It was added in r193752 as part of bug #137692 to fix an infinite loop in network process that happened in EFL because they didn't enable the network cache at that time. I think that was actually a workaround, and it was added without any comment so it has stayed there even when EFL enabled disk cache, and now that is gone. Looking at current code I see no reason why buffering can't work with the disk cache disabled, so I think it's time to remove that workaround.
Created attachment 302231 [details] Patch
(In reply to comment #0) > It was added in r193752 as part of bug #137692 to fix an infinite loop in > network process that happened in EFL because they didn't enable the network > cache at that time. I think that was actually a workaround, and it was added > without any comment so it has stayed there even when EFL enabled disk cache, > and now that is gone. Looking at current code I see no reason why buffering > can't work with the disk cache disabled, so I think it's time to remove that > workaround. Have you tried it out with disabled network cache or do you just think that it will work?
(In reply to comment #2) > (In reply to comment #0) > > It was added in r193752 as part of bug #137692 to fix an infinite loop in > > network process that happened in EFL because they didn't enable the network > > cache at that time. I think that was actually a workaround, and it was added > > without any comment so it has stayed there even when EFL enabled disk cache, > > and now that is gone. Looking at current code I see no reason why buffering > > can't work with the disk cache disabled, so I think it's time to remove that > > workaround. > > Have you tried it out with disabled network cache or do you just think that > it will work? Nope, I've just read the buffering code in the network process.
Committed r212816: <http://trac.webkit.org/changeset/212816>