[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150707144916.GA9127@openwall.com>
Date: Tue, 7 Jul 2015 17:49:16 +0300
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: extend SIMD intrinsics
On Tue, Jul 07, 2015 at 10:37:14AM +0800, Lei Zhang wrote:
> Using a union type, the pseudo-intrinsics would be written this way (for AltiVec):
>
> #vadd_epi32(x, y) (vtype)vec_add(x.v32, y.v32)
> #vadd_epi64(x, y) (vtype)vec_add(x.v64, y.v64)
>
> Does the type casting here violate strict aliasing?
I find it weird. Why would you need it? Instead, assign the vec_*()
intrinsics outputs to .v fields.
Also, cast to union hasn't historically been supported in C. Perhaps
it's OK for us to assume C99 already, but let's not over-complicate
things unnecessarily.
> Or further, for a union type defined as
>
> typedef union {
> vtype32 v32;
> uint32_t s32[...];
> } vtype;
> vtype v;
>
> Would there be any difference between using v.v32 and using (vtype32)v ?
Would the latter even compile? While cast to union is a thing, I think
cast to union member type isn't... but I could be wrong?
More importantly, why are you proposing to define separate vector types
for different element sizes? I don't object to this, I merely ask.
There might be a good reason for it.
Alexander
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
