On Wed, Feb 03, 2021 at 01:19:12PM +0800, Zhiyuan Dai wrote:
> Fixed two coding style issues in kernel/trap.c
> 1. spaces required around that '='
This change makes sense to me as a cleanup.
> 2. Missing a blank line after declarations
We're inconsistent w.r.t. this across the tree, and applying this
everywhere would be very churny, but I think it makes sense to be
consistent locally (e.g. within a function).
Currently call_undef_hook() is inconsistent in this respect, but I think
this patch leaves it inconsistent in a different way. Please see below.
> Signed-off-by: Zhiyuan Dai <daizhiy...@phytium.com.cn>
> ---
> arch/arm64/kernel/traps.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c
> index 6895ce7..4001a39 100644
> --- a/arch/arm64/kernel/traps.c
> +++ b/arch/arm64/kernel/traps.c
> @@ -45,7 +45,7 @@
> #include <asm/system_misc.h>
> #include <asm/sysreg.h>
>
> -static const char *handler[]= {
> +static const char *handler[] = {
> "Synchronous Abort",
> "IRQ",
> "FIQ",
> @@ -318,6 +318,7 @@ static int call_undef_hook(struct pt_regs *regs)
> } else if (compat_thumb_mode(regs)) {
> /* 16-bit Thumb instruction */
> __le16 instr_le;
> +
> if (get_user(instr_le, (__le16 __user *)pc))
> goto exit;
> instr = le16_to_cpu(instr_le);
Immediately above this we have:
| if (!user_mode(regs)) {
| __le32 instr_le;
| if (get_kernel_nofault(instr_le, (__force __le32 *)pc))
| goto exit;
| instr = le32_to_cpu(instr_le);
... which we should either change at the same time, or not bother with
this at all.
Thanks,
Mark.
> @@ -332,6 +333,7 @@ static int call_undef_hook(struct pt_regs *regs)
> } else {
> /* 32-bit ARM instruction */
> __le32 instr_le;
> +
> if (get_user(instr_le, (__le32 __user *)pc))
> goto exit;
> instr = le32_to_cpu(instr_le);
> --
> 1.8.3.1
>
