On 3/6/19 3:33 AM, Vasily Averin wrote:
> James, Martin,
> noone replied 2 weeks,
> could you please pick up this patch?
>
> According to Network guru sendpage must not be called for Slab objects.
> Unfortunately this happen in real life, for example when XFS send metadata
> via network block device.
> Some of such cases -- drbd and ceph -- already have PageSlab() check, however
> iscsi still lacks it.
>
> It was triggered host to crash during internal OpenVZ tests,
> fixed kernel passed this test successfully.
>
> This patch forces iscsi_tcp_segment_map() to set up segment->data for Slab
> pages
> and it switches iscsi_sw_tcp_xmit_segment() to use sendmsg instead of
> sendpage.
>
> Thank you,
> Vasily Averin
>
> On 2/21/19 6:23 PM, Vasily Averin wrote:
>> In "XFS over network block device" scenario XFS can create IO requests
>> with slab-based XFS metadata. During processing such requests
>> tcp_sendpage() can merge skb fragments with neighbour slab objects.
>>
>> If receiving side is located on the same host tcp_recvmsg() can trigger
>> BUG_ON in hardening check and crash the host with following message:
>>
>> usercopy: kernel memory exposure attempt detected
>> from XXXXXXXX (kmalloc-512) (1024 bytes)
>>
>> This patch redirect such requests from sednpage to sendmsg path.
>> The problem is similar to one described in recent commit 7e241f647dc7
>> ("libceph: fall back to sendmsg for slab pages")
>>
>> Signed-off-by: Vasily Averin <v...@virtuozzo.com>
>> ---
>> drivers/scsi/libiscsi_tcp.c | 11 ++++++++---
>> 1 file changed, 8 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/scsi/libiscsi_tcp.c b/drivers/scsi/libiscsi_tcp.c
>> index 8a6b1b3f8277..66d97d3bef5a 100644
>> --- a/drivers/scsi/libiscsi_tcp.c
>> +++ b/drivers/scsi/libiscsi_tcp.c
>> @@ -129,12 +129,17 @@ static void iscsi_tcp_segment_map(struct iscsi_segment
>> *segment, int recv)
>> BUG_ON(sg->length == 0);
>>
>> /*
>> + * We always map for the recv path.
>> + *
>> * If the page count is greater than one it is ok to send
>> * to the network layer's zero copy send path. If not we
>> - * have to go the slow sendmsg path. We always map for the
>> - * recv path.
>> + * have to go the slow sendmsg path.
>> + *
>> + * Same goes for slab pages: skb_can_coalesce() allows
>> + * coalescing neighboring slab objects into a single frag which
>> + * triggers one of hardened usercopy checks.
>> */
>> - if (page_count(sg_page(sg)) >= 1 && !recv)
>> + if (!recv && page_count(sg_page(sg)) >= 1 && !PageSlab(sg_page(sg)))
>> return;
>>
>> if (recv) {
>>
>
Reviewed-by: Lee Duncan <ldun...@suse.com>
