[Ecls-list] [ecl][release candidate][windows]

[Daniel Kochmański]

Hey,

I succesfully fixed issue with :lapack. Problem was with interpreting
values typespec (multpile &optional clauses). Changes are pushed to
develop branch (not rc branch).

Tomorrow I'll try to attack second regression.

BR,
Daniel

Anton Vodonosov writes:

> I have tested 15.3.7-rc1 with cl-test-grid - 
> no regressions comparing to 15.2.21.
>
> I also reviewed the regressions comparing to the old release
> 13.5.1 I reported earlier. Only two deserve investigation
> as real regressions:
>
> 1. (ql:quickload :lapack) fails on the recent 15.3.7-rc1
>    despited worked on 13.5.1
>
> 2. (ql:quickload :cl-store-tests) runs test of cl-store system,
>    and on 15.3.7-rc1 it has one more test failure than on 13.5.1.
>    The failed test is named CIRC.8
>
> Other failures are either not regressions (either can be reproduced
> on old release, or not reproduced on the new release; some of them
> are probably caused by external factors, e.g. the URLS accessed by
> testsuites like trivial-http).
>
> For the record, the current reports:
> https://common-lisp.net/project/cl-test-grid/ecl/ecl-diff-7-lisp-to-c.html
> https://common-lisp.net/project/cl-test-grid/ecl/ecl-diff-7-bytecode.html
>
> Best regards,
> - Anton

-- 
Daniel Kochmański | Poznań, Poland
;; aka jackdaniel

"Be the change that you wish to see in the world." - Mahatma Gandhi