MIME-Version: 1.0
References: <CALZpt+FvLb=N5Qygs+dPmh1o9QCwXj8RoznF5n47opOq7CG_0g@mail.gmail.com>
 <20210708111716.GC1339@erisian.com.au>
 <CALZpt+FCCgSiRh2qAL+RM0S9Vm8s-xS3VdTAZhS9VwLcFi_1QQ@mail.gmail.com>
 <20210710014732.GA5164@erisian.com.au>
In-Reply-To: <20210710014732.GA5164@erisian.com.au>
From: Antoine Riard <antoine.riard@gmail.com>
Date: Sun, 11 Jul 2021 20:02:12 -0400
Message-ID: <CALZpt+G0uN_ek5kVre_e38OtSZ5izRUC2qb5qJC3bVkh5FV-Fw@mail.gmail.com>
To: Anthony Towns <aj@erisian.com.au>
Content-Type: multipart/alternative; boundary="0000000000000b445005c6e1d675"
Cc: Bitcoin Protocol Discussion <bitcoin-dev@lists.linuxfoundation.org>
Subject: Re: [bitcoin-dev] A Stroll through Fee-Bumping Techniques :
 Input-Based vs Child-Pay-For-Parent
Precedence: list

--0000000000000b445005c6e1d675
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

> So the sha256 of the span of the group doesn't commit to start and end
> -- it just serializes a vector, so commits to the number of elements,
> the order, and the elements themselves.

Gotcha wasn't clear to me that the new state pair isn't committed as part
of the annex.

Have been confused by "Introduce a new SIGHASH_GROUP flag, as an
alternative to ALL/SINGLE/NONE, that commits to each output i, start <=3D i=
 <
end."

> Does the above resolve that?

I think so. It shouldn't be susceptible to any spend replay attack, as the
state pair prevents output group overlapping though you might still have to
be careful about siphoning ? Something you should already care about if you
use SIGHASH_SINGLE and your x's amount > y's value.

Le ven. 9 juil. 2021 =C3=A0 21:47, Anthony Towns <aj@erisian.com.au> a =C3=
=A9crit :

> On Fri, Jul 09, 2021 at 09:19:45AM -0400, Antoine Riard via bitcoin-dev
> wrote:
> > > The easy way to avoid O(n^2) behaviour in (3) is to disallow partial
> > > overlaps. So let's treat the tx as being distinct bundles of x-inputs
> > > and y-outputs, and we'll use the annex for grouping, since that is
> > > committed to by singatures. Call the annex field "sig_group_count".
> > > When processing inputs, setup a new state pair, (start, end), initial=
ly
> > > (0,0).
> > > When evaluating an input, lookup sig_group_count. If it's not present=
,
> > > then set start :=3D end. If it's present and 0, leave start and end
> > > unchanged. Otherwise, if it's present and greather than 0, set
> > > start :=3D end, and then set end :=3D start + sig_group_count.
> > IIUC the design rationale, the "sig_group_count" lockdowns the hashing =
of
> > outputs for a given input, thus allowing midstate reuse across signatur=
es
> > input.
>
> No midstates, the message being signed would just replace
> SIGHASH_SINGLE's:
>
>   sha_single_output: the SHA256 of the corresponding output in CTxOut
>   format
>
> with
>
>   sha_group_outputs: the SHA256 of the serialization of the group
>   outputs in CTxOut format.
>
> ie, you'd take span<CTxOut>{start,end}, serialize it (same as if it were
> a vector of just those CTxOuts), and sha256 it.
>
> > Let's say you want to combine {x_1, y_1} and {x_2, y_2} where {x, y}
> denotes
> > bundles of Lightning commitment transactions.
> > x_1 is dual-signed by Alice and Bob under the SIGHASH_GROUP flag with
> > `sig_group_count`=3D3.
> > x_2 is dual-signed by Alice and Caroll under the SIGHASH_GROUP flag, wi=
th
> > `sig_group_count`=3D2.
> > y_1 and y_2 are disjunctive.
> > At broadcast, Alice is not able to combine {x_1,y_1} and {x_2, y_2} for
> the
> > reason that x_1, x_2 are colliding on the absolute output position.
>
> So the sha256 of the span of the group doesn't commit to start and end
> -- it just serializes a vector, so commits to the number of elements,
> the order, and the elements themselves. So you're taking serialize(y_1)
> and serialize(y_2), and each of x_1 signs against the former, and each
> of x_2 signs against the latter.
>
> (Note that the annex for x_1_0 specifies sig_group_count=3Dlen(y_1)
> and the annex for x_1_{1..} specifies sig_group_count=3D0, for "reuse
> previous input's group", and the signatures for each input commit to
> the annex anyway)
>
> > One fix could be to skim the "end > num_ouputs" semantic,
>
> That's only there to ensure the span doesn't go out of range, so I don't
> think it makes any sense to skip it?
>
> > I think this SIGHASH_GROUP proposal might solve other use-cases, but if=
 I
> > understand the semantics correctly, it doesn't seem to achieve the batc=
h
> > fee-bumping of multiple Lightning commitment with O(1) onchain footprin=
t
> I was
> > thinking of for IOMAP...
>
> Does the above resolve that?
>
> Cheers,
> aj
>
>

--0000000000000b445005c6e1d675
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">&gt; So the sha256 of the span of the group doesn&#39;t co=
mmit to start and end<br>&gt; -- it just serializes a vector, so commits to=
 the number of elements,<br>&gt; the order, and the elements themselves.<br=
><br>Gotcha wasn&#39;t clear to me that the new state pair isn&#39;t commit=
ted as part of the annex.<br><br>Have been confused by &quot;Introduce a ne=
w SIGHASH_GROUP flag, as an alternative to ALL/SINGLE/NONE, that commits to=
 each output i, start &lt;=3D i &lt; end.&quot;<br><br>&gt; Does the above =
resolve that?<br><br>I think so. It shouldn&#39;t be susceptible to any spe=
nd replay attack, as the state pair prevents output group overlapping thoug=
h you might still have to be careful about siphoning ? Something you should=
 already care about if you use SIGHASH_SINGLE and your x&#39;s amount &gt; =
y&#39;s value.<br></div><br><div class=3D"gmail_quote"><div dir=3D"ltr" cla=
ss=3D"gmail_attr">Le=C2=A0ven. 9 juil. 2021 =C3=A0=C2=A021:47, Anthony Town=
s &lt;<a href=3D"mailto:aj@erisian.com.au">aj@erisian.com.au</a>&gt; a =C3=
=A9crit=C2=A0:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0=
px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">O=
n Fri, Jul 09, 2021 at 09:19:45AM -0400, Antoine Riard via bitcoin-dev wrot=
e:<br>
&gt; &gt; The easy way to avoid O(n^2) behaviour in (3) is to disallow part=
ial<br>
&gt; &gt; overlaps. So let&#39;s treat the tx as being distinct bundles of =
x-inputs<br>
&gt; &gt; and y-outputs, and we&#39;ll use the annex for grouping, since th=
at is<br>
&gt; &gt; committed to by singatures. Call the annex field &quot;sig_group_=
count&quot;.<br>
&gt; &gt; When processing inputs, setup a new state pair, (start, end), ini=
tially<br>
&gt; &gt; (0,0).<br>
&gt; &gt; When evaluating an input, lookup sig_group_count. If it&#39;s not=
 present,<br>
&gt; &gt; then set start :=3D end. If it&#39;s present and 0, leave start a=
nd end<br>
&gt; &gt; unchanged. Otherwise, if it&#39;s present and greather than 0, se=
t<br>
&gt; &gt; start :=3D end, and then set end :=3D start + sig_group_count.<br=
>
&gt; IIUC the design rationale, the &quot;sig_group_count&quot; lockdowns t=
he hashing of<br>
&gt; outputs for a given input, thus allowing midstate reuse across signatu=
res<br>
&gt; input.<br>
<br>
No midstates, the message being signed would just replace<br>
SIGHASH_SINGLE&#39;s:<br>
<br>
=C2=A0 sha_single_output: the SHA256 of the corresponding output in CTxOut<=
br>
=C2=A0 format<br>
<br>
with<br>
<br>
=C2=A0 sha_group_outputs: the SHA256 of the serialization of the group<br>
=C2=A0 outputs in CTxOut format.<br>
<br>
ie, you&#39;d take span&lt;CTxOut&gt;{start,end}, serialize it (same as if =
it were<br>
a vector of just those CTxOuts), and sha256 it.<br>
<br>
&gt; Let&#39;s say you want to combine {x_1, y_1} and {x_2, y_2} where {x, =
y} denotes<br>
&gt; bundles of Lightning commitment transactions.<br>
&gt; x_1 is dual-signed by Alice and Bob under the SIGHASH_GROUP flag with<=
br>
&gt; `sig_group_count`=3D3.<br>
&gt; x_2 is dual-signed by Alice and Caroll under the SIGHASH_GROUP flag, w=
ith<br>
&gt; `sig_group_count`=3D2.<br>
&gt; y_1 and y_2 are disjunctive.<br>
&gt; At broadcast, Alice is not able to combine {x_1,y_1} and {x_2, y_2} fo=
r the<br>
&gt; reason that x_1, x_2 are colliding on the absolute output position.<br=
>
<br>
So the sha256 of the span of the group doesn&#39;t commit to start and end<=
br>
-- it just serializes a vector, so commits to the number of elements,<br>
the order, and the elements themselves. So you&#39;re taking serialize(y_1)=
<br>
and serialize(y_2), and each of x_1 signs against the former, and each<br>
of x_2 signs against the latter.<br>
<br>
(Note that the annex for x_1_0 specifies sig_group_count=3Dlen(y_1)<br>
and the annex for x_1_{1..} specifies sig_group_count=3D0, for &quot;reuse<=
br>
previous input&#39;s group&quot;, and the signatures for each input commit =
to<br>
the annex anyway)<br>
<br>
&gt; One fix could be to skim the &quot;end &gt; num_ouputs&quot; semantic,=
<br>
<br>
That&#39;s only there to ensure the span doesn&#39;t go out of range, so I =
don&#39;t<br>
think it makes any sense to skip it?<br>
<br>
&gt; I think this SIGHASH_GROUP proposal might solve other use-cases, but i=
f I<br>
&gt; understand the semantics correctly, it doesn&#39;t seem to achieve the=
 batch<br>
&gt; fee-bumping of multiple Lightning commitment with O(1) onchain footpri=
nt I was<br>
&gt; thinking of for IOMAP...<br>
<br>
Does the above resolve that?<br>
<br>
Cheers,<br>
aj<br>
<br>
</blockquote></div>

--0000000000000b445005c6e1d675--