Hello,
I am experimenting on implementing Firefox over a realiable,
out-of-sequence delivering transport protocol (SCTP). I.e., similar to
TCP, SCTP provides no loss/duplicates/errors guarantee, but unlike TCP,
SCTP can deliver data out-of-sequence to the receiving application
(Firefox). SCTP achieves this through 'streams' which can be considered
logical one-way pipes with independent sequence space, and all data
arriving within a stream are delivered in-order to the application. More
info can be found in SCTP RFC2960.
sendmsg and recvmsg calls over an SCTP socket can take an SCTP stream
number as argument so that an application can send or receive on a
particular SCTP stream. The major factor to be considered for modifying
Firefox over SCTP streams is the following: Different responses can and
will be read interleaved (out-of-order) from the SCTP socket. If
requests are pipelined, Firefox should not assume that data from
subsequent read()s on the socket belong to the same response. With SCTP,
subsequent read() on the same SCTP stream belong to the same response.
Now, I've been a little bit successful in implementing a part Firefox
over SCTP streams. I've modified NSPR to write different HTTP requests
on different SCTP streams, and read HTTP responses from SCTP streams.
Note that this implementation assumes that each Write() on a
nsOutputStream is for a different request, but I am OK with this
assumption at the moment.
I would like to know if there is a realistic way to do the following:
- NSPR communicates the SCTP stream number on which a request was
written to the HTTP protocol module. The SCTP stream number is unique to
a HTTP Transaction, and should be stored somewhere, perhaps in
nsHttpTransaction. Transfer of this stream number could be via a new
method - WriteOnSCTPStream() (in nsOutputStream) that is very similar to
Write(). WriteOnSCTPStream() just has an extra pointer argument, where
the callee writes the SCTP stream information.
- nsHttpTransaction accepts the response from nsInputStream only if the
response came on the SCTP stream it stored as mentioned above.
I chose nsHttpTransaction as the point to do these modifications since
nsHttpTransaction is an atomic request-response transaction entitiy.
Maybe the modifications should move closer to NSPR than
nsHttpTransaction? Maybe I should take a completely different route to
achieve what I need?
I would appreciate any comments/critics on the above.
I also have the following questions regarding pipelining:
- In case of pipelining n requests, I assume n instances of
nsHttpTransaction are created. Is that right?
- Does the implementation use the Content-Length field in the resonse
header to demarcate the end of a response?
Any help is very much appreciated.
Thanks,
Preethi
_______________________________________________
dev-tech-network mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-tech-network
