Re: RISC-V Vector patch

John Bowler <[email protected]> Wed, 21 Jun 2023 18:50:20 -0700
Newsgroups gmane.comp.graphics.png.devel
Message-ID <CAP7U399Wrz+B8e5=gSM2B+0i4CzkEGMyNw6eya=f+rWiZLG_Vw@mail.gmail.com>
--===============1565578915404127057==
Content-Type: multipart/alternative; boundary="0000000000000bf03b05feae1cea"

--0000000000000bf03b05feae1cea
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

I did not examine @Drago=C8=99's change more than superficially; I looked
at @mschlaegl's
first because it was the original and seemed to present itself as
semi-official.  I got distracted by other issues (not PNG); it was
definitely on my todo list though all of us know that most of the things on
that list fall off the bottom.

My impression, based on @cosmin's analysis is that both have equal merit.
I do not count libpng delay as a reason for dismissing prior wisdom; as we
all know we only get so far by beating our head against brick walls.

I suggest reaching out to the RISC-V community as a whole; isn't this the
way libpng has always worked?  (Yes, I know the approach has issues.)

--0000000000000bf03b05feae1cea
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>I did not examine=C2=A0<a class=3D"gmail_plusreply" i=
d=3D"plusReplyChip-0">@</a><span style=3D"color:rgb(80,0,80)">Drago=C8=99&#=
39;s change more than superficially; I looked at @</span>mschlaegl&#39;s fi=
rst because it was the original and seemed to present itself as semi-offici=
al.=C2=A0 I got distracted by other issues (not PNG); it was definitely on =
my todo list though all of us know that most of the things on that list fal=
l off the bottom.</div><div><br></div><div>My impression, based on=C2=A0@co=
smin&#39;s analysis is that both have equal merit.=C2=A0 I do not count lib=
png delay as a reason for dismissing prior wisdom; as we all know we only g=
et so far by beating our head against brick walls.</div><div><br></div><div=
>I suggest reaching out to the RISC-V community as a whole; isn&#39;t this =
the way libpng has always worked?=C2=A0 (Yes, I know the approach has issue=
s.)</div></div>

--0000000000000bf03b05feae1cea--


--===============1565578915404127057==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline


--===============1565578915404127057==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
png-mng-implement mailing list
png-mng-implement-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/png-mng-implement

--===============1565578915404127057==--