lkml.org 
[lkml]   [2022]   [Dec]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
SubjectRe: [PATCH] kobject: Fix slab-out-of-bounds in fill_kobj_path()
From

在 2022/12/19 22:54, Greg KH 写道:
> On Mon, Dec 19, 2022 at 10:41:03PM +0800, Wang Hai wrote:
>> In kobject_get_path(), if kobj->name is changed between calls
>> get_kobj_path_length() and fill_kobj_path() and the length becomes
>> longer, then fill_kobj_path() will have an out-of-bounds bug.
>>
>> The actual current problem occurs when the ixgbe probe.
>>
>> In ixgbe_mii_bus_init(), if the length of netdev->dev.kobj.name
>> length becomes longer, out-of-bounds will occur.
>>
>> cpu0 cpu1
>> ixgbe_probe
>> register_netdev(netdev)
>> netdev_register_kobject
>> device_add
>> kobject_uevent // Sending ADD events
>> systemd-udevd // rename netdev
>> dev_change_name
>> device_rename
>> kobject_rename
>> |
>> ixgbe_mii_bus_init |
>> mdiobus_register |
>> __mdiobus_register |
>> device_register |
>> device_add |
>> kobject_uevent |
>> kobject_get_path |
>> len = get_kobj_path_length // old name |
>> path = kzalloc(len, gfp_mask); |
>> kobj->name = name;
>> /* name length becomes
>> * longer
>> */
>> fill_kobj_path /* kobj path length is
>> * longer than path,
>> * resulting in out of
>> * bounds when filling path
>> */
>>
>> This is the kasan report:
>>
>> ==================================================================
>> BUG: KASAN: slab-out-of-bounds in fill_kobj_path+0x50/0xc0
>> Write of size 7 at addr ff1100090573d1fd by task kworker/28:1/673
>>
>> Workqueue: events work_for_cpu_fn
>> Call Trace:
>> <TASK>
>> dump_stack_lvl+0x34/0x48
>> print_address_description.constprop.0+0x86/0x1e7
>> print_report+0x36/0x4f
>> kasan_report+0xad/0x130
>> kasan_check_range+0x35/0x1c0
>> memcpy+0x39/0x60
>> fill_kobj_path+0x50/0xc0
>> kobject_get_path+0x5a/0xc0
>> kobject_uevent_env+0x140/0x460
>> device_add+0x5c7/0x910
>> __mdiobus_register+0x14e/0x490
>> ixgbe_probe.cold+0x441/0x574 [ixgbe]
>> local_pci_probe+0x78/0xc0
>> work_for_cpu_fn+0x26/0x40
>> process_one_work+0x3b6/0x6a0
>> worker_thread+0x368/0x520
>> kthread+0x165/0x1a0
>> ret_from_fork+0x1f/0x30
>>
>> This reproducer triggers that bug:
>>
>> while:
>> do
>> rmmod ixgbe
>> sleep 0.5
>> modprobe ixgbe
>> sleep 0.5
>>
>> When calling fill_kobj_path() to fill path, if the name length of
>> kobj becomes longer, return failure and retry.
>>
>> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
>> Signed-off-by: Wang Hai <wanghai38@huawei.com>
>> ---
>> lib/kobject.c | 14 ++++++++++++--
>> 1 file changed, 12 insertions(+), 2 deletions(-)
> Nice find, and a much better description than was reported here:
> https://lore.kernel.org/all/20221216031320.2634-1-alice.chao@mediatek.com/
>
> Alice, does this patch resolve your problem as well?
>
>> diff --git a/lib/kobject.c b/lib/kobject.c
>> index a0b2dbfcfa23..d129f437b200 100644
>> --- a/lib/kobject.c
>> +++ b/lib/kobject.c
>> @@ -112,7 +112,7 @@ static int get_kobj_path_length(struct kobject *kobj)
>> return length;
>> }
>>
>> -static void fill_kobj_path(struct kobject *kobj, char *path, int length)
>> +static bool fill_kobj_path(struct kobject *kobj, char *path, int length)
> Just return an int and an error value, bool isn't a normal return value,
> right?

Hi, Greg.

Thanks for the suggestion, I will change it to int.

>> {
>> struct kobject *parent;
>>
>> @@ -121,12 +121,16 @@ static void fill_kobj_path(struct kobject *kobj, char *path, int length)
>> int cur = strlen(kobject_name(parent));
>> /* back up enough to print this name with '/' */
>> length -= cur;
>> + if (length < 0)
>> + return false;
> Return -EINVAL or something?
>
> What if length ends up 0 here, is that going to be ok?
If length is 0, there is not enough space for '/', which is also
problematic. Thank you for the reminder.
>
>> memcpy(path + length, kobject_name(parent), cur);
>> *(path + --length) = '/';
>> }
>>
>> pr_debug("kobject: '%s' (%p): %s: path = '%s'\n", kobject_name(kobj),
>> kobj, __func__, path);
>> +
>> + return true;
> return 0;
>
>> }
>>
>> /**
>> @@ -140,14 +144,20 @@ char *kobject_get_path(struct kobject *kobj, gfp_t gfp_mask)
>> {
>> char *path;
>> int len;
>> + bool ret;
> No need for this variable
>
>>
>> +retry:
>> len = get_kobj_path_length(kobj);
>> if (len == 0)
>> return NULL;
>> path = kzalloc(len, gfp_mask);
>> if (!path)
>> return NULL;
>> - fill_kobj_path(kobj, path, len);
>> + ret = fill_kobj_path(kobj, path, len);
>> + if (!ret) {
> Make this one line:
>
> if (fill_kobj_path(kobj, path, len)) {
>
> But, you now have a loop, what guarantees that you can get out of it?

fill_kobj_path() may fail only when the name of kobj or parent is changed,
this change does not happen forever.

And when retrying, the length of the current kobj path is recalculated,
which can guarantee that the length is satisfied next time.


Can I send v2 like this?

diff --git a/lib/kobject.c b/lib/kobject.c
index a0b2dbfcfa23..3f97d903266a 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -112,7 +112,7 @@ static int get_kobj_path_length(struct kobject *kobj)
        return length;
 }

-static void fill_kobj_path(struct kobject *kobj, char *path, int length)
+static int fill_kobj_path(struct kobject *kobj, char *path, int length)
 {
        struct kobject *parent;

@@ -121,12 +121,16 @@ static void fill_kobj_path(struct kobject *kobj,
char *path, int length)
                int cur = strlen(kobject_name(parent));
                /* back up enough to print this name with '/' */
                length -= cur;
+               if (length <= 0)
+                       return -EINVAL;
                memcpy(path + length, kobject_name(parent), cur);
                *(path + --length) = '/';
        }

        pr_debug("kobject: '%s' (%p): %s: path = '%s'\n",
kobject_name(kobj),
                 kobj, __func__, path);
+
+       return 0;
 }

 /**
@@ -141,13 +145,17 @@ char *kobject_get_path(struct kobject *kobj, gfp_t
gfp_mask)
        char *path;
        int len;

+retry:
        len = get_kobj_path_length(kobj);
        if (len == 0)
                return NULL;
        path = kzalloc(len, gfp_mask);
        if (!path)
                return NULL;
-       fill_kobj_path(kobj, path, len);
+       if (fill_kobj_path(kobj, path, len)) {
+               kfree(path);
+               goto retry;
+       }

        return path;
 }
>
> thanks,
>
> greg k-h
> .

--
Wang Hai

\
 
 \ /
  Last update: 2022-12-19 16:28    [W:1.737 / U:0.172 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site