visorbus: fix error return code in visorchipset_init()
authorZhen Lei <thunder.leizhen@huawei.com>
Fri, 28 May 2021 08:26:14 +0000 (16:26 +0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 14 Jul 2021 14:56:41 +0000 (16:56 +0200)
[ Upstream commit ce52ec5beecc1079c251f60e3973b3758f60eb59 ]

Commit 1366a3db3dcf ("staging: unisys: visorbus: visorchipset_init clean
up gotos") assigns the initial value -ENODEV to the local variable 'err',
and the first several error branches will return this value after "goto
error". But commit f1f537c2e7f5 ("staging: unisys: visorbus: Consolidate
controlvm channel creation.") overwrites 'err' in the middle of the way.
As a result, some error branches do not successfully return the initial
value -ENODEV of 'err', but return 0.

In addition, when kzalloc() fails, -ENOMEM should be returned instead of
-ENODEV.

Fixes: f1f537c2e7f5 ("staging: unisys: visorbus: Consolidate controlvm channel creation.")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Link: https://lore.kernel.org/r/20210528082614.9337-1-thunder.leizhen@huawei.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drivers/visorbus/visorchipset.c

index cb1eb7e..5668cad 100644 (file)
@@ -1561,7 +1561,7 @@ schedule_out:
 
 static int visorchipset_init(struct acpi_device *acpi_device)
 {
-       int err = -ENODEV;
+       int err = -ENOMEM;
        struct visorchannel *controlvm_channel;
 
        chipset_dev = kzalloc(sizeof(*chipset_dev), GFP_KERNEL);
@@ -1584,8 +1584,10 @@ static int visorchipset_init(struct acpi_device *acpi_device)
                                 "controlvm",
                                 sizeof(struct visor_controlvm_channel),
                                 VISOR_CONTROLVM_CHANNEL_VERSIONID,
-                                VISOR_CHANNEL_SIGNATURE))
+                                VISOR_CHANNEL_SIGNATURE)) {
+               err = -ENODEV;
                goto error_delete_groups;
+       }
        /* if booting in a crash kernel */
        if (is_kdump_kernel())
                INIT_DELAYED_WORK(&chipset_dev->periodic_controlvm_work,