Bug 1382628 - [mozjs45] test failures on s390x
Summary: [mozjs45] test failures on s390x
Keywords:
Status: CLOSED RAWHIDE
Alias: None
Product: Fedora
Classification: Fedora
Component: mozjs45
Version: 25
Hardware: s390x
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Marek Skalický
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks: ZedoraTracker
TreeView+ depends on / blocked
 
Reported: 2016-10-07 09:02 UTC by Dan Horák
Modified: 2016-12-13 08:24 UTC (History)
3 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2016-12-13 08:24:44 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)

Description Dan Horák 2016-10-07 09:02:34 UTC
The build of mozjs45-45.4.0-1 fails on s390x with test failures

...
REGRESSIONS
    js1_8_5/extensions/clone-v1-typed-array.js
    js1_8_5/extensions/typedarray-subarray-of-subarray.js
    js1_8_5/extensions/typedarray-set-neutering.js
    js1_8_5/extensions/clone-complex-object.js
    js1_8_5/extensions/typedarray.js
    js1_8_5/extensions/file-mapped-arraybuffers.js
    js1_8_5/extensions/clone-transferables.js
    js1_8_5/extensions/clone-typed-array.js
    ecma_6/Array/fill.js
    ecma_6/TypedArray/slice.js
    ecma_6/TypedArray/entries.js
    ecma_6/TypedArray/reverse.js
    ecma_6/TypedArray/from_basics.js
    ecma_6/TypedArray/find-and-findIndex.js
    ecma_6/TypedArray/values.js
    ecma_6/TypedArray/reduce-and-reduceRight.js
    ecma_6/TypedArray/from_mapping.js
    ecma_6/TypedArray/map-and-filter.js
    ecma_6/TypedArray/join.js
    ecma_6/TypedArray/from_proxy.js
    ecma_6/TypedArray/forEach.js
    ecma_6/TypedArray/uint8clamped-constructor.js
    ecma_6/TypedArray/from_this.js
    ecma_6/TypedArray/from_iterable.js
    ecma_6/TypedArray/object-defineproperty.js
    ecma_6/TypedArray/from_realms.js
    ecma_6/TypedArray/set-same-buffer-different-source-target-types.js
    ecma_6/TypedArray/every-and-some.js
    ecma_6/TypedArray/indexOf-and-lastIndexOf.js
    ecma_6/TypedArray/of.js
    ecma_6/TypedArray/seal-and-freeze.js
    ecma_6/TypedArray/includes.js
    ecma_6/TypedArray/from_constructor.js
    ecma_6/TypedArray/element-setting-converts-using-ToNumber.js
    ecma_6/TypedArray/fill.js
    ecma_6/Reflect/get.js
    ecma_6/Math/trunc.js
    ecma_6/Math/sign.js
    ecma_6/Math/atanh-approx.js
    ecma_6/Math/expm1-exact.js
    ecma_6/Math/log10-approx.js
    ecma_6/Math/clz32.js
    ecma_6/Math/log2-approx.js
    ecma_6/Math/fround.js
    ecma_6/Math/tanh-approx.js
    ecma_6/Math/acosh-approx.js
    ecma_6/Math/cosh-approx.js
    ecma_6/Math/20.2.2.ToNumber.js
    ecma_6/Math/cbrt-exact.js
    ecma_6/Math/asinh-approx.js
    ecma_6/Math/tanh-exact.js
    ecma_6/Math/log1p-exact.js
    ecma_6/Math/sinh-approx.js
    ecma_6/Math/sinh-exact.js
    ecma_6/Math/log1p-approx.js
    ecma_6/Math/cbrt-approx.js
    ecma_6/Math/cosh-exact.js
    ecma_6/Math/log10-exact.js
    ecma_6/Math/expm1-approx.js
    ecma_6/Math/acosh-exact.js
    ecma_6/Math/log2-exact.js
    ecma_6/Math/asinh-exact.js
    ecma_6/Math/atanh-exact.js
    ecma_6/Symbol/typed-arrays.js
    ecma_5/extensions/reviver-mutates-holder-array-nonnative.js
    ecma_5/extensions/reviver-mutates-holder-object-nonnative.js
FAIL
error: Bad exit status from /var/tmp/rpm-tmp.vLNDdt (%check)
    Bad exit status from /var/tmp/rpm-tmp.vLNDdt (%check)
...


for complete logs please see http://s390.koji.fedoraproject.org/koji/taskinfo?taskID=2367267

Interestingly there is no such issue on ppc64 (big endian as s390x), so it looks as a s390x specific issue.


Version-Release number of selected component (if applicable):
mozjs45-45.4.0-1.fc25

Comment 1 Dan Horák 2016-12-05 10:54:46 UTC
still the same issues with mozjs45-45.5.1-1.fc26 (http://s390.koji.fedoraproject.org/koji/taskinfo?taskID=2417826)

Comment 2 Marek Skalický 2016-12-05 11:19:33 UTC
Thanks for retesting it...

Could you provide please gcc preprocessor code to check all Fedora non-primary architectures?

I am not sure sure what is the value of XY for all secondary in code:
#if defined(XY)

(I need it to update a patch correctly)

Comment 3 Dan Horák 2016-12-06 11:40:24 UTC
s390(x) defines
__s390x__ and __s390__ for the 64-bit s390x
__s390__ for 32-bit s390

ppc defines
__powerpc64__ and __powerpc__ for ppc64/ppcle (needs additional endian check via __BYTE_ORDER__

Comment 4 Dan Horák 2016-12-06 12:32:17 UTC
Than, have you made any progress with this issue?

Comment 5 Than Ngo 2016-12-06 12:47:46 UTC
(In reply to Dan Horák from comment #4)
> Than, have you made any progress with this issue?

Hi Dan, i was busy with other stuff in qt/kde due to openssl-1.1. so no solution atm, i will continue to work on it today or tomorrow after the openssl-1.1 issue in qt/kde is resolved.

Comment 7 Than Ngo 2016-12-08 15:41:57 UTC
i can confirm the fix. It's built fine on s390x.

http://s390.koji.fedoraproject.org/koji/taskinfo?taskID=2419458

Comment 8 Marek Skalický 2016-12-13 08:24:44 UTC
Thanks for feedback.


Note You need to log in before you can comment on or make changes to this bug.