RISC-V: Report vector unaligned access speed hwprobe

Browse source

Detect if vector misaligned accesses are faster or slower than
equivalent vector byte accesses. This is useful for usermode to know
whether vector byte accesses or vector misaligned accesses have a better
bandwidth for operations like memcpy.

Signed-off-by: Jesse Taube <jesse@rivosinc.com>
Reviewed-by: Charlie Jenkins <charlie@rivosinc.com>
Link: https://lore.kernel.org/r/20241017-jesse_unaligned_vector-v10-5-5b33500160f8@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>

...

This commit is contained in:

Jesse Taube 2024-10-17 12:00:22 -07:00 • committed by Palmer Dabbelt

parent d1703dc7bc

commit e7c9d66e31

No known key found for this signature in database

GPG key ID: 2E1319F35FBB1889

6 changed files with 228 additions and 3 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

18

arch/riscv/Kconfig

View file

3

arch/riscv/kernel/Makefile

View file

5

arch/riscv/kernel/copy-unaligned.h

View file

6

arch/riscv/kernel/sys_hwprobe.c

View file

141

arch/riscv/kernel/unaligned_access_speed.c

View file

58

arch/riscv/kernel/vec-copy-unaligned.S

View file