stress-ng: clone system calls randomly fail with -EINVAL on aarch64

[Colin Ian King]

[SRU] Xenial

Getting -EINVAL on stressors that call clone() on aarch64. Turns out that the clone() thread's stack has to be 16 byte aligned.

[REPODUCER + FIX]
For example, run:

stress-ng --vm-rw 1 -v

Without the fix, this can randomly exit with -EINVAL when the stack being passed to the clone() system call is not aligned on a 16 byte boundary.

With the fix, the clone() system call works fine and the stressor will run.

[REGRESSION POTENIAL]
Minimal. I have exhaustively tested this on multiple platforms/architectures and the fix ensures that the stack is aligned correctly. stress-ng is a leaf project in universe and this fix only touches 3 of the stress-ng stressors that use the clone() system call.

[Colin Ian King (colin-king)]

Fix committed, http://kernel.ubuntu.com/git/cking/stress-ng.git/commit/?id=4887d3b4a4031aab6b108e57007c3934ca5aab74

[Colin Ian King (colin-king)]

SRU: package: https://launchpad.net/~colin-king/+archive/ubuntu/stress-ng-xenial-sru/+packages

Debdiff: https://launchpadlibrarian.net/255152031/stress-ng_0.05.23-1_0.05.23-1ubuntu1.diff.gz

[Martin Pitt (pitti)]

Hello Colin, or anyone else affected,

Accepted stress-ng into xenial-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/stress-ng/0.05.23-1ubuntu1 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, and change the tag from verification-needed to verification-done. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed. In either case, details of your testing will help us make a better decision.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance!

[Colin Ian King (colin-king)]

This failed my testing, looks like I botched the mask and committed in a buggy fix. I've got an additional patch to address this issue:

http://kernel.ubuntu.com/git/cking/stress-ng.git/commit/?id=d12036f3b0bb62b2bc9203d1ca4dae032d50092e

Going to upload the fixed version ASAP.

[Colin Ian King (colin-king)]

..I got bitten that the original code worked on the machines I tested because I got lucky with random stack alignment. Gah.

[Martin Pitt (pitti)]

Hello Colin, or anyone else affected,

Accepted stress-ng into xenial-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/stress-ng/0.05.23-1ubuntu2 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, and change the tag from verification-needed to verification-done. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed. In either case, details of your testing will help us make a better decision.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance!

[Colin Ian King (colin-king)]

Tested, works on aarm64 dragonboard.

ubuntu@dragon410c:~$ dpkg -l | grep stress-ng
ii stress-ng 0.05.23-1ubuntu2 arm64 tool to load and stress a computer

ubuntu@dragon410c:~$ stress-ng --vm-rw 0 -v -t 10 --times
stress-ng: debug: [966] 4 processors online, 4 processors configured
stress-ng: info: [966] dispatching hogs: 4 vm-rw
stress-ng: debug: [966] cache allocate: reducing cache level from L3 (too high) to L2
stress-ng: info: [966] cache allocate: default cache size: 512K
stress-ng: debug: [966] starting stressors
stress-ng: debug: [967] stress-ng-vm-rw: started [967] (instance 0)
stress-ng: debug: [968] stress-ng-vm-rw: started [968] (instance 1)
stress-ng: debug: [966] 4 stressors spawned
stress-ng: debug: [969] stress-ng-vm-rw: started [969] (instance 2)
stress-ng: debug: [971] stress-ng-vm-rw: started [971] (instance 3)
stress-ng: debug: [968] stress-ng-vm-rw: exited [968] (instance 1)
stress-ng: debug: [967] stress-ng-vm-rw: exited [967] (instance 0)
stress-ng: debug: [966] process [967] terminated
stress-ng: debug: [966] process [968] terminated
stress-ng: debug: [969] stress-ng-vm-rw: exited [969] (instance 2)
stress-ng: debug: [966] process [969] terminated
stress-ng: debug: [971] stress-ng-vm-rw: exited [971] (instance 3)
stress-ng: debug: [966] process [971] terminated
stress-ng: info: [966] successful run completed in 10.11s
stress-ng: info: [966] for a 10.11s run time:
stress-ng: info: [966] 40.45s available CPU time
stress-ng: info: [966] 0.01s user time ( 0.02%)
stress-ng: info: [966] 40.20s system time ( 99.37%)
stress-ng: info: [966] 40.21s total time ( 99.40%)
stress-ng: info: [966] load average: 0.67 0.25 0.10

[Launchpad Janitor (janitor)]

This bug was fixed in the package stress-ng - 0.05.23-1ubuntu2

---------------
stress-ng (0.05.23-1ubuntu2) xenial; urgency=medium

  * Fix alignment mask to ensure stacks are 16 byte aligned (LP: #1573117)
    - incorrect mask used in previous fix, now using correct mask

stress-ng (0.05.23-1ubuntu1) xenial; urgency=medium

  * Ensure all clone() calls are 16 byte aligned for aarch64 (LP: #1573117)
  * stress-mmap: handle SIGBUS signals (LP: #1569468)

 -- Colin King <email address hidden> Tue, 26 Apr 2016 12:16:47 +0100

[Chris J Arges (arges)]

The verification of the Stable Release Update for stress-ng has completed successfully and the package has now been released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regressions.