Re: [Part2 PATCH v5.1 12.6/31] crypto: ccp: Implement SEV_PDH_GEN ioctl command

[Borislav Petkov]

On Thu, Oct 12, 2017 at 03:21:04PM -0500, Brijesh Singh wrote:
> We need to follow the platform state machine logic defined in SEV spec
> section 5.1.2. The PEK_GEN can not be issued when platform is in WORKING
> state because the command actually re-generate the identity of the
> platform itself (in other words re-generate the Platform Endorsement
> Key). Whereas, the PDH_GEN command is used for re-generating Platform
> Diffie-Hellman Key which can be changed while the guest is running.

I see.

So the proposition to carve out and split the platform *init commands
might come in handy here too...

-- 
Regards/Gruss,
Boris.

SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 
(AG Nürnberg)
-- 

Re: [Part2 PATCH v5.1 12.6/31] crypto: ccp: Implement SEV_PDH_GEN ioctl command

[Brijesh Singh]

On 10/12/17 1:48 PM, Borislav Petkov wrote:
...
> On Fri, Oct 06, 2017 at 08:06:04PM -0500, Brijesh Singh wrote:
>> The SEV_PDH_GEN command is used to re-generate the Platform
>> Diffie-Hellman (PDH) key. The command is defined in SEV spec section
>> 5.9.
>>
>> Cc: Paolo Bonzini 
>> Cc: "Radim Krčmář" 
>> Cc: Borislav Petkov 
>> Cc: Herbert Xu 
>> Cc: Gary Hook 
>> Cc: Tom Lendacky 
>> Cc: linux-crypto@vger.kernel.org
>> Cc: k...@vger.kernel.org
>> Cc: linux-ker...@vger.kernel.org
>> Signed-off-by: Brijesh Singh 
>> ---
>>  drivers/crypto/ccp/psp-dev.c | 32 
>>  1 file changed, 32 insertions(+)
>>
>> diff --git a/drivers/crypto/ccp/psp-dev.c b/drivers/crypto/ccp/psp-dev.c
>> index 03d7bd03ad58..28efb7a9245a 100644
>> --- a/drivers/crypto/ccp/psp-dev.c
>> +++ b/drivers/crypto/ccp/psp-dev.c
>> @@ -271,6 +271,34 @@ static int sev_ioctl_pek_gen(struct sev_issue_cmd *argp)
>>  return ret;
>>  }
>>  
>> +static int sev_ioctl_pdh_gen(struct sev_issue_cmd *argp)
>> +{
>> +int ret, state, do_shutdown = 0;
>> +
>> +/*
>> + * PDH_GEN command can be issued when platform is in INIT or WORKING
>> + * state. If we are in UNINIT state then transition in INIT state
>> + * before issuing the command.
>> + */
>> +ret = sev_platform_get_state(, >error);
>> +if (ret)
>> +return ret;
>> +
> Why isn't this function doing:
>
> if (state == SEV_STATE_WORKING) {
> return -EBUSY;
>
> like the PEK_GEN one?

We need to follow the platform state machine logic defined in SEV spec
section 5.1.2. The PEK_GEN can not be issued when platform is in WORKING
state because the command actually re-generate the identity of the
platform itself (in other words re-generate the Platform Endorsement
Key). Whereas, the PDH_GEN command is used for re-generating Platform
Diffie-Hellman Key which can be changed while the guest is running.

> Because if so, you can convert it and the PEK_GEN one into a single
> function doing the work and wrappers handing in the command to avoid the
> code duplication.
>
>> +if (state == SEV_STATE_UNINIT) {
>> +ret = sev_firmware_init(>error);
>> +if (ret)
>> +return ret;
>> +do_shutdown = 1;
>> +}
>> +
>> +ret = sev_handle_cmd(SEV_CMD_PDH_GEN, 0, >error);
>> +
>> +if (do_shutdown)
>> +sev_handle_cmd(SEV_CMD_SHUTDOWN, 0, NULL);
>> +
>> +return ret;
>> +}
>> +
>>  static long sev_ioctl(struct file *file, unsigned int ioctl, unsigned long 
>> arg)
>>  {
>>  void __user *argp = (void __user *)arg;
>> @@ -300,6 +328,10 @@ static long sev_ioctl(struct file *file, unsigned int 
>> ioctl, unsigned long arg)
>>  ret = sev_ioctl_pek_gen();
>>  break;
>>  }
>> +case SEV_PDH_GEN: {
>> +ret = sev_ioctl_pdh_gen();
>> +break;
>> +}
> And those curly braces can go, as before.
>

Re: [Part2 PATCH v5.1 12.6/31] crypto: ccp: Implement SEV_PDH_GEN ioctl command

[Borislav Petkov]

On Fri, Oct 06, 2017 at 08:06:04PM -0500, Brijesh Singh wrote:
> The SEV_PDH_GEN command is used to re-generate the Platform
> Diffie-Hellman (PDH) key. The command is defined in SEV spec section
> 5.9.
> 
> Cc: Paolo Bonzini 
> Cc: "Radim Krčmář" 
> Cc: Borislav Petkov 
> Cc: Herbert Xu 
> Cc: Gary Hook 
> Cc: Tom Lendacky 
> Cc: linux-crypto@vger.kernel.org
> Cc: k...@vger.kernel.org
> Cc: linux-ker...@vger.kernel.org
> Signed-off-by: Brijesh Singh 
> ---
>  drivers/crypto/ccp/psp-dev.c | 32 
>  1 file changed, 32 insertions(+)
> 
> diff --git a/drivers/crypto/ccp/psp-dev.c b/drivers/crypto/ccp/psp-dev.c
> index 03d7bd03ad58..28efb7a9245a 100644
> --- a/drivers/crypto/ccp/psp-dev.c
> +++ b/drivers/crypto/ccp/psp-dev.c
> @@ -271,6 +271,34 @@ static int sev_ioctl_pek_gen(struct sev_issue_cmd *argp)
>   return ret;
>  }
>  
> +static int sev_ioctl_pdh_gen(struct sev_issue_cmd *argp)
> +{
> + int ret, state, do_shutdown = 0;
> +
> + /*
> +  * PDH_GEN command can be issued when platform is in INIT or WORKING
> +  * state. If we are in UNINIT state then transition in INIT state
> +  * before issuing the command.
> +  */
> + ret = sev_platform_get_state(, >error);
> + if (ret)
> + return ret;
> +

Why isn't this function doing:

if (state == SEV_STATE_WORKING) {
return -EBUSY;

like the PEK_GEN one?

Because if so, you can convert it and the PEK_GEN one into a single
function doing the work and wrappers handing in the command to avoid the
code duplication.

> + if (state == SEV_STATE_UNINIT) {
> + ret = sev_firmware_init(>error);
> + if (ret)
> + return ret;
> + do_shutdown = 1;
> + }
> +
> + ret = sev_handle_cmd(SEV_CMD_PDH_GEN, 0, >error);
> +
> + if (do_shutdown)
> + sev_handle_cmd(SEV_CMD_SHUTDOWN, 0, NULL);
> +
> + return ret;
> +}
> +
>  static long sev_ioctl(struct file *file, unsigned int ioctl, unsigned long 
> arg)
>  {
>   void __user *argp = (void __user *)arg;
> @@ -300,6 +328,10 @@ static long sev_ioctl(struct file *file, unsigned int 
> ioctl, unsigned long arg)
>   ret = sev_ioctl_pek_gen();
>   break;
>   }
> + case SEV_PDH_GEN: {
> + ret = sev_ioctl_pdh_gen();
> + break;
> + }

And those curly braces can go, as before.

-- 
Regards/Gruss,
Boris.

SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 
(AG Nürnberg)
--