On (25/03/05 11:45), Herbert Xu wrote: > > IIUC, what Herbert is suggesting is that we rework all of this to use SG > > lists to reduce copies, but I am not sure which copies can go away? We > > have one copy in the compression path that probably cannot go away. > > After the zsmalloc changes (and ignoring highmem), we have one copy in > > the decompression path for when objects span two pages. I think this > > will still happen with SG lists, except internally in the crypto API. > > It's the decompression copy when the object spans two pages that > will disappear. Because I have added SG support to LZO: Hmm at a price of diverging code bases? Is it really worth it? And it's not guaranteed that all algorithms will switch over to SG, so the linerisation will stay around.
