Messages in this thread | | | Date | Tue, 24 Jun 2008 11:31:33 -0700 (PDT) | From | Linus Torvalds <> | Subject | Re: [rfc patch 3/4] splice: remove confirm from pipe_buf_operations |
| |
On Tue, 24 Jun 2008, Miklos Szeredi wrote: > > OK. But currently we have an implementation that > > 1) doesn't do any of this, unless readahead is disabled
Sure. But removing even the conceptual support? Not a good idea.
> And in addition, splice-in and splice-out can return a short count or > even zero count if the filesystem invalidates the cached pages during > the splicing (data became stale for example). Are these the right > semantics? I'm not sure.
What does that really have with splice() and removing the features? Why don't you just fix that issue?
Linus
| |