RFR(xxxs): 8182034: os::Posix::is_valid_signal() should, if it uses sigaddset, initialize the signal set

[Kim Barrett]

> On Jun 21, 2017, at 9:32 AM, Thomas Stüfe <thomas.stuefe at gmail.com> wrote:
> 
> Thank you Christoph!
> 
> ..Thomas

I thought I’d reviewed this already, but apparently not.

Looks good to me too.  I can sponsor it through jprt for you.

> On Wed, Jun 21, 2017 at 2:35 PM, Langer, Christoph <christoph.langer at sap.com
>> wrote:
> 
>> Hi Thomas,
>> 
>> this one line change looks fine, reviewed :)
>> 
>> Best regards
>> Christoph
>> 
>>> -----Original Message-----
>>> From: hotspot-runtime-dev [mailto:hotspot-runtime-dev-
>>> bounces at openjdk.java.net] On Behalf Of Thomas Stüfe
>>> Sent: Dienstag, 13. Juni 2017 08:54
>>> To: hotspot-runtime-dev at openjdk.java.net
>>> Subject: RFR(xxxs): 8182034: os::Posix::is_valid_signal() should, if it
>> uses
>>> sigaddset, initialize the signal set
>>> 
>>> Hi all,
>>> 
>>> May I please have a review and a sponsor for this very tiny fix. Kim
>> Barret
>>> found this while reviewing "8181503: Can't compile hotspot with c++11".
>>> 
>>> We are using sigaddset() to find out if a given signal number is a valid
>>> signal number. The signal set used was not initialized probably. This
>> never
>>> led to an error (the function in its current form is tested and used at
>> SAP
>>> for years) but it is still invalid according to Posix.
>>> 
>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8182034
>>> Webrev:
>>> http://cr.openjdk.java.net/~stuefe/webrevs/8182034-os-posix-
>>> is_valid_signal-sigemptyset/webrev.00/webrev/
>>> 
>>> Thanks and Kind Regards, Thomas