ACK: [X][SRU][PATCH 0/1] test_bpf: flag tests that cannot be jited on s390x

Andrea Righi andrea.righi at canonical.com
Fri Mar 20 11:37:02 UTC 2020


On Fri, Mar 20, 2020 at 07:03:49PM +0800, Po-Hsu Lin wrote:
> == SRU Justification ==
> In the ubuntu_bpf_jit test, it will insert the test_bpf module to perform
> the test. There will be 4 failures reported on s390x instances by the test,
> 3 of them were simply because they cannot be jited on s390x:
>   * test_bpf: #243 BPF_MAXINSNS: Ctx heavy transformations FAIL to prog_create err=-524 len=4096
>   * test_bpf: #244 BPF_MAXINSNS: Call heavy transformations FAIL to prog_create err=-524 len=4096
>   * test_bpf: #249 BPF_MAXINSNS: ld_abs+get_processor_id FAIL to prog_create err=-524 len=4096
> 
> (For another failure "ld_abs+vlan_push/pop" please check bug 1868243)
> 
> == Fix ==
> * 3203c901 (test_bpf: flag tests that cannot be jited on s390)
> Only Xenial needs this patch, and it needs to be backported to Xenial
> as some test case does not exist.
> 
> == Test ==
> Test kernel could be found here:
> https://people.canonical.com/~phlin/kernel/lp-1768452-bpf-jit/
> 
> The test will be marked as pass with the patched kernel:
> test_bpf: #243 BPF_MAXINSNS: Ctx heavy transformations PASS
> test_bpf: #244 BPF_MAXINSNS: Call heavy transformations PASS
> test_bpf: #249 BPF_MAXINSNS: ld_abs+get_processor_id PASS
> 
> == Regression Potential ==
> Low, change limited to the test_bpf module itself and specific to s390x,
> no impact to real functions.

Looks good to me.

Acked-by: Andrea Righi <andrea.righi at canonical.com>



More information about the kernel-team mailing list