Revision 9f339caf8454f0c21983111350ede93983db4340 authored by Bojan Smojver on 25 November 2010, 22:41:39 UTC, committed by Rafael J. Wysocki on 06 December 2010, 22:38:29 UTC
This is a fix for reading LZO compressed image using async I/O. Essentially, instead of having just one page into which we keep reading blocks from swap, we allocate enough of them to cover the largest compressed size and then let block I/O pick them all up. Once we have them all (and here we wait), we decompress them, as usual. Obviously, the very first block we still pick up synchronously, because we need to know the size of the lot before we pick up the rest. Also fixed the copyright line, which I've forgotten before. Signed-off-by: Bojan Smojver <bojan@rexursive.com> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
1 parent 698fd6a
Kconfig
#
# DCA server configuration
#
config DCA
tristate
Computing file changes ...