Re: [Xen-devel] [PATCH 2/3] xtf: avoid shifting a negative value

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

On 06/05/16 15:57, Andrew Cooper wrote:
> On 06/05/16 15:25, Roger Pau Monne wrote:
>> Because it's undefined behaviour.
>>
>> Signed-off-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
>> ---
>> Cc: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
>> ---
>>  include/arch/x86/processor.h | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/include/arch/x86/processor.h b/include/arch/x86/processor.h
>> index c9f253f..841953c 100644
>> --- a/include/arch/x86/processor.h
>> +++ b/include/arch/x86/processor.h
>> @@ -134,7 +134,7 @@
>>  /* Segment-based Error Code - supplemental constants. */
>>  #define X86_EC_TABLE_MASK (3  << 1)
>>  #define X86_EC_SEL_SHIFT  3
>> -#define X86_EC_SEL_MASK   (-1 << X86_EC_SEL_SHIFT)
>> +#define X86_EC_SEL_MASK   (-1U << X86_EC_SEL_SHIFT)
> The resulting expression needs to be signed, to DTRT when implicitly
> promoted.  I presume this was clang complaining?
>
> Does using the constant (-9) work?

Or indeed perhaps ~(1 << X86_EC_SEL_SHIFT) which (iirc) will keep the
signed-ness of 1 even through bitwise inversion.

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel