RFR: 8301869: Regression ~14% in J2dBench-bimg_misc-* in 21-b5 only on linux-aarch64 [v2]

[Phil Race]

On Mon, 27 Feb 2023 16:16:50 GMT, Jayathirth D V <jdv at openjdk.org> wrote:

>> Under https://bugs.openjdk.org/browse/JDK-8264846 we moved to -O3 level of gcc optimizations from -O1 level for libawt build. This improved our J2DBench performance numbers in some options considerably.
>> 
>> Recent changes done under https://bugs.openjdk.org/browse/JDK-8299337 causes difference in generated code by gcc and this is resulting in performance regression for bimg_misc-* J2DBench options in our performance servers. Under https://bugs.openjdk.org/browse/JDK-8299337 we have just removed unused variables and it is a cleanup task.
>> 
>> We can force gcc to generate position independent code by using -fpic option.Also i have removed -fgcse-after-reload option for gcc, because this is by default covered under -O3 level of optimization introduced under https://bugs.openjdk.org/browse/JDK-8264846.
>> 
>> With this change bimg_misc-* J2DBench option performance regression is resolved and there are no regression in other options of J2DBench or SwingMark and it is verified in our performance servers.
>
> Jayathirth D V has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Update comment

make/modules/java.desktop/lib/Awt2dLibraries.gmk line 126:

> 124: # about initializing a declared 'extern' cannot be turned off individually. Only
> 125: # applies to debug builds. This limitation in gcc is tracked in
> 126: # https://gcc.gnu.org/bugzilla/show_bug.cgi?id=45977

The comment above needs to be moved inside the if block because it only applies to the first setting, not the second.

make/modules/java.desktop/lib/Awt2dLibraries.gmk line 130:

> 128:   BUILD_LIBAWT_debug_mem.c_CFLAGS := -w
> 129:   # This option improves performance of FillRect/FillOval in Java2D by 15% for some gcc
> 130:   LIBAWT_CFLAGS += -fpic

So .. acc. to the eval in JBS we can get rid of -fgcse-after-reload because it is implied by -O3
And Erik says -fPIC is implied by -fpic 

So how is -fpic different than -fPIC in a way that helps here ?
So far as I can tell the -fpic are using just limits the size of the global offset table that -fPIC will generate.
Since we had an unlimited offset table before, then -fpic won't change anything. 
So what's the explanation ?

-------------

PR: https://git.openjdk.org/jdk/pull/12761