RE: [PATCH v2 2/5] Drivers: hv: vmbus: Track decrypted status in vmbus_gpadl

From: Michael Kelley <mhklinux@outlook.com>
To: Kuppuswamy Sathyanarayanan
	<sathyanarayanan.kuppuswamy@linux.intel.com>,
	"rick.p.edgecombe@intel.com" <rick.p.edgecombe@intel.com>,
	"kys@microsoft.com" <kys@microsoft.com>,
	"haiyangz@microsoft.com" <haiyangz@microsoft.com>,
	"wei.liu@kernel.org" <wei.liu@kernel.org>,
	"decui@microsoft.com" <decui@microsoft.com>,
	"gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
	"davem@davemloft.net" <davem@davemloft.net>,
	"edumazet@google.com" <edumazet@google.com>,
	"kuba@kernel.org" <kuba@kernel.org>,
	"pabeni@redhat.com" <pabeni@redhat.com>,
	"kirill.shutemov@linux.intel.com"
	<kirill.shutemov@linux.intel.com>,
	"dave.hansen@linux.intel.com" <dave.hansen@linux.intel.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-hyperv@vger.kernel.org" <linux-hyperv@vger.kernel.org>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	"linux-coco@lists.linux.dev" <linux-coco@lists.linux.dev>
Cc: "elena.reshetova@intel.com" <elena.reshetova@intel.com>
Subject: RE: [PATCH v2 2/5] Drivers: hv: vmbus: Track decrypted status in vmbus_gpadl
Date: Thu, 14 Mar 2024 13:56:27 +0000	[thread overview]
Message-ID: <SN6PR02MB4157CC7305AEDE7520565A0DD4292@SN6PR02MB4157.namprd02.prod.outlook.com> (raw)
In-Reply-To: <4e6627b2-30cd-4c50-bf2f-24cf845cd4bc@linux.intel.com>

From: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
> 
> >>> @@ -886,6 +901,8 @@ int vmbus_teardown_gpadl(struct vmbus_channel *channel, struct vmbus_gpadl *gpad
> >>>  	if (ret)
> >>>  		pr_warn("Fail to set mem host visibility in GPADL teardown %d.\n", ret);
> >>
> >> Will this be called only if vmbus_establish_gpad() is successful? If not, you
> >> might want to skip set_memory_encrypted() call for decrypted = false case.
> >
> > It's only called if vmbus_establish_gpadl() is successful.  I agree
> > we don't want to call set_memory_encrypted() if the
> > set_memory_decrypted() wasn't executed or it failed.  But
> > vmbus_teardown_gpadl() is never called with decrypted = false.
> 
> Since you rely on  vmbus_teardown_gpadl() callers, personally I think it
> is better to add that check. It is up to you.
> 

In my judgment, a check isn't really necessary.  The structure of the GPADL
code has been stable for a long time, and I'm not aware of anything
pending that would motivate a change.  And if something did change
to call vmbus_teardown_gpadl() with the memory still encrypted,
the call to set_memory_encrypted() will cause an immediate error and
a WARN_ONCE from Rick's patch to __set_memory_enc_pgtable().
The problem won't go unnoticed.

Michael