Subject: Re: [boost] [random] parallel random number generators
From: John Salmon (john_at_[hidden])
Date: 2012-01-29 16:53:51
On Sun, Jan 29, 2012 at 2:09 PM, Jeremiah Willcock <jewillco_at_[hidden]>wrote:
> On Sun, 29 Jan 2012, John Salmon wrote:
> By the way, the code in aes_hasaesni.hpp that trashes %ebx, and is
>> therefore problematic with -m32 -fPIC is conditionalized on #ifdef
>> BOOST_HAS_AESNI. While I agree that the #ifdef logic to set
>> BOOST_HAS_AESNI needs work, I think it's safe to say that BOOST_HAS_AESNI
>> should never be set when -m32 is in effect.
> Is there a reason for that? AESNI instructions use SSE registers, so they
> are independent of 32-bit vs. 64-bit mode.
> You're correct. I mistakenly assumed that you couldn't use AESNI
instructions in 32-bit code. I'm not sure why anyone would want to, but it
*is* possible. I'll add the code to push and pop %ebx around cpuid.
-- Jeremiah Willcock
> Unsubscribe & other changes: http://lists.boost.org/**
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk