.

Quoted 4 lines

Peter Xu just pointed out on IRC that it's actually only the zero-copy + TLS configuration that will be allowed with the simpler fix. I think it's best we go that route instead and hold this patch until after the release.

rc4 has already been tagged and I really don't want to put anything more into 11.0 unless it is a super high priority release-blocker kind of a bug. The fact that this is a fix for a commit that's been in git for pretty much the whole of the 11.0 cycle without anybody noticing earlier suggests it probably isn't that important ?

Quoted 7 lines

If we want to put it in then the commit message needs to clearly state the severity (e.g. "everybody trying to use migration hits a segfault") that makes it release-critical.

Otherwise we should put this into 11.1 and with the usual cc:stable so it is backported to the 11.0 stable series.

Quoted 7 lines

Makes sense.

@Peter Xu, do you agree with leaving this to stable? Given that it needs multifd and zero-copy-send to be both set, I'd say it doesn't have that large of a user base.

@Maciej, what's the impact for you?

Minimal - I can just run my tests with patched QEMU.

Thanks,
Maciej

#6 lore raw
In reply to Fabiano Rosas
Quoted 3 lines · Fabiano Rosas

@Peter Xu, do you agree with leaving this to stable? Given that it needs multifd and zero-copy-send to be both set, I'd say it doesn't have that large of a user base.

Yes.

Signature · 1 line
Peter Xu
#7 lore raw
In reply to Fabiano Rosas
Quoted 25 lines · Fabiano Rosas

Use QAPI_CLONE_MEMBERS instead of making an assignment. The QAPI method makes the handling of the TLS strings more intuitive because it clones them as well.

This also fixes a segfault when a NULL TLS option is accessed as part of a validation check for another option (e.g. in the zero-copy + multifd compression case). Details follow:

Currently, after copying s->parameters to the temporary MigrationParameters object before migrate_params_check(), the references in temporary object to the TLS options are dropped, either because:

a) the user set a new option, in which case that's fine as s->parameters still holds the reference to the old memory or,

b) the user did not set a new option, in which case keeping the references in the temporary object would later cause them to be freed along with it, leading to double-free when s->parameters is also freed later on.

In this second case, it was overlooked that the TLS options can be accessed already during migrate_params_check() as part of validation of another option. Those pointers should not have been cleared.

Quoted 8 lines

Using QAPI_CLONE_MEMBERS fixes the issue because the temporary object is not stealing a reference from s->parameters anymore.

Queued and added cc:stable.

Signature · 1 line
Peter Xu