1 |
On Monday, March 23, 2015 6:48:39 PM Mike Gilbert wrote: |
2 |
> On Mon, Mar 23, 2015 at 6:41 PM, Fernando Rodriguez |
3 |
> <frodriguez.developer@×××××××.com> wrote: |
4 |
> > On Monday, March 23, 2015 6:18:46 PM Mike Gilbert wrote: |
5 |
> >> On Mon, Mar 23, 2015 at 9:51 PM, Walter Dnes <waltdnes@××××××××.org> |
6 |
wrote: |
7 |
> >> > On Sun, Mar 22, 2015 at 09:25:53PM -0400, Fernando Rodriguez wrote |
8 |
> >> > |
9 |
> >> >> I guess gcc devs are careful when using the model numbers (Intel |
10 |
> >> >> lists 3 for Atoms, gcc uses only two so that may account for the |
11 |
> >> >> models I mentioned) but the chance of error is there. The -mno-xxx |
12 |
> >> >> flags would safeguard against it. |
13 |
> >> > |
14 |
> >> > I have one of the earliest Atom chips. Some people have a hard time |
15 |
> >> > believing this, but it's a 32-bit-only chip; a couple of lines from |
16 |
> >> > /proc/cpuinfo |
17 |
> >> > |
18 |
> >> > model name : Intel(R) Atom(TM) CPU Z520 @ 1.33GHz |
19 |
> >> > address sizes : 32 bits physical, 32 bits virtual |
20 |
> >> > |
21 |
> >> > Intel gives the CPU's specs at... |
22 |
> >> > |
23 |
> >> > http://ark.intel.com/products/35466/Intel-Atom-Processor-Z520-512K-Cache-1_33-GHz-533-MHz-FSB |
24 |
> >> > |
25 |
> >> > ...where it specifically says... |
26 |
> >> > |
27 |
> >> > Intel 64 # No |
28 |
> >> > |
29 |
> >> > I want to make absolutely certain that "illegal instructions" are not |
30 |
> >> > compiled for it. |
31 |
> >> |
32 |
> >> You will probably need to add -m32 to CFLAGS to avoid building 64-bit |
33 |
> >> objects on the 64-bit machine. |
34 |
> >> |
35 |
> > |
36 |
> > Your CPU is an example of what I'm saying, not just because it doesn't |
37 |
have 64 |
38 |
> > bit extensions but because it doesn't have MMX (at least according to the |
39 |
> > specs) and according to the GCC manual -march=atom means: "Intel Atom CPU |
40 |
with |
41 |
> > 64-bit extensions, MOVBE, MMX, SSE, SSE2, SSE3 and SSSE3 instruction set |
42 |
> > support." So I guess it's more common than I thought. |
43 |
> > |
44 |
> > So you may also want to add -mno-mmx to be sure. GCC does check for mmx |
45 |
but it |
46 |
> > doesn't not use it on the output (probably a bug?). |
47 |
> > |
48 |
> |
49 |
> It's much more likely that Intel's website doesn't bother including |
50 |
> MMX because it is so damn old that nobody cares. |
51 |
> |
52 |
> /proc/cpuinfo would be a more reliable source of data. |
53 |
> |
54 |
|
55 |
I agree that's very likely, that's why I said if the specs are right... |
56 |
This one doesn't list any SIMD extensions at all: |
57 |
http://ark.intel.com/products/85475/Intel-Atom-x7-Z8700-Processor-2M-Cache-up-to-2_40-GHz |
58 |
|
59 |
|
60 |
-- |
61 |
Fernando Rodriguez |