Re: kern/56355: dtrace triggers double-fault in supervisor mode

[bch <brad.harder%gmail.com@localhost>]

The following reply was made to PR kern/56355; it has been noted by GNATS.

From: bch <brad.harder%gmail.com@localhost>
To: Taylor R Campbell <riastradh%netbsd.org@localhost>
Cc: gnats-bugs%netbsd.org@localhost
Subject: Re: kern/56355: dtrace triggers double-fault in supervisor mode
Date: Sun, 15 Aug 2021 09:05:52 -0700

 --000000000000eecb1f05c99b42f0
 Content-Type: text/plain; charset="UTF-8"
 Content-Transfer-Encoding: quoted-printable
 
 On Sun, Aug 15, 2021 at 08:10 Taylor R Campbell <riastradh%netbsd.org@localhost>
 wrote:
 
 > > Date: Sun, 15 Aug 2021 15:08:11 +0000
 > > From: Taylor R Campbell <riastradh%NetBSD.org@localhost>
 > >
 > > > Date: Tue, 10 Aug 2021 13:26:08 -0700
 > > > From: bch <brad.harder%gmail.com@localhost>
 > > >
 > > > @@ -251,8 +251,9 @@
 > > >         /*
 > > >          * Exclude some more symbols which can be called from probe
 > context.
 > > >          */
 > > > -       if (strcmp(name, "x86_curcpu") =3D=3D 0 ||
 > > > -           strcmp(name, "x86_curlwp") =3D=3D 0) {
 > > > +       if (strncmp(name, "x86_curcpu", 10) =3D=3D 0 ||
 > > > +           strncmp(name, "trap", 4) =3D=3D 0 ||
 > > > +           strncmp(name, "x86_curlwp", 10) =3D=3D 0) {
 > > >                 return 0;
 > >
 > > This patch changes the semantics of the code so that it excludes any
 > > names that _start_ with `trap', `x86_curcpu', or `x86_curlwp', such as
 > > `trapsignal'.  The use of strncmp instead of strcmp should be restored
 > > (and the list should maybe be sorted lexicographically).
 >
 > Errr, I mean: this code should use strcmp, not strncmp.
 
 
 If you read through to the MI exclusion code at fbt_exclude(), it=E2=80=99s=
  using
 strncmp(). Should that be changed too, or otherwise, why pick strcmp() v
 strncmp()?
 
 
 -bch
 
 
 >
 
 --000000000000eecb1f05c99b42f0
 Content-Type: text/html; charset="UTF-8"
 Content-Transfer-Encoding: quoted-printable
 
 <div><br></div><div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=
 =3D"gmail_attr">On Sun, Aug 15, 2021 at 08:10 Taylor R Campbell &lt;<a href=
 =3D"mailto:riastradh%netbsd.org@localhost";>riastradh%netbsd.org@localhost</a>&gt; wrote:<br></d=
 iv><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;bord=
 er-left-width:1px;border-left-style:solid;padding-left:1ex;border-left-colo=
 r:rgb(204,204,204)" dir=3D"auto">&gt; Date: Sun, 15 Aug 2021 15:08:11 +0000=
 <br>
 &gt; From: Taylor R Campbell &lt;riastradh%NetBSD.org@localhost&gt;<br>
 &gt; <br>
 &gt; &gt; Date: Tue, 10 Aug 2021 13:26:08 -0700<br>
 &gt; &gt; From: bch &lt;<a href=3D"mailto:brad.harder%gmail.com@localhost"; target=3D"=
 _blank">brad.harder%gmail.com@localhost</a>&gt;<br>
 &gt; &gt; <br>
 &gt; &gt; @@ -251,8 +251,9 @@<br>
 &gt; &gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0/*<br>
 &gt; &gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 * Exclude some more symbols whi=
 ch can be called from probe context.<br>
 &gt; &gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 */<br>
 &gt; &gt; -=C2=A0 =C2=A0 =C2=A0 =C2=A0if (strcmp(name, &quot;x86_curcpu&quo=
 t;) =3D=3D 0 ||<br>
 &gt; &gt; -=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0strcmp(name, &quot;x86_=
 curlwp&quot;) =3D=3D 0) {<br>
 &gt; &gt; +=C2=A0 =C2=A0 =C2=A0 =C2=A0if (strncmp(name, &quot;x86_curcpu&qu=
 ot;, 10) =3D=3D 0 ||<br>
 &gt; &gt; +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0strncmp(name, &quot;tra=
 p&quot;, 4) =3D=3D 0 ||<br>
 &gt; &gt; +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0strncmp(name, &quot;x86=
 _curlwp&quot;, 10) =3D=3D 0) {<br>
 &gt; &gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0retu=
 rn 0;<br>
 &gt; <br>
 &gt; This patch changes the semantics of the code so that it excludes any<b=
 r>
 &gt; names that _start_ with `trap&#39;, `x86_curcpu&#39;, or `x86_curlwp&#=
 39;, such as<br>
 &gt; `trapsignal&#39;.=C2=A0 The use of strncmp instead of strcmp should be=
  restored<br>
 &gt; (and the list should maybe be sorted lexicographically).<br>
 <br>
 Errr, I mean: this code should use strcmp, not strncmp.=C2=A0</blockquote><=
 div dir=3D"auto"><br></div><div dir=3D"auto">If you read through to the MI =
 exclusion code at fbt_exclude(), it=E2=80=99s using strncmp(). Should that =
 be changed too, or otherwise, why pick strcmp() v strncmp()?</div><div dir=
 =3D"auto"><br></div><div dir=3D"auto"><br></div><div dir=3D"auto">-bch</div=
 ><div dir=3D"auto"><br></div><blockquote class=3D"gmail_quote" style=3D"mar=
 gin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;padding=
 -left:1ex;border-left-color:rgb(204,204,204)" dir=3D"auto"><br>
 </blockquote></div></div>
 
 --000000000000eecb1f05c99b42f0--