On Thu, Sep 18, 2025 at 4:35 PM Sebastian Andrzej Siewior <[email protected]> wrote: > > The current openssl package fails to build in unstable/ experimental due > to a failure in the testuite. The unstable build has been retried three > times with the same outcome. I tried ti build it on the porterbox is > succeeded on the first attempt. > Based on db.d.o I would assume that eberlin and mipsel-osuosl-03/ 05 are > both Loongson 3A4000 and therfore identical. One has 4GiB ram and the > other 16GiB so maybe not identical.
I might have my wires crossed, but should Debian be building the FIPS version of the library? >From ><https://buildd.debian.org/status/fetch.php?pkg=openssl&arch=mips64el&ver=3.6.0%7E%7Ealpha1-1&stamp=1757461246&raw=0>: make[4]: Entering directory '/build/reproducible-path/openssl-3.6.0~~alpha1/build_shared' ( SRCTOP=.. \ BLDTOP=. \ PERL="/usr/bin/perl" \ FIPSKEY="f4556650ac31d35461610bac4ed81b1a181b2d8a43ea2854cbae22ca74560813" \ EXE_EXT= \ /usr/bin/perl ../test/run_tests.pl ) 00-prep_fipsmodule_cnf.t .. # The results of this test will end up in test-runs/prep_fipsmodule 1..1 HMAC : (KAT_Integrity) : Pass HMAC : (Module_Integrity) : Pass ... KEM_Encap : (KAT_KEM) : Pass KEM_Decap : (KAT_KEM) : Pass KEM_Decap_Reject : (KAT_KEM) : Pass Failed to load FIPS module INSTALL FAILED C0EA0FF1FF000000:error:1C8000D4:Provider routines:SELF_TEST_post:invalid state:../providers/fips/self_test.c:336: C0EA0FF1FF000000:error:1C8000D8:Provider routines:OSSL_provider_init_int:self test post failure:../providers/fips/fipsprov.c:1004: C0EA0FF1FF000000:error:078C0105:common libcrypto routines:provider_init:init fail:../crypto/provider_core.c:1045:name=fips ../../util/wrap.pl ../../apps/openssl fipsinstall -pedantic -module ../../providers/fips.so -provider_name fips -section_name fips_sect -out ../../test/fipsmodule.cnf => 1 not ok 1 - fips install # Failed test 'fips install' # at ../test/recipes/00-prep_fipsmodule_cnf.t line 33. # Looks like you failed 1 test of 1. Dubious, test returned 1 (wstat 256, 0x100) Failed 1/1 subtests > Anyway since it fails and the buildd and succeeds on the porterbox, what > could be done next? Jeff
