Re: [kvm-unit-tests PATCH 1/4] Make vmx_tests.c compilable with -Wmissing-prototypes and -Wstrict-prototypes

[Thomas Huth <thuth@xxxxxxxxxx>]

On 27.06.2018 10:55, Andrew Jones wrote:
> On Wed, Jun 27, 2018 at 10:02:05AM +0200, Thomas Huth wrote:
>> -extern void insn_hlt();
>> -extern void insn_invlpg();
>> -extern void insn_mwait();
>> -extern void insn_rdpmc();
>> -extern void insn_rdtsc();
>> -extern void insn_cr3_load();
>> -extern void insn_cr3_store();
>> +extern void insn_hlt(void);
>> +extern void insn_invlpg(void);
>> +extern void insn_mwait(void);
>> +extern void insn_rdpmc(void);
>> +extern void insn_rdtsc(void);
>> +extern void insn_cr3_load(void);
>> +extern void insn_cr3_store(void);
>>  #ifdef __x86_64__
>> -extern void insn_cr8_load();
>> -extern void insn_cr8_store();
>> +extern void insn_cr8_load(void);
>> +extern void insn_cr8_store(void);
>>  #endif
>> -extern void insn_monitor();
>> -extern void insn_pause();
>> -extern void insn_wbinvd();
>> -extern void insn_sgdt();
>> -extern void insn_lgdt();
>> -extern void insn_sidt();
>> -extern void insn_lidt();
>> -extern void insn_sldt();
>> -extern void insn_lldt();
>> -extern void insn_str();
>> -extern void insn_cpuid();
>> -extern void insn_invd();
>> -extern void insn_rdrand();
>> -extern void insn_rdseed();
>> +extern void insn_monitor(void);
>> +extern void insn_pause(void);
>> +extern void insn_wbinvd(void);
>> +extern void insn_sgdt(void);
>> +extern void insn_lgdt(void);
>> +extern void insn_sidt(void);
>> +extern void insn_lidt(void);
>> +extern void insn_sldt(void);
>> +extern void insn_lldt(void);
>> +extern void insn_str(void);
>> +extern void insn_cpuid(void);
>> +extern void insn_invd(void);
>> +extern void insn_rdrand(void);
>> +extern void insn_rdseed(void);
>>
> 
> Another nice prototype cleanup would be to remove the
> unnecessary use of 'extern' everywhere.  We have around
> 250 useless uses. You could get the majority of the x86
> ones while you're touching them here.
> 
> s390x only has one and powerpc only 16. So they don't
> require much churn. I have 85 in arm code. Churning
> code just to drop the use is a bit lame, but continuing
> to add more in order to maintain consistency is also
> getting pretty lame.

Ok, I agree, sounds like a good idea to clean that up while I'm changing
the lines anyway. I'll sent a v2 for this patch when I got some spare
time again (likely not very soon...).

 Thomas