[dpdk-dev] [Bug 675] net/ice: possible integer overflow when computing max_pkt_len

bugzilla at dpdk.org bugzilla at dpdk.org
Wed Apr 14 08:39:07 CEST 2021


https://bugs.dpdk.org/show_bug.cgi?id=675

            Bug ID: 675
           Summary: net/ice: possible integer overflow when computing
                    max_pkt_len
           Product: DPDK
           Version: 20.11
          Hardware: All
                OS: All
            Status: UNCONFIRMED
          Severity: normal
          Priority: Normal
         Component: ethdev
          Assignee: dev at dpdk.org
          Reporter: tudor.cornea at gmail.com
  Target Milestone: ---

Created attachment 155
  --> https://bugs.dpdk.org/attachment.cgi?id=155&action=edit
Patch with possible fix

Greetings,

While working with the Intel ice driver (E810 NIC), I have stumbled into, what
I think to be a bug, which is causing all Jumbo Frames to get dropped. 

I think that the len variable, used in the computation of max_pkt_len could
overflow, if used to store the result of the following computation:

ICE_SUPPORT_CHAIN_NUM * rxq->rx_buf_len

Since, we could define the mbuf size to have a large value (i.e 13312),
and ICE_SUPPORT_CHAIN_NUM is defined as 5, the computation mentioned above,
could potentially result in a value which might be bigger than MAX_USHORT.

The result will be that Jumbo Frames will not work properly

Could you please review the following pull request (I have also added the patch
as an attachment) ?


[1] https://github.com/DPDK/dpdk/pull/29

-- 
You are receiving this mail because:
You are the assignee for the bug.


More information about the dev mailing list