Re: [Discuss-gnuradio] Volk library invalid opcode exception

[Tom Rondeau]

On Sun, Apr 15, 2012 at 8:24 PM, Nick Foster <address@hidden> wrote:
> Attached is a patch with one further check -- to make sure the check that
> AVX is enabled by the OS, is enabled by the OS.
>
> No kidding.
>
> --n

Wonderful, Nick, thanks.

Joanna, let us know how this works for you. I'm going to be on the
road for the next few days, so I'll be sparsely available.

Tom


> On Sun, Apr 15, 2012 at 4:09 PM, Nick Foster <address@hidden> wrote:
>>
>> On Sun, Apr 15, 2012 at 2:26 PM, Joanna Rutkowska
>> <address@hidden> wrote:
>> <snip>
>>
>>> Another potential explanations of why this doesn't work I could come up
>>> with:
>>>
>>> 1) Perhaps volk somehow erroneously interprets cpuid info and assumes
>>> that AVX is present, while it is no...? Tom, can you point out the
>>> specific code in volk that is responsible for deciding whether to use
>>> AVX or not?
>>
>>
>> Your CPU has AVX capability, no doubt about it. I agree with Tom that it's
>> likely that Xen is disabling AVX support with XSETVB -- I'm not sure why it
>> does that. Normal people do not disable extended instruction sets on new
>> processors. It's just turning off silicon you paid for, after all. =)
>>
>> Attached is a patch for Volk which performs the additional step of
>> verifying AVX with XGETBV to determine that the OS is not turning off useful
>> things. This doesn't fix the fact that Xen is busted, it just won't run AVX
>> instructions when the instructions are disabled.
>>
>> Joanna, please test this patch for me and verify that your Volk machine
>> enumerates as sse4_2_64. Thanks!
>>
>> Tom, the patch is available (based on latest master) at
>> github/bistromath:gnuradio.git on the xgetbv branch.
>>
>> --n
>
>