How about appending -Wno-pointer-sign and -Wno-implicit. That should  
get rid of both warnings causing errors and change no files (so we can  
drop in a replacement). Clint, could you see if that works for you?


On a somewhat related topic, env.Copy() prints a deprecated warning  
and will go away within a year. env.Clone() is the new syntax, however  
that wasn't added until scons 0.96.93. Currently we're depending on  
scons 0.96.91. Should we change it? We could put it an a try/except  
block to support both.

Ali

On Oct 14, 2008, at 9:16 PM, Clint Smullen wrote:

> Though it does not fix the underlying concerns. It makes all the
> warnings go away without having to disable -Werror.
>
> The change in elf_begin.c is to fix a warning about fruncate being
> implicitly defined. I saw the warning on both OSX and Linux. Other
> files in libelf include unistd.h, so it should not introduce any new
> compatibility issues.
>
>       - Clint Smullen
>
> On Oct 14, 2008, at 9:11 PM, Clint Smullen wrote:
>
>> # HG changeset patch
>> # User Clint Smullen <[EMAIL PROTECTED]>
>> # Date 1224032340 14400
>> # Node ID b57a19be4155e6a364e582f582106f8eb4b3b391
>> # Parent  733318abb7b1a4995cad43f94b5b0f7c7f547189
>> imported patch libelf_simplefix
>>
>> diff --git a/ext/libelf/SConscript b/ext/libelf/SConscript
>> --- a/ext/libelf/SConscript
>> +++ b/ext/libelf/SConscript
>> @@ -88,8 +88,7 @@
>> ElfFile('libelf_msize.c')
>>
>> m4env = env.Copy()
>> -del m4env['CCFLAGS']
>> -del m4env['CPPPATH']
>> +m4env.Append(CCFLAGS='-Wno-pointer-sign')
>>
>> # If we have gm4 use it
>> if m4env.Detect('gm4'):
>> diff --git a/ext/libelf/elf_begin.c b/ext/libelf/elf_begin.c
>> --- a/ext/libelf/elf_begin.c
>> +++ b/ext/libelf/elf_begin.c
>> @@ -33,6 +33,7 @@
>> #include <ar.h>
>> #include <ctype.h>
>> #include "libelf.h"
>> +#include <unistd.h>
>>
>> #include "_libelf.h"
>>
>
> _______________________________________________
> m5-dev mailing list
> [email protected]
> http://m5sim.org/mailman/listinfo/m5-dev
>

_______________________________________________
m5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/m5-dev

Reply via email to