kernel/git/next/linux-next.git/lib/raid, branch master

raid6: add missing vzeroupper to AVX-512 code

2026-09-06T03:23:25+00:00

Since the AVX-512 optimized RAID6 code uses ZMM registers, execute vzeroupper before returning from it. This is needed to avoid degrading the performance of any later SSE code that may happen to be executed. Link: https://lore.kernel.org/20260831212316.213896-1-ebiggers@kernel.org Fixes: e0a491c12968 ("lib/raid6: Add AVX512 optimized gen_syndrome functions") Fixes: 13c520b2993c ("lib/raid6: Add AVX512 optimized recovery functions") Signed-off-by: Eric Biggers Cc: Christoph Hellwig Cc: Signed-off-by: Andrew Morton

raid6: add missing vzeroupper to AVX2 code

2026-09-06T03:23:25+00:00

Since the AVX2 optimized RAID6 code uses YMM registers, execute vzeroupper before returning from it. This is needed to avoid degrading the performance of any later SSE code that may happen to be executed. Link: https://lore.kernel.org/20260831212308.213855-1-ebiggers@kernel.org Fixes: 2c935842bdb4 ("lib/raid6: Add AVX2 optimized gen_syndrome functions") Fixes: 7056741fd9fc ("lib/raid6: Add AVX2 optimized recovery functions") Signed-off-by: Eric Biggers Cc: Christoph Hellwig Cc: Signed-off-by: Andrew Morton

xor: add missing vzeroupper to AVX code

2026-09-06T03:23:24+00:00

Since the AVX optimized XOR code uses YMM registers, execute vzeroupper before returning from it. This is needed to avoid degrading the performance of any later SSE code that may happen to be executed. Link: https://lore.kernel.org/20260831212248.213805-1-ebiggers@kernel.org Fixes: ea4d26ae24e5 ("raid5: add AVX optimized RAID5 checksumming") Signed-off-by: Eric Biggers Cc: Christoph Hellwig Cc: Signed-off-by: Andrew Morton

raid/kunit: enable RAID6 PQ and XOR benchmarks if KUNIT_ALL_TESTS=m

2026-09-06T03:23:18+00:00

Enabling the (possibly long-running benchmarks) by default may cause a big delay in boot time in case of built-in tests. However, they can still safely be enabled by default if all tests are modular, as they would only run when requested explicitly by the system administrator. Link: https://lore.kernel.org/64c6e0191bd8ccef0074ffbb09bd0584680d710b.1787584360.git.geert@linux-m68k.org Signed-off-by: Geert Uytterhoeven Reviewed-by: Christoph Hellwig Acked-by: Ard Biesheuvel Cc: Eric Biggers Signed-off-by: Andrew Morton

raid6/kunit: add a benchmark

2026-08-04T04:10:16+00:00

Add a benchmark to test the RAID 6 P/Q functions for more representative block sizes and numbers of disks. This splits the maximum alloc size used for the benchmark from the max size for the kunit test because recovery is currently limited to a single page at a time. Hopefully this will be fixed soon. The runtime numbers are reported in GB/s as the numbers of modern implementations are basically unreadable as MB/s. This means retro-architectures could report 0, but that is an easy tradeoff. Link: https://lore.kernel.org/20260715144825.95432-9-hch@lst.de Signed-off-by: Christoph Hellwig Cc: Eric Biggers Signed-off-by: Andrew Morton

raid6: improve the runtime selection benchmark

2026-08-04T04:10:16+00:00

Use plain ktime_get_ns for the timing, use 8 + 2 disks for a realistic load, and report the throughput on the data disks as that is what storage systems are measured on. Link: https://lore.kernel.org/20260715144825.95432-8-hch@lst.de Signed-off-by: Christoph Hellwig Cc: Eric Biggers Signed-off-by: Andrew Morton

raid6: defer implementation selection when built-in

2026-08-04T04:10:16+00:00

Don't hold up early boot and defer the selection just like we've been doing for the RAID5 XOR code since commit 524ccdbdfb52 ("crypto: xor - defer load time benchmark to a later time"). This will also allow full use of the timing subsystem for benchmarking. Link: https://lore.kernel.org/20260715144825.95432-7-hch@lst.de Signed-off-by: Christoph Hellwig Cc: Eric Biggers Signed-off-by: Andrew Morton

raid6: enable lock context analysis

2026-08-04T04:10:16+00:00

The code doesn't have any locking, so this is trivial. Link: https://lore.kernel.org/20260715144825.95432-6-hch@lst.de Signed-off-by: Christoph Hellwig Cc: Eric Biggers Signed-off-by: Andrew Morton

xor/kunit: add a benchmark

2026-08-04T04:10:15+00:00

Add a benchmark to test the XOR functions for more representative block sizes and numbers of disks. Including 64k would be useful here, but increasing the test buffer size increases the runtime of the functional kunit test too much unfortunately. The runtime numbers are reported in GB/s as the numbers of modern implementations are basically unreadable as MB/s. This means retro-architectures could report 0, but that is an easy tradeoff. Link: https://lore.kernel.org/20260715144825.95432-5-hch@lst.de Signed-off-by: Christoph Hellwig Cc: Eric Biggers Signed-off-by: Andrew Morton

xor/kunit: fix a spelling error

2026-08-04T04:10:15+00:00

Link: https://lore.kernel.org/20260715144825.95432-4-hch@lst.de Signed-off-by: Christoph Hellwig Cc: Eric Biggers Signed-off-by: Andrew Morton