linux-kernel - Re: [PATCH v2] x86/sev: Initialize ctxt variable and zero fi

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: <fc2202de-595b-f561-dc59-08f32c56ff73@amd.com>
Date: Tue, 19 Nov 2024 08:23:14 -0600
From: Tom Lendacky <thomas.lendacky@....com>
To: Ragavendra <ragavendra.bn@...il.com>, tglx@...utronix.de,
 mingo@...hat.com, bp@...en8.de, dave.hansen@...ux.intel.com, hpa@...or.com,
 ardb@...nel.org, tzimmermann@...e.de, bhelgaas@...gle.com
Cc: x86@...nel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] x86/sev: Initialize ctxt variable and zero fi

On 11/18/24 16:58, Ragavendra wrote:
> Updating the ctxt value to {} in the svsm_perform_ghcb_protocol as
> it was not initialized. Updating memory to zero for the ctxt->fi
> variable in verify_exception_info when ES_EXCEPTION is returned.
> 
> Fixes: 34ff65901735 x86/sev: Use kernel provided SVSM Calling Areas
> Signed-off-by: Ragavendra Nagraj <ragavendra.bn@...il.com>
> ---
>  arch/x86/coco/sev/shared.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/x86/coco/sev/shared.c b/arch/x86/coco/sev/shared.c
> index 71de53194089..5e0f6fbf4dd2 100644
> --- a/arch/x86/coco/sev/shared.c
> +++ b/arch/x86/coco/sev/shared.c
> @@ -239,6 +239,8 @@ static enum es_result verify_exception_info(struct ghcb *ghcb, struct es_em_ctxt
>  		if ((info & SVM_EVTINJ_VALID) &&
>  		    ((v == X86_TRAP_GP) || (v == X86_TRAP_UD)) &&
>  		    ((info & SVM_EVTINJ_TYPE_MASK) == SVM_EVTINJ_TYPE_EXEPT)) {
> +			memset(&ctxt->fi, 0, sizeof(ctxt->fi));
> +
>  			ctxt->fi.vector = v;
>  
>  			if (info & SVM_EVTINJ_VALID_ERR)
> @@ -335,7 +337,7 @@ static int svsm_perform_msr_protocol(struct svsm_call *call)
>  
>  static int svsm_perform_ghcb_protocol(struct ghcb *ghcb, struct svsm_call *call)
>  {
> -	struct es_em_ctxt ctxt;
> +	struct es_em_ctxt ctxt = {};

This isn't necessary if you are doing the memset.

Thanks,
Tom

>  	u8 pending = 0;
>  
>  	vc_ghcb_invalidate(ghcb);