signal06 from ubuntu_ltp_syscalls failed on Eoan 5.3 GCP

[Po-Hsu Lin]

Issue found on 5.3.0-1010 Eoan GCP

 startup='Sat Dec 7 10:43:06 2019'
 signal06 0 TINFO : loop = 10771
 signal06 1 TFAIL : signal06.c:87: Bug Reproduced!
 tag=signal06 stime=1575715386 dur=1 exit=exited stat=1 core=no cu=1 cs=3

This is not a regression as this failure can be found in 5.3.0-1009.10-gcp as well.

[Po-Hsu Lin (cypressyew)]

This issue can be found on E-AWS, but just on the following instances:
  * m4.large
  * m5a.large
  * t2.small
  * t3.medium

[Po-Hsu Lin (cypressyew)]

From the test case, it's a regression-test for the following two commits:
df24fb85 (x86, fpu: __restore_xstate_sig()->math_state_restore() needs preempt_disable())
66463db4 (x86, fpu: shift drop_init_fpu() from save_xstate_sig() to handle_signal())

They're all available in Eoan GCP tree.
Further investigation is needed.

[Po-Hsu Lin (cypressyew)]

Didn't see this on E-AWS (5.3.0-1024.26) with those instances mentioned in comment #1

[Po-Hsu Lin (cypressyew)]

Didn't see this on Eoan GCP as well. Closing this.
5.3.0-1027.29-gcp