MIPS: Correct the 64-bit DSP accumulator register size
authorMaciej W. Rozycki <macro@mips.com>
Tue, 15 May 2018 22:33:26 +0000 (23:33 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 5 Sep 2018 07:26:41 +0000 (09:26 +0200)
commit62c59b1ddbdc62ca9cf775ecc0d77ebb380105cf
tree60dfc65f028ece8a805f0a2cefb5d14423a41113
parent4bdf9c17598014d0f10c7500edeefc25e0f501c7
MIPS: Correct the 64-bit DSP accumulator register size

commit f5958b4cf4fc38ed4583ab83fb7c4cd1ab05f47b upstream.

Use the `unsigned long' rather than `__u32' type for DSP accumulator
registers, like with the regular MIPS multiply/divide accumulator and
general-purpose registers, as all are 64-bit in 64-bit implementations
and using a 32-bit data type leads to contents truncation on context
saving.

Update `arch_ptrace' and `compat_arch_ptrace' accordingly, removing
casts that are similarly not used with multiply/divide accumulator or
general-purpose register accesses.

Signed-off-by: Maciej W. Rozycki <macro@mips.com>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Fixes: e50c0a8fa60d ("Support the MIPS32 / MIPS64 DSP ASE.")
Patchwork: https://patchwork.linux-mips.org/patch/19329/
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: James Hogan <jhogan@kernel.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: linux-fsdevel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Cc: stable@vger.kernel.org # 2.6.15+
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch/mips/include/asm/processor.h
arch/mips/kernel/ptrace.c
arch/mips/kernel/ptrace32.c