Bug 1218954

Summary:	[Build 20240117] ltp: openQA test fails in readahead01
Product:	[openSUSE] openSUSE Tumbleweed	Reporter:	Dominique Leuenberger <dimstar>
Component:	Kernel	Assignee:	Cyril Hrubis <chrubis>
Status:	RESOLVED INVALID	QA Contact:	E-mail List <qa-bugs>
Severity:	Normal
Priority:	P5 - None	CC:	chrubis, guillaume.gardet, jack, petr.vorel
Version:	Current
Target Milestone:	---
Hardware:	Other
OS:	Other
URL:	https://openqa.opensuse.org/tests/3877952/modules/readahead01/steps/8
Whiteboard:
Found By:	openQA	Services Priority:
Business Priority:		Blocker:	Yes
Marketing QA Status:	---	IT Deployment:	---

Description Dominique Leuenberger 2024-01-18 15:18:02 UTC

## Observation

tst_test.c:1709: TINFO: LTP version: 20230929
tst_test.c:1593: TINFO: Timeout per run is 0h 00m 30s
readahead01.c:36: TPASS: readahead() with fd = -1 : EBADF (9)
readahead01.c:43: TPASS: readahead() with invalid fd : EBADF (9)
readahead01.c:62: TPASS: readahead() on O_PATH file : EBADF (9)
readahead01.c:62: TPASS: readahead() on directory : EINVAL (22)
readahead01.c:62: TPASS: readahead() on /dev/zero : EINVAL (22)
readahead01.c:62: TPASS: readahead() on pipe read end : EINVAL (22)
readahead01.c:62: TPASS: readahead() on pipe write end : EBADF (9)
readahead01.c:62: TPASS: readahead() on unix socket : EINVAL (22)
readahead01.c:62: TPASS: readahead() on inet socket : EINVAL (22)
readahead01.c:62: TPASS: readahead() on epoll : EINVAL (22)
readahead01.c:62: TPASS: readahead() on eventfd : EINVAL (22)
readahead01.c:62: TPASS: readahead() on signalfd : EINVAL (22)
readahead01.c:62: TPASS: readahead() on timerfd : EINVAL (22)
readahead01.c:62: TPASS: readahead() on pidfd : EINVAL (22)
readahead01.c:62: TPASS: readahead() on fanotify : EINVAL (22)
readahead01.c:62: TPASS: readahead() on inotify : EINVAL (22)
readahead01.c:62: TPASS: readahead() on userfaultfd : EINVAL (22)
readahead01.c:62: TPASS: readahead() on perf event : EINVAL (22)
readahead01.c:62: TPASS: readahead() on io uring : EINVAL (22)
readahead01.c:62: TPASS: readahead() on bpf map : EINVAL (22)
readahead01.c:62: TPASS: readahead() on fsopen : EINVAL (22)
readahead01.c:62: TPASS: readahead() on fspick : EINVAL (22)
readahead01.c:62: TPASS: readahead() on open_tree : EBADF (9)
readahead01.c:62: TFAIL: readahead() on memfd secret succeeded

Summary:
passed   23
failed   1
broken   0
skipped  0
warnings 0
### TEST readahead01 COMPLETE >>> 1.


openQA test in scenario opensuse-Tumbleweed-JeOS-for-kvm-and-xen-x86_64-jeos-ltp-syscalls@uefi_virtio-2G fails in
[readahead01](https://openqa.opensuse.org/tests/3877952/modules/readahead01/steps/8)

## Test suite description



## Reproducible

Fails since (at least) Build [20240117](https://openqa.opensuse.org/tests/3877750)


## Expected result

Last good: [20240116](https://openqa.opensuse.org/tests/3876036) (or more recent)


## Further details

Always latest result in this scenario: [latest](https://openqa.opensuse.org/tests/latest?arch=x86_64&distri=opensuse&flavor=JeOS-for-kvm-and-xen&machine=uefi_virtio-2G&test=jeos-ltp-syscalls&version=Tumbleweed)

Comment 1 Petr Vorel 2024-01-18 19:22:29 UTC

Cyril, test fails due recent change, can you please have a look?

TFAIL: readahead() on memfd secret succeeded

Since:
https://github.com/linux-test-project/ltp/commit/ecf81d72963e03e1e0ec2a7aa0ce30eb2da2b770

I haven't found any behavior change in mainline mm/readahead.c, I sent a patchset to just whitelist the problem, but maybe I'm wrong and it would cover bug:

https://lore.kernel.org/ltp/20240118191353.2815915-1-pvorel@suse.cz/

Comment 2 Jan Kara 2024-01-19 14:19:52 UTC

Yeah, memfd_secret() is the same as ordinary memfd for readahead purposes so it should be accepted for readahead test.

Comment 3 Petr Vorel 2024-01-21 19:34:38 UTC

Fix merged in LTP upstream, test result should be fixed in next openQA build

https://github.com/linux-test-project/ltp/commit/60ebbf3f6b9ba2110f2143447af8d6ca0f57f64b

Bug in the test => resolved as invalid.