[Pkg-javascript-devel] Bug#1112009: Bug#1112009: node-cipher-base: autopkgtest regression on s390x: endianness issue

[Yadd]

On 8/25/25 07:52, Paul Gevers wrote:
> Source: node-cipher-base
> Version: 1.0.6-1
> Severity: serious
> User: debian-ci at lists.debian.org
> Usertags: regression
> X-Debbugs-CC: debian-s390 at lists.debian.org
> User: debian-s390 at lists.debian.org
> Usertags: s390x
> 
> Dear maintainer(s),
> 
> With a recent upload of node-cipher-base the autopkgtest of node-cipher- 
> base fails in testing when that autopkgtest is run with the binary 
> packages of node-cipher-base from unstable. Looking at the output, it 
> seems that the fix for CVE-2025-9287 has endianness problem (bytes are 
> swapped). I have no idea if this is a test only problem, or the patch 
> has a problem. I copied some of the output at the bottom of this report.
> 
> Currently this regression is blocking the migration to testing [1]. Can 
> you please investigate the situation and fix it?
> 
> Paul
> 
> [1] https://qa.debian.org/excuses.php?package=node-cipher-base
> 
> https://ci.debian.net/data/autopkgtest/testing/s390x/n/node-cipher- 
> base/63663875/log.gz
> 
> 
>   84s not ok 17 should be strictly equal
>   84s   ---
>   84s     operator: equal
>   84s     expected: 'd2040002'
>   84s     actual:   '04d20200'
>   84s     at: Test.<anonymous> (/tmp/autopkgtest-lxc.7hdyz_hz/downtmp/ 
> autopkgtest_tmp/smokegJXUWQ/test/index.js:215:5)
>   84s     stack: |-
>   84s       Error: should be strictly equal
>   84s           at Test.assert [as _assert] (/usr/share/nodejs/tape/lib/ 
> test.js:312:48)
>   84s           at Test.bound [as _assert] (/usr/share/nodejs/tape/lib/ 
> test.js:95:17)
>   84s           at Test.strictEqual (/usr/share/nodejs/tape/lib/ 
> test.js:476:7)
>   84s           at Test.bound [as equals] (/usr/share/nodejs/tape/lib/ 
> test.js:95:17)
>   84s           at Test.<anonymous> (/tmp/autopkgtest-lxc.7hdyz_hz/ 
> downtmp/autopkgtest_tmp/smokegJXUWQ/test/index.js:215:5)
>   84s           at Test.bound [as _cb] (/usr/share/nodejs/tape/lib/ 
> test.js:95:17)
>   84s           at Test.run (/usr/share/nodejs/tape/lib/test.js:115:28)
>   84s           at Test.bound [as run] (/usr/share/nodejs/tape/lib/ 
> test.js:95:17)
>   84s           at Immediate.next [as _onImmediate] (/usr/share/nodejs/ 
> tape/lib/results.js:157:7)
>   84s           at process.processImmediate (node:internal/timers:483:21)
>   84s   ...
>   84s  84s 1..17
>   84s # tests 17
>   84s # pass  16
>   84s # fail  1
>   84s  84s autopkgtest [18:12:07]: test pkg-js-autopkgtest
It's not really a regression, but a new test added in last version that 
shows a different behavior between big-endian and little-endian platforms.
I asked to upstream what should be fixed; the test or the code