8:48 p.m.
Hi Raymond --
Thanks for keeping up the buildbot slaves. I note that recently, as visible
at http://lidell.nu/xemacs-buildbot/grid , they’ve given up with the
configure test:
configure: error: C compiler cannot create executables
See `config.log' for more details
This might be because the C compiler is no longer installed, or there’s a
flag that no longer works. Could you have a look and see?
Best,
Aidan
--
‘As I sat looking up at the Guinness ad, I could never figure out /
How your man stayed up on the surfboard after forty pints of stout’
(C. Moore)
6:39 a.m.
On 3/15/17 1:48 PM, Aidan Kehoe wrote:
Hi Raymond --
Thanks for keeping up the buildbot slaves. I note that recently, as visible
at http://lidell.nu/xemacs-buildbot/grid , they’ve given up with the
configure test:
configure: error: C compiler cannot create executables
See `config.log' for more details
This might be because the C compiler is no longer installed, or there’s a
flag that no longer works. Could you have a look and see?
Oh, thanks for letting me know. (I don't check the status very often).
I'll take a look soon. It's probably related to the fact that I finally
updated my mac to macOS 10.12 a short time ago.
6:53 a.m.
On 3/18/17 11:39 PM, Raymond Toy wrote:
On 3/15/17 1:48 PM, Aidan Kehoe wrote:
> Hi Raymond --
>
> Thanks for keeping up the buildbot slaves. I note that recently, as visible
> at http://lidell.nu/xemacs-buildbot/grid , they’ve given up with the
> configure test:
>
> configure: error: C compiler cannot create executables
> See `config.log' for more details
>
> This might be because the C compiler is no longer installed, or there’s a
> flag that no longer works. Could you have a look and see?
>
Oh, thanks for letting me know. (I don't check the status very often).
I'll take a look soon. It's probably related to the fact that I finally
updated my mac to macOS 10.12 a short time ago.
I also (accidentally) upgraded to Xcode 8.2.1 and I needed to agree to
the license terms again.
I think it should be working again. I'd test, but I don't remember how
to retrigger a build.
If it's still not working, please let me know and I'll take another look.
--
Ray
11:30 p.m.
Ar an t-ochtú lá déag de mí Márta, scríobh Raymond Toy:
I also (accidentally) upgraded to Xcode 8.2.1 and I needed to agree
to
the license terms again.
I think it should be working again. I'd test, but I don't remember how
to retrigger a build.
Thanks. I’ve just pushed a changeset and will have a look in the morning to
make sure they’re all building.
If it's still not working, please let me know and I'll take
another look.
Sure!
--
‘As I sat looking up at the Guinness ad, I could never figure out /
How your man stayed up on the surfboard after forty pints of stout’
(C. Moore)
7:22 a.m.
Hi Raymond,
Raymond Toy writes:
I think it should be working again. I'd test, but I don't remember how
to retrigger a build.
I don't think you can trigger a build locally. I can do it. Please let me know
when you want to retrigger a build. I can't promise immediate action but in a
day or two.
But with Aidan now being so active, great to see you hacking along Aidan!, I
think you are going to get some opportunities to trouble shoot your
environment!
Yours
--
%% Mats
8:39 a.m.
Ar an fichiú lá de mí Márta, scríobh Mats Lidell:
Hi Raymond,
> Raymond Toy writes:
> I think it should be working again. I'd test, but I don't remember how
> to retrigger a build.
I don't think you can trigger a build locally. I can do it. Please let me know
when you want to retrigger a build. I can't promise immediate action but in a
day or two.
But with Aidan now being so active, great to see you hacking along Aidan!, I
think you are going to get some opportunities to trouble shoot your
environment!
Yes, there are a couple of his builds currently failing because of absent
bignum and freetype libraries, but Raymond’s bare-bones build is working OK.
--
‘As I sat looking up at the Guinness ad, I could never figure out /
How your man stayed up on the surfboard after forty pints of stout’
(C. Moore)
7:34 p.m.
On Mon, Mar 20, 2017 at 1:39 AM, Aidan Kehoe <kehoea(a)parhasard.net> wrote:
Ar an fichiú lá de mí Márta, scríobh Mats Lidell:
> Hi Raymond,
>
> > Raymond Toy writes:
> > I think it should be working again. I'd test, but I don't remember
how
> > to retrigger a build.
>
> I don't think you can trigger a build locally. I can do it. Please let
me know
> when you want to retrigger a build. I can't promise immediate action
but in a
> day or two.
>
> But with Aidan now being so active, great to see you hacking along
Aidan!, I
> think you are going to get some opportunities to trouble shoot your
> environment!
Yes, there are a couple of his builds currently failing because of absent
bignum and freetype libraries, but Raymond’s bare-bones build is working
OK.
I'll look into that soon. I don't know why the libraries aren't there. I
haven't changed them. Although I think I got them from MacPorts, so
perhaps something is messed up there in the paths or something.
1:54 a.m.
On Tue, Mar 21, 2017 at 12:34 PM, Raymond Toy <toy.raymond(a)gmail.com> wrote:
Yes, there are a couple of his builds currently failing because of absent
> bignum and freetype libraries, but Raymond’s bare-bones build is working
> OK.
I'll look into that soon. I don't know why the libraries aren't there. I
haven't changed them. Although I think I got them from MacPorts, so
perhaps something is messed up there in the paths or something.
Not sure what happened, but freetype2 is now in
/opt/local/include/freetype2.
libfreetype and libgmp are all in /opt/local/lib
It also seems as if my local build of xemacs uses the X11 libraries from
/opt/local/lib as well.
That's not really what I want. I want to use the libraries included in
XQuartz in /usr/X11/lib, because I hate having to rebuild xemacs on the few
occasions I update MacPorts.
But I guess for the bots using either locations for X11 will be fine.
I don't remember how to fix these myself (if possible), but I hope that's
enough info to fix the bots.
--
Ray
10:47 p.m.
Hi Raymond,
Sorry for late answer.
Raymond Toy writes:
On Tue, Mar 21, 2017 at 12:34 PM, Raymond Toy <toy.raymond(a)gmail.com> wrote:
Yes, there are a couple of his builds currently failing because of absent
bignum and freetype libraries, but Raymond’s bare-bones build is working OK.
I'll look into that soon. I don't know why the libraries aren't there. I
haven't
changed them. Although I think I got them from MacPorts, so perhaps something is
messed up there in the paths or something.
Not sure what happened, but freetype2 is now in /opt/local/include/freetype2.
libfreetype and libgmp are all in /opt/local/lib
It also seems as if my local build of xemacs uses the X11 libraries from /opt/local/lib
as well.
That's not really what I want. I want to use the libraries included in XQuartz in
/usr/X11/lib, because I hate having to rebuild xemacs on the few occasions I update
MacPorts.
But I guess for the bots using either locations for X11 will be fine.
I don't remember how to fix these myself (if possible), but I hope that's enough
info
to fix the bots.
I had set it up before so that configure would use
'--with-site-includes=/usr/X11R6/include/freetype2'. You see all that is passed to
configure in the configure logs in the build bot. I suppose since that folder is not on
your machine any more I should simply remove that extra piece to configure. I will do that
shortly.
Yours
--
%% Mats
8:47 a.m.
Hi Raymond,
Mats Lidell writes:
I had set it up before so that configure would use
'--with-site-includes=/usr/X11R6/include/freetype2'. You see all that
is passed to configure in the configure logs in the build bot. I
suppose since that folder is not on your machine any more I should
simply remove that extra piece to configure. I will do that shortly.
I have removed the with-site-include for your builders and triggered new builds. That
brought them up to the testing phase! However there they got stuck in the mule tests so
that is something to look into.
Does "make check" work for you when you build locally?
Yours
--
%% Mats
2:53 a.m.
Sorry for the delay.
That's great news that it builds now.
I'm running "make check" now and it seems to be running ok, but I get lots
of tests fail like this on:
SUMMARY for
/Volumes/share2/src/xemacs/hg/xemacs-21.5/tests/automated/lisp-tests.el:
12846 passes
78 assertion failures
4 errors that should have been generated, but weren't
16 wrong-error failures
0 missing-message failures
0 other failures
Are these expected?
I also noticed when I did a build on Ubuntu with ToT, xemacs crashes with
an assertion when I run gnus. I'll try to get gdb backtrace for that.
Something to with hash tables, I think.
I also note that when I build ToT on Fedora 25, everything works, and gnus
appears to work fine. But the menu pulldowns are really horribly mangled
with weird effects all over. I'll fill a separate report on that.
However, my OSX build works fine.
On Sun, Apr 2, 2017 at 12:47 AM, Mats Lidell <mats.lidell(a)cag.se> wrote:
Hi Raymond,
> Mats Lidell writes:
> I had set it up before so that configure would use
> '--with-site-includes=/usr/X11R6/include/freetype2'. You see all that
> is passed to configure in the configure logs in the build bot. I
> suppose since that folder is not on your machine any more I should
> simply remove that extra piece to configure. I will do that shortly.
I have removed the with-site-include for your builders and triggered new
builds. That brought them up to the testing phase! However there they got
stuck in the mule tests so that is something to look into.
Does "make check" work for you when you build locally?
Yours
--
%% Mats
--
Ray
2:38 a.m.
Here are the results from make check, run from my account, not from the
buildbot account:
SUMMARY for
/Volumes/share2/src/xemacs/hg/xemacs-21.5/tests/automated/weak-tests.el:
140 passes
0 assertion failures
0 errors that should have been generated, but weren't
0 wrong-error failures
0 missing-message failures
0 other failures
base64-tests.el: 1234 of 1234 tests successful (100%).
byte-compiler-tests.el: 66 of 66 tests successful (100%).
c-tests.el: 146 of 146 tests successful (100%).
case-tests.el: 9436 of 9436 tests successful (100%).
ccl-tests.el: 9082 of 9082 tests successful (100%).
completion-tests.el: 126 of 126 tests successful (100%).
database-tests.el: 20 of 20 tests successful (100%).
dired-tests.el: 32 of 32 tests successful (100%).
extent-tests.el: 684 of 686 tests successful ( 99%).
face-tests.el: 20 of 20 tests successful (100%).
file-tests.el: 18 of 18 tests successful (100%).
files-tests.el: 20 of 20 tests successful (100%).
format-tests.el: 286 tests completed (aborted).
hash-table-tests.el: 11894 of 11894 tests successful (100%).
iso-ir-196-test.el: 2 of 2 tests successful (100%).
keymap-tests.el: 36 of 36 tests successful (100%).
lisp-reader-tests.el: 1208 tests completed (aborted).
lisp-tests.el: 12944 tests completed (aborted).
md5-tests.el: 56 of 56 tests successful (100%).
mule-tests.el: 9424 tests completed (aborted).
os-tests.el: 12 of 12 tests successful (100%).
process-tests.el: 30 of 30 tests successful (100%).
query-coding-tests.el: 3200 of 3200 tests successful (100%).
regexp-tests.el: 6706 of 6708 tests successful ( 99%).
region-tests.el: 28 of 28 tests successful (100%).
register-tests.el: 12 of 12 tests successful (100%).
search-tests.el: 116 of 120 tests successful ( 96%).
symbol-tests.el: 252 of 252 tests successful (100%).
syntax-ppss-tests.el: 8 of 8 tests successful (100%).
syntax-tests.el: 148 of 154 tests successful ( 96%).
tag-tests.el: 6 of 6 tests successful (100%).
weak-tests.el: 140 of 140 tests successful (100%).
***** There were 8 unexpected test suite failures in files:
/Volumes/share2/src/xemacs/hg/xemacs-21.5/tests/automated/mule-tests.el
/Volumes/share2/src/xemacs/hg/xemacs-21.5/tests/automated/lisp-tests.el
/Volumes/share2/src/xemacs/hg/xemacs-21.5/tests/automated/lisp-reader-tests.el
/Volumes/share2/src/xemacs/hg/xemacs-21.5/tests/automated/format-tests.el
Done
real 33m26.652s
user 33m9.459s
sys 0m6.069s
On Tue, Apr 11, 2017 at 6:53 PM, Raymond Toy <toy.raymond(a)gmail.com> wrote:
Sorry for the delay.
That's great news that it builds now.
I'm running "make check" now and it seems to be running ok, but I get lots
of tests fail like this on:
SUMMARY for /Volumes/share2/src/xemacs/hg/xemacs-21.5/tests/automated/
lisp-tests.el:
12846 passes
78 assertion failures
4 errors that should have been generated, but weren't
16 wrong-error failures
0 missing-message failures
0 other failures
Are these expected?
I also noticed when I did a build on Ubuntu with ToT, xemacs crashes with
an assertion when I run gnus. I'll try to get gdb backtrace for that.
Something to with hash tables, I think.
I also note that when I build ToT on Fedora 25, everything works, and gnus
appears to work fine. But the menu pulldowns are really horribly mangled
with weird effects all over. I'll fill a separate report on that.
However, my OSX build works fine.
On Sun, Apr 2, 2017 at 12:47 AM, Mats Lidell <mats.lidell(a)cag.se> wrote:
> Hi Raymond,
>
> > Mats Lidell writes:
> > I had set it up before so that configure would use
> > '--with-site-includes=/usr/X11R6/include/freetype2'. You see all that
> > is passed to configure in the configure logs in the build bot. I
> > suppose since that folder is not on your machine any more I should
> > simply remove that extra piece to configure. I will do that shortly.
>
> I have removed the with-site-include for your builders and triggered new
> builds. That brought them up to the testing phase! However there they got
> stuck in the mule tests so that is something to look into.
>
> Does "make check" work for you when you build locally?
>
> Yours
> --
> %% Mats
>
--
Ray
--
Ray
2563
days inactive
2592
days old
11 comments
3 participants
participants (3)
-
Aidan Kehoe -
Mats Lidell -
Raymond Toy