Imported Upstream version 11.3

TODO

@@ -17,6 +17,18 @@ Not checked disks should be allowed to be missing.
 * Add an option to ignore subsecond timestamp.
 Like when you copy data to a filesystem with less timestamp precision.
 
+* Use threads to scan all the disks at the same time.
+- After 7.0 Windows changes it seems fast enough even
+with a mono thread implementation with 100.0000 files.
++ But if you have millions of files, it could take minutes.
+
+* Support more parity levels
+It can be done with a generic computation function, using
+intrinsic for SSSE3 and AVX instructions.
+It would be intersting to compare performance with the hand-written
+assembler functions. Eventially we can convert them to use intrinsic also.
+https://sourceforge.net/p/snapraid/discussion/1677233/thread/9dbd7581/
+
 * Extend haspdeep to support the SnapRAID hash :
 https://github.com/jessek/hashdeep/
 https://sourceforge.net/p/snapraid/discussion/1677233/thread/90b0e9b2/?limit=25
@@ -222,10 +234,6 @@ and if we manage to keep it in the cache, we should save time.
 - We now hash first the faster disks, and this could
 reduce performance as we'll have to wait for all disks.
 
-* Use threads to scan all the disks at the same time.
-- After 7.0 Windows changes it seems fast enough even
-with a mono thread implementation.
-
 * Enable storing of creation time NTFS, crtime/birth time EXT4.
 But see: http://unix.stackexchange.com/questions/50177/birth-is-empty-on-ext4
 coreutils stat has an example, but it doesn't work in Linux (see lib/stat-time.h)