http://www.riscos.info/bugzilla3/show_bug.cgi?id=243
--- Comment #4 from John Tytgat <John.Tytgat@aaug.net> 2012-09-02 06:54:34 PDT ---
Alternatively, I've made a test build of gccsdk 4.1.2 Release 2 available at
http://www.riscos.info/downloads/gccsdk/testing/4.1.2/
--
Configure bugmail: http://www.riscos.info/bugzilla3/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
Sunday, 2 September 2012
[gccsdk] [Bug 243] Segfault compiling innocent looking for loop in 4.1.2 r1
http://www.riscos.info/bugzilla3/show_bug.cgi?id=243
--- Comment #3 from Lee <leenoar@sky.com> 2012-09-02 06:51:22 PDT ---
(In reply to comment #2)
> (In reply to comment #1)
> > This is caused by tail call optimisations clashing with the stack extension
> > routines in UnixLib. It should be fixed with r5990 and has already been fixed
> > at trunk.
>
> Is there somewhere to get a binary with r5990 in from?
> Alternatively do any of the earlier releases generate ARMv7 compatible
> loads/stores but simultaneously not have this bug?
I've placed a copy of the cc1 binary for you to try here:
<http://leenoar.yolasite.com/resources/gcc/cc1.zip>
This seems to be the main culprit and IIRC, the only binary in GCCSDK that I
have seen fail due to this (obviously cc1plus is a possibility too). It's
possible that TCOs in libunixlib may trigger the problem, but I think it's
unlikely.
It would be useful to know that this does in fact fix your problem.
Its location is !GCC/libexec/gcc/arm-unknown-riscos/4.1.2/cc1,e1f
As for previous versions, well I was hoping that this problem didn't go back
further that 4.6, but unfortunately, it looks like it may have been present
from the beginning. Perhaps the increasing complexity of GCC has made the rare
runtime conditions under which the problem occurs more likely.
--
Configure bugmail: http://www.riscos.info/bugzilla3/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
--- Comment #3 from Lee <leenoar@sky.com> 2012-09-02 06:51:22 PDT ---
(In reply to comment #2)
> (In reply to comment #1)
> > This is caused by tail call optimisations clashing with the stack extension
> > routines in UnixLib. It should be fixed with r5990 and has already been fixed
> > at trunk.
>
> Is there somewhere to get a binary with r5990 in from?
> Alternatively do any of the earlier releases generate ARMv7 compatible
> loads/stores but simultaneously not have this bug?
I've placed a copy of the cc1 binary for you to try here:
<http://leenoar.yolasite.com/resources/gcc/cc1.zip>
This seems to be the main culprit and IIRC, the only binary in GCCSDK that I
have seen fail due to this (obviously cc1plus is a possibility too). It's
possible that TCOs in libunixlib may trigger the problem, but I think it's
unlikely.
It would be useful to know that this does in fact fix your problem.
Its location is !GCC/libexec/gcc/arm-unknown-riscos/4.1.2/cc1,e1f
As for previous versions, well I was hoping that this problem didn't go back
further that 4.6, but unfortunately, it looks like it may have been present
from the beginning. Perhaps the increasing complexity of GCC has made the rare
runtime conditions under which the problem occurs more likely.
--
Configure bugmail: http://www.riscos.info/bugzilla3/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
Re: [gccsdk] cross compiler building errors (under Linux)
In message <DUB002-W3374BEE9DB6DD4EE650E45BCA50@phx.gbl>
Chris Syntichakis <csynt@hotmail.com> wrote:
> Hello,
>
>
> > I see that you do have flex and bison installed, but if this was done as
> > a result of a failed build, then you may need to reconfigure GCC as John
> > says above.
> >
> > Lee.
> >
>
>
> Thank you Lee,
>
> I followed John's advice but I got the same error,
That advice was for the GCCSDK 4.1.2 build, with GCCSDK 4.7.x currently
at trunk you need to remove buildstepsdir/src-gcc-copied (as this will
trigger the gcc recofinguring step and rebuilds the gcc part from scratch
while reusing whatever was built as binutils, gmp, mpc, mpfr and ppl).
> then I issued a "make clean all" and started rebuild...
That's the alternative, a full rebuild.
> after a while I got a "fatal error: archive.h: No such file or directory"
> so I completely deleted the gccsdk folder , got the files again (svn .....) and started
> the build-world script, but this, too, stops on the same error:
>
> /home/gccsdk/gcc4/riscos/asasm/elftoolchain/current-trunk/ar/ar.c:65:21: fatal error: archive.h: No such file or directory
>
> so I assume something is broken there ...
The good news is that the cross compiler is now fully built and installed.
This is the asasm assembler part failing (asasm is not needed if you're
going to use gcc+binutils only) which I'm changing to use
elftoolchain/libdwarf. So a fallout of my last asasm changes. Note trunk
is still in development but of course we do our best to keep everything
buildable and usable.
This build error is because of a missing host library, called libarchive
(Ubuntu, Debian, etc this is the libarchive-dev package). So install the
libarchive library + headers and re-issue 'make'.
BTW, libarchive is only needed for testing asasm and I will do some changes
to avoid this libarchive dependency for those just building and using
gccsdk.
Hope this helps,
John.
--
John Tytgat, in his comfy chair at home BASS
John.Tytgat@aaug.net ARM powered, RISC OS driven
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
Chris Syntichakis <csynt@hotmail.com> wrote:
> Hello,
>
>
> > I see that you do have flex and bison installed, but if this was done as
> > a result of a failed build, then you may need to reconfigure GCC as John
> > says above.
> >
> > Lee.
> >
>
>
> Thank you Lee,
>
> I followed John's advice but I got the same error,
That advice was for the GCCSDK 4.1.2 build, with GCCSDK 4.7.x currently
at trunk you need to remove buildstepsdir/src-gcc-copied (as this will
trigger the gcc recofinguring step and rebuilds the gcc part from scratch
while reusing whatever was built as binutils, gmp, mpc, mpfr and ppl).
> then I issued a "make clean all" and started rebuild...
That's the alternative, a full rebuild.
> after a while I got a "fatal error: archive.h: No such file or directory"
> so I completely deleted the gccsdk folder , got the files again (svn .....) and started
> the build-world script, but this, too, stops on the same error:
>
> /home/gccsdk/gcc4/riscos/asasm/elftoolchain/current-trunk/ar/ar.c:65:21: fatal error: archive.h: No such file or directory
>
> so I assume something is broken there ...
The good news is that the cross compiler is now fully built and installed.
This is the asasm assembler part failing (asasm is not needed if you're
going to use gcc+binutils only) which I'm changing to use
elftoolchain/libdwarf. So a fallout of my last asasm changes. Note trunk
is still in development but of course we do our best to keep everything
buildable and usable.
This build error is because of a missing host library, called libarchive
(Ubuntu, Debian, etc this is the libarchive-dev package). So install the
libarchive library + headers and re-issue 'make'.
BTW, libarchive is only needed for testing asasm and I will do some changes
to avoid this libarchive dependency for those just building and using
gccsdk.
Hope this helps,
John.
--
John Tytgat, in his comfy chair at home BASS
John.Tytgat@aaug.net ARM powered, RISC OS driven
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
[gccsdk] [Bug 243] Segfault compiling innocent looking for loop in 4.1.2 r1
http://www.riscos.info/bugzilla3/show_bug.cgi?id=243
--- Comment #2 from info@sprow.co.uk 2012-09-02 05:12:40 PDT ---
(In reply to comment #1)
> This is caused by tail call optimisations clashing with the stack extension
> routines in UnixLib. It should be fixed with r5990 and has already been fixed
> at trunk.
Is there somewhere to get a binary with r5990 in from?
Alternatively do any of the earlier releases generate ARMv7 compatible
loads/stores but simultaneously not have this bug?
--
Configure bugmail: http://www.riscos.info/bugzilla3/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
--- Comment #2 from info@sprow.co.uk 2012-09-02 05:12:40 PDT ---
(In reply to comment #1)
> This is caused by tail call optimisations clashing with the stack extension
> routines in UnixLib. It should be fixed with r5990 and has already been fixed
> at trunk.
Is there somewhere to get a binary with r5990 in from?
Alternatively do any of the earlier releases generate ARMv7 compatible
loads/stores but simultaneously not have this bug?
--
Configure bugmail: http://www.riscos.info/bugzilla3/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
[gccsdk] [Bug 243] Segfault compiling innocent looking for loop in 4.1.2 r1
http://www.riscos.info/bugzilla3/show_bug.cgi?id=243
Lee <leenoar@sky.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
CC| |leenoar@sky.com
Resolution| |FIXED
--- Comment #1 from Lee <leenoar@sky.com> 2012-09-02 04:23:39 PDT ---
This is caused by tail call optimisations clashing with the stack extension
routines in UnixLib. It should be fixed with r5990 and has already been fixed
at trunk. Unfortunately, there is no work around that will allow the existing
release to work. The only solution is to rebuild GCCSDK to remove all the TCOs.
--
Configure bugmail: http://www.riscos.info/bugzilla3/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
Lee <leenoar@sky.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
CC| |leenoar@sky.com
Resolution| |FIXED
--- Comment #1 from Lee <leenoar@sky.com> 2012-09-02 04:23:39 PDT ---
This is caused by tail call optimisations clashing with the stack extension
routines in UnixLib. It should be fixed with r5990 and has already been fixed
at trunk. Unfortunately, there is no work around that will allow the existing
release to work. The only solution is to rebuild GCCSDK to remove all the TCOs.
--
Configure bugmail: http://www.riscos.info/bugzilla3/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
Saturday, 1 September 2012
Re: [-SPAM-] Re: Zap and the Raspberry Pi
On 28 Aug, James Aylett <james@tartarus.org> wrote:
> On 28 Aug 2012, at 17:01, Steve (ROOL) <srevill@riscosopen.org> wrote:
> > We at ROOL [1] are preparing disc images for the Raspberry Pi computer
> > [2], containing RISC OS and a selection of the best free software
> > available for the platform. These will be distributed electronically by
> > the Raspberry Pi Foundation (a registered charity) as well as via
> > file-sharing.
> >
> > We would like to include Zap (et al) in these disc images, and would
> > appreciate permission to do so. Please accept my apologies if you have
> > already been contacted on this matter by others, but any permission you
> > may have given has not been passed on to us.
>
> I think we're generally happy with this kind of thing. Darren / Christian,
> any objections?
OK, thanks. We'll hold off until everyone is happy.
> (Has stock Zap been confirmed to run on Raspberry Pi? I know we've had
> various patches and re-distros floating around for some other RO targets
> over the last few years.)
Well, I've been using it on the RPi here but it's the one Chris Hall put
into a prototype distro and I've no idea what its provenance is. However,
the RPi hardware shouldn't look a lot different to Iyonix from Zap's PoV so
I'd expect the latest Zap to be OK.
Could you point me at what you guys would consider to be the 'current and
correct' version?
Thanks,
Steve
--
Stephen Revill, Director www.riscosopen.org
RISC OS Open Ltd
Sovereign House, 22 Shelley Road,
Worthing, West Sussex, BN11 1TU
--
_______________________________________________
zap-technical maillist - zap-technical@zap.tartarus.org
http://lists.tartarus.org/mailman/listinfo/zap-technical
> On 28 Aug 2012, at 17:01, Steve (ROOL) <srevill@riscosopen.org> wrote:
> > We at ROOL [1] are preparing disc images for the Raspberry Pi computer
> > [2], containing RISC OS and a selection of the best free software
> > available for the platform. These will be distributed electronically by
> > the Raspberry Pi Foundation (a registered charity) as well as via
> > file-sharing.
> >
> > We would like to include Zap (et al) in these disc images, and would
> > appreciate permission to do so. Please accept my apologies if you have
> > already been contacted on this matter by others, but any permission you
> > may have given has not been passed on to us.
>
> I think we're generally happy with this kind of thing. Darren / Christian,
> any objections?
OK, thanks. We'll hold off until everyone is happy.
> (Has stock Zap been confirmed to run on Raspberry Pi? I know we've had
> various patches and re-distros floating around for some other RO targets
> over the last few years.)
Well, I've been using it on the RPi here but it's the one Chris Hall put
into a prototype distro and I've no idea what its provenance is. However,
the RPi hardware shouldn't look a lot different to Iyonix from Zap's PoV so
I'd expect the latest Zap to be OK.
Could you point me at what you guys would consider to be the 'current and
correct' version?
Thanks,
Steve
--
Stephen Revill, Director www.riscosopen.org
RISC OS Open Ltd
Sovereign House, 22 Shelley Road,
Worthing, West Sussex, BN11 1TU
--
_______________________________________________
zap-technical maillist - zap-technical@zap.tartarus.org
http://lists.tartarus.org/mailman/listinfo/zap-technical
Re: [gccsdk] cross compiler building errors (under Linux)
Hello,
> I see that you do have flex and bison installed, but if this was done as
> a result of a failed build, then you may need to reconfigure GCC as John
> says above.
>
> Lee.
>
Thank you Lee,
I followed John's advice but I got the same error, then I issued
a "make clean all" and started rebuild...
after a while I got a "fatal error: archive.h: No such file or directory"
so I completely deleted the gccsdk folder , got the files again (svn .....) and started
the build-world script, but this, too, stops on the same error:
/home/gccsdk/gcc4/riscos/asasm/elftoolchain/current-trunk/ar/ar.c:65:21: fatal error: archive.h: No such file or directory
so I assume something is broken there ...
Thanks anyway..
> I see that you do have flex and bison installed, but if this was done as
> a result of a failed build, then you may need to reconfigure GCC as John
> says above.
>
> Lee.
>
Thank you Lee,
I followed John's advice but I got the same error, then I issued
a "make clean all" and started rebuild...
after a while I got a "fatal error: archive.h: No such file or directory"
so I completely deleted the gccsdk folder , got the files again (svn .....) and started
the build-world script, but this, too, stops on the same error:
/home/gccsdk/gcc4/riscos/asasm/elftoolchain/current-trunk/ar/ar.c:65:21: fatal error: archive.h: No such file or directory
so I assume something is broken there ...
Thanks anyway..
Subscribe to:
Posts (Atom)