[polyml] 5.7 Release

Discussion:

David Matthews

2017-05-12 14:50:05 UTC

I've finally got round to releasing version 5.7. The source code was
actually tagged as 5.7 on Github a while ago and Github considered that
a release. There are Windows installers for 32-bit and 64-bit as well
as the normal source code for everything else.

I'm aware that there are still some unresolved issues with this version
but it seems to be stable enough for release.

David

Ramana Kumar

2017-05-16 03:33:53 UTC

Permalink

Hi David,

Is there going to be a fixes-5.7 branch as with previous releases?

Cheers,
Ramana

Post by David Matthews
I've finally got round to releasing version 5.7. The source code was
actually tagged as 5.7 on Github a while ago and Github considered that a
release. There are Windows installers for 32-bit and 64-bit as well as the
normal source code for everything else.
I'm aware that there are still some unresolved issues with this version
but it seems to be stable enough for release.
David
_______________________________________________
polyml mailing list
http://lists.inf.ed.ac.uk/mailman/listinfo/polyml

David Matthews

2017-05-16 14:43:00 UTC

Permalink

Hi Ramana,
I was going to do that at some stage, perhaps when there was something
to include, but I've done it now.
Regards,
David

Post by Ramana Kumar
Hi David,
Is there going to be a fixes-5.7 branch as with previous releases?
Cheers,
Ramana

Jerry James

2017-05-16 03:47:14 UTC

Permalink

On Fri, May 12, 2017 at 8:50 AM, David Matthews

I attempted to build this version for Fedora. It succeeded on all
architectures except aarch64:

https://koji.fedoraproject.org/koji/taskinfo?taskID=19526061

I don't understand the failure, though. We run "make check" after the
build to gain some confidence that the build is good. Here is an
extract from the build log
(https://kojipkgs.fedoraproject.org//work/tasks/6133/19526133/build.log):

echo "val () = use \"./Tests/RunTests\"; val () = OS.Process.exit(if
runTests \"./Tests\" then OS.Process.success else
OS.Process.failure):unit;" | ./poly
Poly/ML 5.7 Release
val runTests = fn: string -> bool
Test137.ML => Passed
Test101.ML => Passed
Test032.ML => Passed
Test056.ML => Passed
Test080.ML => Passed
Test136.ML => Passed
Test127.ML => Passed
Test065.ML => Passed
Test100.ML => Passed
Test145.ML => Passed
Test025.ML => Passed
Test003.ML => Passed
Test144.ML => Passed
Test109.ML => Passed
Test128.ML => Passed
Test019.ML => Passed
Test015.ML => Passed
Test009.ML => Passed
Test037.ML => Passed
Test113.ML => Passed
Test162.ML => Passed
Test072.ML => Passed
Test061.ML => Passed
Test118.ML => Passed
Test051.ML => Passed
Test103.ML => Passed
Test083.ML => Passed
Test134.ML => Passed
Test159.ML => Passed
Test126.ML => Passed
Test077.ML => Passed
Test020.ML => Passed
Test026.ML => Passed
Test131.ML => Passed
Test021.ML => Passed
Test013.ML => Passed
Test052.ML => Passed
Test169.ML => Passed
Test115.ML => Passed
Test170.ML => Passed
Test116.ML => Passed
Test148.ML => Passed
Test151.ML => Passed
Test117.ML => Passed
Test066.ML => Passed
Test147.ML => Passed
Test097.ML => Passed
Test007.ML => Passed
Test016.ML => Passed
Test139.ML => Passed
Test120.ML => Passed
Test168.ML => Passed
Test047.ML => Passed
Test158.ML => Passed
Test157.ML => Passed
Test091.ML => Passed
Test140.ML => Passed
Test022.ML => Passed
Test001.ML => Passed
Test138.ML => Passed
Test040.ML => Passed
Test028.ML => Passed
Test011.ML => Passed
Test106.ML => Passed
Test074.ML => Passed
Test010.ML => Passed
Test081.ML => Passed
Test038.ML => Passed
Test031.ML => Passed
Test014.ML => Passed
Test164.ML => Passed
Test057.ML => Passed
Test048.ML => Passed
Test107.ML => Passed
Test006.ML => Passed
Test063.ML => Passed
Test142.ML => Passed
Test073.ML => Passed
Test033.ML => Passed
Test133.ML => Done
Passed
Test036.ML => Passed
Test018.ML => Passed
Test035.ML => Passed
Test129.ML => Passed
Test024.ML => Passed
Test004.ML => Passed
Test090.ML => Passed
Test112.ML => Passed
Test084.ML => Passed
Test123.ML => Passed
Test155.ML => Passed
Test076.ML => Passed
Test165.ML => Passed
Test104.ML => Passed
Test046.ML => Passed
Test030.ML => Passed
Test089.ML => Passed
Test023.ML => Passed
Test088.ML => Passed
Test064.ML => Passed
Test163.ML => Passed
Test122.ML => Passed
Test111.ML => Passed
Test045.ML => Passed
Test161.ML => Passed
Test095.ML => Passed
Test152.ML => Passed
Test049.ML => Passed
Test166.ML => Passed
Test062.ML => Passed
Test094.ML => Passed
Test012.ML => Passed
Test079.ML => Passed
Test087.ML => Passed
Test058.ML => Passed
Test059.ML => Passed
Test135.ML => Passed
Test132.ML => Passed
Test039.ML => Passed
Test002.ML => Passed
Test160.ML => Passed
Test119.ML => Passed
Test093.ML => Passed
Test124.ML => Passed
Test167.ML => Passed
Test105.ML => Passed
Test146.ML => Passed
Test078.ML => Passed
Test102.ML => Passed
Test092.ML => Passed
Test098.ML => Passed
Test068.ML => Passed
Test005.ML => Passed
Test125.ML => Passed
Test041.ML => Passed
Test108.ML => Passed
Test050.ML => Passed
Test034.ML => Passed
Test149.ML => Passed
Test027.ML => Passed
Test143.ML => Passed
Test055.ML => Passed
Test150.ML => Passed
Test008.ML => Passed
Test096.ML => Passed
Test121.ML => Passed
Test070.ML => Passed
Test082.ML => Passed
Test029.ML => Passed
Test154.ML => Passed
Test086.ML => Passed
Test130.ML => Passed
Test042.ML => Passed
Test075.ML => Passed
Test141.ML => Passed
Test156.ML => Passed
Test017.ML => Passed
Test067.ML => Passed
Test071.ML => Passed
Test114.ML => Passed
Test053.ML => Passed
Test110.ML => Passed
Test054.ML => Passed
Test069.ML => Passed
Test085.ML => Passed
Test060.ML => Passed
Test153.ML => Passed
Test043.ML => Passed
Test044.ML => Passed
Test099.ML => Passed
Test032.ML => Passed
Test056.ML => Passed
Test065.ML => Passed
Test025.ML => Passed
Test003.ML => Passed
Test019.ML => Passed
Test015.ML => Passed
Test009.ML => Passed
Test037.ML => Passed
Test072.ML => Passed
Test061.ML => Passed
Test051.ML => Passed
Test020.ML => Passed
Test026.ML => Passed
Test021.ML => Passed
Test013.ML => Passed
Test052.ML => Passed
Test066.ML => Passed
Test007.ML => Passed
Test016.ML => Passed
Test047.ML => Passed
Test022.ML => Passed
Test001.ML => Passed
Test040.ML => Passed
Test028.ML => Passed
Test011.ML => Passed
Test074.ML => Passed
Test010.ML => Passed
Test038.ML => Passed
Test031.ML => Passed
Test014.ML => Passed
Test057.ML => Passed
Test048.ML => Passed
Test006.ML => Passed
Test063.ML => Passed
Test073.ML => Passed
Test033.ML => Passed
Test036.ML => Passed
Test018.ML => Passed
Test035.ML => Passed
Test024.ML => Passed
Test004.ML => Passed
Test046.ML => Passed
Test030.ML => Passed
Test023.ML => Passed
Test064.ML => Passed
Test045.ML => Passed
Test062.ML => Passed
Test012.ML => Passed
Test058.ML => Passed
Test059.ML => Passed
Test039.ML => Passed
Test002.ML => Passed
Test068.ML => Passed
Test005.ML => Passed
Test041.ML => Passed
Test050.ML => Passed
Test034.ML => Passed
Test027.ML => Passed
Test055.ML => Passed
Test008.ML => Passed
Test070.ML => Passed
Test029.ML => Passed
Test042.ML => Passed
Test067.ML => Passed
Test071.ML => Passed
Test053.ML => Passed
Test054.ML => Passed
Test069.ML => Passed
Test060.ML => Passed
Test043.ML => Passed
Test044.ML => Passed
make[2]: Leaving directory '/builddir/build/BUILD/polyml-5.7'
make[2]: *** [Makefile:1175: check-local] Error 1
make[1]: Leaving directory '/builddir/build/BUILD/polyml-5.7'
make[1]: *** [Makefile:997: check-am] Error 2
make: *** [Makefile:706: check-recursive] Error 1

It looks like every individual test passed, but the test suite as a
whole failed. Do you have any idea what might cause this? Thank you,

--
Jerry James
http://www.jamezone.org/

David Matthews

2017-05-16 14:49:40 UTC

Permalink

Post by Jerry James
I attempted to build this version for Fedora. It succeeded on all
https://koji.fedoraproject.org/koji/taskinfo?taskID=19526061
I don't understand the failure, though. We run "make check" after the
build to gain some confidence that the build is good. Here is an
extract from the build log
Test044.ML => Passed
make[2]: Leaving directory '/builddir/build/BUILD/polyml-5.7'
make[2]: *** [Makefile:1175: check-local] Error 1
make[1]: Leaving directory '/builddir/build/BUILD/polyml-5.7'
make[1]: *** [Makefile:997: check-am] Error 2
make: *** [Makefile:706: check-recursive] Error 1
It looks like every individual test passed, but the test suite as a
whole failed. Do you have any idea what might cause this? Thank you,

The only thing that occurs to me is that one or other of the tests
messed up the internal state so that when the poly process came to exit
it crashed. I don't have access to any machine running aarch64 (that's
64-bit ARM isn't it?) so there's no way I can test any of this.

Regards,
David

Ramana Kumar

2017-11-11 07:15:44 UTC

Permalink

Hi David,

Are you still updating the fixes-5.7 branch?
It seems to me that this commit might be a bug fix worth including, though
I'm not sure if it depends on new developments.
https://github.com/polyml/polyml/commit/5ba42e588373bd3cd63b8cadb8c4cb8a349c7fff#diff-27af5828c397c28b2fdeacab709ccf57

Cheers,
Ramana

The only thing that occurs to me is that one or other of the tests messed
up the internal state so that when the poly process came to exit it
crashed. I don't have access to any machine running aarch64 (that's 64-bit
ARM isn't it?) so there's no way I can test any of this.
Regards,
David
_______________________________________________
polyml mailing list
http://lists.inf.ed.ac.uk/mailman/listinfo/polyml

David Matthews

2017-11-11 08:37:18 UTC

Permalink

Hi Ramana,
The fixes branches are useful in the early phase of the development
cycle if bugs are found soon after a release. At some point it is no
longer viable to keep back-porting especially if there is a risk that
the fix itself may introduce bugs. This particular change was one of a
number to make the thread system more resilient. At this point I'd
suggest switching to Git master.

Regards,
David

Post by Ramana Kumar
Hi David,
Are you still updating the fixes-5.7 branch?
It seems to me that this commit might be a bug fix worth including, though
I'm not sure if it depends on new developments.
https://github.com/polyml/polyml/commit/5ba42e588373bd3cd63b8cadb8c4cb8a349c7fff#diff-27af5828c397c28b2fdeacab709ccf57
Cheers,
Ramana

The only thing that occurs to me is that one or other of the tests messed
up the internal state so that when the poly process came to exit it
crashed. I don't have access to any machine running aarch64 (that's 64-bit
ARM isn't it?) so there's no way I can test any of this.
Regards,
David
_______________________________________________
polyml mailing list
http://lists.inf.ed.ac.uk/mailman/listinfo/polyml

Ramana Kumar

2017-11-11 10:41:56 UTC

Permalink

OK, thanks.

My impression of Git master is that there is often work-in-progress that
should not be expected to work.
Is it useful to report bugs while development is still in progress?
For example, trying it now I saw this:
Fail "Exception- InternalError: codeToPRegRev raised while compiling"

Post by David Matthews
Hi Ramana,
The fixes branches are useful in the early phase of the development cycle
if bugs are found soon after a release. At some point it is no longer
viable to keep back-porting especially if there is a risk that the fix
itself may introduce bugs. This particular change was one of a number to
make the thread system more resilient. At this point I'd suggest switching
to Git master.
Regards,
David

Post by Ramana Kumar
Hi David,
Are you still updating the fixes-5.7 branch?
It seems to me that this commit might be a bug fix worth including, though
I'm not sure if it depends on new developments.
https://github.com/polyml/polyml/commit/5ba42e588373bd3cd63b
8cadb8c4cb8a349c7fff#diff-27af5828c397c28b2fdeacab709ccf57
Cheers,
Ramana

Post by Jerry James
I attempted to build this version for Fedora. It succeeded on all

Post by Jerry James
https://koji.fedoraproject.org/koji/taskinfo?taskID=19526061
I don't understand the failure, though. We run "make check" after the
build to gain some confidence that the build is good. Here is an
extract from the build log
(https://kojipkgs.fedoraproject.org//work/tasks/6133/19526133/build.log
Test044.ML => Passed
make[2]: Leaving directory '/builddir/build/BUILD/polyml-5.7'
make[2]: *** [Makefile:1175: check-local] Error 1
make[1]: Leaving directory '/builddir/build/BUILD/polyml-5.7'
make[1]: *** [Makefile:997: check-am] Error 2
make: *** [Makefile:706: check-recursive] Error 1
It looks like every individual test passed, but the test suite as a
whole failed. Do you have any idea what might cause this? Thank you,

The only thing that occurs to me is that one or other of the tests messed
up the internal state so that when the poly process came to exit it
crashed. I don't have access to any machine running aarch64 (that's 64-bit
ARM isn't it?) so there's no way I can test any of this.
Regards,
David
_______________________________________________
polyml mailing list
http://lists.inf.ed.ac.uk/mailman/listinfo/polyml

David Matthews

2017-11-11 11:12:24 UTC

Permalink

Post by Ramana Kumar
OK, thanks.
My impression of Git master is that there is often work-in-progress that
should not be expected to work.
Is it useful to report bugs while development is still in progress?
Fail "Exception- InternalError: codeToPRegRev raised while compiling"

That is often true but at this point I'm trying to get the bugs out of
it in order to prepare the release. So any bug reports are welcome.

Michael Norrish reported this exception on the mailing list a few days
ago and I've haven't had time to look into it. If you have a piece of
code that exhibits it then I would like to see it especially if you can
cut it down to something manageable. Please send anything to me
directly rather than through the mailing list.

Regards,
David