<html><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"></head><body>Those tests were not useful, because I ran them against references I generated from the latest commit. It'll be several days before I can try something more useful.<br><br>-------- Original message --------<br>From: darxus@chaosreigns.com <br>Date: 07/08/2016 8:58 AM (GMT-05:00) <br>To: Bryce Harrington <bryce@osg.samsung.com> <br>Cc: cairo@cairographics.org <br>Subject: Re: [cairo] What does it take to get a make check to pass with the
xcb target, using CAIRO_REF_DIR? <br><br>On 07/07, Bryce Harrington wrote:<br>> When doing releases I also try to do a run and compare with the prior<br>> release to spot regressions. I'd really love to be able to do a<br>> thorough git bisect search to locate which changes caused which tests to<br>> fail, but as the suite takes a while to run it'd be pretty time<br>> consuming.<br><br>I think bisecting sounds like great fun, except for the irregularity of<br>problems. I wrote a script to build a random commit, back through 2010<br>(because they seemed to not be successfully building before that). All of<br>them either gave me no test-suite.log, or, for the 22 that gave me a log,<br>the test suite had at least 6 failures. <br><br>Nothing passed all tests. Does anybody have any ideas when all tests for<br>the xcb target might have passed? Is this evidence that the FAILs are<br>caused by an external library?<br><br>This is the output, plus the perl script I wrote:<br>http://www.chaosreigns.com/cairo/tests.tar.gz<br>(The script is kind of crude.)<br><br><br>This is a filename with a unix timestamp (git log -1 --pretty=format:%ct),<br>an underscore, the commit hash, a colon, the number of FAILs, some spaces,<br>then a GMT date calculated from the unix timestamp:<br><br>darxus@dancer:~/source/cairolog$ grep -l test * | xargs grep -c ' FAIL'<br>1264201313_c0458b456007f718747be7fd690e674df5026059:7 Fri Jan 22 23:01:53 2010<br>1269204078_247ef2bd63e755963f7930ac06c79b95aed2adb4:6 Sun Mar 21 20:41:18 2010<br>1271277994_54670ec13d64efa94f552b5473c1f15a9db1cecd:145 Wed Apr 14 20:46:34 2010<br>1273693338_094f0e0fa0153f290061635eed51e8d1dbe2cf4a:5 Wed May 12 19:42:18 2010<br>1276288178_d31e2418d6d9920205a154f21becca12a7b84c2a:7 Fri Jun 11 20:29:38 2010<br>1278883932_fbb0a260b707cb5f02a14cc368c6f2f0d63564c3:146 Sun Jul 11 21:32:12 2010<br>1286969384_90d50cd92315d6760069ad8062aba5e297370b20:6 Wed Oct 13 11:29:44 2010<br>1292229969_d9ba8337ab456ae0e232d3c603cb41cea984ebea:32 Mon Dec 13 08:46:09 2010<br>1292427140_e4b78424ac82588bcb9b855d5b6d5872050d33f9:7 Wed Dec 15 15:32:20 2010<br>1297614844_c3605bd3adcae1f12731230a1ea599d15e3c8cad:7 Sun Feb 13 16:34:04 2011<br>1311937629_58df191946b7d275c5706dc40e1cf3dffa7cceae:8 Fri Jul 29 11:07:09 2011<br>1313177159_bb24f165ff57973347b34956a371c6b33d2d9b59:55 Fri Aug 12 19:25:59 2011<br>1342113299_3ed581149f82b653e1987e746467ad8ae9ce3ca2:7 Thu Jul 12 17:14:59 2012<br>1348565464_31291611625c21bc9ec2827cf4ccc8a1f18392ed:55 Tue Sep 25 09:31:04 2012<br>1405128883_d5ffe67008afe46c030b31afc0803321b6479990:7 Sat Jul 12 01:34:43 2014<br>1413893554_f032133e6d5ad05157fc46609d8c63103028342c:7 Tue Oct 21 12:12:34 2014<br>1423145626_3cf862f6d973755cd9824c2224fbe0a623c47ff1:7 Thu Feb 5 14:13:46 2015<br>1425713393_58df191946b7d275c5706dc40e1cf3dffa7cceae:7 Sat Mar 7 07:29:53 2015<br>1434656331_ae608035c7b7133826a608d45e067c3875a1aceb:7 Thu Jun 18 19:38:51 2015<br>1435360353_70cc8f250b5669e757b4f044571ba0f71e3dea9e:7 Fri Jun 26 23:12:33 2015<br>1465125216_caa4c9fdeb3aecd9a4288114e75d24ec931cd01b:6 Sun Jun 5 11:13:36 2016<br>1467845160_f9b65ae1fc91bc558a01c2ad7be5a121c6f10818:7 Wed Jul 6 22:46:00 2016<br><br>Of the ones that had fewer than 10 FAILs, the test names mostly started<br>with "record", with a few "subsurface". <br><br>The most recent one with no test-suite.log was <br>1466331752_bf5adaf3942388e58ad3bda30173e53b214df885 Sun Jun 19 10:22:32 2016 <br>Which I manually checked, and it did fail to build:<br>test/cairo-test-runner.c:130: undefined reference to `cairo_boilerplate_xmalloc'<br><br>The second most recent one with no test-suite.log was:<br>1465125216_afe6f4f0519606c4bc7e9b705b0cae75692d7af2 Sun Jun 5 11:13:36 2016 <br>It also failed to build:<br>./autogen.sh: running `./configure --cache-file=config.cache --disable-static --enable-test-surfaces --prefix=/home/darxus/install --enable-xcb'<br>configure: creating cache config.cache<br>configure: error: cannot find install-sh, install.sh, or shtool in "." "./.." "./../.."<br><br>(Which, so far, confirms my expectation that empty output files are due to<br>failures to build.)<br><br><br>This is how many I attempted to build per year that produced no<br>test-suite.log. The ones from before 2010 were from before I started to<br>restrict the date range (and go to sleep):<br><br>$ grep -L test * | cut -d'_' -f1 | ~/bin/unixtime.pl | awk '{print $5}' | uniq -c<br> 1 2002<br> 3 2003<br> 3 2004<br> 16 2005<br> 14 2006<br> 15 2007<br> 23 2008<br> 6 2009<br> 100 2010<br> 80 2011<br> 40 2012<br> 22 2013<br> 18 2014<br> 7 2015<br> 2 2016<br><br>This is how many total I ran:<br><br>darxus@dancer:~/source/cairolog$ ls *_* | cut -d'_' -f1 | ~/bin/unixtime.pl | awk '{print $5}' | uniq -c<br> 1 2002<br> 3 2003<br> 3 2004<br> 17 2005<br> 14 2006<br> 15 2007<br> 23 2008<br> 6 2009<br> 109 2010<br> 84 2011<br> 42 2012<br> 22 2013<br> 20 2014<br> 11 2015<br> 4 2016<br><br></body></html>