Re: bin/57735: mtree: FreeBSD unsubmitted patches

[Jose Luis Duran <jlduran%gmail.com@localhost>]

The following reply was made to PR bin/57735; it has been noted by GNATS.

From: Jose Luis Duran <jlduran%gmail.com@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: 
Subject: Re: bin/57735: mtree: FreeBSD unsubmitted patches
Date: Sat, 2 Dec 2023 12:16:09 -0300

 --00000000000002492b060b885e2e
 Content-Type: text/plain; charset="UTF-8"
 Content-Transfer-Encoding: quoted-printable
 
 On Sat, Dec 2, 2023 at 10:40=E2=80=AFAM Christos Zoulas <christos%zoulas.co@localhost=
 m> wrote:
 
 >  Thank you! I don't understand why we need to set errno to 0 before
 > calling fts_read()
 >  because it is supposed to do it so I did not. Is it necessary?
 >
 
 Reading the commit message:
 
 > This is addressing cases such as fts_read(3) encountering an [EIO]
 > from fchdir(2) when FTS_NOCHDIR is not set. That would otherwise be
 > seen as a successful traversal in some of these cases while silently
 > discarding expected work.
 
 > As noted in
 > [
 https://github.com/freebsd/freebsd-src/commit/73a0af46b854be94a05fcc4d46036=
 19a00ac5707
 ]
 > fts_read() does not set errno to 0 on a successful
 > EOF so it needs to be set before calling it. Otherwise we might see
 > a random error from one of the iterations.
 
 ...made me take a look at fts.c in NetBSD, and your commit:
 https://github.com/NetBSD/src/commit/b53fea69d008d28365e2dd90e145beaaa53528=
 78
 gave me an idea of what I should do next.
 
 In short, for NetBSD, there should be no need, the commit is good. For
 FreeBSD, there is the corner case described in the commit message above, so
 I need to fix a few things before (or at least try)! In the meantime, I'll
 keep it just for FreeBSD.
 
 Thank you!
 
 --00000000000002492b060b885e2e
 Content-Type: text/html; charset="UTF-8"
 Content-Transfer-Encoding: quoted-printable
 
 <div dir=3D"ltr"><div dir=3D"ltr">On Sat, Dec 2, 2023 at 10:40=E2=80=AFAM C=
 hristos Zoulas &lt;<a href=3D"mailto:christos%zoulas.com@localhost";>christos@zoulas.c=
 om</a>&gt; wrote:</div><div class=3D"gmail_quote"><blockquote class=3D"gmai=
 l_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border-lef=
 t-style:solid;border-left-color:rgb(204,204,204);padding-left:1ex">
 =C2=A0Thank you! I don&#39;t understand why we need to set errno to 0 befor=
 e calling fts_read()<br>
 =C2=A0because it is supposed to do it so I did not. Is it necessary?<br></b=
 lockquote><div><br></div><div>Reading the commit message:<br></div><div><br=
 ></div>&gt; This is addressing cases such as fts_read(3) encountering an [E=
 IO]<br>&gt; from fchdir(2) when FTS_NOCHDIR is not set. That would otherwis=
 e be<br>&gt; seen as a successful traversal in some of these cases while si=
 lently<br>&gt; discarding expected work.<br><br>&gt; As noted in</div><div =
 class=3D"gmail_quote">&gt; [<a href=3D"https://github.com/freebsd/freebsd-s=
 rc/commit/73a0af46b854be94a05fcc4d4603619a00ac5707">https://github.com/free=
 bsd/freebsd-src/commit/73a0af46b854be94a05fcc4d4603619a00ac5707</a>]</div><=
 div class=3D"gmail_quote">&gt; fts_read() does not set errno to 0 on a succ=
 essful<br>&gt; EOF so it needs to be set before calling it. Otherwise we mi=
 ght see<br><div>&gt; a random error from one of the iterations.</div><div><=
 br></div><div>...made me take a look at fts.c in NetBSD, and your commit: <=
 a href=3D"https://github.com/NetBSD/src/commit/b53fea69d008d28365e2dd90e145=
 beaaa5352878">https://github.com/NetBSD/src/commit/b53fea69d008d28365e2dd90=
 e145beaaa5352878</a> gave me an idea of what I should do next.</div><div><b=
 r></div><div>In short, for NetBSD, there should be no need, the commit is g=
 ood. For FreeBSD, there is the corner case described in the commit message =
 above, so I need to fix a few things before (or at least try)! In the meant=
 ime, I&#39;ll keep it just for FreeBSD.</div><div><br></div><div>Thank you!=
 </div></div></div>
 
 --00000000000002492b060b885e2e--