| #
0e833e69 |
|
30-Jun-2015 |
Ard Biesheuvel <ardb@kernel.org> |
md/raid6: delta syndrome for ARM NEON
This implements XOR syndrome calculation using NEON intrinsics.
As before, the module can be built for ARM and arm64 from the
same source.
Relative performance on a Cortex-A57 based system:
raid6: int64x1 gen() 905 MB/s
raid6: int64x1 xor() 881 MB/s
raid6: int64x2 gen() 1343 MB/s
raid6: int64x2 xor() 1286 MB/s
raid6: int64x4 gen() 1896 MB/s
raid6: int64x4 xor() 1321 MB/s
raid6: int64x8 gen() 1773 MB/s
raid6: int64x8 xor() 1165 MB/s
raid6: neonx1 gen() 1834 MB/s
raid6: neonx1 xor() 1278 MB/s
raid6: neonx2 gen() 2528 MB/s
raid6: neonx2 xor() 1942 MB/s
raid6: neonx4 gen() 2888 MB/s
raid6: neonx4 xor() 2334 MB/s
raid6: neonx8 gen() 2957 MB/s
raid6: neonx8 xor() 2232 MB/s
raid6: using algorithm neonx8 gen() 2957 MB/s
raid6: .... xor() 2232 MB/s, rmw enabled
Cc: Markus Stockhausen <stockhausen@collogia.de>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: NeilBrown <neilb@suse.com>
|
| #
fe5cbc6e |
|
14-Dec-2014 |
Markus Stockhausen <stockhausen@collogia.de> |
md/raid6 algorithms: delta syndrome functions
v3: s-o-b comment, explanation of performance and descision for
the start/stop implementation
Implementing rmw functionality for RAID6 requires optimized syndrome
calculation. Up to now we can only generate a complete syndrome. The
target P/Q pages are always overwritten. With this patch we provide
a framework for inplace P/Q modification. In the first place simply
fill those functions with NULL values.
xor_syndrome() has two additional parameters: start & stop. These
will indicate the first and last page that are changing during a
rmw run. That makes it possible to avoid several unneccessary loops
and speed up calculation. The caller needs to implement the following
logic to make the functions work.
1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source
blocks inside P/Q between (and including) start and end.
2) modify any block with start <= block <= stop
3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of
source blocks into P/Q between (and including) start and end.
Pages between start and stop that won't be changed should be filled
with a pointer to the kernel zero page. The reasons for not taking NULL
pages are:
1) Algorithms cross the whole source data line by line. Thus avoid
additional branches.
2) Having a NULL page avoids calculating the XOR P parity but still
need calulation steps for the Q parity. Depending on the algorithm
unrolling that might be only a difference of 2 instructions per loop.
The benchmark numbers of the gen_syndrome() functions are displayed in
the kernel log. Do the same for the xor_syndrome() functions. This
will help to analyze performance problems and give an rough estimate
how well the algorithm works. The choice of the fastest algorithm will
still depend on the gen_syndrome() performance.
With the start/stop page implementation the speed can vary a lot in real
life. E.g. a change of page 0 & page 15 on a stripe will be harder to
compute than the case where page 0 & page 1 are XOR candidates. To be not
to enthusiatic about the expected speeds we will run a worse case test
that simulates a change on the upper half of the stripe. So we do:
1) calculation of P/Q for the upper pages
2) continuation of Q for the lower (empty) pages
Signed-off-by: Markus Stockhausen <stockhausen@collogia.de>
Signed-off-by: NeilBrown <neilb@suse.de>
|