On 8 November 2013 22:34, Will Deacon <will.dea...@arm.com> wrote:
> On Thu, Oct 17, 2013 at 12:17:49PM +0100, Sandeepa Prabhu wrote:
>> AArch64 ISA does not instructions to pop PC register value
>> from stack(like ARM v7 has ldmia {...,pc}) without using
>> one of the general purpose registers. This means return probes
>> cannot return to the actual return address directly without
>> modifying register context, and without trapping into debug exception.
>>
>> So like many other architectures, we prepare a global routine
>> with NOPs, which serve as trampoline to hack away the
>> function return address, by placing an extra kprobe on the
>> trampoline entry.
>>
>> The pre-handler of this special trampoline' kprobe execute return
>> probe handler functions and restore original return address in ELR_EL1,
>> this way, saved pt_regs still hold the original register context to be
>> carried back to the probed kernel function.
>>
>> Signed-off-by: Sandeepa Prabhu <sandeepa.pra...@linaro.org>
>> ---
>> arch/arm64/Kconfig | 1 +
>> arch/arm64/include/asm/kprobes.h | 1 +
>> arch/arm64/include/asm/ptrace.h | 5 ++
>> arch/arm64/kernel/kprobes.c | 125
>> ++++++++++++++++++++++++++++++++++++++-
>> 4 files changed, 129 insertions(+), 3 deletions(-)
>>
>> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
>> index 2e89059..73eff55 100644
>> --- a/arch/arm64/Kconfig
>> +++ b/arch/arm64/Kconfig
>> @@ -28,6 +28,7 @@ config ARM64
>> select HAVE_MEMBLOCK
>> select HAVE_PERF_EVENTS
>> select HAVE_KPROBES if !XIP_KERNEL
>> + select HAVE_KRETPROBES if (HAVE_KPROBES)
>
> Don't need the brackets.
OK.
>
>> select IRQ_DOMAIN
>> select MODULES_USE_ELF_RELA
>> select NO_BOOTMEM
>> diff --git a/arch/arm64/include/asm/kprobes.h
>> b/arch/arm64/include/asm/kprobes.h
>> index 9b491d0..eaca849 100644
>> --- a/arch/arm64/include/asm/kprobes.h
>> +++ b/arch/arm64/include/asm/kprobes.h
>> @@ -55,5 +55,6 @@ void arch_remove_kprobe(struct kprobe *);
>> int kprobe_fault_handler(struct pt_regs *regs, unsigned int fsr);
>> int kprobe_exceptions_notify(struct notifier_block *self,
>> unsigned long val, void *data);
>> +void kretprobe_trampoline(void);
>>
>> #endif /* _ARM_KPROBES_H */
>> diff --git a/arch/arm64/include/asm/ptrace.h
>> b/arch/arm64/include/asm/ptrace.h
>> index 89f1727..58b2589 100644
>> --- a/arch/arm64/include/asm/ptrace.h
>> +++ b/arch/arm64/include/asm/ptrace.h
>> @@ -166,6 +166,11 @@ static inline int valid_user_regs(struct user_pt_regs
>> *regs)
>> #define instruction_pointer(regs) (regs)->pc
>> #define stack_pointer(regs) ((regs)->sp)
>>
>> +static inline long regs_return_value(struct pt_regs *regs)
>> +{
>> + return regs->regs[0];
>> +}
>
> This is also being implemented by another patch series (I think the audit
> stuff?).
Not sure, I did not see this being implemented in audit(audit adds for
'syscallno', not for return value in x0)
I can rebase my code if this change is implemented and queued in other patchset.
>
> Will
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
