Re: Fix libjava bootstrap failure around IUCLC

[Mark Wielaard]

On Mon, 2012-03-26 at 10:01 +0100, Andrew Haley wrote:
> On 03/25/2012 05:08 PM, Gerald Pfeifer wrote:
> > 2012-03-25  Gerald Pfeifer  <gerald@...>
> >
> >         PR libgcj/52694
> >         * java/io/natVMConsole.cc (IUCLC): Define, if undefined.
>
> Sure.  WTF is IUCLC anyway?  :-)

(map) Input (characters) UpperCase (to) Lower Case.
It is a non-posix termios extension.

patch should also go into upstream classpath.

This has been introduced by

  2012-03-16  Andrew John Hughes  <ahughes@...>

        * Makefile.am: Add natVMConsole.cc.
        :
        * java/io/VMConsole.h: Initial generation.
        * java/io/VMConsole.java: Copied from classpath/vm/reference.
        * java/io/natVMConsole.cc: Implemented readPassword.

where now bootstrap on FreeBSD, Darwin and possibly others is broken.

Tested on amd64-unknown-freebsd8.3.  Okay?

Gerald


2012-03-25  Gerald Pfeifer  <gerald@...>

        PR libgcj/52694
        * java/io/natVMConsole.cc (IUCLC): Define, if undefined.

Index: java/io/natVMConsole.cc
===================================================================
--- java/io/natVMConsole.cc (revision 185766)
+++ java/io/natVMConsole.cc (working copy)
@@ -19,6 +19,10 @@
 #include <java/io/Console.h>
 #include <java/io/VMConsole.h>
 
+#ifndef IUCLC
+#define IUCLC 0
+#endif
+
 #define TERMIOS_ECHO_IFLAGS (IUCLC|IXON|IXOFF|IXANY)
 #define TERMIOS_ECHO_LFLAGS (ECHO|ECHOE|ECHOK|ECHONL|TOSTOP)
 

[Pekka Enberg-4]

On Mon, Mar 26, 2012 at 1:11 PM, Mark Wielaard <mark@...> wrote:

> On Mon, 2012-03-26 at 10:01 +0100, Andrew Haley wrote:
>> On 03/25/2012 05:08 PM, Gerald Pfeifer wrote:
>> > 2012-03-25  Gerald Pfeifer  <gerald@...>
>> >
>> >         PR libgcj/52694
>> >         * java/io/natVMConsole.cc (IUCLC): Define, if undefined.
>>
>> Sure.  WTF is IUCLC anyway?  :-)
>
> (map) Input (characters) UpperCase (to) Lower Case.
> It is a non-posix termios extension.
>
> patch should also go into upstream classpath.

Yes, please.

Btw, why are people not fixing GNU Classpath upstream-first?

[Andrew Haley]

On 03/26/2012 11:47 AM, Pekka Enberg wrote:

> On Mon, Mar 26, 2012 at 1:11 PM, Mark Wielaard <mark@...> wrote:
>> On Mon, 2012-03-26 at 10:01 +0100, Andrew Haley wrote:
>>> On 03/25/2012 05:08 PM, Gerald Pfeifer wrote:
>>>> 2012-03-25  Gerald Pfeifer  <gerald@...>
>>>>
>>>>         PR libgcj/52694
>>>>         * java/io/natVMConsole.cc (IUCLC): Define, if undefined.
>>>
>>> Sure.  WTF is IUCLC anyway?  :-)
>>
>> (map) Input (characters) UpperCase (to) Lower Case.
>> It is a non-posix termios extension.
>>
>> patch should also go into upstream classpath.
>
> Yes, please.
>
> Btw, why are people not fixing GNU Classpath upstream-first?

Because, I guess, no-one is using it. At least, if they were, this
bug would have been noticed.

Andrew.

[Pekka Enberg-4]

On Mon, Mar 26, 2012 at 1:50 PM, Andrew Haley <aph@...> wrote:
>> Btw, why are people not fixing GNU Classpath upstream-first?
>
> Because, I guess, no-one is using it. At least, if they were, this
> bug would have been noticed.

Right. I'm developing and testing with Fedora and Ubuntu on x86 and I
verify GNU Classpath bootstrap on Darwin at times. I guess Buildbot
would help here.

                        Pekka

[Mark Wielaard]

On Mon, 2012-03-26 at 13:47 +0300, Pekka Enberg wrote:

> On Mon, Mar 26, 2012 at 1:11 PM, Mark Wielaard <mark@...> wrote:
> > On Mon, 2012-03-26 at 10:01 +0100, Andrew Haley wrote:
> >> On 03/25/2012 05:08 PM, Gerald Pfeifer wrote:
> >> > 2012-03-25  Gerald Pfeifer  <gerald@...>
> >> >
> >> >         PR libgcj/52694
> >> >         * java/io/natVMConsole.cc (IUCLC): Define, if undefined.
> >>
> >> Sure.  WTF is IUCLC anyway?  :-)
> >
> > (map) Input (characters) UpperCase (to) Lower Case.
> > It is a non-posix termios extension.
> >
> > patch should also go into upstream classpath.
>
> Yes, please.
>
> Btw, why are people not fixing GNU Classpath upstream-first?

In this case, because libgcj doesn't use the JNI interface for some of
its native implementations. So this was found in the CNI .cc variant.
The patch should be similar though for the jni .c variant.

Cheers,

Mark

[Mark Wielaard]

On Mon, 2012-03-26 at 13:53 +0300, Pekka Enberg wrote:
> On Mon, Mar 26, 2012 at 1:50 PM, Andrew Haley <aph@...> wrote:
> >> Btw, why are people not fixing GNU Classpath upstream-first?
> >
> > Because, I guess, no-one is using it. At least, if they were, this
> > bug would have been noticed.
>
> Right. I'm developing and testing with Fedora and Ubuntu on x86 and I
> verify GNU Classpath bootstrap on Darwin at times. I guess Buildbot
> would help here.

It is there on GNU/Linux, OpenBSD and Solaris at least, but there might
be termios out there (apparently freebsd at least) that don't have
IUCLC.

Buildbot is on the TODO list :) Of course even with that we need slaves
for some of the more obscure platforms.

Cheers,

Mark