The five failures were 3 separate bugs in ASDF due to changes that broke #+ecl code that I hadn't tested recently in the plan branch. They are now fixed.
Now ECL passes all these tests... but since I upgraded ECL to the latest from git (branch develop), test-run-program-unix has started failing with: "The function :STOPPED is undefined." in the terminate-process tests. I suspect an ECL bug. The test passes with ECL 16.1.3 (release tarball) from NixOS or 15.3.7 (also release I suppose) from Ubuntu. —♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org (map()(lambda(x)(format t"~31R "x))'(595756 9556552524 643802 496307950)) On Fri, Jul 21, 2017 at 1:43 AM, Daniel Kochmański <dan...@turtleware.eu> wrote: > I don't think this is a case unless ASDF calls new ECL process itself. AFAIR > ECL doesn't call itself anywhere (except `make check` target in build > process, but that's by design). > > Regards, > > Daniel > > > > On 21.07.2017 03:01, Robert Goldman wrote: >> >> On 7/20/17 Jul 20 -7:00 PM, Robert Goldman wrote: >>> >>> Our new version of ASDF has 5 failures for me on Linux, testing against >>> the master branch. >>> >>> Is that what I should be testing, or should I be testing against the dev >>> branch? >>> >>> The tests which are failing are: >>> >>> ecl-make-image.script ecl-prebuilt-systems.script test-bundle.script >>> test-program.script test-require.script >>> >>> Thanks, >>> r >>> >> Quick follow-up: Fare suspects that what might be happening is that ECL >> is spawning a sub-process that gets the built-in ASDF instead of the one >> that we loaded. Is there any chance that is what's happening? If so, >> any idea how to ensure that if one upgrades the ASDF in an image, one >> gets the same ASDF version in all ECL processes started by that image? >> > >
