[PATCH v10 02/43] arm64: RME: Handle Granule Protection Faults (GPFs)

[Steven Price]

On 29/08/2025 12:38, Catalin Marinas wrote:
> On Wed, Aug 20, 2025 at 03:55:22PM +0100, Steven Price wrote:
>> If the host attempts to access granules that have been delegated for use
>> in a realm these accesses will be caught and will trigger a Granule
>> Protection Fault (GPF).
>>
>> A fault during a page walk signals a bug in the kernel and is handled by
>> oopsing the kernel. A non-page walk fault could be caused by user space
>> having access to a page which has been delegated to the kernel and will
>> trigger a SIGBUS to allow debugging why user space is trying to access a
>> delegated page.
>>
>> Reviewed-by: Suzuki K Poulose <suzuki.poulose at arm.com>
>> Reviewed-by: Gavin Shan <gshan at redhat.com>
>> Signed-off-by: Steven Price <steven.price at arm.com>
>> ---
>> Changes since v2:
>>  * Include missing "Granule Protection Fault at level -1"
>> ---
>>  arch/arm64/mm/fault.c | 31 +++++++++++++++++++++++++------
>>  1 file changed, 25 insertions(+), 6 deletions(-)
>>
>> diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
>> index d816ff44faff..e4237637cd8f 100644
>> --- a/arch/arm64/mm/fault.c
>> +++ b/arch/arm64/mm/fault.c
>> @@ -854,6 +854,25 @@ static int do_tag_check_fault(unsigned long far, unsigned long esr,
>>  	return 0;
>>  }
>>  
>> +static int do_gpf_ptw(unsigned long far, unsigned long esr, struct pt_regs *regs)
>> +{
>> +	const struct fault_info *inf = esr_to_fault_info(esr);
>> +
>> +	die_kernel_fault(inf->name, far, esr, regs);
>> +	return 0;
>> +}
> 
> This is fine, it's irrelevant whether the fault happened at EL0 or EL1.
> 
>> +static int do_gpf(unsigned long far, unsigned long esr, struct pt_regs *regs)
>> +{
>> +	const struct fault_info *inf = esr_to_fault_info(esr);
>> +
>> +	if (!is_el1_instruction_abort(esr) && fixup_exception(regs, esr))
>> +		return 0;
>> +
>> +	arm64_notify_die(inf->name, regs, inf->sig, inf->code, far, esr);
>> +	return 0;
>> +}
> 
> The end result is somewhat similar but why not just return 1 and avoid
> the arm64_notify_die() call? Let do_mem_abort() handle the oops vs user
> signal. With die_kernel_fault() we print the "Unable to handle
> kernel..." message and some more information.
> 

Yes, that makes sense - something has gone very wrong if the kernel hits
a GPF but that's no reason not to output a (more) useful message.

Thanks,
Steve