RFR: 8210416: [linux] Poor StrictMath performance due to non-optimized compilation

Erik Joelsson erik.joelsson at oracle.com
Mon Sep 10 16:29:57 UTC 2018


I see. I was not aware of that issue, but we clearly need to file a bug 
for it and fix it. In this case I think it's fine to us the macro however.

/Erik


On 2018-09-10 03:29, Severin Gehwolf wrote:
> Hi Erik,
>
> On Fri, 2018-09-07 at 10:19 -0700, Erik Joelsson wrote:
>> Instead of checking for compiler version, you could simply use the
>> FLAGS_COMPILER_CHECK_ARGUMENTS macro. This would apply to both gcc and
>> clang. We usually prefer this over version checks (unless a version is
>> just known to be unstable with a certain flag, in which case a
>> capability check will not work).
> I see. Note JDK-8157358, though, which seems to have changed[1] the
> build system to not use FLAGS_COMPILER_CHECK_ARGUMENTS as it seems to
> break cross-compilation. It's not used at the moment. Do you still want
> me to use it?
>
>> CoreLibrarkes.gmk:
>> 43: I think "with" is superfluous
> Thanks, fixed locally.
>
>> Otherwise I think this looks like the right direction.
> Glad to hear that.
>
> Thanks,
> Severin
>
> [1] http://hg.openjdk.java.net/jdk9/jdk9/rev/f51004322fbe
>




More information about the build-dev mailing list