Re: [Xen-devel] [PATCH v8 26/50] x86emul: support remaining AVX512BW legacy-equivalent insns

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 21.05.19 at 15:08, <andrew.cooper3@xxxxxxxxxx> wrote:
> On 15/03/2019 10:54, Jan Beulich wrote:
>> --- a/xen/arch/x86/x86_emulate/x86_emulate.c
>> +++ b/xen/arch/x86/x86_emulate/x86_emulate.c
>> @@ -435,7 +435,10 @@ static const struct ext0f38_table {
>>      disp8scale_t d8s:4;
>>  } ext0f38_table[256] = {
>>      [0x00] = { .simd_size = simd_packed_int, .d8s = d8s_vl },
>> -    [0x01 ... 0x0b] = { .simd_size = simd_packed_int },
>> +    [0x01 ... 0x03] = { .simd_size = simd_packed_int },
>> +    [0x04] = { .simd_size = simd_packed_int, .d8s = d8s_vl },
>> +    [0x05 ... 0x0b] = { .simd_size = simd_packed_int },
>> +    [0x0b] = { .simd_size = simd_packed_int, .d8s = d8s_vl },
> 
> It doesn't look as if you mean 0x0b twice here, although its quite
> possible that GCC elides it silently (Clang definitely won't).

Indeed, I failed to edit the upper bound of the range in question -
with gcc the last initializer for an array slot wins. Fixed for v9.

Jan



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel