On 18-Apr-05, at 9:43 AM, Greg Ames wrote:
if the network data for pipelined request n+1 isn't already present during ap_read_request for request n, it is highly unlikely to arrive by the time we start writing the response for request n.
It may well arrive during the time we're writing the response, though, if it is based on information from the response.
yes, for example if it's the main .html page with imbedded graphics/whatever and network latency is low.
if a new request does arrive during that window it's no big deal. we might write one small packet to the network and nothing breaks.
I agree with this. I doubt whether the optimization actually buys enough to make it worth the trouble. Also, if we ever get it wrong and leave unflushed data at the end of a request, it will look like the request is stalled.
correct again.
to be clear, what I'd like to see is this: when in doubt, flush. if we have a cheap way of knowing that stashed input data (i.e. the next request) is already present, maybe don't flush.
Greg
