Re: [PATCH v5] x86/hvm: Implement hvmemul_write() using real mappings

From: "Jan Beulich" <JBeulich@suse.com>
To: Alexandru Stefan ISAILA <aisaila@bitdefender.com>,
	Andrew Cooper <andrew.cooper3@citrix.com>
Cc: "paul.durrant@citrix.com" <paul.durrant@citrix.com>,
	"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>
Subject: Re: [PATCH v5] x86/hvm: Implement hvmemul_write() using real mappings
Date: Wed, 27 Sep 2017 06:40:32 -0600	[thread overview]
Message-ID: <59CBB860020000780018015E@prv-mh.provo.novell.com> (raw)
In-Reply-To: <acd6903e-6e35-6012-47fa-02e3eb181bc4@citrix.com>

>>> On 27.09.17 at 14:11, <andrew.cooper3@citrix.com> wrote:
> On 27/09/17 12:48, Alexandru Stefan ISAILA wrote:
>> On Mi, 2017-09-27 at 09:38 +0100, Andrew Cooper wrote:
>>> On 27/09/2017 09:04, Alexandru Isaila wrote:
>>>> From: Andrew Cooper <andrew.cooper3@citrix.com>
>>>>
>>>>
>>>> -        return X86EMUL_EXCEPTION;
>>>> -    case HVMTRANS_bad_gfn_to_mfn:
>>>> -        return hvmemul_linear_mmio_write(addr, bytes, p_data,
>>>> pfec, hvmemul_ctxt, 0);
>>> Where has the if ( !mapping ) test gone?  The HVMTRANS_bad_gfn_to_mfn
>>> case needs handling.
>> There was a comment form Jan in V2. NOTE: "v1
>> comment:'Pointless"else".'"
> 
> That means that the "else " text is pointless, not the clause.  (Jan:
> Please do try to be clearer when stating "pointless else", because it
> really is ambiguous and this mistake is a valid interpretation of your
> statement.)

How would one word this in an unambiguous way? And anyway -
it should have occurred to Alexandru that deleting not the "else"
alone has an unwanted side effect. If this wasn't enough for the
ambiguity to be resolved, he should have asked back.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel