Re: [perl #132390] multiconcat breaks blead on VMS

--Apple-Webmail-42--be8afb8f-08bd-4c30-80af-4bf3bb923f52
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
    charset=utf-8;
    format=flowed

=0A=0A=0AOn Nov 13, 2017, at 06:37 AM, Dave Mitchell <davem@iabyn.com> wro=
te:=0A=0A=0AOn Fri, Nov 10, 2017 at 08:59:56AM -0600, Craig A. Berry wrote=
:=0A=0AI've now finally gotten back to this and the fix is easy:=0A=0A=0A-=
-- pp_hot.c;-0 2017-11-04 06:24:52 -0500=0A+++ pp_hot.c 2017-11-09 20:41:4=
1 -0600=0A@@ -391,7 +391,7 @@ PP(pp_multiconcat)=0AUNOP_AUX_item *aux; /* =
PL_op->op_aux buffer */=0AUNOP_AUX_item *const_lens; /* the segment length=
 array part of aux */=0Aconst char *const_pv; /* the current segment of th=
e const string buf */=0A- UV nargs; /* how many args were expected */=0A+ =
IV nargs; /* how many args were expected; IV=0A=0AI've now pushed a more g=
eneral fix:=0A=0Acommit ca84e88ece180337b1ea0b8a2b9d4211b1089878=0AAuthor:=
 David Mitchell <davem@iabyn.com>=0AAuthorDate: Mon Nov 13 11:50:14 2017 +=
0000=0ACommit: David Mitchell <davem@iabyn.com>=0ACommitDate: Mon Nov 13 1=
2:23:24 2017 +0000=0A=0Achange OP_MULTICONCAT nargs from UV to SSize_t=0A=0A=
Change it from unsigned to unsigned since it makes the SP-adjusting code=0A=
in pp_multiconcat easier without hitting undefined behaviour (RT #132390);=
=0Aand change its size from UV to SSize_t since it represents the number=0A=
of args on the stack.=0A=0A=0AThanks.=C2=A0 That got things building again=
 and I think this ticket can be closed.=0A
--Apple-Webmail-42--be8afb8f-08bd-4c30-80af-4bf3bb923f52
Content-Type: multipart/related;
    type="text/html";
    boundary=Apple-Webmail-86--be8afb8f-08bd-4c30-80af-4bf3bb923f52


--Apple-Webmail-86--be8afb8f-08bd-4c30-80af-4bf3bb923f52
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
    charset=utf-8;

<html><body><div><br></div><div><br>On Nov 13, 2017, at 06:37 AM, Dave Mitchell &lt;da=
vem@iabyn.com&gt; wrote:<br><br><div><blockquote type=3D"cite"><div class=3D=
"msg-quote"><div class=3D"_stretch"><span class=3D"body-text-content">On F=
ri, Nov 10, 2017 at 08:59:56AM -0600, Craig A. Berry wrote:<br><blockquote=
 type=3D"cite" class=3D"quoted-plain-text">I've now finally gotten back to=
 this and the fix is easy:</blockquote><blockquote type=3D"cite" class=3D"=
quoted-plain-text"><br></blockquote><blockquote type=3D"cite" class=3D"quo=
ted-plain-text">--- pp_hot.c;-0 2017-11-04 06:24:52 -0500</blockquote><blo=
ckquote type=3D"cite" class=3D"quoted-plain-text">+++ pp_hot.c 2017-11-09 =
20:41:41 -0600</blockquote><blockquote type=3D"cite" class=3D"quoted-plain=
-text">@@ -391,7 +391,7 @@ PP(pp_multiconcat)</blockquote><blockquote type=
=3D"cite" class=3D"quoted-plain-text">UNOP_AUX_item *aux; /* PL_op-&gt;op_=
aux buffer */</blockquote><blockquote type=3D"cite" class=3D"quoted-plain-=
text">UNOP_AUX_item *const_lens; /* the segment length array part of aux *=
/</blockquote><blockquote type=3D"cite" class=3D"quoted-plain-text">const =
char *const_pv; /* the current segment of the const string buf */</blockqu=
ote><blockquote type=3D"cite" class=3D"quoted-plain-text">- UV nargs; /* h=
ow many args were expected */</blockquote><blockquote type=3D"cite" class=3D=
"quoted-plain-text">+ IV nargs; /* how many args were expected; IV</blockq=
uote><br>I've now pushed a more general fix:<br><br>commit ca84e88ece18033=
7b1ea0b8a2b9d4211b1089878<br>Author: David Mitchell &lt;davem@iabyn.com&gt=
;<br>AuthorDate: Mon Nov 13 11:50:14 2017 +0000<br>Commit: David Mitchell =
&lt;davem@iabyn.com&gt;<br>CommitDate: Mon Nov 13 12:23:24 2017 +0000<br><=
br> change OP_MULTICONCAT nargs from UV to SSize_t<br> <br> Change it from=
 unsigned to unsigned since it makes the SP-adjusting code<br> in pp_multi=
concat easier without hitting undefined behaviour (RT #132390);<br> and ch=
ange its size from UV to SSize_t since it represents the number<br> of arg=
s on the stack.</span></div></div></blockquote></div><div><span><br data-m=
ce-bogus=3D"1"></span></div><div><span>Thanks.&nbsp; That got things build=
ing again and I think this ticket can be closed. <br></span></div></div></body></html>
--Apple-Webmail-86--be8afb8f-08bd-4c30-80af-4bf3bb923f52--

--Apple-Webmail-42--be8afb8f-08bd-4c30-80af-4bf3bb923f52--
0
craigberry
11/13/2017 6:32:51 PM
perl.perl5.porters 46715 articles. 0 followers. Follow

0 Replies
1 Views

Similar Articles

[PageSpeed] 17

Reply: