When you turn it off, the kernel is unusable, so get rid of the option
and always allow unaligned access.
The Octeon specific memcpy intentionally does unaligned accesses and it
must not fault.
Signed-off-by: David Daney <david.daney@cavium.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5303/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
with this option to be run at the same time as one built without this
option.
-config CAVIUM_OCTEON_HW_FIX_UNALIGNED
- bool "Enable hardware fixups of unaligned loads and stores"
- default "y"
- help
- Configure the Octeon hardware to automatically fix unaligned loads
- and stores. Normally unaligned accesses are fixed using a kernel
- exception handler. This option enables the hardware automatic fixups,
- which requires only an extra 3 cycles. Disable this option if you
- are running code that relies on address exceptions on unaligned
- accesses.
-
config CAVIUM_OCTEON_CVMSEG_SIZE
int "Number of L1 cache lines reserved for CVMSEG memory"
range 0 54
ori v0, CONFIG_CAVIUM_OCTEON_CVMSEG_SIZE
dmtc0 v0, CP0_CVMMEMCTL_REG # Write the cavium mem control register
dmfc0 v0, CP0_CVMCTL_REG # Read the cavium control register
-#ifdef CONFIG_CAVIUM_OCTEON_HW_FIX_UNALIGNED
# Disable unaligned load/store support but leave HW fixup enabled
+ # Needed for octeon specific memcpy
or v0, v0, 0x5001
xor v0, v0, 0x1001
-#else
- # Disable unaligned load/store and HW fixup support
- or v0, v0, 0x5001
- xor v0, v0, 0x5001
-#endif
# Read the processor ID register
mfc0 v1, CP0_PRID_REG
# Disable instruction prefetching (Octeon Pass1 errata)