Re: [PATCH v2] cifs: fix possible uninitialized access and race on iface_list

[Steve French <smfrench@gmail.com>]

tentatively merged into cifs-2.6.git for-next and buildbot github tree
pending testing

On Wed, Dec 4, 2019 at 9:59 AM Paulo Alcantara <pc@cjr.nz> wrote:
>
> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
>
> On December 4, 2019 12:14:54 PM GMT-03:00, Aurelien Aptel <aaptel@suse.com> wrote:
>>
>> iface[0] was accessed regardless of the count value and without
>> locking.
>>
>> * check count before accessing any ifaces
>> * make copy of iface list (it's a simple POD array) and use it without
>>   locking.
>>
>> Signed-off-by: Aurelien Aptel <aaptel@suse.com>
>> ________________________________
>> changes since v1:
>> * remove unecessary kfree()
>>
>>  fs/cifs/sess.c | 29 ++++++++++++++++++++++++++---
>>  1 file changed, 26 insertions(+), 3 deletions(-)
>>
>> diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c
>> index fb3bdc44775c..396400cf2800 100644
>> --- a/fs/cifs/sess.c
>> +++ b/fs/cifs/sess.c
>> @@ -77,6 +77,8 @@ int cifs_try_adding_channels(struct cifs_ses *ses)
>>   int i = 0;
>>   int rc = 0;
>>   int tries = 0;
>> + struct cifs_server_iface *ifaces = NULL;
>> + size_t iface_count;
>>
>>   if (left <= 0) {
>>   cifs_dbg(FYI,
>> @@ -90,6 +92,26 @@ int cifs_try_adding_channels(struct cifs_ses *ses)
>>   return 0;
>>   }
>>
>> + /*
>> + * Make a copy of the iface list at the time and use that
>> + * instead so as to not hold the iface spinlock for opening
>> + * channels
>> + */
>> + spin_lock(&ses->iface_lock);
>> + iface_count = ses->iface_count;
>> + if (iface_count <= 0) {
>> + spin_unlock(&ses->iface_lock);
>> + cifs_dbg(FYI, "no iface list available to open channels\n");
>> + return 0;
>> + }
>> + ifaces = kmemdup(ses->iface_list, iface_count*sizeof(*ifaces),
>> + GFP_ATOMIC);
>> + if (!ifaces) {
>> + spin_unlock(&ses->iface_lock);
>> + return 0;
>> + }
>> + spin_unlock(&ses->iface_lock);
>> +
>>   /*
>>   * Keep connecting to same, fastest, iface for all channels as
>>   * long as its RSS. Try next fastest one if not RSS or channel
>> @@ -105,9 +127,9 @@ int cifs_try_adding_channels(struct cifs_ses *ses)
>>   break;
>>   }
>>
>> - iface = &ses->iface_list[i];
>> + iface = &ifaces[i];
>>   if (is_ses_using_iface(ses, iface) && !iface->rss_capable) {
>> - i = (i+1) % ses->iface_count;
>> + i = (i+1) % iface_count;
>>   continue;
>>   }
>>
>> @@ -115,7 +137,7 @@ int cifs_try_adding_channels(struct cifs_ses *ses)
>>   if (rc) {
>>   cifs_dbg(FYI, "failed to open extra channel on iface#%d rc=%d\n",
>>   i, rc);
>> - i = (i+1) % ses->iface_count;
>> + i = (i+1) % iface_count;
>>   continue;
>>   }
>>
>> @@ -124,6 +146,7 @@ int cifs_try_adding_channels(struct cifs_ses *ses)
>>   left--;
>>   }
>>
>> + kfree(ifaces);
>>   return ses->chan_count - old_chan_count;
>>  }
>>
>
>
> --
> Sent from my p≡p for Android.



-- 
Thanks,

Steve