On Sun, May 24, 2020 at 04:11:53AM +0100, Ken Moffat wrote:
>
> This is LFS r11848 cross-chap5 branch from 13th May and I'm trying
> to understand why tests fail (although I don't always achieve any
> understanding).
>
> So, I gave vim's tests another try, but they failed somewhat badly:
>
> Executed: 2667 Tests
> Skipped: 47 Tests
> FAILED: 3 Tests
>
>
> Failures:
> From test_excmd.vim:
> Found errors in Test_redir_cmd():
> function RunTheTest[40]..Test_redir_cmd line 20: command did not
> fail: redir! > Xfile
> From test_help.vim:
> Found errors in Test_helptag_cmd():
> function RunTheTest[40]..Test_helptag_cmd line 20: command did not
> fail: r-xr--r--
> function RunTheTest[40]..Test_helptag_cmd line 31: command did not
> fail: -w-------
> From test_quickfix.vim:
> Found errors in Test_switchbuf():
> function RunTheTest[40]..Test_switchbuf line 131: Expected 'split'
> but got ''
> function RunTheTest[40]..Test_switchbuf line 136: Expected 'usetab'
> but got 'useopen'
> function RunTheTest[40]..Test_switchbuf line 141: Expected '' but got
> 'useopen'
>
I didn't have any older versions of vim handy, but looking at the
9.1 book we were using vim-8.2.0190, so I downloaded that. With
gcc-10.1.0 it too fails, although not quite as badly:
Executed: 2407 Tests
Skipped: 49 Tests
FAILED: 1 Tests
Failures:
From test_quickfix.vim:
Found errors in Test_switchbuf():
function RunTheTest[40]..Test_switchbuf line 131: Expected 'split' but
got ''
function RunTheTest[40]..Test_switchbuf line 136: Expected 'usetab' but
got 'useopen'
function RunTheTest[40]..Test_switchbuf line 141: Expected '' but got
'useopen'
Hmm, my local copy of the book shows vim is now 8.2.0716 but
I can't see that in the changelog.
And then I noticed that at some point I accidentally started running
the tests as root, which obvously isn't going to help.
With 8.2.0716 the tests run as user nobody take a very long time,
and eventually terminate, probably early, with
Flaky test failed too often, giving up
Found errors in Test_textprop_with_syntax():
Run 1:
function
RunTheTest[39]..Test_textprop_with_syntax[15]..VerifyScreenDump line 58: See
dump file difference: call
term_dumpdiff("testdir/failed/Test_textprop_syn_1.dump",
"testdir/dumps/Test_textprop_syn_1.dump"); difference in line 1:
">(+0&#ffffff0|a|b|c|)| @69"; difference in line 2: "|~| @73"; difference in
line 6: "@57|1|,|1| @10|A|l@1| "
Run 2:
function
RunTheTest[39]..Test_textprop_with_syntax[15]..VerifyScreenDump line 58: See
dump file difference: call
term_dumpdiff("testdir/failed/Test_textprop_syn_1.dump",
"testdir/dumps/Test_textprop_syn_1.dump"); difference in line 1:
">(+0&#ffffff0|a|b|c|)| @69"; difference in line 2: "|~| @73"; difference in
line 6: "@57|1|,|1| @10|A|l@1| "
Flaky test failed too often, giving up
I found a gentoo post along the way, about problems with gcc-10
including vim where somebody had a crash, related to vim NOT
applying its normal -D_FORTIFY_SOURCE=1 because the gcc major varion
now has two digits, not one :
[ https://trofi.github.io/posts/213-gcc-10-in-gentoo.html ]
Unfortunately, my logs show that the versions of vim which I have
tried DO enable that.
I'di then taken a look at fedora, who are using 8.2.0806 with
gcc-10.1, although I don't think they run vims testsuite. Latest
when I looked was 8.2.0814. That too appears to take for ever to
run its tests, eventually ending similarly to 8.2.0716.
If you'll excuse my klatchian (or even if you won't!) "Sod this for
a game of soldiers!".
ĸen
--
Do you not know that, what you belittle by the name tree is but the
mere four-dimensional analogue of a whole multidimensional universe
which - no, I can see you do not. -- Druellae (a Dryad)
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Do not top post on this list.
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
http://en.wikipedia.org/wiki/Posting_style
