linux-kernel - Re: [PATCH v2 1/9] crypto: ccp: Move dev_info/err messages for SEV/SNP initialization

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <08045113-a4a1-4dd6-be66-0c794559d212@amd.com>
Date: Fri, 27 Dec 2024 19:58:38 +1100
From: Alexey Kardashevskiy <aik@....com>
To: Ashish Kalra <Ashish.Kalra@....com>, seanjc@...gle.com,
 pbonzini@...hat.com, tglx@...utronix.de, mingo@...hat.com, bp@...en8.de,
 dave.hansen@...ux.intel.com, x86@...nel.org, hpa@...or.com,
 thomas.lendacky@....com, john.allen@....com, herbert@...dor.apana.org.au,
 davem@...emloft.net
Cc: michael.roth@....com, dionnaglaze@...gle.com, kvm@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-crypto@...r.kernel.org,
 linux-coco@...ts.linux.dev
Subject: Re: [PATCH v2 1/9] crypto: ccp: Move dev_info/err messages for
 SEV/SNP initialization

On 17/12/24 10:57, Ashish Kalra wrote:
> From: Ashish Kalra <ashish.kalra@....com>
> 
> Remove dev_info and dev_err messages related to SEV/SNP initialization
> from callers and instead move those inside __sev_platform_init_locked()
> and __sev_snp_init_locked().

It is not actually removing anything, only adding.

> 
> This allows both _sev_platform_init_locked() and various SEV/SNP ioctls
> to call __sev_platform_init_locked() and __sev_snp_init_locked() for
> implicit SEV/SNP initialization and shutdown without additionally
> printing any errors/success messages.
> 
> Signed-off-by: Ashish Kalra <ashish.kalra@....com>
> ---
>   drivers/crypto/ccp/sev-dev.c | 23 ++++++++++++++++++-----
>   1 file changed, 18 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/crypto/ccp/sev-dev.c b/drivers/crypto/ccp/sev-dev.c
> index af018afd9cd7..1c1c33d3ed9a 100644
> --- a/drivers/crypto/ccp/sev-dev.c
> +++ b/drivers/crypto/ccp/sev-dev.c
> @@ -1177,19 +1177,27 @@ static int __sev_snp_init_locked(int *error)
>   
>   	rc = __sev_do_cmd_locked(cmd, arg, error);
>   	if (rc)
> -		return rc;
> +		goto err;

here ...

>   
>   	/* Prepare for first SNP guest launch after INIT. */
>   	wbinvd_on_all_cpus();
>   	rc = __sev_do_cmd_locked(SEV_CMD_SNP_DF_FLUSH, NULL, error);
>   	if (rc)
> -		return rc;
> +		goto err;
>   

... and here are different calls, and the message below is going to say 
"failed to INIT" when it actually failed to SEV_CMD_SNP_DF_FLUSH in this 
case. I'd like separate dev_err() for both. Other errors in this 
function do have own dev_err() already.


>   	sev->snp_initialized = true;
>   	dev_dbg(sev->dev, "SEV-SNP firmware initialized\n");
>   
> +	dev_info(sev->dev, "SEV-SNP API:%d.%d build:%d\n", sev->api_major,
> +		 sev->api_minor, sev->build);
> +
>   	sev_es_tmr_size = SNP_TMR_SIZE;
>   
> +	return 0;
> +
> +err:
> +	dev_err(sev->dev, "SEV-SNP: failed to INIT rc %d, error %#x\n",
> +		rc, *error);
>   	return rc;
>   }
>   
> @@ -1268,7 +1276,7 @@ static int __sev_platform_init_locked(int *error)
>   
>   	rc = __sev_platform_init_handle_init_ex_path(sev);
>   	if (rc)
> -		return rc;
> +		goto err;
>   
>   	rc = __sev_do_init_locked(&psp_ret);
>   	if (rc && psp_ret == SEV_RET_SECURE_DATA_INVALID) {
> @@ -1288,7 +1296,7 @@ static int __sev_platform_init_locked(int *error)
>   		*error = psp_ret;
>   
>   	if (rc)
> -		return rc;
> +		goto err;
>   
>   	sev->state = SEV_STATE_INIT;
>   
> @@ -1296,7 +1304,7 @@ static int __sev_platform_init_locked(int *error)
>   	wbinvd_on_all_cpus();
>   	rc = __sev_do_cmd_locked(SEV_CMD_DF_FLUSH, NULL, error);
>   	if (rc)
> -		return rc;
> +		goto err;
>   
>   	dev_dbg(sev->dev, "SEV firmware initialized\n");
>   
> @@ -1304,6 +1312,11 @@ static int __sev_platform_init_locked(int *error)
>   		 sev->api_minor, sev->build);
>   
>   	return 0;
> +
> +err:
> +	dev_err(sev->dev, "SEV: failed to INIT error %#x, rc %d\n",
> +		psp_ret, rc);

The same comment here. For example, I saw the "invalid page state" error 
from the PSP soooo many times so I believe any command can return it :) 
Thanks,


> +	return rc;
>   }
>   
>   static int _sev_platform_init_locked(struct sev_platform_init_args *args)

-- 
Alexey