Re: [PATCH v4 02/15] firmware/psci: Add psci_early_test_conduit()

[Steven Price]

Hi Will!

On 09/07/2024 11:48, Will Deacon wrote:
> On Mon, Jul 01, 2024 at 10:54:52AM +0100, Steven Price wrote:
>> From: Jean-Philippe Brucker <jean-philippe@xxxxxxxxxx>
>>
>> Add a function to test early if PSCI is present and what conduit it
>> uses. Because the PSCI conduit corresponds to the SMCCC one, this will
>> let the kernel know whether it can use SMC instructions to discuss with
>> the Realm Management Monitor (RMM), early enough to enable RAM and
>> serial access when running in a Realm.
>>
>> Signed-off-by: Jean-Philippe Brucker <jean-philippe@xxxxxxxxxx>
>> Signed-off-by: Steven Price <steven.price@xxxxxxx>
>> ---
>> v4: New patch
>> ---
>>  drivers/firmware/psci/psci.c | 25 +++++++++++++++++++++++++
>>  include/linux/psci.h         |  5 +++++
>>  2 files changed, 30 insertions(+)
>>
>> diff --git a/drivers/firmware/psci/psci.c b/drivers/firmware/psci/psci.c
>> index d9629ff87861..a40dcaf17822 100644
>> --- a/drivers/firmware/psci/psci.c
>> +++ b/drivers/firmware/psci/psci.c
>> @@ -13,6 +13,7 @@
>>  #include <linux/errno.h>
>>  #include <linux/linkage.h>
>>  #include <linux/of.h>
>> +#include <linux/of_fdt.h>
>>  #include <linux/pm.h>
>>  #include <linux/printk.h>
>>  #include <linux/psci.h>
>> @@ -767,6 +768,30 @@ int __init psci_dt_init(void)
>>  	return ret;
>>  }
>>  
>> +/*
>> + * Test early if PSCI is supported, and if its conduit matches @conduit
>> + */
>> +bool __init psci_early_test_conduit(enum arm_smccc_conduit conduit)
>> +{
>> +	int len;
>> +	int psci_node;
>> +	const char *method;
>> +	unsigned long dt_root;
>> +
>> +	/* DT hasn't been unflattened yet, we have to work with the flat blob */
>> +	dt_root = of_get_flat_dt_root();
>> +	psci_node = of_get_flat_dt_subnode_by_name(dt_root, "psci");
>> +	if (psci_node <= 0)
>> +		return false;
>> +
>> +	method = of_get_flat_dt_prop(psci_node, "method", &len);
>> +	if (!method)
>> +		return false;
>> +
>> +	return  (conduit == SMCCC_CONDUIT_SMC && strncmp(method, "smc", len) == 0) ||
>> +		(conduit == SMCCC_CONDUIT_HVC && strncmp(method, "hvc", len) == 0);
>> +}
> 
> Hmm, I don't think this is sufficient to check for SMCCC reliably.
> Instead, I think you need to do something more involved:
> 
> 1. Check for PSCI in the DT
> 2. Check that the PSCI major version is >= 1
> 3. Use PSCI_FEATURES to check that you have SMCCC
> 4. Use SMCCC_VERSION to find out which version of SMCCC you have
> 
> That's roughly what the PSCI driver does, so we should avoid duplicating
> that logic.

Hmm, you have a point. This was an improvement over the previous (assume
that making an SMC call is going to work well enough to return at error
for non-realms), but I guess it's technically possible for a system to
have PSCI but not implement enough of SMCCC to reliably get an error
code back.

Do you have any pointers on how pKVM detects it is a guest? Currently we
have a couple of things we have to do very early as a realm guest:

 * Mark memory as RIPAS_RAM. My preference would be to punt this problem
to the VMM (or UEFI) and make it a boot requirement. But it gets in the
way for attestation purposes as you can't have the same attestation
report for VMs which differ only by available RAM in that case.

 * earlycon - we need to know if the serial device is in unprotected
MMIO as that needs mapping with the top IPA bit set.

(I think that's all that's really early - everything else I can think of
should be after the usual PSCI probe should have happened)

Thanks for taking a look!

Steve