linux.git/lib/raid/raid6/algos.c, branch v7.2-rc1

raid6: use kmalloc() in raid6_select_algo()

2026-06-04T21:49:28+00:00

raid6_select_algo() allocates 8 pages for buffer that is used
as a scratch area for selection of the best algorithm.

This buffer can be allocated with kmalloc() as there's nothing special
about it to go directly to the page allocator.

kmalloc() provides a better API than ancient __get_free_pages().
kmalloc() does not require ugly casts and kfree() does not need to know the
size of the freed object.

There is no performance difference because kmalloc() redirects allocations
of such size to the page allocator.

Replace __get_free_pages() call with kmalloc().

Link: https://lore.kernel.org/all/635405e4-9423-4a25-a6e7-e03c8ea0bcbe@redhat.com
Link: https://lore.kernel.org/20260528-lib-v4-2-4e3ad1277279@kernel.org
Signed-off-by: Mike Rapoport (Microsoft) 
Reviewed-by: Christoph Hellwig 
Cc: Christoph Hellwig 
Cc: Hannes Reinecke 
Cc: Li Nan 
Cc: Song Liu 
Signed-off-by: Andrew Morton

raid6: update top of file comments

2026-05-29T04:24:55+00:00

Drop the pointless mention of the file name, and use standard formatting
for the top of file comments.

Link: https://lore.kernel.org/20260518051804.462141-14-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Ard Biesheuvel 
Tested-by: Ard Biesheuvel  # kunit only on arm64
Cc: Albert Ou 
Cc: Alexander Gordeev 
Cc: Alexandre Ghiti 
Cc: Arnd Bergmann 
Cc: "Borislav Petkov (AMD)" 
Cc: Catalin Marinas 
Cc: Chris Mason 
Cc: Christian Borntraeger 
Cc: Dan Williams 
Cc: David Sterba 
Cc: Heiko Carstens 
Cc: Herbert Xu 
Cc: "H. Peter Anvin" 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Li Nan 
Cc: Madhavan Srinivasan 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Palmer Dabbelt 
Cc: Song Liu 
Cc: Sven Schnelle 
Cc: Vasily Gorbik 
Cc: WANG Xuerui 
Cc: Will Deacon 
Signed-off-by: Andrew Morton

raid6: use static_call for raid6_recov_2data and raid6_recov_datap

2026-05-29T04:24:55+00:00

Avoid expensive indirect calls for the recovery routines as well.

Link: https://lore.kernel.org/20260518051804.462141-13-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Ard Biesheuvel 
Tested-by: Ard Biesheuvel  # kunit only on arm64
Cc: Albert Ou 
Cc: Alexander Gordeev 
Cc: Alexandre Ghiti 
Cc: Arnd Bergmann 
Cc: "Borislav Petkov (AMD)" 
Cc: Catalin Marinas 
Cc: Chris Mason 
Cc: Christian Borntraeger 
Cc: Dan Williams 
Cc: David Sterba 
Cc: Heiko Carstens 
Cc: Herbert Xu 
Cc: "H. Peter Anvin" 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Li Nan 
Cc: Madhavan Srinivasan 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Palmer Dabbelt 
Cc: Song Liu 
Cc: Sven Schnelle 
Cc: Vasily Gorbik 
Cc: WANG Xuerui 
Cc: Will Deacon 
Signed-off-by: Andrew Morton

raid6: use static_call for gen_syndrom and xor_syndrom

2026-05-29T04:24:55+00:00

Avoid indirect calls for P/Q parity generation.

Link: https://lore.kernel.org/20260518051804.462141-12-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Ard Biesheuvel 
Tested-by: Ard Biesheuvel  # kunit only on arm64
Cc: Albert Ou 
Cc: Alexander Gordeev 
Cc: Alexandre Ghiti 
Cc: Arnd Bergmann 
Cc: "Borislav Petkov (AMD)" 
Cc: Catalin Marinas 
Cc: Chris Mason 
Cc: Christian Borntraeger 
Cc: Dan Williams 
Cc: David Sterba 
Cc: Heiko Carstens 
Cc: Herbert Xu 
Cc: "H. Peter Anvin" 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Li Nan 
Cc: Madhavan Srinivasan 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Palmer Dabbelt 
Cc: Song Liu 
Cc: Sven Schnelle 
Cc: Vasily Gorbik 
Cc: WANG Xuerui 
Cc: Will Deacon 
Signed-off-by: Andrew Morton

raid6: rework registration of optimized algorithms

2026-05-29T04:24:54+00:00

Replace the static array of algorithms with a call to an architecture
helper to register algorithms.  This serves two purposes: it avoid having
to register all algorithms in a single central place, and it removes the
need for the priority field by just registering the algorithms that the
architecture considers suitable for the currently running CPUs.

[hch@lst.de: register avx512 after avx2]
  Link: https://lore.kernel.org/20260527074539.2292913-3-hch@lst.de
Link: https://lore.kernel.org/20260518051804.462141-11-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Ard Biesheuvel 
Tested-by: Ard Biesheuvel  # kunit only on arm64
Cc: Albert Ou 
Cc: Alexander Gordeev 
Cc: Alexandre Ghiti 
Cc: Arnd Bergmann 
Cc: "Borislav Petkov (AMD)" 
Cc: Catalin Marinas 
Cc: Chris Mason 
Cc: Christian Borntraeger 
Cc: Dan Williams 
Cc: David Sterba 
Cc: Heiko Carstens 
Cc: Herbert Xu 
Cc: "H. Peter Anvin" 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Li Nan 
Cc: Madhavan Srinivasan 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Palmer Dabbelt 
Cc: Song Liu 
Cc: Sven Schnelle 
Cc: Vasily Gorbik 
Cc: WANG Xuerui 
Cc: Will Deacon 
Signed-off-by: Andrew Morton

raid6: hide internals

2026-05-29T04:24:54+00:00

Split out two new headers from the public pq.h:

 - lib/raid/raid6/algos.h contains the algorithm lists private to
   lib/raid/raid6
 - include/linux/raid/pq_tables.h contains the tables also used by
   async_tx providers.

The public include/linux/pq.h is now limited to the public interface for
the consumers of the RAID6 PQ API.

[hch@lst.de: remove duplicate ccflags-y line]
  Link: https://lore.kernel.org/20260527074539.2292913-2-hch@lst.de
Link: https://lore.kernel.org/20260518051804.462141-10-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Ard Biesheuvel 
Tested-by: Ard Biesheuvel  # kunit only on arm64
Cc: Albert Ou 
Cc: Alexander Gordeev 
Cc: Alexandre Ghiti 
Cc: Arnd Bergmann 
Cc: "Borislav Petkov (AMD)" 
Cc: Catalin Marinas 
Cc: Chris Mason 
Cc: Christian Borntraeger 
Cc: Dan Williams 
Cc: David Sterba 
Cc: Heiko Carstens 
Cc: Herbert Xu 
Cc: "H. Peter Anvin" 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Li Nan 
Cc: Madhavan Srinivasan 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Palmer Dabbelt 
Cc: Song Liu 
Cc: Sven Schnelle 
Cc: Vasily Gorbik 
Cc: WANG Xuerui 
Cc: Will Deacon 
Signed-off-by: Andrew Morton

raid6: warn when using less than four devices

2026-05-29T04:24:54+00:00

Quoting H.  Peter Anvin who came up with the RAID6 P/Q algorithm, and who
wrote the initial implementation, then still part of the md driver:

  The RAID-6 code has *never* supported only 3 units, and if it ever
  worked for *any* of the implementations it was purely by accident.
  Speaking as the original author I should know; this was deliberate as
  in some cases the degenerate case (3) would have required extra trays
  in the code to no user benefit.

While md never allowed less than 4 devices, btrfs does.  This new warning
will trigger for such file systems, but given how it already causes havoc
that is a good thing.  If btrfs wants to fix third, it should switch to
transparently use three-way mirroring underneath, which will work as P and
Q are copies of the single data device by the definition of the Linux RAID
6 P/Q algorithm.

Link: https://lore.kernel.org/20260518051804.462141-9-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Ard Biesheuvel 
Tested-by: Ard Biesheuvel  # kunit only on arm64
Cc: Albert Ou 
Cc: Alexander Gordeev 
Cc: Alexandre Ghiti 
Cc: Arnd Bergmann 
Cc: "Borislav Petkov (AMD)" 
Cc: Catalin Marinas 
Cc: Chris Mason 
Cc: Christian Borntraeger 
Cc: Dan Williams 
Cc: David Sterba 
Cc: Heiko Carstens 
Cc: Herbert Xu 
Cc: "H. Peter Anvin" 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Li Nan 
Cc: Madhavan Srinivasan 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Palmer Dabbelt 
Cc: Song Liu 
Cc: Sven Schnelle 
Cc: Vasily Gorbik 
Cc: WANG Xuerui 
Cc: Will Deacon 
Signed-off-by: Andrew Morton

raid6: improve the public interface

2026-05-29T04:24:54+00:00

Stop directly calling into function pointers from users of the RAID6 PQ
API, and provide exported functions with proper documentation and API
guarantees asserts where applicable instead.

Link: https://lore.kernel.org/20260518051804.462141-8-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Ard Biesheuvel 
Tested-by: Ard Biesheuvel  # kunit only on arm64
Cc: Albert Ou 
Cc: Alexander Gordeev 
Cc: Alexandre Ghiti 
Cc: Arnd Bergmann 
Cc: "Borislav Petkov (AMD)" 
Cc: Catalin Marinas 
Cc: Chris Mason 
Cc: Christian Borntraeger 
Cc: Dan Williams 
Cc: David Sterba 
Cc: Heiko Carstens 
Cc: Herbert Xu 
Cc: "H. Peter Anvin" 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Li Nan 
Cc: Madhavan Srinivasan 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Palmer Dabbelt 
Cc: Song Liu 
Cc: Sven Schnelle 
Cc: Vasily Gorbik 
Cc: WANG Xuerui 
Cc: Will Deacon 
Signed-off-by: Andrew Morton

raid6: move to lib/raid/

2026-05-29T04:24:53+00:00

Move the raid6 code to live in lib/raid/ with the XOR code, and change the
internal organization so that each architecture has a subdirectory similar
to the CRC, crypto and XOR libraries, and fix up the Makefile to only
build files actually needed.

Also move the kunit test case from the history test/ subdirectory to
tests/ and use the normal naming scheme for it.

Link: https://lore.kernel.org/20260518051804.462141-4-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Ard Biesheuvel 
Tested-by: Ard Biesheuvel  # kunit only on arm64
Cc: Albert Ou 
Cc: Alexander Gordeev 
Cc: Alexandre Ghiti 
Cc: Arnd Bergmann 
Cc: "Borislav Petkov (AMD)" 
Cc: Catalin Marinas 
Cc: Chris Mason 
Cc: Christian Borntraeger 
Cc: Dan Williams 
Cc: David Sterba 
Cc: Heiko Carstens 
Cc: Herbert Xu 
Cc: "H. Peter Anvin" 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Li Nan 
Cc: Madhavan Srinivasan 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Palmer Dabbelt 
Cc: Song Liu 
Cc: Sven Schnelle 
Cc: Vasily Gorbik 
Cc: WANG Xuerui 
Cc: Will Deacon 
Signed-off-by: Andrew Morton