RFR (T): 8218060: JDK-8217786 breaks build due to remaining unused function

David Holmes david.holmes at oracle.com
Wed Jan 30 22:40:06 UTC 2019


Yes the fix is wrong. :(

I'll restore the S390 fix if noone else has gotten to it yet.

David

On 31/01/2019 1:22 am, Stefan Karlsson wrote:
> I think this is the wrong fix. It's used here on s390:
> 
> 2185 void os::Linux::print_virtualization_info(outputStream* st) {
> 2186 #if defined(S390)
> 2187   // /proc/sysinfo contains interesting information about
> 2188   // - LPAR
> 2189   // - whole "Box" (CPUs )
> 2190   // - z/VM / KVM (VM<nn>); this is not available in an LPAR-only 
> setup
> 2191   const char* kw[] = { "LPAR", "CPUs", "VM", NULL };
> 2192
> 2193   if (! print_matching_lines_from_sysinfo_file(st, kw)) {
> 2194     st->print_cr("  </proc/sysinfo Not Available>");
> 2195   }
> 2196 #endif
> 2197 }
> 
> StefanK
> 
> 
> 
> On 2019-01-30 16:06, Thomas Schatzl wrote:
>> Hi all,
>>
>>    can I have quick reviews for this change that fixes a compilation
>> error due to an unused function?
>>
>> I.e.
>>
>> Compiling the repo after JDK-8217786 gives the following error:
>>
>> .../vmshare/jdk10/hs/open/src/hotspot/os/linux/os_linux.cpp:1860:13:
>> error: 'bool print_matching_lines_from_sysinfo_file(outputStream*,
>> const char**)' defined but not used [-Werror=unused-function]
>>   static bool print_matching_lines_from_sysinfo_file(outputStream* st,
>> const char* keywords_to_match[]) {
>>               ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> Compiling 305 files for jdk.javadoc
>> cc1plus: all warnings being treated as errors
>> make[3]: *** [[...]variant-server/libjvm/objs/os_linux.o] Error 1
>> lib/CompileJvm.gmk:172: recipe for target '[...]/linux-
>> x64/hotspot/variant-server/libjvm/objs/os_linux.o' failed
>>
>> The change simply removes that unused function.
>>
>> CR:
>> https://bugs.openjdk.java.net/browse/JDK-8218060
>> Webrev:
>> http://cr.openjdk.java.net/~tschatzl/8218060/webrev/
>> Testing:
>> local compilation
>>
>> Thanks,
>>    Thomas
>>
>>
> 


More information about the hotspot-dev mailing list