On Tuesday 13 October 2015 11:06:44 Jongsung Kim wrote: > On 10/12/2015 06:27 PM, Arnd Bergmann wrote: > > How about changing the zcomp code to pass the page pointer instead of the > > kernel space pointer? That would avoid having to do the kmap_atomic, which > > can itself be expensive on 32-bit machines and should not be needed here if > > you have a HW DMA engine doing the compression. Arnd > > Mainline zram uses lzo / lz4 library functions as backend. Using kmap_atomic > and passing address look reasonable.
Yes, I know, but changing this to do the kmap_atomic in the backend seems better here if you are adding another backend that doesn't want or need the kmap_atomic. Arnd -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/