Issue metadata
Sign in to add a comment
|
Code with confusing fast-path semantic [QuicConnection::SendStreamData]
Reported by
leanderz...@gmail.com,
Jan 19 2017
|
||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.95 Safari/537.36 Steps to reproduce the problem: 1. This is a bug reported by our static analysis tool. 2. 3. What is the expected behavior? What went wrong? In function QuicConsumedData QuicConnection::SendStreamData, a fast path is implemented to process packets which fit into a standard buffer and don't need padding. The fastpath function ConsumeDataFastPath then calls "CreateAndSerializeStreamFrame" to serialize and encrypt the packet. However, CreateAndSerializeStreamFrame can return with the "AppendPacketHead failed" error. When this error happens, the code is expected to switch back to execute its slow path "packet_generator_.ConsumeData". But the current code does not correctly follow this design. Should we change the code to prevent any regression here? Did this work before? Yes Does this work in other browsers? Yes Chrome version: 55.0.2883.95 Channel: stable OS Version: OS X 10.12.2 Flash Version: Shockwave Flash 24.0 r0
,
Feb 6 2017
Not Hotlist-Interop, adding QUIC component.
,
Feb 12 2018
Issue has not been modified or commented on in the last 365 days, please re-open or file a new bug if this is still an issue. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by rbasuvula@chromium.org
, Jan 19 2017Labels: TE-NeedsTriageHelp