2017-09-13 16:00:56

by Vince Weaver

Subject: perf: weird si_code results on overflow signal

Hello

I just compiled up a fresh git kernel and all of the perf_event_test
overflow tests are failing.

The reason is that instead of getting POLL_IN or POLL_HUP sources as
expected, they are getting weird results in si_code of "-5".

I haven't had time to bisect this, but I do notice that some major changes
to si_code handling went in with
64a76d0d64bea159da997c002a916ffc03f98bfc

Vince

2017-09-19 11:40:38

by Vince Weaver

Subject: Re: perf: weird si_code results on overflow signal

On Wed, 13 Sep 2017, Vince Weaver wrote:

> I just compiled up a fresh git kernel and all of the perf_event_test
> overflow tests are failing.
>
> The reason is that instead of getting POLL_IN or POLL_HUP sources as
> expected, they are getting weird results in si_code of "-5".
>
> I haven't had time to bisect this, but I do notice that some major changes
> to si_code handling went in with
> 64a76d0d64bea159da997c002a916ffc03f98bfc

resending in case everyone was away at plumbers...

Vince

2017-09-19 18:23:01

by Eric W. Biederman

Subject: Re: perf: weird si_code results on overflow signal

Vince Weaver <[email protected]> writes:

> On Wed, 13 Sep 2017, Vince Weaver wrote:
>
>> I just compiled up a fresh git kernel and all of the perf_event_test
>> overflow tests are failing.
>>
>> The reason is that instead of getting POLL_IN or POLL_HUP sources as
>> expected, they are getting weird results in si_code of "-5".
>>
>> I haven't had time to bisect this, but I do notice that some major changes
>> to si_code handling went in with
>> 64a76d0d64bea159da997c002a916ffc03f98bfc
>
> resending in case everyone was away at plumbers...

Apologies it looks like I missed this one the first time through.

I will push ou tthe fix in just a minute.

Eric